Christoph Schiller

MOTION MOUNTAIN
the adventure of physics

www.motionmountain.net
 Christoph Schiller

Motion Mountain

The Adventure of Physics

available free of charge at

www.motionmountain.net
Editio undevicesima.

Proprietas scriptoris © Christophori Schiller

secundo anno Olympiadis vicesimae sextae
tertio anno Olympiadis vicesimae octavae.

Omnia proprietatis iura reservantur et vindicantur.

Imitatio prohibita sine auctoris permissione.
Non licet pecuniam expetere pro aliquo, quod
partem horum verborum continet; liber
pro omnibus semper gratuitus erat et manet.

Nineteenth revision.

Copyright © 1997–2006 by Christoph Schiller,

between the second year of the 26th olympiad
and the third year of the 28th olympiad.

All rights reserved. Commercial reproduction,

distribution or use, in whole or in part, is not allowed
without the written consent of the copyright owner.
You are not allowed to charge money for anything
containing any part of this text; it was and remains
free to read for everybody. Details of the cover
photographs are on page 1295.
 To Esther

τῷ ἐµοὶ δαὶµονι
Die Menschen stärken, die Sachen klären.
 Contents
Preface 20
A request 21
An appetizer 23

First Part : Cl assical Physics – How Do Things and Images Move?

Ch a pter I Galilean Motion 28 Dvipsbugw

1. Why should we care about motion? 28
2. Galilean physics – motion in everyday life 42
3. Global descriptions of motion – the simplicity of complexity 173
4. From the limitations of physics to the limits of motion 263
Ch a pter II Special R el ativity 275
5. Maximum speed, observers at rest, and motion of light 275

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ch a pter III Gravitation and R el ativity 349
6. Maximum force – general relativity in one statement 349
7. The new ideas on space, time and gravity 377
8. Motion in general relativity – bent light and wobbling vacuum 397
9. Why can we see the stars? – Motion in the universe 437
10. Black holes – falling forever 476
11. Does space differ from time? 491
12. General relativity in ten points – a summary for the layman 497
Ch a pter IV Cl assical Electrodynamics 517
13. Liquid electricity, invisible fields and maximum speed 517
14. What is light? 558
15. Charges are discrete – the limits of classical electrodynamics 589
16. Electromagnetic effects and challenges 592
17. Classical physics in a nutshell – one and a half steps out of three 614
Intermezzo The Brain, L anguage and the Human Condition 631

Second Part : Quantum Theory – What Is Matter? What Are Interac-

tions? Copyright © Christoph Schiller November 1997–May 2006

Ch a pter V Quanta of Light and Matter 704

18. Minimum action – quantum theory for poets and lawyers 704
19. Light – the strange consequences of the quantum of action 717
20. Motion of matter – beyond classical physics 735
21. Colours and other interactions between light and matter 753
Ch a pter VI Permu tation of Particles 771
22. Are particles like gloves? 771
23. Rotations and statistics – visualizing spin 780
Ch a pter VII Details of Quantum Theory and Electromagnetism 793
24. Superpositions and probabilities – quantum theory without ideology 793
25. Applied quantum mechanics – life, pleasure and the means to achieve them 816

Dvipsbugw
 26. Quantum electrodynamics – the origin of virtual reality 854
27. Quantum mechanics with gravitation – the first approach 868
Ch a pter VIII Inside the Nucleus 897
28. The structure of the nucleus – the densest clouds 897
29. The strong nuclear interaction and the birth of matter 921
30. The weak nuclear interaction and the handedness of nature 935
31. The standard model of elementary particle physics – as seen on television 940
32. Grand unification – a simple dream 941
Dvipsbugw
Ch a pter IX Advanced Quantum Theory (not yet avail able) 949
Ch a pter X Quantum Physics in a Nu tshell 951
Intermezzo Bacteria, Flies and Knots 969

Third Part : Motion Withou t Motion – What Are Space, Time and
Particles?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ch a pter XI General R el ativity Versus Quantum Mechanics 996
33. Does matter differ from vacuum? 998
34. Nature at large scales – is the universe something or nothing? 1032
35. The physics of love – a summary of the first two and a half parts 1056
36. Maximum force and minimum distance – physics in limit statements 1068
37. The shape of points – extension in nature 1101
38. String theory – a web of dualities 1128
Ch a pter XII Unification (not yet avail able) 1137
Ch a pter XIII The Top of the Mountain (not yet avail able) 1139

Fourth Part : Appendices

Appendix A Notation and Conventions 1142

Appendix B Units, Measurements and Constants 1154
Appendix C Particle Properties 1174
Appendix D Numbers and Spaces 1194
Copyright © Christoph Schiller November 1997–May 2006

Appendix E Sources of Information on Motion 1227

Appendix F Challenge Hints and Solu tions 1233
Appendix G List of Illustrations 1284
Appendix H List of Tables 1298
Appendix I Name Index 1301
Appendix J Subject Index 1326

Dvipsbugw
 Detailed Contents
20 Preface
21 A request
23 An appetizer

First Part : Cl assical Physics Dvipsbugw

How Do Things and Images Move?

28 Ch a pter I Galilean Motion

28 1. Why should we care about motion?
Does motion exist? 30 • How should we talk about motion? 32 • What are the
types of motion? 33 • Perception, permanence and change 36 • Does the world

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

need states? 39 • Curiosities and fun challenges about motion 40
42 2. Galilean physics – motion in everyday life
What is velocity? 43 • What is time? 45 • Why do clocks go clockwise? 48 • Does
time flow? 49 • What is space? 49 • Are space and time absolute or relative? 52 •
Size – why area exists, but volume does not 53 • What is straight? 56 • A hollow
Earth? 56 • Curiosities and fun challenges about everyday space and time 57
64 How to describe motion – kinematics
Throwing and shooting 66 • What is rest? 68 • Objects and point particles 71 •
Legs and wheels 73
74 Objects and images
Motion and contact 75 • What is mass? 77 • Is motion eternal? 82 • More on
conservation – energy 84 • Is velocity absolute? – The theory of everyday relativ-
ity 86 • Rotation 88 • Rolling wheels 91 • How do we walk? 92 • Is the Earth
rotating? 94 • How does the Earth rotate? 98 • Does the Earth move? 100 • Is
rotation relative? 104 • Curiosities and fun challenges about everyday motion 104
• Legs or wheels? – Again 114
116 Dynamics due to gravitation
Properties of gravitation 119 • Dynamics – how do things move in various dimen-
sions? 123 • Gravitation in the sky 123 • The Moon 124 • Orbits 127 • Tides 129
• Can light fall? 132 • What is mass? – Again 133 • Curiosities and fun challenges
about gravitation 135
146 What is classical mechanics?
Copyright © Christoph Schiller November 1997–May 2006

Should one use force? 147 • Complete states – initial conditions 152 • Do surprises
exist? Is the future determined? 154 • A strange summary about motion 157
158 Bibliography
173 3. Global descriptions of motion – the simplicity of complexity
176 Measuring change with action
The principle of least action 179 • Why is motion so often bounded? 183 • Curios-
ities and fun challenges about Lagrangians 186
189 Motion and symmetry
Why can we think and talk? 189 • Viewpoints 190 • Symmetries and groups 192
• Representations 192 • Symmetries, motion and Galilean physics 195 • Reprodu-
cibility, conservation and Noether’s theorem 198 • Curiosities and fun challenges
about motion symmetry 203

Dvipsbugw
10 contents

203 Simple motions of extended bodies – oscillations and waves

Waves and their motion 205 • Why can we talk to each other? – Huygens’ prin-
ciple 210 • Signals 210 • Solitary waves and solitons 212 • Curiosities and fun
challenges about waves and extended bodies 214
218 Do extended bodies exist?
Mountains and fractals 219 • Can a chocolate bar last forever? 219 • How high can
animals jump? 221 • Felling trees 221 • The sound of silence 222 • Little hard
balls 223 • Curiosities and fun challenges about fluids and solids 225
233 What can move in nature? Dvipsbugw
234 How do objects get warm?
235 Temperature
Entropy 237 • Flow of entropy 239 • Do isolated systems exist? 240 • Why do
balloons take up space? – The end of continuity 240 • Brownian motion 242 •
Entropy and particles 244 • The minimum entropy of nature – the quantum of in-
formation 245 • Why can’t we remember the future? 246 • Is everything made of
particles? 247 • Why stones can be neither smooth nor fractal, nor made of little

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

hard balls 248 • Curiosities and fun challenges about heat 249
255 Self-organization and chaos
Curiosities and fun challenges about self-organization 261
263 4. From the limitations of physics to the limits of motion
Research topics in classical dynamics 263 • What is contact? 264 • Precision and
accuracy 264 • Can all of nature be described in a book? 265 • Why is measurement
possible? 265 • Is motion unlimited? 266
267 Bibliography

275 Ch a pter II Special Relativity

275 5. Maximum speed, observers at rest, and motion of light
Can one play tennis using a laser pulse as the ball and mirrors as rackets? 280 • Spe-
cial relativity in a few lines 283 • Acceleration of light and the Doppler effect 284
• The difference between light and sound 287 • Can one shoot faster than one’s
shadow? 287 • The composition of velocities 289 • Observers and the principle
of special relativity 290 • What is space-time? 294 • Can we travel to the past? –
Time and causality 295
297 Curiosities of special relativity
Faster than light: how far can we travel? 297 • Synchronization and time travel –
can a mother stay younger than her own daughter? 297 • Length contraction 300
• Relativistic films – aberration and Doppler effect 302 • Which is the best seat in a
Copyright © Christoph Schiller November 1997–May 2006

bus? 305 • How fast can one walk? 306 • Is the speed of shadow greater than the
speed of light? 306 • Parallel to parallel is not parallel – Thomas rotation 309 • A
never-ending story – temperature and relativity 310
310 Relativistic mechanics
Mass in relativity 310 • Why relativistic snooker is more difficult 312 • Mass is
concentrated energy 313 • Collisions, virtual objects and tachyons 315 • Systems
of particles – no centre of mass 317 • Why is most motion so slow? 317 • The
history of the mass–energy equivalence formula of de Pretto and Einstein 318 • 4-
vectors 319 • 4-momentum 322 • 4-force 323 • Rotation in relativity 324 • Wave
motion 325 • The action of a free particle – how do things move? 326 • Conformal
transformations – why is the speed of light constant? 327
329 Accelerating observers

Dvipsbugw
contents 11

Acceleration for inertial observers 330 • Accelerating frames of reference 331 •

Event horizons 335 • Acceleration changes colours 336 • Can light move faster than
c? 337 • What is the speed of light? 338 • Limits on the length of solid bodies 339
340 Special relativity in four sentences
Could the speed of light vary? 340 • What happens near the speed of light? 341
342 Bibliography

349 Ch a pter III Gravitation and Relativity

349 6. Maximum force – general relativity in one statement Dvipsbugw
The maximum force and power limits 350 • The experimental evidence 352 • Dedu-
cing general relativity 353 • Space-time is curved 357 • Conditions of validity of the
force and power limits 359 • Gedanken experiments and paradoxes about the force
limit 359 • Gedanken experiments with the power limit and the mass flow limit 364
• Hide and seek 367 • An intuitive understanding of general relativity 368 • An
intuitive understanding of cosmology 370 • Experimental challenges for the third
millennium 371 • A summary of general relativity 372 • Acknowledgement 373

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

374 Bibliography
377 7. The new ideas on space, time and gravity
Rest and free fall 377 • What is gravity? – A second answer 378 • What tides tell us
about gravity 381 • Bent space and mattresses 382 • Curved space-time 384 • The
speed of light and the gravitational constant 386 • Why does a stone thrown into the
air fall back to Earth? – Geodesics 388 • Can light fall? 390 • Curiosities and fun
challenges about gravitation 391 • What is weight? 395 • Why do apples fall? 396
397 8. Motion in general relativity – bent light and wobbling vacuum
397 Weak fields
The Thirring effects 397 • Gravitomagnetism 399 • Gravitational waves 402 •
Bending of light and radio waves 409 • Time delay 410 • Effects on orbits 411
• The geodesic effect 414 • Curiosities and fun challenges about weak fields 414
415 How is curvature measured?
Curvature and space-time 418 • Curvature and motion in general relativity 420 •
Universal gravity 421 • The Schwarzschild metric 421 • Curiosities and fun chal-
lenges about curvature 422
422 All observers – heavier mathematics
The curvature of space-time 422 • The description of momentum, mass and en-
ergy 424 • Hilbert’s action – how things fall? 425 • The symmetries of general
relativity 426 • Einstein’s field equations 427 • More on the force limit 430 • De-
ducing universal gravity 430 • Deducing linearized general relativity 431 • How to
Copyright © Christoph Schiller November 1997–May 2006

calculate the shape of geodesics 431 • Mass in general relativity 433 • Is gravity an
interaction? 433 • The essence of general relativity 434 • Riemann gymnastics 435
• Curiosities and fun challenges about general relativity 437
437 9. Why can we see the stars? – Motion in the universe
Which stars do we see? 438 • What do we see at night? 439 • What is the uni-
verse? 445 • The colour and the motion of the stars 447 • Do stars shine every
night? 449 • A short history of the universe 450 • The history of space-time 454
• Why is the sky dark at night? 458 • Is the universe open, closed or marginal? 460
• Why is the universe transparent? 461 • The big bang and its consequences 462
• Was the big bang a big bang? 463 • Was the big bang an event? 463 • Was the
big bang a beginning? 463 • Does the big bang imply creation? 464 • Why can we
see the Sun? 465 • Why are the colours of the stars different? 466 • Are there dark

Dvipsbugw
12 contents

stars? 467 • Are all stars different? – Gravitational lenses 467 • What is the shape
of the universe? 469 • What is behind the horizon? 470 • Why are there stars all
over the place? – Inflation 471 • Why are there so few stars? – The energy and en-
tropy content of the universe 471 • Why is matter lumped? 473 • Why are stars so
small compared with the universe? 473 • Are stars and galaxies moving apart or is
the universe expanding? 473 • Is there more than one universe? 473 • Why are the
stars fixed? – Arms, stars and Mach’s principle 474 • At rest in the universe 475 •
Does light attract light? 475 • Does light decay? 476
476 10. Black holes – falling forever Dvipsbugw
Why study black holes? 476 • Horizons 477 • Orbits 479 • Hair and entropy 482
• Black holes as energy sources 484 • Curiosities and fun challenges about black
holes 485 • Formation of and search for black holes 488 • Singularities 489 • A
quiz – is the universe a black hole? 490
491 11. Does space differ from time?
Can space and time be measured? 492 • Are space and time necessary? 493 •
Do closed timelike curves exist? 494 • Is general relativity local? – The hole argu-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ment 494 • Is the Earth hollow? 495 • Are space, time and mass independent? 496
497 12. General relativity in ten points – a summary for the layman
The accuracy of the description 498 • Research in general relativity and cosmo-
logy 499 • Could general relativity be different? 501 • The limits of general relativ-
ity 502
503 Bibliography

517 Ch a pter IV Classical Electrodynamics

517 13. Liquid electricity, invisible fields and maximum speed
517 Amber, lodestone and mobile phones
How can one make lightning? 520 • Electric charge and electric fields 522 • Can
we detect the inertia of electricity? 526 • Feeling electric fields 528 • Magnets 529
• Can humans feel magnetic fields? 530 • How can one make a motor? 531 • Mag-
netic fields 532 • How motors prove relativity to be right 536 • Curiosities and fun
challenges about things electric and magnetic 537
545 The description of electromagnetic field evolution
Colliding charged particles 547 • The gauge field – the electromagnetic vector poten-
tial 548 • Energy and linear and angular momentum of the electromagnetic field 552
• The Lagrangian of electromagnetism 552 • Symmetries – the energy–momentum
tensor 554 • What is a mirror? 554 • What is the difference between electric and
magnetic fields? 556
Copyright © Christoph Schiller November 1997–May 2006

557 Electrodynamic challenges and curiosities

Could electrodynamics be different? 557 • The toughest challenge for electrodyna-
mics 558
558 14. What is light?
The slowness of progress in physics 566 • How does the world look when riding on
a light beam? 567 • Does light travel in a straight line? 567 • The concentration of
light 571 • Can one touch light? 572 • War, light and lies 574 • What is colour? 574
• What is the speed of light? – Again 577 • 200 years too late – negative refraction
indices 579 • Signals and predictions 580 • Does the aether exist? 581
582 Curiosities and fun challenges about light
How to prove you’re holy 582 • Do we see what exists? 583 • How does one make
pictures of the inside of the eye? 585 • How does one make holograms and other

Dvipsbugw
contents 13

three-dimensional images? 586 • Imaging 588 • Light as weapon? 589

589 15. Charges are discrete – the limits of classical electrodynamics
How fast do charges move? 590 • Challenges and curiosities about charge discrete-
ness 591
592 16. Electromagnetic effects and challenges
Is lightning a discharge? – Electricity in the atmosphere 597 • Does gravity make
charges radiate? 600 • Research questions 600 • Levitation 602 • Matter, levitation
and electromagnetic effects 604 • Why can we see each other? 612 • A summary
of classical electrodynamics and of its limits 614 Dvipsbugw
614 17. Classical physics in a nutshell – one and a half steps out of three
The future of planet Earth 616 • The essence of classical physics – the infinitely small
implies the lack of surprises 618 • Why have we not yet reached the top of the moun-
tain? 618
620 Bibliography
631 Intermezzo The Brain, L anguage and the Human Condition

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Evolution 632
632 Children and physics
Why a brain? 635 • What is information? 636 • What is memory? 637 • The capa-
city of the brain 639
641 What is language?
What is a concept? 644 • What are sets? What are relations? 646 • Infinity 648
• Functions and structures 650 • Numbers 651 • Why use mathematics? 656 •
Is mathematics a language? 657 • Curiosities and fun challenges about mathemat-
ics 657
659 Physical concepts, lies and patterns of nature
Are physical concepts discovered or created? 660 • How do we find physical patterns
and rules? 661 • What is a lie? 662 • Is this statement true? 667 • Curiosities and
fun challenges about lies 668
670 Observations
Have enough observations been recorded? 670 • Are all physical observables
known? 671 • Do observations take time? 673 • Is induction a problem in phys-
ics? 673
675 The quest for precision and its implications
What are interactions? – No emergence 676 • What is existence? 677 • Do things
exist? 678 • Does the void exist? 679 • Is nature infinite? 680 • Is the uni-
verse a set? 681 • Does the universe exist? 682 • What is creation? 683 • Is
nature designed? 685 • What is a description? 686 • Reason, purpose and explan-
Copyright © Christoph Schiller November 1997–May 2006

ation 687 • Unification and demarcation 688 • Pigs, apes and the anthropic prin-
ciple 689 • Does one need cause and effect in explanations? 691 • Is consciousness
required? 691 • Curiosity 692 • Courage 694
696 Bibliography

Second Part : Quantum Theory

What Is Matter? What Are Interactions?

704 Ch a pter V Quanta of Light and Matter

704 18. Minimum action – quantum theory for poets and lawyers

Dvipsbugw
14 contents

The effects of the quantum of action on motion 705 • The consequences of the
quantum of action for objects 706 • What does ‘quantum’ mean? 708 • Quantum
surprises 710 • Waves 714 • Information 715 • Curiosities and fun challenges
about the quantum of action 716
717 19. Light – the strange consequences of the quantum of action
What is colour? 717 • What is light? – Again 720 • The size of photons 721 •
Are photons countable? – Squeezed light 722 • The position of photons 724 • Are
photons necessary? 726 • How can a wave be made up of particles? 727 • Can light
move faster than light? – Virtual photons 733 • Indeterminacy of electric fields 734 Dvipsbugw
• Curiosities and fun challenges about photons 734
735 20. Motion of matter – beyond classical physics
Wine glasses and pencils 735 • Cool gas 736 • No rest 736 • Flows and the
quantization of matter 737 • Quantons 737 • The motion of quantons – matter
as waves 738 • Rotation and the lack of North Poles 740 • Silver, Stern and Ger-
lach 742 • The language of quantum theory and its description of motion 743 •
The state – or wave function – and its evolution 745 • Why are atoms not flat? Why

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

do shapes exist? 747 • Rest – spread and the quantum Zeno effect 747 • Tunnel-
ling, hills and limits on memory 748 • Spin and motion 749 • Relativistic wave
equations 750 • Maximum acceleration 751 • Curiosities and fun challenges about
quantum theory 751
753 21. Colours and other interactions between light and matter
What are stars made of? 753 • What determines the colour of atoms? 754 • Relativ-
istic hydrogen 756 • Relativistic wave equations – again 757 • Antimatter 759 •
Virtual particles and QED diagrams 759 • Compositeness 760 • Curiosities and
fun challenges about colour 761 • The strength of electromagnetism 762
764 Bibliography

771 Ch a pter VI Permutation of Particles

771 22. Are particles like gloves?
Why does indistinguishability appear in nature? 773 • Can particles be counted? 773
• What is permutation symmetry? 774 • Indistinguishability and symmetry 775
• The behaviour of photons 776 • The energy dependence of permutation sym-
metry 776 • Indistinguishability in quantum field theory 777 • How accurately
is permutation symmetry verified? 778 • Copies, clones and gloves 778
780 23. Rotations and statistics – visualizing spin
The belt trick 783 • The Pauli exclusion principle and the hardness of matter 785 •
Integer spin 786 • Is spin a rotation about an axis? 787 • Why is fencing with laser
beams impossible? 787 • Rotation requires antiparticles 788 • Limits and open
Copyright © Christoph Schiller November 1997–May 2006

questions of quantum statistics 789

791 Bibliography

793 Ch a pter VII Details of Quantum Theory and Electromagnetism

793 24. Superpositions and probabilities – quantum theory without ideology
794 Why are people either dead or alive?
Conclusions on decoherence, life and death 799
800 What is a system? What is an object?
Is quantum theory non-local? – A bit about the Einstein-Podolsky-Rosen para-
dox 801 • Curiosities and fun challenges about superpositions 803
805 What is all the fuss about measurements in quantum theory?
Hidden variables 809

Dvipsbugw
contents 15

811 Conclusions on probabilities and determinism

What is the difference between space and time? 813 • Are we good observers? 814
• What connects information theory, cryptology and quantum theory? 814 • Does
the universe have a wave function? And initial conditions? 815
816 25. Applied quantum mechanics – life, pleasure and the means to achieve them
816 Biology
Reproduction 817 • Quantum machines 818 • How do we move? – Molecular mo-
tors 820 • Curiosities and fun challenges about biology 822
825 The physics of pleasure Dvipsbugw
The nerves and the brain 827 • Clocks in quantum mechanics 827 • Do clocks
exist? 828 • Living clocks 830 • Metre sticks 831 • Why are predictions so difficult,
especially of the future? 831 • Decay and the golden rule 832 • Zeno and the present
in quantum theory 833 • What is motion? 834 • Consciousness – a result of the
quantum of action 834 • Why can we observe motion? 835 • Curiosities and fun
challenges about quantum experience 835
837 Chemistry – from atoms to DNA

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ribonucleic acid and Deoxyribonucleic acid 838 • Curiosities and fun challenges
about chemistry 839
840 Materials science
Why does the floor not fall? 840 • Rocks and stones 841 • How can one look
through matter? 842 • What is necessary to make matter invisible? 842 • How does
matter behave at the lowest temperatures? 844 • Curiosities and fun challenges about
materials science 844
846 Quantum technology
Motion without friction – superconductivity and superfluidity 847 • Quantized con-
ductivity 848 • The fractional quantum Hall effect 849 • Lasers and other spin-one
vector boson launchers 849 • Can two photons interfere? 851 • Can two electron
beams interfere? 852 • Challenges and dreams about quantum technology 853
854 26. Quantum electrodynamics – the origin of virtual reality
Ships, mirrors and the Casimir effect 854 • The Banach–Tarski paradox for va-
cuum 857 • The Lamb shift 857 • The QED Lagrangian 857 • Interactions and
virtual particles 857 • Vacuum energy 858 • Moving mirrors 858 • Photons hitting
photons 858 • Is the vacuum a bath? 859 • Renormalization – why is an electron
so light? 860
860 Curiosities and fun challenges of quantum electrodynamics
How can one move on perfect ice? – The ultimate physics test 864
865 Summary of quantum electrodynamics
Copyright © Christoph Schiller November 1997–May 2006

Open questions in QED 867

868 27. Quantum mechanics with gravitation – the first approach
Corrections to the Schrödinger equation 869 • A rephrased large number hypo-
thesis 869 • Is quantum gravity necessary? 870 • Limits to disorder 870 • Measur-
ing acceleration with a thermometer – Fulling–Davies–Unruh radiation 871
872 Black holes aren’t black
Gamma ray bursts 875 • Material properties of black holes 877 • How do black
holes evaporate? 878 • The information paradox of black holes 878 • More para-
doxes 879
880 Quantum mechanics of gravitation
The gravitational Bohr atom 880 • Decoherence of space-time 881 • Do gravitons
exist? 881 • Space-time foam 882 • No particles 883 • No science fiction 883 •

Dvipsbugw
16 contents

Not cheating any longer 883

884 Bibliography

897 Ch a pter VIII Inside the Nucleus

897 28. The structure of the nucleus – the densest clouds
A physical wonder – magnetic resonance imaging 897 • The size of nuclei 898 •
Nuclei are composed 902 • Nuclei can move alone – cosmic rays 904 • Nuclei de-
cay 908 • Why is hell hot? 910 • Nuclei can form composites 911 • Nuclei have
colours and shapes 911 • Motion in the nuclear domain – four types of motion 913 Dvipsbugw
• Nuclei react 913 • Bombs and nuclear reactors 914 • The Sun 915 • Curiosities
and fun challenges on radioactivity 917
921 29. The strong nuclear interaction and the birth of matter
Why do the stars shine? 921 • Why are fusion reactors not common yet? 923 •
Where do our atoms come from? 926 • The weak side of the strong interaction 926
• Bound motion, the particle zoo and the quark model 927 • The mass, shape and
colour of protons 928 • Experimental consequences of the quark model 929 • The

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Lagrangian of quantum chromodynamics 930 • The sizes and masses of quarks 933
• Confinement and the future of the strong interaction 933 • Curiosities about the
strong interactions 934
935 30. The weak nuclear interaction and the handedness of nature
Curiosities about the weak interactions 936 • Mass, the Higgs boson and a ten thou-
sand million dollar lie 937 • Neutrinium and other curiosities on the electroweak
interaction 938
940 31. The standard model of elementary particle physics – as seen on television
Conclusion and open questions about the standard model 940
941 32. Grand unification – a simple dream
Experimental consequences 941 • The state of grand unification 942
945 Bibliography

949 Ch a pter IX Advanced Quantum Theory (not yet available)

951 Ch a pter X Quantum Physics in a Nutshell

951 Quantum theory’s essence – the lack of the infinitely small
Achievements in precision 951 • Physical results of quantum theory 953 • Results
of quantum field theory 955 • Is quantum theory magic? 956 • The dangers of
buying a can of beans 957
959 The essence and the limits of quantum theory
Copyright © Christoph Schiller November 1997–May 2006

What is unexplained by quantum theory and general relativity? 959 • How to delude
oneself that one has reached the top of Motion Mountain 961 • What awaits us? 964
967 Bibliography

969 Intermezzo Bacteria, Flies and Knots

Bumblebees and other miniature flying systems 969 • Swimming 972 • Falling cats
and the theory of shape change 976 • Turning a sphere inside out 976 • Knots,
links and braids 978 • Knots in nature and on paper 980 • Clouds 983 • A
one-dimensional classical analogue of the Schrödinger equation 983 • Fluid space-
time 987 • Solid space-time 987 • Swimming in curved space 989 • Curiosities
and fun challenges on wobbly entities 990 • Outlook 991
992 Bibliography

Dvipsbugw
contents 17

Third Part : Motion Withou t Motion

What Are Space, Time and Particles?

996 Ch a pter XI General Relativity Versus Quantum Mechanics

The contradictions 996
998 33. Does matter differ from vacuum?
Planck scales 999 • Farewell to instants of time 1001 • Farewell to points in
Dvipsbugw
space 1003 • Farewell to the space-time manifold 1005 • Farewell to observables
and measurements 1008 • Can space-time be a lattice? – A glimpse of quantum geo-
metry 1009 • Farewell to particles 1010 • Farewell to mass 1013 • Curiosities and
fun challenges on Planck scales 1016 • Farewell to the big bang 1020 • The baggage
left behind 1020
1021 Some experimental predictions
1026 Bibliography

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1032 34. Nature at large scales – is the universe something or nothing?
Cosmological scales 1032 • Maximum time 1033 • Does the universe have a defin-
ite age? 1033 • How precise can age measurements be? 1034 • Does time exist? 1035
• What is the error in the measurement of the age of the universe? 1036 • Maximum
length 1039 • Is the universe really a big place? 1040 • The boundary of space-time
– is the sky a surface? 1041 • Does the universe have initial conditions? 1042 • Does
the universe contain particles and stars? 1042 • Does the universe contain masses
and objects? 1043 • Do symmetries exist in nature? 1045 • Does the universe have
a boundary? 1045 • Is the universe a set? – Again 1046 • Curiosities and fun chal-
lenges about the universe 1048 • Hilbert’s sixth problem settled 1049 • Does the
universe make sense? 1049 • A concept without a set eliminates contradictions 1051
• Extremal scales and open questions in physics 1051 • Is extremal identity a prin-
ciple of nature? 1051
1054 Bibliography
1056 35. The physics of love – a summary of the first two and a half parts
1067 Bibliography
1068 36. Maximum force and minimum distance – physics in limit statements
1068 Fundamental limits to all observables
Special relativity in one statement 1068 • Quantum theory in one statement 1069 •
General relativity in one statement 1071 • Deducing general relativity 1072 • De-
ducing universal gravitation 1075 • The size of physical systems in general relativ-
ity 1075 • A mechanical analogy for the maximum force 1075
Copyright © Christoph Schiller November 1997–May 2006

1076 Units and limit values for all physical observables

Limits to space and time 1078 • Mass and energy limits 1078 • Virtual particles –
a new definition 1079 • Limits in thermodynamics 1079 • Electromagnetic limits
and units 1080 • Vacuum and mass–energy – two sides of the same coin 1081 •
Curiosities and fun challenges about Planck limits 1082
1085 Upper and lower limits to observables
Size and energy dependence 1085 • Angular momentum, action and speed 1085 •
Force, power and luminosity 1086 • Acceleration 1087 • Momentum 1087 • Life-
time, distance and curvature 1088 • Mass change 1088 • Mass and density 1088
• The strange charm of the entropy bound 1089 • Temperature 1090 • Electro-
magnetic observables 1091 • Curiosities and fun challenges about limits to observ-
ables 1092

Dvipsbugw
18 contents

1093 Limits to measurement precision and their challenge to thought

Measurement precision and the existence of sets 1093 • Why are observers
needed? 1094 • A solution to Hilbert’s sixth problem 1095 • Outlook 1095
1097 Bibliography
1101 37. The shape of points – extension in nature
Introduction – vacuum and particles 1102 • How else can we show that matter and
vacuum cannot be distinguished? 1103
1104 Argument 1: The size and shape of elementary particles
Do boxes exist? 1105 • Can the Greeks help? – The limits of knives 1105 • Are Dvipsbugw
cross-sections finite? 1106 • Can one take a photograph of a point? 1106 • What is
the shape of an electron? 1108 • Is the shape of an electron fixed? 1109
1110 Argument 2: The shape of points in vacuum
Measuring the void 1111 • What is the maximum number of particles that fits inside
a piece of vacuum? 1111
1112 Argument 3: The large, the small and their connection
Small is large? 1113 • Unification and total symmetry 1113

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1115 Argument 4: Does nature have parts?
Does the universe contain anything? 1117 • An amoeba 1117
1118 Argument 5: The entropy of black holes
1120 Argument 6: Exchanging space points or particles at Planck scales
1121 Argument 7: The meaning of spin
1122 Present research
Conceptual checks of extension 1123 • Experimental falsification of models based
on extended entities 1124 • Possibilities for confirmation of extension models 1124
• Curiosities and fun challenges about extension 1125 • An intermediate status re-
port 1127 • Sexual preferences in physics 1127 • A physical aphorism 1127
1128 38. String theory – a web of dualities
1129 Strings and membranes – why string theory is so difficult
1129 Matrix models and M-theory
Masses and couplings 1130 • Outlook 1130
1132 Bibliography
1137 Ch a pter XII Unification (not yet available)

1139 Ch a pter XIII The Top of the Mountain (not yet available)

Fourth Part : Appendices

Copyright © Christoph Schiller November 1997–May 2006

1142 Appendix A Notation and Conventions

The Latin alphabet 1142 • The Greek alphabet 1143 • The Hebrew alphabet and
other scripts 1146 • Digits and numbers 1146 • The symbols used in the text 1147
• Calendars 1149 • Abbreviations and eponyms or concepts? 1151
1152 Bibliography
1154 Appendix B Units, Measurements and Constants
Planck’s natural units 1157 • Other unit systems 1158 • Curiosities and fun chal-
lenges about units 1160 • Precision and accuracy of measurements 1164 • Basic
physical constants 1165 • Useful numbers 1170
1171 Bibliography

Dvipsbugw
contents 19

1174 Appendix C Particle Properties

1192 Bibliography
1194 Appendix D Numbers and Spaces
1194 Numbers as mathematical structures
Complex numbers 1196 • Quaternions 1197 • Octonions 1202 • Other types of
numbers 1204 • Grassmann numbers 1204
1204 Vector spaces
1207 Algebras Dvipsbugw
Lie algebras 1209 • Classification of Lie algebras 1210 • Lie superalgebras 1211 •
The Virasoro algebra 1212 • Kac–Moody algebras 1212
1213 Topology – what shapes exist?
Topological spaces 1213 • Manifolds 1214 • Holes, homotopy and homology 1216
1217 Types and classification of groups
Lie groups 1218 • Connectedness 1219 • Compactness 1220
1224 Mathematical curiosities and fun challenges

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1226 Bibliography

1227 Appendix E S ources of Information on Motion

1233 Appendix F Challenge Hints and S olutions

1284 Appendix G List of Illustrations

Picture credits 1295
1298 Appendix H List of Tables

1301 Appendix I Name Index

1326 Appendix J Subject Index

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
Preface

Primum movere, deinde docere.

“ Antiquity

The intensity with which small children explore their environment suggests that there is ”
a drive to grasp the way the world works, a ‘physics instinct’, built into each of us. What
would happen if this drive, instead of being stifled during school education, as it usually
is, were allowed to thrive in an environment without bounds, reaching from the atoms Dvipsbugw
to the stars? Probably most adolescents would then know more about nature than most
senior physics teachers today. This text tries to provide this possibility to the reader. It
acts as a guide in an exploration, free of all limitations, of physics, the science of motion.
The project is the result of a threefold aim I have pursued since 1990: to present the basics
of motion in a way that is simple, up to date and vivid.
In order to be simple, the text focuses on concepts, while keeping mathematics to the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

necessary minimum. Understanding the concepts of physics is given precedence over
using formulae in calculations. The whole text is within the reach of an undergraduate. It
presents simple summaries of the main domains of physics.
There are three main stages in the physical description of motion. First, there is every-
day physics, or classical continuum physics. It is based on the existence of the infinitely
small and the infinitely large. In the second stage, each domain of physics is centred
around a basic inequality for the main observable. Thus, statistical thermodynamics lim-
its entropy by S E k~2; special relativity limits speeds by v D c; general relativity limits
force by F D c 4 ~4G; quantum theory limits action by L E ħ~2; and quantum electrody-
namics limits change of charge by ∆q E e. These results, though not so well known, are
proved rigorously. It is shown that within each domain, the principal equations follow
from the relevant limit. Basing the domains of physics on limit principles allows them to
be introduced in a simple, rapid and intuitive way. The third and final stage is the unific-
ation of all these limits in a single description of motion. This unusual way of learning
physics should reward the curiosity of every reader – whether student or professional.
In order to be up to date, the text includes discussions of quantum gravity, string theory
and M theory. Meanwhile, the standard topics – mechanics, electricity, light, quantum
theory, particle physics and general relativity – are enriched by the many gems – both
theoretical and empirical – that are scattered throughout the scientific literature.
In order to be vivid, a text must be challenging, questioning and daring. This text tries
Copyright © Christoph Schiller November 1997–May 2006

to startle the reader as much as possible. Reading a book on general physics should be like
going to a magic show. We watch, we are astonished, we do not believe our eyes, we think,
and finally – maybe – we understand the trick. When we look at nature, we often have
the same experience. The text tries to intensify this by following a simple rule: on each
page, there should be at least one surprise or provocation for the reader to think about.
Numerous interesting challenges are proposed. Hints or answers to these are given in an
appendix.
The strongest surprises are those that seem to contradict everyday experience. Most
of the surprises in this text are taken from daily life: in particular, from the the things
one experiences when climbing a mountain. Observations about trees, stones, the Moon,
the sky and people are used wherever possible; complex laboratory experiments are men-

Dvipsbugw
 preface 21

tioned only where necessary. These surprises are organized so as to lead in a natural way
to the most extreme conclusion of all, namely that continuous space and time do not exist.
The concepts of space and time, useful as they may be in everyday life, are only approxim-
ations. Indeed, they turn out to be mental crutches that hinder the complete exploration
of the world.
Giving full rein to one’s curiosity and thought leads to the development of a strong
and dependable character. The motto of the text, a famous statement by Harmut von
Hentig on pedagogy, translates as: ‘To clarify things, to fortify people.’ Exploring any limit
requires courage; and courage is also needed to abandon space and time as tools for the Dvipsbugw
description of the world. Changing habits of thought produces fear, often hidden by anger;
but we grow by overcoming our fears. Achieving a description of the world without the
use of space and time may be the most beautiful of all adventures of the mind.

Eindhoven and other places, 1 May 2006

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

A request

The text is and remains free for everybody. In exchange for getting the file for free, please
send me a short email on the following issues:
— What was unclear?
— What did you miss?
Challenge 1 ny — What should be improved or corrected?
Copyright © Christoph Schiller November 1997–May 2006

Feedback on the specific points listed on the http://www.motionmountain.net/project.

html web page is most welcome of all. On behalf of myself and all other readers, thank you
in advance for your input. For a particularly useful contribution you will be mentioned
– if you want – in the acknowledgements, receive a reward, or both. But above all, enjoy
the reading.

C. Schiller
[email protected]

Dvipsbugw
 Acknowledgements

Many people who have kept their gift of curiosity alive have helped to make this project
come true. Most of all, Saverio Pascazio has been – present or not – a constant reference
for this project. Fernand Mayné, Anna Koolen, Ata Masafumi, Roberto Crespi, Luca Bom-
belli, Herman Elswijk, Marcel Krijn, Marc de Jong, Martin van der Mark, Kim Jalink, my
parents Peter and Isabella Schiller, Mike van Wijk, Renate Georgi, Paul Tegelaar, Barbara
and Edgar Augel, M. Jamil, Ron Murdock, Carol Pritchard and, most of all, my wife Britta
have all provided valuable advice and encouragement. Dvipsbugw
Many people have helped with the project and the collection of material. Most use-
ful was the help of Mikael Johansson, Bruno Barberi Gnecco, Lothar Beyer, the numer-
ous improvements by Bert Sierra, the detailed suggestions by Claudio Farinati, the many
improvements by Eric Sheldon, the continuous help and advice of Jonatan Kelu, and in
particular the extensive, passionate and conscientious help of Adrian Kubala.
Important material was provided by Bert Peeters, Anna Wierzbicka, William Beaty,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Jim Carr, John Merrit, John Baez, Frank DiFilippo, Jonathan Scott, Jon Thaler, Luca
Bombelli, Douglas Singleton, George McQuarry, Tilman Hausherr, Brian Oberquell, Peer
Zalm, Martin van der Mark, Vladimir Surdin, Julia Simon, Antonio Fermani, Don Page,
Stephen Haley, Peter Mayr, Allan Hayes, Igor Ivanov, Doug Renselle, Wim de Muynck,
Steve Carlip, Tom Bruce, Ryan Budney, Gary Ruben, Chris Hillman, Olivier Glassey,
Jochen Greiner, squark, Martin Hardcastle, Mark Biggar, Pavel Kuzin, Douglas Breb-
ner, Luciano Lombardi, Franco Bagnoli, Lukas Fabian Moser, Dejan Corovic, Paul Van-
noni, John Haber, Saverio Pascazio, Klaus Finkenzeller, Leo Volin, Jeff Aronson, Roggie
Boone, Lawrence Tuppen, Quentin David Jones, Arnaldo Uguzzoni, Frans van Nieuw-
poort, Alan Mahoney, Britta Schiller, Petr Danecek, Ingo Thies, Vitaliy Solomatin, Carl
Offner, Nuno Proença, Elena Colazingari, Paula Henderson, Daniel Darre, Wolfgang
Rankl, John Heumann, Joseph Kiss, Martha Weiss, Antonio González, Antonio Martos,
John Heumann, André Slabber, Ferdinand Bautista, Zoltán Gácsi, Pat Furrie, Michael
Reppisch, Enrico Pasi, Thomas Köppe, Martin Rivas, Herman Beeksma, Tom Helmond,
John Brandes, Vlad Tarko, Nadia Murillo, Ciprian Dobra, Romano Perini, Harald van Lin-
tel, Andrea Conti, François Belfort, Dirk Van de Moortel, Heinrich Neumaier, Jarosław
Królikowski, John Dahlman, Fathi Namouni, Elmar Bartel plus a number of people who
wanted to remain unnamed.
The software tools were refined with extensive help on fonts and typesetting by Mi-
chael Zedler and Achim Blumensath and with the repeated and valuable support of
Copyright © Christoph Schiller November 1997–May 2006

Donald Arseneau; help came also from Ulrike Fischer, Piet van Oostrum, Gerben Wi-
erda, Klaus Böhncke, Craig Upright, Herbert Voss, Andrew Trevorrow, Danie Els, Heiko
Oberdiek, Sebastian Rahtz, Don Story, Vincent Darley, Johan Linde, Joseph Hertzlinger,
Rick Zaccone and John Warkentin.
Many illustrations in the text were made available by the copyright holders. A warm
Page 1295 thank you to all of them; they are mentioned in Appendix G. In particular, Luca Gastaldi
and Antonio Martos produced images specifically for this text. The improvement in the
typesetting is due to the professional typographic consulting of Ulrich Dirr. The design
of the book and the website owe much to the suggestions and support of my wife Britta.

Dvipsbugw
 An appetizer
Die Lösung des Rätsels des Lebens in Raum und

“ Zeit liegt außerhalb von Raum und Zeit.*

Ludwig Wittgenstein, Tractatus, 6.4312

”
W hat is the most daring and amazing journey we can make in a lifetime?
e can travel to remote places, like adventurers, explorers or cosmonauts;
e can look at even more distant places, like astronomers; we can visit the past, like
historians, archaeologists, evolutionary biologists or geologists; or we can delve deeply
Dvipsbugw
into the human soul, like artists or psychologists. All these voyages lead either to other
places or to other times. However, we can do better.
The most daring trip of all is not the one leading to the most inaccessible place, but
the one leading to where there is no place at all. Such a journey implies leaving the prison
of space and time and venturing beyond it, into a domain where there is no position, no
present, no future and no past, where we are free of the restrictions imposed by space

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

and time, but also of the mental reassurance that these concepts provide. In this domain,
many new discoveries and new adventures await us. Almost nobody has ever been there;
humanity’s journey there has so far taken at least 2500 years, and is still not complete.
To venture into this domain, we need to be curious about the essence of travel itself.
The essence of travel is motion. By exploring motion we will be led to the most fascinating
adventures in the universe.
The quest to understand motion in all its details and limitations can be pursued behind
a desk, with a book, some paper and a pen. But to make the adventure more vivid, this
text uses the metaphor of a mountain ascent. Every step towards the top corresponds to
a step towards higher precision in the description of motion. In addition, with each step
the scenery will become more delightful. At the top of the mountain we shall arrive in a
domain where ‘space’ and ‘time’ are words that have lost all meaning and where the sight
of the world’s beauty is overwhelming and unforgettable.
Thinking without time or space is difficult but fascinating. In order to get a taste of
the issues involved, try to respond to the following questions without referring to either
Challenge 2 n space or time:**
— Can you prove that two points extremely close to each other always leave room for a
third point in between?
— Can you describe the shape of a knot over the telephone?
Copyright © Christoph Schiller November 1997–May 2006

— Can you explain on the telephone what ‘right’ and ‘left’ mean, or what a mirror is?
— Can you make a telephone appointment with a friend without using any terms of time
or position, such as ‘clock’, ‘hour’, ‘place’, ‘where’, ‘when’, ‘at’, ‘near’, ‘before’, ‘after’, ‘near’,
‘upon’, ‘under’, ‘above’, ‘below’?
— Can you describe the fall of a stone without using the language of space or time?
— Do you know of any observation at all that you can describe without concepts from
the domains of ‘space’, ‘time’ or ‘object’?

* ‘The solution of the riddle of life in space and time lies outside space and time.’
** Solutions to, and comments on, challenges are either given on page 1233 or later on in the text. Challenges
are classified as research level (r), difficult (d), normal student level (n) and easy (e). Challenges for which
no solution has yet been included in the book are marked (ny).

Dvipsbugw
24 an appetizer

— Can you explain what time is? And what clocks are?
— Can you imagine a finite history of the universe, but without a ‘first instant of time’?
— Can you imagine a domain of nature where matter and vacuum are indistinguishable?
— Have you ever tried to understand why motion exists?
This book explains how to achieve these and other feats, bringing to completion an an-
cient dream of the human spirit, namely the quest to describe every possible aspect of
motion.
Why do your shoelaces remain tied? They do so because space has three dimensions. Dvipsbugw
Why not another number? The question has taxed researchers for thousands of years. The
answer was only found by studying motion down to its smallest details, and by exploring
its limits.
Why do the colours of objects differ? Why does the Sun shine? Why does the Moon not
fall out of the sky? Why is the sky dark at night? Why is water liquid but fire not? Why
is the universe so big? Why is it that birds can fly but men can’t? Why is lightning not

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

straight? Why are atoms neither square, nor the size of cherries? These questions seem to
have little in common – but they are related. They are all about motion – about its details
and its limitations. Indeed, they all appear, and are answered, in this text. Studying the lim-
its of motion, we discover that when a mirror changes its speed it emits light. We also dis-
cover that gravity can be measured with a thermometer. We find that there are more cells
in the brain than stars in the galaxy, giving substance to the idea that people have a whole
universe in their head. Exploring any detail of motion is already an adventure in itself.
By exploring the properties of motion we will find that, despite appearance, motion
never stops. We will find out why the floor cannot fall. We will understand why com-
puters cannot be made arbitrarily fast. We will see that perfect memory cannot exist. We
will understand that nothing can be perfectly black. We will learn that every clock has a
certain probability of going backwards. We will discover that time does not exist. We will
find that all objects in the world are connected. We will learn that matter cannot be distin-
guished precisely from empty space. We will learn that we are literally made of nothing.
We will learn quite a few things about our destiny. And we will understand why the world
is the way it is.
The quest to understand motion, together with all its details and all its limits, involves
asking and answering three specific questions.
How do things move? Motion usually defined as is an object changing position over
time. This seemingly mundane definition actually encompasses general relativity, one of
Copyright © Christoph Schiller November 1997–May 2006

the most amazing descriptions of nature ever imagined. We will find that space is warped,
that light does not usually travel in a straight line, and that time is not the same for every-
body. We will discover that there is a maximum force of gravity, and that gravity is not an
interaction, but rather the change of time with position. We will see how the blackness
of the sky at night proves that the universe has a finite age. We will also discover that
there is a smallest entropy in nature, which prevents us from knowing everything about
a physical system. In addition, we will discover the smallest electrical charge. These and
other strange properties and phenomena of motion are summarized in the first part of
this text, whose topic is classical physics. It leads directly to the next question.
What are things? Things are composites of particles. Not only tangible things, but all
interactions and forces – those of the muscles, those that make the Sun burn, those that

Dvipsbugw
 an appetizer 25

make the Earth turn, those that determine the differences between attraction, repulsion,
friction, creation and annihilation – are made of particles as well. The growth of trees,
the colours of the sky, the burning of fire, the warmth of a human body, the waves of the
sea and the mood changes of people are all composed of particles in motion. This story is
told in more detail in the second part of the text, which deals with quantum mechanics.
Here we will learn that there is a smallest change in nature. This minimum value forces
everything to keep changing. In particular, we will learn that it is impossible to completely
fill a glass of wine, that eternal life is impossible, and that light can be transformed into
matter. If you find this boring, you can read about the substantial dangers involved in Dvipsbugw
Page 957 buying a can of beans.
The first two parts of this text can be summarized with the help of a few limit principles:

statistical thermodynamics limits entropy: S E k~2

special relativity limits speed: v D c
F D c 4 ~4G

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

general relativity limits force:
quantum theory limits action: L E ħ~2
quantum electrodynamics limits change of charge: ∆q E e .

In other words, each of the constants of nature k~2, c, c 4 ~4G, ħ~2 and e that appear above
is a limit value. We will discover in each case that the equations of the corresponding
domain of physics follow from this limit property. After these results, the path is prepared
for the final part of our mountain ascent.
What are particles, position and time? The recent results of an age-long search are
making it possible to start answering this question. One just needs to find a description
that explains all limit principles at the same time. This third part is not yet complete,
because the necessary research results are not yet available. Nevertheless, some of the
intermediate results are striking:
— It is known already that space and time are not continuous; that – to be precise –
neither points nor particles exist; and that there is no way to distinguish space from
Page 998 time, nor vacuum from matter, nor matter from radiation.
— It is known already that nature is not simply made of particles and vacuum.
— It seems that position, time and particles are aspects of a complex, extended entity that
is incessantly varying in shape.
Copyright © Christoph Schiller November 1997–May 2006

— Among the mysteries that should be cleared up in the coming years are the origin of
the three dimensions of space, the origin of time and the details of the big bang.
— Current research indicates that motion is an intrinsic property of matter and radiation
and that, as soon as we introduce these two concepts in our description of nature,
motion appears automatically. Indeed, it is impossible not to introduce these concepts,
because they necessarily appear when we divide nature into parts, an act we cannot
avoid because of the mechanisms of our senses and therefore of our thinking.
— Current research also indicates that the final, completely precise, description of nature
does not use any form of infinity. We find, step by step, that all infinities appearing in
the human description of nature – both the infinitely large and the infinitely small –
result from approximations. ‘Infinity’ turns out to be merely a conceptual convenience

Dvipsbugw
26 an appetizer

that has no place in nature. However, we find that the precise description does not in-
clude any finite quantities either! These and many other astonishing results of modern
physics appear in the third part of this text.
This third and final part of the text thus describes the present state of the search for a
unified theory encompassing general relativity and quantum mechanics. To achieve such
a description, the secrets of space, time, matter and forces have to be unravelled. It is a
fascinating story, assembled piece by piece by thousands of researchers. At the end of the
ascent, at the top of the mountain, the idea of motion will have undergone a complete Dvipsbugw
transformation. Without space and time, the world will look magical, incredibly simple
and fascinating: pure beauty.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
First Part

Classical Physics:
How Do Things and Images Move?

Where the experience of hiking and other motion

leads us to introduce, for its description,
the concepts of velocity, time, length, mass and charge,
as well as action, field and manifold,
allowing us to discover limits to
speed, entropy, force and charge,
and thus to understand – among other things –
why we have legs instead of wheels,
how empty space can bend, wobble and move,
what love has to do with magnets and amber,
and why we can see the stars.
 C h a pter I

GALI LEAN MOTION

Dvipsbugw

W
ham! The lightning striking the tree nearby violently disrupts our quiet forest
alk and causes our hearts to suddenly beat faster. In the top of the tree
e see the fire start and fade again. The gentle wind moving the leaves around
us helps to restore the calmness of the place. Nearby, the water in a small river follows
its complicated way down the valley, reflecting on its surface the ever-changing shapes of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the clouds.

1. why should we care abou t motion?

All motion is an illusion.

“
Motion is everywhere: friendly and threatening, terrible and
beautiful. It is fundamental to our human existence. We need
Zeno of Elea*

”
motion for growing, for learning, for thinking and for enjoying
life. We use motion for walking through a forest, for listening
to its noises and for talking about all this. Like all animals, we
rely on motion to get food and to survive dangers. Plants by
contrast cannot move (much); for their self-defence, they de-
veloped poisons. Examples of such plants are the stinging nettle,
the tobacco plant, digitalis, belladonna and poppy; poisons in-
clude caffeine, nicotine, curare and many others. Poisons such
as these are at the basis of most medicines. Therefore, most
medicines exist essentially because plants have no legs. Like all
Copyright © Christoph Schiller November 1997–May 2006

living beings, we need motion to reproduce, to breathe and to

digest; like all objects, motion keeps us warm.
Motion is the most fundamental observation about nature F I G U R E 1 An example of
at large. It turns out that everything that happens in the world is motion observed in nature
some type of motion. There are no exceptions. Motion is such
a basic part of our observations that even the origin of the word is lost in the darkness
of Indo-European linguistic history. The fascination of motion has always made it a fa-
vourite object of curiosity. By the fifth century bce in ancient Greece, its study had been
Ref. 1 given a name: physics.

* Zeno of Elea (c. 450 bce), one of the main exponents of the Eleatic school of philosophy.

Dvipsbugw
 why should we care abou t motion? 29

ASTRONOMY

tom
MATERIAL SCIENCES

CHEMISTRY
part III: mt GEOSCIENCES
MEDICINE
part II: qt Motion Dvipsbugw
Mountain

part I: clm, gr & em

BIOLOGY
PHYSICS

emotion bay

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

MATHEMATICS
THE HUMANITIES

social sea

F I G U R E 2 Experience Island, with Motion Mountain and the trail to be followed (clm: classical mechanics, gr:
general relativity, em: electromagnetism, qt: quantum theory, mt: M-theory, tom: the theory of motion)

Motion is also important to the human condition. Who are we? Where do we come
from? What will we do? What should we do? What will the future bring? Where do people
come from? Where do they go? What is death? Where does the world come from? Where
does life lead? All these questions are about motion. The study of motion provides an-
swers that are both deep and surprising.
Ref. 3 Motion is mysterious. Though found everywhere – in the stars, in the tides, in our
Copyright © Christoph Schiller November 1997–May 2006

eyelids – neither the ancient thinkers nor myriads of others in the 25 centuries since then
have been able to shed light on the central mystery: what is motion? We shall discover
that the standard reply, ‘motion is the change of place in time’, is inadequate. Just recently
an answer has finally been found. This is the story of the way to find it.
Motion is a part of human experience. If we imagine human experience as an island,
then destiny, symbolized by the waves of the sea, carried us to its shore. Near the centre of
the island an especially high mountain stands out. From its top we can see over the whole
landscape and get an impression of the relationships between all human experiences, in
particular between the various examples of motion. This is a guide to the top of what I
have called Motion Mountain. The hike is one of the most beautiful adventures of the
human mind. Clearly, the first question to ask is:

Dvipsbugw
 30 i galilean motion • 1. why should we care abou t motion?

Dvipsbugw
F I G U R E 3 Illusions of motion: look at the figure on the left and slightly move the page, or look at the
white dot at the centre of the figure on the right and move your head back and forward

Does motion exist?

Das Rätsel gibt es nicht. Wenn sich eine Frage

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

überhaupt stellen läßt, so kann sie beantwortet
werden.*
Ludwig Wittgenstein, Tractatus, 6.5

Ref. 2
To sharpen the mind for the issue of motion’s existence, have a look at Figure 3 and follow
the instructions. In both cases the figures seem to rotate. One can experience similar
effects if one walks over Italian cobblestone in wave patterns or if one looks at the illusions
”
on the webpage www.ritsumei.ac.jp/~akitaoka/. How can one make sure that real motion
Challenge 3 n is different from these or other similar illusions?**
Many scholars simply argued that motion does not exist at all. Their arguments deeply
Ref. 4 influenced the investigation of motion. For example, the Greek philosopher Parmenides
(born c. 515 bce in Elea, a small town near Naples, in southern Italy) argued that since
nothing comes from nothing, change cannot exist. He underscored the permanence of
Ref. 5 nature and thus consistently maintained that all change and thus all motion is an illusion.
Heraclitus (c. 540 to c. 480 bce ) held the opposite view. He expressed it in his famous
statement πάντα ῥεῖ ‘panta rhei’ or ‘everything flows’.*** He saw change as the essence
of nature, in contrast to Parmenides. These two equally famous opinions induced many
scholars to investigate in more detail whether in nature there are conserved quantities or
whether creation is possible. We will uncover the answer later on; until then, you might
Challenge 4 n ponder which option you prefer.
Parmenides’ collaborator Zeno of Elea (born c. 500 bce) argued so intensely against
Copyright © Christoph Schiller November 1997–May 2006

motion that some people still worry about it today. In one of his arguments he claims –
in simple language – that it is impossible to slap somebody, since the hand first has to
travel halfway to the face, then travel through half the distance that remains, then again
so, and so on; the hand therefore should never reach the face. Zeno’s argument focuses
on the relation between infinity and its opposite, finitude, in the description of motion.
Ref. 6 In modern quantum theory, a similar issue troubles many scientists up to this day.
Zeno also maintained that by looking at a moving object at a single instant of time,

* The riddle does not exist. If a question can be put at all, it can be answered.
** Solutions to challenges are given either on page 1233 or later on in the text. Challenges are classified as
research level (r), difficult (d), normal student level (n) and easy (e). Challenges with no solution as yet are

Dvipsbugw
why should we care abou t motion? 31

Dvipsbugw

F I G U R E 4 How much water is required to make a bucket hang vertically? At what angle does the
pulled reel change direction of motion? (© Luca Gastaldi)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

one cannot maintain that it moves. Zeno argued that at a single instant of time, there is
no difference between a moving and a resting body. He then deduced that if there is no
difference at a single time, there cannot be a difference for longer times. Zeno therefore
questioned whether motion can clearly be distinguished from its opposite, rest. Indeed,
in the history of physics, thinkers switched back and forward between a positive and a
negative answer. It was this very question that led Albert Einstein to the development of
general relativity, one of the high points of our journey. We will follow the main answers
given in the past. Later on, we will be even more daring: we will ask whether single in-
stants of time do exist at all. This far-reaching question is central to the last part of our
adventure.
When we explore quantum theory, we will discover that motion is indeed – to a certain
extent – an illusion, as Parmenides claimed. More precisely, we will show that motion is
observed only due to the limitations of the human condition. We will find that we exper-
ience motion only because we evolved on Earth, with a finite size, made of a large but
finite number of atoms, with a finite but moderate temperature, electrically neutral, large
compared with a black hole of our same mass, large compared with our quantum mech-
anical wavelength, small compared with the universe, with a limited memory, forced by
our brain to approximate space and time as continuous entities, and forced by our brain
Copyright © Christoph Schiller November 1997–May 2006

to describe nature as made of different parts. If any one of these conditions were not ful-
filled, we would not observe motion; motion, then, would not exist. Each of these results
can be uncovered most efficiently if we start with the following question:

marked (ny).
*** Appendix A explains how to read Greek text.

Dvipsbugw
 32 i galilean motion • 1. why should we care abou t motion?

Anaximander Empedocles Eudoxus Ctesibius Strabo Frontinus Cleomedes

Anaximenes Aristotle Archimedes Varro Maria Artemidor
the Jew
Pythagoras Heraclides Konon Athenaius Josephus Sextus Empiricus
Almaeon Philolaos Theophrastus Chrysippos Eudoxus Pomponius Dionysius Athenaios Diogenes
of Kyz. Mela Periegetes of Nauc. Laertius
Heraclitus Zeno Autolycus Eratosthenes Sosigenes Marinus
Xenophanes Anthistenes Euclid Dositheus Virgilius Menelaos Philostratus
Thales Parmenides Archytas Epicure Biton Polybios Horace Nicomachos Apuleius

Alexander Ptolemaios II Ptolemaios VIII Caesar Nero Trajan

Dvipsbugw
600 BCE 500 400 300 200 100 1 100 200

Socrates Plato Ptolemaios I Cicero Seneca

Anaxagoras Aristarchus Asclepiades Livius Dioscorides Ptolemy

Leucippus Pytheas Archimedes Seleukos Vitruvius Geminos Epictetus
Protagoras Erasistratus Diocles Manilius Demonax Diophantus
Oenopides Aristoxenus Aratos Philo Dionysius Diodorus Valerius Theon Alexander

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of Byz. Thrax Siculus Maximus of Smyrna of Aphr.
Hippocrates Berossos
Herodotus Herophilus Apollonius Theodosius Plinius Rufus Galen
Senior
Democritus Straton Hipparchus Lucretius Aetius Arrian
Hippasos Speusippos Dikaiarchus Poseidonius Heron Plutarch Lucian

F I G U R E 5 A time line of scientific and political personalities in antiquity (the last letter of the name is
aligned with the year of death)

How should we talk about motion?

Je hais le mouvement, qui déplace les lignes,

“ Et jamais je ne pleure et jamais je ne ris.

Like any science, the approach of physics is twofold: we advance with precision and with
Charles Baudelaire, La Beauté.*

”
curiosity. Precision makes meaningful communication possible, and curiosity makes it
worthwhile.** Whenever one talks about motion and aims for increased precision or for
more detailed knowledge, one is engaged, whether knowingly or not, in the ascent of
Motion Mountain. With every increase in the precision of the description, one gains some
height. The examples of Figure 4 make the point. When you fill a bucket with a small
amount of water, it does not hang vertically. (Why?) If you continue adding water, it starts
Copyright © Christoph Schiller November 1997–May 2006

Challenge 5 ny to hang vertically at a certain moment. How much water is necessary? When you pull a
thread from a reel in the way shown, the reel will move either forwards or backwards,
depending on the angle at which you pull. What is the limiting angle between the two
possibilities?
High precision means going into fine details. This method actually increases the pleas-
ure of the adventure.*** The higher we get on Motion Mountain, the further we can see

* Charles Baudelaire (b. 1821 Paris, d. 1867 Paris) Beauty: ‘I hate movement, which changes shapes, and
Ref. 7 never do I weep and never do I laugh.’ Beauty.
Ref. 8 ** For a collection of interesting examples of motion in everyday life, see the excellent book by Walker.
Challenge 6 n *** Distrust anybody who wants to talk you out of investigating details. He is trying to deceive you. Details
are important. Also, be vigilant also during this walk.

Dvipsbugw
 why should we care abou t motion? 33

TA B L E 1 Content of books about motion found in a public library

Motion topics Motion topics

motion pictures motion as therapy for cancer, diabetes, acne and de-
pression
motion perception Ref. 21 motion sickness
motion for fitness and wellness motion for meditation
motion control in sport motion ability as health check Dvipsbugw
perpetual motion motion in dance, music and other arts
motion as proof of various gods Ref. 10 motion of stars and angels Ref. 11
economic efficiency of motion the connection between motional and emotional
habits
motion as help to overcome trauma motion in psychotherapy Ref. 12
locomotion of insects, horses and robots commotion

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

motions in parliament movements in art, sciences and politics
movements in watches movements in the stock market
movement teaching and learning movement development in children
musical movements troop movements Ref. 13
religious movements bowel movements
moves in chess cheating moves in casinos Ref. 14
connection between gross national product and citizen mobility

and the more our curiosity is rewarded. The views offered are breathtaking, especially
from the very top. The path we will follow – one of the many possible routes – starts from
Ref. 9 the side of biology and directly enters the forest that lies at the foot of the mountain.
Intense curiosity drives us to go straight to the limits: understanding motion requires
exploration of the largest distances, the highest velocities, the smallest particles, the
strongest forces and the strangest concepts. Let us begin.

What are the types of motion?

Every movement is born of a desire for change.

“ Antiquity

”
Copyright © Christoph Schiller November 1997–May 2006

The best place to obtain a general overview on the types of motion is a large library; this
is shown in Table 1. The domains in which motion, movements and moves play a role
are indeed varied. Already in ancient Greece people had the suspicion that all types of
motion, as well as many other types of change, are related. It is usual to distinguish at
least three categories.
The first category of change is that of material transport, such as a person walking or
a leaf falling from a tree. Transport is the change of position or orientation of objects. To
a large extent, the behaviour of people also falls into this category.
A second category of change groups observations such as the dissolution of salt in wa-
ter, the formation of ice by freezing, the putrefaction of wood, the cooking of food, the

Dvipsbugw
 34 i galilean motion • 1. why should we care abou t motion?

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 6 An example of transport

coagulation of blood, and the melting and alloying of metals. These changes of colour,
brightness, hardness, temperature and other material properties are all transformations.
Transformations are changes not visibly connected with transport. To this category, a few
ancient thinkers added the emission and absorption of light. In the twentieth century,
these two effects were proven to be special cases of transformations, as were the newly
discovered appearance and disappearance of matter, as observed in the Sun and in radio-
activity. Mind change, such as change of mood, of health, of education and of character,
Ref. 15 is also (mostly) a type of transformation.
Ref. 16 The third and especially important category of change is growth; it is observed for
animals, plants, bacteria, crystals, mountains, stars and even galaxies. In the nineteenth
century, changes in the population of systems, biological evolution, and in the twentieth
century, changes in the size of the universe, cosmic evolution, were added to this category.
Traditionally, these phenomena were studied by separate sciences. Independently they all
arrived at the conclusion that growth is a combination of transport and transformation.
The difference is one of complexity and of time scale.
At the beginnings of modern science during the Renaissance, only the study of trans-
Copyright © Christoph Schiller November 1997–May 2006

port was seen as the topic of physics. Motion was equated to transport. The other two do-
mains were neglected by physicists. Despite this restriction, the field of enquiry remains
large, covering a large part of Experience Island. The obvious temptation is to structure
the field by distinguishing types of transport by their origin. Movements such as those
of the legs when walking are volitional, because they are controlled by one’s will, whereas
movements of external objects, such as the fall of a snowflake, which one cannot influ-
ence by will-power, are called passive. Children are able to make this distinction by about
the age of six, and this marks a central step in the development of every human towards
a precise description of the environment.* From this distinction stems the historical but
* Failure to pass this stage completely can result in a person having various strange beliefs, such as believing

Dvipsbugw
 why should we care abou t motion? 35

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 7 Transport, growth and transformation (© Philip Plisson)

now outdated definition of physics as the science of the motion of non-living things.
Then, one day, machines appeared. From that moment, the distinction between vo-
litional and passive motion was put into question. Like living beings, machines are self-
moving and thus mimic volitional motion. However, careful observation shows that every
part in a machine is moved by another, so that their motion is in fact passive. Are living
beings also machines? Are human actions examples of passive motion as well? The ac-
cumulation of observations in the last 100 years made it clear that volitional movement*
indeed has the same physical properties as passive motion in non-living systems. (Of
course, from the emotional viewpoint, the differences are important; for example, grace
Ref. 17 can only be ascribed to volitional movements.) The distinction between the two types is
thus not necessary and is dropped in the following. Since passive and volitional motion
have the same properties, through the study of motion of non-living objects we can learn
something about the human condition. This is most evident when touching the topics of
determinism, causality, probability, infinity, time and sex, to name but a few of the themes
we will encounter on the way.
With the accumulation of observations in the nineteenth and twentieth centuries, even
Copyright © Christoph Schiller November 1997–May 2006

more of the historical restrictions on the study of motion were put into question. Extens-
ive observations showed that all transformations and all growth phenomena, including
behaviour change and evolution, are also examples of transport. In other words, over
2 000 years of studies have shown that the ancient classification of observations was use-

in the ability to influence roulette balls, as found in compulsive players, or in the ability to move other bod-
ies by thought, as found in numerous otherwise healthy-looking people. An entertaining and informative
account of all the deception and self-deception involved in creating and maintaining these beliefs is given
by James R andi, The Faith Healers, Prometheus Books, 1989. A professional magician, he presents many
similar topics in several of his other books. See also his http://www.randi.org website for more details.
* The word ‘movement’ is rather modern; it was imported into English from the old French and became
popular only at the end of the eighteenth century. It is never used by Shakespeare.

Dvipsbugw
 36 i galilean motion • 1. why should we care abou t motion?

less: all change is transport. In the middle of the twentieth century this culminated in the
confirmation of an even more specific idea already formulated in ancient Greece: every
type of change is due to the motion of particles. It takes time and work to reach this con-
clusion, which appears only when one relentlessly pursues higher and higher precision
in the description of nature. The first two parts of this adventure retrace the path to this
Challenge 7 n result. (Do you agree with it?)
The last decade of the twentieth century changed this view completely. The particle
idea turns out to be wrong. This new result, already suggested by advanced quantum
theory, is reached in the third part of our adventure through a combination of careful Dvipsbugw
observation and deduction. But we still have some way to go before we reach there.
At present, at the beginning of our walk, we simply note that history has shown that
classifying the various types of motion is not productive. Only by trying to achieve max-
imum precision can we hope to arrive at the fundamental properties of motion. Precision,
not classification is the path to follow. As Ernest Rutherford said: ‘All science is either
physics or stamp collecting.’

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

To achieve precision in our description of motion, we need to select specific examples
of motion and study them fully in detail. It is intuitively obvious that the most precise
description is achievable for the simplest possible examples. In everyday life, this is the
case for the motion of any non-living, solid and rigid body in our environment, such as
a stone thrown through the air. Indeed, like all humans, we learned to throw objects long
Ref. 18 before we learned to walk. Throwing is one of the first physical experiment we performed
by ourselves.* During our early childhood, by throwing stones, toys and other objects
until our parents feared for every piece of the household, we explored the perception and
the properties of motion. We do the same.
Die Welt ist unabhängig von meinem Willen.**

Perception, permanence and change

“ Ludwig Wittgenstein, Tractatus, 6.373

”
Only wimps study only the general case; real

“ scientists pursue examples.

Beresford Parlett

Human beings enjoy perceiving. Perception starts before birth, and we continue enjoying
”
it for as long as we can. That is why television, even when devoid of content, is so success-
ful. During our walk through the forest at the foot of Motion Mountain we cannot avoid
Copyright © Christoph Schiller November 1997–May 2006

perceiving. Perception is first of all the ability to distinguish. We use the basic mental act
of distinguishing in almost every instant of life; for example, during childhood we first
learned to distinguish familiar from unfamiliar observations. This is possible in combin-
ation with another basic ability, namely the capacity to memorize experiences. Memory
gives us the ability to experience, to talk and thus to explore nature. Perceiving, classify-

* The importance of throwing is also seen from the terms derived from it: in Latin, words like subject or
‘thrown below’, object or ‘thrown in front’, and interjection or ‘thrown in between’; in Greek, it led to terms
like symbol or ‘thrown together’, problem or ‘thrown forward’, emblem or ‘thrown into’, and – last but not
least – devil or ‘thrown through’.
** The world is independent of my will.

Dvipsbugw
 why should we care abou t motion? 37

ing and memorizing together form learning. Without any one of these three abilities, we
could not study motion.
Children rapidly learn to distinguish permanence from variability. They learn to recog-
nize human faces, even though a face never looks exactly the same each time it is seen.
From recognition of faces, children extend recognition to all other observations. Recog-
nition works pretty well in everyday life; it is nice to recognize friends, even at night, and
even after many beers (not a challenge). The act of recognition thus always uses a form of
generalization. When we observe, we always have a general idea in our mind. We specify
the main ones. Dvipsbugw
All forests can remind us of the essence of perception. Sitting on the grass in a clear-
ing of the forest at the foot of Motion Mountain, surrounded by the trees and the silence
typical of such places, a feeling of calmness and tranquillity envelops us. Suddenly, some-
thing moves in the bushes; immediately our eyes turn and our attention focuses. The
nerve cells that detect motion are part of the most ancient part of our brain, shared with
Ref. 19 birds and reptiles: the brain stem. Then the cortex, or modern brain, takes over to ana-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

lyse the type of motion and to identify its origin. Watching the motion across our field
of vision, we observe two invariant entities: the fixed landscape and the moving animal.
After we recognize the animal as a deer, we relax again.
How did we distinguish between landscape and deer? Several steps in the eye and in
the brain are involved. Motion plays an essential part in them, as is best deduced from
Ref. 20 the flip movie shown in the lower left corners of these pages. Each image shows only a
rectangle filled with a mathematically-random pattern. But when the pages are scanned,
one discerns a shape moving against a fixed background. At any given instant, the shape
cannot be distinguished from the background; there is no visible object at any given in-
stant of time. Nevertheless it is easy to perceive its motion.* Perception experiments such
as this one have been performed in many variations. In one, it was found that detecting
such a window is nothing special to humans; flies have the same ability, as do, in fact, all
animals that have eyes.
The flip movie in the lower left corner, like many similar experiments, shows two cent-
ral connections. First, motion is perceived only if an object can be distinguished from a
background or environment. Many motion illusions focus on this point.** Second, motion
is required to define both the object and the environment, and to distinguish them from
each other. In fact, the concept of space is – among others – an abstraction of the idea of
background. The background is extended; the moving entity is localized. Does this seem
boring? It is not; just wait a second.
Copyright © Christoph Schiller November 1997–May 2006

We call the set of localized aspects that remain invariant or permanent during motion,
such as size, shape, colour etc., taken together, a (physical) object or a (physical) body.
We will tighten the definition shortly, since otherwise images would be objects as well. In
other words, right from the start we experience motion as a relative process; it is perceived

* The human eye is rather good at detecting motion. For example, the eye can detect motion of a point of
light even if the change of angle is smaller than that which can be distinguished in a fixed image. Details of
Ref. 21 this and similar topics for the other senses are the domain of perception research.
** The topic of motion perception is full of interesting aspects. An excellent introduction is chapter 6 of the
beautiful text by Donald D. Hoffman, Visual Intelligence – How We Create What We See, W.W. Norton
& Co., 1998. His collection of basic motion illusions can be experienced and explored on the associated
http://aris.ss.uci.edu/cogsci/personnel/hoffman/hoffman.html website.

Dvipsbugw
 38 i galilean motion • 1. why should we care abou t motion?

TA B L E 2 Family tree of the basic physical concepts

motion
the basic type of change

parts relations background

permanent variable permanent
Dvipsbugw
bounded unbounded extended
shaped unshaped measurable

objects images states interactions phase space space-time

impenetrable penetrable global local composed simple

The corresponding aspects:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

mass intensity instant source dimension curvature
size colour position domain distance topology
charge appearance momentum strength volume distance
spin disappearance energy direction subspaces area
etc. etc. etc. etc. etc. etc.

world – nature – universe – cosmos

the collection of all parts, relations and backgrounds

in relation and in opposition to the environment. The concept of an object is therefore also
a relative concept. But the basic conceptual distinction between localized, isolable objects
and the extended environment is not trivial or unimportant. First, it has the appearance
Challenge 8 n of a circular definition. (Do you agree?) This issue will keep us very busy later on. Second,
we are so used to our ability of isolating local systems from the environment that we take
it for granted. However, as we will see in the third part of our walk, this distinction turns
Page 1018 out to be logically and experimentally impossible!* Our walk will lead us to discover the
reason for this impossibility and its important consequences. Finally, apart from moving
Copyright © Christoph Schiller November 1997–May 2006

entities and the permanent background, we need a third concept, as shown in Table 2.
Wisdom is one thing: to understand the thought

Ref. 22
“ which steers all things through all things.
Heraclitus of Ephesus

”
* Contrary to what is often read in popular literature, the distinction is possible in quantum theory. It be-
comes impossible only when quantum theory is unified with general relativity.

Dvipsbugw
 why should we care abou t motion? 39

Does the world need states?

Das Feste, das Bestehende und der Gegenstand

“ sind Eins. Der Gegenstand ist das Feste,

Bestehende; die Konfiguration ist das
Wechselnde, Unbeständige.*
Ludwig Wittgenstein, Tractatus, 2.027 – 2.0271

What distinguishes the various patterns in the lower left corners of this text? In everyday
life we would say: the situation or configuration of the involved entities. The situation
somehow describes all those aspects that can differ from case to case. It is customary to
” Dvipsbugw

call the list of all variable aspects of a set of objects their (physical) state of motion, or
simply their state.
The situations in the lower left corners differ first of all in time. Time is what makes
opposites possible: a child is in a house and the same child is outside the house. Time
describes and resolves this type of contradiction. But the state not only distinguishes situ-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ations in time: the state contains all those aspects of a system (i.e., of a group of objects)
that set it apart from all similar systems. Two objects can have the same mass, shape, col-
our, composition and be indistinguishable in all other intrinsic properties; but at least
they will differ in their position, or their velocity, or their orientation. The state pinpoints
the individuality of a physical system,** and allows us to distinguish it from exact copies
of itself. Therefore, the state also describes the relation of an object or a system with re-
spect to its environment. Or in short: the state describes all aspects of a system that depend
on the observer. These properties are not boring – just ponder this: does the universe have
Challenge 10 n a state?
Describing nature as a collection of permanent entities and changing states is the start-
ing point of the study of motion. The various aspects of objects and of their states are
called observables. All these rough, preliminary definitions will be refined step by step in
the following. Using the terms just introduced, we can say that motion is the change of
state of objects.***
States are required for the description of motion. In order to proceed and to achieve
a complete description of motion, we thus need a complete description of objects and a
complete description of their possible states. The first approach, called Galilean physics,
consists in specifying our everyday environment as precisely as possible.
Copyright © Christoph Schiller November 1997–May 2006

* Objects, the unalterable, and the subsistent are one and the same. Objects are what is unalterable and
subsistent; their configuration is what is changing and unstable.
** A physical system is a localized entity of investigation. In the classification of Table 2, the term ‘physical
system’ is (almost) the same as ‘object’ or ‘physical body’. Images are usually not counted as physical systems
Challenge 9 n (though radiation is one). Are holes physical systems?
*** The exact separation between those aspects belonging to the object and those belonging to the state
depends on the precision of observation. For example, the length of a piece of wood is not permanent; wood
shrinks and bends with time, due to processes at the molecular level. To be precise, the length of a piece of
wood is not an aspect of the object, but an aspect of its state. Precise observations thus shift the distinction
between the object and its state; the distinction itself does not disappear – at least not for quite while.

Dvipsbugw
 40 i galilean motion • 1. why should we care abou t motion?

Dvipsbugw

F I G U R E 8 A block and tackle

and a differential pulley

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Curiosities and fun challenges about motion
Motion is not always a simple topic.*
**
Challenge 11 n Is the motion of a ghost an example of motion?

**
A man climbs a mountain from 9 a.m. to 1 p.m. He sleeps on the top and comes down
the next day, taking again from 9 am to 1 pm for the descent. Is there a place on the path
Challenge 12 n that he passes at the same time on the two days?
**
Challenge 13 n Can something stop moving? If yes: how would you show it? If not: does this mean that
nature is infinite?
**
Challenge 14 n Can the universe move? Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 15 n To talk about precision with precision, we need to measure it. How would you do that?

**
Challenge 16 n Would we observe motion if we had no memory?
**
Challenge 17 n What is the lowest speed you have observed? Is there a lowest speed in nature?

* Sections entitled ‘curiosities’ are collections of topics and problems that allow one to check and to expand
the usage of concepts already introduced.

Dvipsbugw
 why should we care abou t motion? 41

**
According to legend, Sessa ben Zahir, the Indian inventor of the game of chess, demanded
from King Shirham the following reward for his invention: he wanted one grain of rice
for the first square, two for the second, four for the third, eight for the fourth, and so on.
Challenge 18 n How much time would all the rice fields of the world take to produce the necessary rice?
**
When a burning candle is moved, the flame lags behind the candle. How does the flame
Challenge 19 n behave if the candle is inside a glass, still burning, and the glass is accelerated? Dvipsbugw

**
A good way to make money is to build motion detectors. A motion detector is a small box
with a few wires. The box produces an electrical signal whenever the box moves. What
types of motion detectors can you imagine? How cheap can you make such a box? How
Challenge 20 d precise?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
A perfectly frictionless and spherical ball lies near the edge of a perfectly flat and hori-
Challenge 21 d zontal table. What happens? In what time scale?
**
You step into a closed box without windows. The box is moved by outside forces unknown
Challenge 22 n to you. Can you determine how you move from inside the box?

**
What is the length of rope one has to pull in order to lift a mass by a height h with a block
Challenge 23 n and tackle with four wheels, as shown in Figure 8?
**
When a block is rolled over the floor over a set of cylinders, how are the speed of the
Challenge 24 n block and that of the cylinders related?
**
Ref. 15 Do you dislike formulae? If you do, use the following three-minute method to change
Challenge 25 n the situation. It is worth trying it, as it will make you enjoy this book much more. Life is
Copyright © Christoph Schiller November 1997–May 2006

short; as much of it as possible, like reading this text, should be a pleasure.

1 - Close your eyes and recall an experience that was absolutely marvellous, a situation
when you felt excited, curious and positive.
2 - Open your eyes for a second or two and look at page 321 – or any other page that
contains many formulae.
3 - Then close your eyes again and return to your marvellous experience.
4 - Repeat the observation of the formulae and the visualization of your memory –
steps 2 and 3 – three more times.
Then leave the memory, look around yourself to get back into the here and now, and test
yourself. Look again at page 321. How do you feel about formulae now?

Dvipsbugw
 42 i galilean motion • 2. galilean physics – motion in everyday life

**
In the sixteenth century, Niccolò Tartaglia* proposed the following problem. Three young
couples want to cross a river. Only a small boat that can carry two people is available. The
men are extremely jealous, and would never leave their brides alone with another man.
Challenge 26 n How many journeys across the river are necessary?

2. galilean physics – motion in everyday life

Dvipsbugw
Physic ist wahrlich das eigentliche Studium des

“ Menschen.**
Georg Christoph Lichtenberg

The simplest description of motion is the one we all, like cats or monkeys, use uncon- ”
sciously in everyday life: only one thing can be at a given spot at a given time. This general
description can be separated into three assumptions: matter is impenetrable and moves,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

time is made of instants, and space is made of points. Without these three assumptions (do
Challenge 27 n you agree with them?) it is not possible to define velocity in everyday life. This description
of nature is called Galilean or Newtonian physics.
Galileo Galilei (1564–1642), Tuscan professor of mathematics,
was a founder of modern physics and is famous for advocating the
importance of observations as checks of statements about nature. By
requiring and performing these checks throughout his life, he was
led to continuously increase the accuracy in the description of mo-
tion. For example, Galileo studied motion by measuring change of
position with a self-constructed stopwatch. His approach changed
the speculative description of ancient Greece into the experimental
physics of Renaissance Italy.***
The English alchemist, occultist, theologian, physicist and politi- Galileo Galilei
cian Isaac Newton (1643–1727) was one of the first to pursue with
vigour the idea that different types of motion have the same properties, and he made
important steps in constructing the concepts necessary to demonstrate this idea.****

Copyright © Christoph Schiller November 1997–May 2006

* Niccolò Fontana Tartaglia (1499–1557), important Venetian mathematician.

** ‘Physics truly is the proper study of man.’ Georg Christoph Lichtenberg (1742–1799) was an important
physicist and essayist.
*** The best and most informative book on the life of Galileo and his times is by Pietro Redondi (see the
footnote on page 220). Galileo was born in the year the pencil was invented. Before his time, it was impossible
to do paper and pencil calculations. For the curious, the http://www.mpiwg-berlin.mpg.de website allows
you to read an original manuscript by Galileo.
**** Newton was born a year after Galileo died. Newton’s other hobby, as master of the Mint, was to supervise
Ref. 23 personally the hanging of counterfeiters. About Newton’s infatuation with alchemy, see the books by Dobbs.
Among others, Newton believed himself to be chosen by god; he took his Latin name, Isaacus Neuutonus,
and formed the anagram Jeova sanctus unus. About Newton and his importance for classical mechanics, see
Ref. 24 the text by Clifford Truesdell.

Dvipsbugw
 galilean physics – motion in everyday life 43

TA B L E 3 Properties of everyday – or Galilean – velocity

Ve l o c i t i e s Physical M at h e m at i c a l Definition
can propert y name
Be distinguished distinguishability element of set Page 646
Change gradually continuum real vector space Page 69, Page 1214
Point somewhere direction vector space, dimensionality Page 69
Be compared measurability metricity Page 1205 Dvipsbugw
Be added additivity vector space Page 69
Have defined angles direction Euclidean vector space Page 69
Exceed any limit infinity unboundedness Page 647

What is velocity?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

There is nothing else like it.

“ Jochen Rindt*

Velocity fascinates. To physicists, not only car races are interesting, but any moving entity ”
is. Therefore they first measure as many examples as possible. A selection is given in
Table 4.
Everyday life teaches us a lot about motion: objects can overtake each other, and they
can move in different directions. We also observe that velocities can be added or changed
smoothly. The precise list of these properties, as given in Table 3, is summarized by math-
ematicians in a special term; they say that velocities form a Euclidean vector space.** More
Page 69 details about this strange term will be given shortly. For now we just note that in describ-
ing nature, mathematical concepts offer the most accurate vehicle.
When velocity is assumed to be an Euclidean vector, it is called Galilean velocity. Ve-
locity is a profound concept. For example, velocity does not need space and time meas-
urements to be defined. Are you able to find a means of measuring velocities without
Challenge 28 d measuring space and time? If so, you probably want to skip to page 275, jumping 2000
years of enquiries. If you cannot do so, consider this: whenever we measure a quantity we
assume that everybody is able to do so, and that everybody will get the same result. In
other words, we define measurement as a comparison with a standard. We thus implicitly
assume that such a standard exists, i.e. that an example of a ‘perfect’ velocity can be found.
Copyright © Christoph Schiller November 1997–May 2006

Historically, the study of motion did not investigate this question first, because for many
centuries nobody could find such a standard velocity. You are thus in good company.
Some researchers have specialized in the study of the lowest velocities found in nature:

* Jochen Rindt (1942–1970), famous Austrian Formula One racing car driver, speaking about speed.
** It is named after Euclid, or Eukleides, the great Greek mathematician who lived in Alexandria around
300 bce. Euclid wrote a monumental treatise of geometry, the Στοιχεῖα or Elements, which is one of the
milestones of human thought. The text presents the whole knowledge on geometry of that time. For the first
time, Euclid introduces two approaches that are now in common use: all statements are deduced from a
small number of basic ‘axioms’ and for every statement a ‘proof ’ is given. The book, still in print today, has
been the reference geometry text for over 2000 years. On the web, it can be found at http://aleph0.clarku.
edu/~djoyce/java/elements/elements.html.

Dvipsbugw
44 i galilean motion • 2. galilean physics – motion in everyday life

TA B L E 4 Some measured velocity values

O b s e r va t i o n Ve l o c i t y

Stalagmite growth 0.3 pm~s

Can you find something slower? Challenge 29 n
Growth of deep sea manganese crust 80 am~s
Lichen growth down to 7 pm~s
Typical motion of continents 10 mm~a = 0.3 nm~s Dvipsbugw
Human growth during childhood, hair growth 4 nm~s
Tree growth up to 30 nm~s
Electron drift in metal wire 1 µm~s
Sperm motion 60 to 160 µm~s
Speed of light at Sun’s centre 0.1 mm~s
Ketchup motion 1 mm~s

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Slowest speed of light measured in matter on Earth 0.3 m~s Ref. 25
Speed of snowflakes 0.5 m~s to 1.5 m~s
Signal speed in human nerve cells 0.5 m~s to 120 m~s Ref. 26
Wind speed at 1 Beaufort (light air) below 1.5 m~s
Speed of rain drops, depending on radius 2 m~s to 8 m~s
Fastest swimming fish, sailfish (Istiophorus platypterus) 22 m~s
Fastest running animal, cheetah (Acinonyx jubatus) 30 m~s
Wind speed at 12 Beaufort (hurricane) above 33 m~s
Speed of air in throat when sneezing 42 m~s
Fastest measured throw: cricket ball 45 m~s
Freely falling human 50 to 90 m~s
Fastest bird, diving Falco peregrinus 60 m~s
Fastest badminton serve 70 m~s
Average speed of oxygen molecule in air at room temperature 280 m~s
Speed of sound in dry air at sea level and standard temperature 330 m~s
Cracking whip’s end 750 m~s
Speed of a rifle bullet 3 km~s
Speed of crack propagation in breaking silicon 5 km~s
Copyright © Christoph Schiller November 1997–May 2006

Highest macroscopic speed achieved by man – the Voyager satellite 14 km~s

Average (and peak) speed of lightning tip 600 km~s (50 000 km~s)
Speed of Earth through universe 370 km~s
Highest macroscopic speed measured in our galaxy 0.97 ë 108 m~s Ref. 27
Speed of electrons inside a colour TV 1 ë 108 m~s
Speed of radio messages in space 299 972 458 m~s
Highest ever measured group velocity of light 10 ë 108 m~s
Speed of light spot from a light tower when passing over the Moon 2 ë 109 m~s
Highest proper velocity ever achieved for electrons by man 7 ë 1013 m~s
Highest possible velocity for a light spot or shadow infinite

Dvipsbugw
 galilean physics – motion in everyday life 45

Ref. 28 they are called geologists. Do not miss the opportunity to walk across a landscape while
listening to one of them.
Velocity is a profound subject for a second reason: we will discover that all properties of
Table 3 are only approximate; none is actually correct. Improved experiments will uncover
limits in every property of Galilean velocity. The failure of the last three properties will
lead us to special and general relativity, the failure of the middle two to quantum theory
and the failure of the first two properties to the unified description of nature. But for now,
we’ll stick with Galilean velocity, and continue with another Galilean concept derived
from it: time. Dvipsbugw
Without the concepts place, void and time,

“ change cannot be. [...] It is therefore clear [...]

that their investigation has to be carried out, by
studying each of them separately.
Aristotle* Physics, Book III, part 1.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What is time?
Time does not exist in itself, but only through

“ the perceived objects, from which the concepts

of past, of present and of future ensue.
Lucrece,** De rerum natura, lib. 1, v. 460 ss.

Ref. 18
In their first years of life, children spend a lot of time throwing objects around. The term
‘object’ is a Latin word meaning ‘that which has been thrown in front.’ Developmental
psychology has shown experimentally that from this very experience children extract
”
the concepts of time and space. Adult physicists do the same when studying motion at
university.
When we throw a stone through the air, we can define a se-
quence of observations. Our memory and our senses give us this
ability. The sense of hearing registers the various sounds during
the rise, the fall and the landing of the stone. Our eyes track
the location of the stone from one point to the next. All obser-
vations have their place in a sequence, with some observations
preceding them, some observations simultaneous to them, and
still others succeeding them. We say that observations are per-
ceived to happen at various instants and we call the sequence of
all instants time.
Copyright © Christoph Schiller November 1997–May 2006

An observation that is considered the smallest part of a se- F I G U R E 9 A typical path

quence, i.e. not itself a sequence, is called an event. Events are followed by a stone
central to the definition of time; in particular, starting or stop- thrown through the air
Challenge 30 n ping a stopwatch are events. (But do events really exist? Keep
this question in the back of your head as we move on.)
Sequential phenomena have an additional property known as stretch, extension or
duration. Some measured values are given in Table 5.*** Duration expresses the idea that

* Aristotle (384/3–322), Greek philosopher and scientist.

** Lucretius Carus (c. 95 to c. 55 bce ), Roman scholar and poet.
*** A year is abbreviated a (Latin ‘annus’).

Dvipsbugw
 46 i galilean motion • 2. galilean physics – motion in everyday life

TA B L E 5 Selected time measurements

O b s e r va t i o n Ti m e

Shortest measurable time 10−44 s

Shortest time ever measured 10−23 s
Time for light to cross a typical atom 10−181 s
Period of caesium ground state hyperfine transition 108.782 775 707 78 ps
Beat of wings of fruit fly 1 ms Dvipsbugw
Period of pulsar (rotating neutron star) PSR 1913+16 0.059 029 995 271(2) s
Human ‘instant’ 20 ms
Shortest lifetime of living being 0.3 d
Average length of day 400 million years ago 79 200 s
Average length of day today 86 400.002(1) s
From birth to your 1000 million seconds anniversary 31.7 a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Age of oldest living tree 4600 a
Use of human language 2 ë 105 a
Age of Himalayas 35 to 55 ë 106 a
Age of Earth 4.6 ë 109 a
Age of oldest stars 13.7 Ga
Age of most protons in your body 13.7 Ga
Lifetime of tantalum nucleus 180 Ta 1015 a
Lifetime of bismuth 209 Bi nucleus 1.9(2) ë 1019 a

sequences take time. We say that a sequence takes time to express that other sequences
can take place in parallel with it.
How exactly is the concept of time, including sequence and duration, deduced from ob-
servations? Many people have looked into this question: astronomers, physicists, watch-
makers, psychologists and philosophers. All find that time is deduced by comparing mo-
tions. Children, beginning at a very young age, develop the concept of ‘time’ from the
Ref. 18 comparison of motions in their surroundings. Grown-ups take as a standard the motion
of the Sun and call the resulting type of time local time. From the Moon they deduce a
lunar calendar. If they take a particular village clock on a European island they call it the
Copyright © Christoph Schiller November 1997–May 2006

universal time coordinate (UTC), once known as ‘Greenwich mean time.’*Astronomers use
the movements of the stars and call the result ephemeris time. An observer who uses his
personal watch calls the reading his proper time; it is often used in the theory of relativity.
Not every movement is a good standard for time. In the year 2000 an Earth rotation
Page 1161 did not take 86 400 seconds any more, as it did in the year 1900, but 86 400.002 seconds.
Can you deduce in which year your birthday will have shifted by a whole day from the
Challenge 32 n time predicted with 86 400 seconds?

* Official UTC time is used to determine power grid phase, phone companies’ bit streams and the signal
to the GPS system used by many navigation systems around the world, especially in ships, aeroplanes and
lorries. For more information, see the http://www.gpsworld.com website. The time-keeping infrastructure
Challenge 31 n is also important for other parts of the modern economy. Can you spot the most important ones?

Dvipsbugw
 galilean physics – motion in everyday life 47

All methods for the definition of time are thus based on comparisons of motions. In
order to make the concept as precise and as useful as possible, a standard reference motion
is chosen, and with it a standard sequence and a standard duration is defined. The device
that performs this task is called a clock. We can thus answer the question of the section
title: time is what we read from a clock. Note that all definitions of time used in the various
branches of physics are equivalent to this one; no ‘deeper’ or more fundamental definition
is possible.* Note that the word ‘moment’ is indeed derived from the word ‘movement’.
Language follows physics in this case. Astonishingly, the definition of time just given is
final; it will never be changed, not even at the top of Motion Mountain. This is surprising Dvipsbugw
at first sight, because many books have been written on the nature of time. Instead, they
should investigate the nature of motion! But this is the aim of our walk anyhow. We are
thus set to discover all the secrets of time as a side result of our adventure. Every clock
reminds us that in order to understand time, we need to understand motion.
A clock is a moving system whose position can be read. Of course, a precise clock is
a system moving as regularly as possible, with as little outside disturbance as possible. Is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

there a perfect clock in nature? Do clocks exist at all? We will continue to study these ques-
tions throughout this work and eventually reach a surprising conclusion. At this point,
however, we state a simple intermediate result: since clocks do exist, somehow there is in
Challenge 33 n nature an intrinsic, natural and ideal way to measure time. Can you see it?
Time is not only an aspect of observations, it is also a facet of personal experience.
Even in our innermost private life, in our thoughts, feelings and dreams, we experience
sequences and durations. Children learn to relate this internal experience of time with ex-
ternal observations, and to make use of the sequential property of events in their actions.
Studies of the origin of psychological time show that it coincides – apart from its lack
of accuracy – with clock time.** Every living human necessarily uses in his daily life the
concept of time as a combination of sequence and duration; this fact has been checked
Ref. 30 in numerous investigations. For example, the term ‘when’ exists in all human languages.
Time is a concept necessary to distinguish between observations. In any sequence, we
observe that events succeed each other smoothly, apparently without end. In this context,
‘smoothly’ means that observations that are not too distant tend to be not too different.
Yet between two instants, as close as we can observe them, there is always room for other
events. Durations, or time intervals, measured by different people with different clocks
agree in everyday life; moreover, all observers agree on the order of a sequence of events.
Time is thus unique.
The mentioned properties of everyday time, listed in Table 6, correspond to the precise
Copyright © Christoph Schiller November 1997–May 2006

version of our everyday experience of time. It is called Galilean time; all the properties can
be expressed simultaneously by describing time with real numbers. In fact, real numbers
have been constructed to have exactly the same properties as Galilean time, as explained
Page 655 in the Intermezzo. Every instant of time can be described by a real number, often abbre-
viated t, and the duration of a sequence of events is given by the difference between the
values for the final and the starting event.
* The oldest clocks are sundials. The science of making them is called gnomonics. An excellent and complete
introduction into this somewhat strange world can be found at the http://www.sundials.co.uk website.
Page 830 ** The brain contains numerous clocks. The most precise clock for short time intervals, the internal interval
timer, is more accurate than often imagined, especially when trained. For time periods between a few tenths
Ref. 29 of a second, as necessary for music, and a few minutes, humans can achieve accuracies of a few per cent.

Dvipsbugw
 48 i galilean motion • 2. galilean physics – motion in everyday life

TA B L E 6 Properties of Galilean time

I n s ta n t s o f t i m e Physical M at h e m at i c a l Definition
propert y name

Can be distinguished distinguishability element of set Page 646

Can be put in order sequence order Page 1214
Define duration measurability metricity Page 1205
Can have vanishing duration continuity denseness, completeness Page 1214 Dvipsbugw
Allow durations to be added additivity metricity Page 1205
Don’t harbour surprises translation invariance homogeneity Page 154
Don’t end infinity unboundedness Page 647
Are equal for all observers absoluteness uniqueness

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

When Galileo studied motion in the seventeenth century, there were as yet no stop-
watches. He thus had to build one himself, in order to measure times in the range between
Challenge 34 n a fraction and a few seconds. Can you guess how he did it?
We will have quite some fun with Galilean time in the first two chapters. However,
hundreds of years of close scrutiny have shown that every single property of time just
listed is approximate, and none is strictly correct. This story is told in the subsequent
chapters.

Why do clocks go clockwise?

What time is it at the North Pole now?

“ ”
Challenge 35 n

All rotational motions in our society, such as athletic races, horse, bicycle or ice skat-
ing races, turn anticlockwise. Likewise, every supermarket leads its guests anticlockwise
through the hall. Mathematicians call this the positive rotation sense. Why? Most people
are right-handed, and the right hand has more freedom at the outside of a circle. There-
Copyright © Christoph Schiller November 1997–May 2006

fore thousands of years ago chariot races in stadia went anticlockwise. As a result, all
races still do so to this day. That is why runners move anticlockwise. For the same reason,
helical stairs in castles are built in such a way that defending right-handers, usually from
above, have that hand on the outside.
On the other hand, the clock imitates the shadow of sundials; obviously, this is true
on the northern hemisphere only, and only for sundials on the ground, which were the
most common ones. (The old trick to determine south by pointing the hour hand of an
horizontal watch to the Sun and halving the angle between it and the direction of 12 o’clock
does not work on the southern hemisphere.) So every clock implicitly continues to state
on which hemisphere it was invented. In addition, it also tells us that sundials on walls
came in use much later than those on the floor.

Dvipsbugw
 galilean physics – motion in everyday life 49

Does time flow?

Wir können keinen Vorgang mit dem ‘Ablauf

“ der Zeit’ vergleichen – diesen gibt es nicht –,

sondern nur mit einem anderen Vorgang (etwa
dem Gang des Chronometers).*
Ludwig Wittgenstein, Tractatus, 6.3611

The expression ‘the flow of time’ is often used to convey that in nature change follows
after change, in a steady and continuous manner. But though the hands of a clock ‘flow’,
time itself does not. Time is a concept introduced specially to describe the flow of events
” Dvipsbugw

around us; it does not itself flow, it describes flow. Time does not advance. Time is neither
linear nor cyclic. The idea that time flows is as hindering to understanding nature as is
Page 554 the idea that mirrors exchange right and left.
The misleading use of the expression ‘flow of time’, propagated first by some Greek
Ref. 31 thinkers and then again by Newton, continues. Aristotle (384/3–322 bce ), careful to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

think logically, pointed out its misconception, and many did so after him. Nevertheless,
expressions such as ‘time reversal’, the ‘irreversibility of time’, and the much-abused ‘time’s
Challenge 36 e arrow’ are still common. Just read a popular science magazine chosen at random. The fact
is: time cannot be reversed, only motion can, or more precisely, only velocities of objects;
time has no arrow, only motion has; it is not the flow of time that humans are unable
to stop, but the motion of all the objects in nature. Incredibly, there are even books writ-
Ref. 32 ten by respected physicists that study different types of ‘time’s arrows’ and compare them
with each other. Predictably, no tangible or new result is extracted. Time does not flow.
In the same manner, colloquial expressions such as ‘the start (or end) of time’ should be
avoided. A motion expert translates them straight away into ‘the start (or end) of motion’.

What is space?
The introduction of numbers as coordinates [...]

“ is an act of violence [...].

Hermann Weyl, Philosophie der Mathematik
und Naturwissenschaft.**

Whenever we distinguish two objects from each other, such as two stars, we first of all dis-
tinguish their positions. Distinguishing positions is the main ability of our sense of sight.
Position is therefore an important aspect of the physical state of an object. A position is
”
taken by only one object at a time. Positions are limited. The set of all available positions,
Copyright © Christoph Schiller November 1997–May 2006

called (physical) space, acts as both a container and a background.

Closely related to space and position is size, the set of positions an objects occupies.
Small objects occupy only subsets of the positions occupied by large ones. We will discuss
size shortly.
How do we deduce space from observations? During childhood, humans (and most
higher animals) learn to bring together the various perceptions of space, namely the
* We cannot compare a process with ‘the passage of time’ – there is no such thing – but only with another
process (such as the working of a chronometer).
** Hermann Weyl (1885–1955) was one of the most important mathematicians of his time, as well as an
important theoretical physicist. He was one of the last universalists in both fields, a contributor to quantum
theory and relativity, father of the term ‘gauge’ theory, and author of many popular texts.

Dvipsbugw
 50 i galilean motion • 2. galilean physics – motion in everyday life

visual, the tactile, the auditory, the kinesthetic, the vestibular etc., into one coherent set
of experiences and description. The result of this learning process is a certain ‘image’ of
space in the brain. Indeed, the question ‘where?’ can be asked and answered in all lan-
guages of the world. Being more precise, adults derive space from distance measurements.
The concepts of length, area, volume, angle and solid angle are all deduced with their help.
Geometers, surveyors, architects, astronomers, carpet salesmen and producers of metre
sticks base their trade on distance measurements. Space is a concept formed to summar-
ize all the distance relations between objects for a precise description of observations.
Metre sticks work well only if they are straight. But when humans lived in the jungle, Dvipsbugw
there were no straight objects around them. No straight rulers, no straight tools, noth-
Challenge 37 n ing. Today, a cityscape is essentially a collection of straight lines. Can you describe how
humans achieved this?
Once humans came out of the jungle with their newly built metre sticks, they collec-
ted a wealth of results. The main ones are listed in Table 7; they are easily confirmed
by personal experience. Objects can take positions in an apparently continuous manner:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

there indeed are more positions than can be counted.* Size is captured by defining the
distance between various positions, called length, or by using the field of view an object
takes when touched, called its surface. Length and surface can be measured with the help
of a metre stick. Selected measurement results are given in Table 8. The length of objects
is independent of the person measuring it, of the position of the objects and of their ori-
entation. In daily life the sum of angles in any triangle is equal to two right angles. There
are no limits in space.
Experience shows us that space has three di-
mensions; we can define sequences of positions
in precisely three independent ways. Indeed,
the inner ear of (practically) all vertebrates has
three semicircular canals that sense the body’s
position in the three dimensions of space, as
shown in Figure 10.** Similarly, each human
eye is moved by three pairs of muscles. (Why
Challenge 38 n three?) Another proof that space has three di-
mensions is provided by shoelaces: if space had
F I G U R E 10 Two proofs of the
more than three dimensions, shoelaces would three-dimensionality of space: a knot and the
not be useful, because knots exist only in three- inner ear of a mammal
dimensional space. But why does space have
Copyright © Christoph Schiller November 1997–May 2006

three dimensions? This is probably the most difficult question of physics; it will be
answered only in the very last part of our walk.
Challenge 39 n It is often said that thinking in four dimensions is impossible. That is wrong. Just try.
For example, can you confirm that in four dimensions knots are impossible?
Like time intervals, length intervals can be described most precisely with the help of
real numbers. In order to simplify communication, standard units are used, so that every-
body uses the same numbers for the same length. Units allow us to explore the general
* For a definition of uncountability, see page 649.
** Note that saying that space has three dimensions implies that space is continuous; the Dutch mathem-
atician and philosopher Luitzen Brouwer (b. 1881 Overschie, d. 1966 Blaricum) showed that dimensionality
is only a useful concept for continuous sets.

Dvipsbugw
 galilean physics – motion in everyday life 51

TA B L E 7 Properties of Galilean space

Points Physical M at h e m at i c a l Defini-

propert y name tion

Can be distinguished distinguishability element of set Page 646

Can be lined up if on one line sequence order Page 1214
Can form shapes shape topology Page 1213
Lie along three independent possibility of knots 3-dimensionality Page 1204 Dvipsbugw
directions
Can have vanishing distance continuity denseness, Page 1214
completeness
Define distances measurability metricity Page 1205
Allow adding translations additivity metricity Page 1205
Define angles scalar product Euclidean space Page 69

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Don’t harbour surprises translation invariance homogeneity
Can beat any limit infinity unboundedness Page 647
Defined for all observers absoluteness uniqueness Page 52

properties of Galilean space experimentally: space, the container of objects, is continuous,

three-dimensional, isotropic, homogeneous, infinite, Euclidean and unique or ‘absolute’.
In mathematics, a structure or mathematical concept with all the properties just men-
tioned is called a three-dimensional Euclidean space. Its elements, (mathematical) points,
are described by three real parameters. They are usually written as

(x, y, z) (1)

and are called coordinates. They specify and order the location of a point in space. (For
the precise definition of Euclidean spaces, see page 69.)
What is described here in just half a page actually took 2000 years to be worked out,
mainly because the concepts of ‘real number’ and ‘coordinate’ had to be discovered first.
The first person to describe points of space in this way was the famous mathematician and
philosopher René Descartes*, after whom the coordinates of expression (1) are named
Cartesian.
Copyright © Christoph Schiller November 1997–May 2006

Like time, space is a necessary concept to describe the world. Indeed, space is auto-
matically introduced when we describe situations with many objects. For example, when
many spheres lie on a billiard table, we cannot avoid using space to describe the relations
between them. There is no way to avoid using spatial concepts when talking about nature.
Even though we need space to talk about nature, it is still interesting to ask why this
is possible. For example, since length measurement methods do exist, there must be a
Challenge 40 n natural or ideal way to measure distances, sizes and straightness. Can you find it?

* René Descartes or Cartesius (1596–1650), French mathematician and philosopher, author of the famous
statement ‘je pense, donc je suis’, which he translated into ‘cogito ergo sum’ – I think therefore I am. In his
view this is the only statement one can be sure of.

Dvipsbugw
52 i galilean motion • 2. galilean physics – motion in everyday life

TA B L E 8 Some measured distance values

O b s e r va t i o n D i s ta n c e

Galaxy Compton wavelength 10−85 m (calculated only)

Planck length, the shortest measurable length 10−32 m
Proton diameter 1 fm
Electron Compton wavelength 2.426 310 215(18) pm
Hydrogen atom size 30 pm Dvipsbugw
Smallest eardrum oscillation detectable by human ear 50 pm
Wavelength of visible light 0.4 to 0.8 µm
Size of small bacterium 5 µm
Point: diameter of smallest object visible with naked eye 20 µm
Diameter of human hair (thin to thick) 30 to 80 µm
Total length of DNA in each human cell 2m

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Largest living thing, the fungus Armillaria ostoyae 3 km
Length of Earth’s Equator 40 075 014.8(6) m
Total length of human nerve cells 8 ë 105 km
Average distance to Sun 149 597 870 691(30) m
Light year 9.5 Pm
Distance to typical star at night 10 Em
Size of galaxy 1 Zm
Distance to Andromeda galaxy 28 Zm
Most distant visible object 125 Ym

As in the case of time, each of the properties of space just lis-

ted has to be checked. And again, careful observations will show
that each property is an approximation. In simpler and more
drastic words, all of them are wrong. This confirms Weyl’s state-
ment at the beginning of this section. In fact, the story about the
violence connected with the introduction of numbers is told by
every forest in the world, and of course also by the one at the foot
of Motion Mountain. To hear it, we need only listen carefully to
Copyright © Christoph Schiller November 1997–May 2006

what the trees have to tell. René Descartes

“
Μέτρον ἄριστον.*
Cleobulus

Are space and time absolute or relative? ”

In everyday life, the concepts of Galilean space and time include two opposing aspects;
the contrast has coloured every discussion for several centuries. On the one hand, space
and time express something invariant and permanent; they both act like big containers for

* ‘Measure is the best (thing).’ Cleobulus (Κλεοβουλος) of Lindos, (c. 620–550 BCE ) was another of the
proverbial seven sages.

Dvipsbugw
 galilean physics – motion in everyday life 53

all the objects and events found in nature. Seen this way, space and time have an existence
of their own. In this sense one can say that they are fundamental or absolute. On the other
hand, space and time are tools of description that allow us to talk about relations between
objects. In this view, they do not have any meaning when separated from objects, and only
result from the relations between objects; they are derived, relational or relative. Which
Challenge 41 e of these viewpoints do you prefer? The results of physics have alternately favoured one
viewpoint or the other. We will repeat this alternation throughout our adventure, until
Ref. 33 we find the solution. And obviously, it will turn out to be a third option.
Dvipsbugw
Size – why area exists, but volume does not
A central aspect of objects is their size. As a small child, under
school age, every human learns how to use the properties of size
and space in their actions. As adults seeking precision, the defin-
ition of distance as the difference between coordinates allows us

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

to define length in a reliable way. It took hundreds of years to dis-
cover that this is not the case. Several investigations in physics
and mathematics led to complications.
The physical issues started with an astonishingly simple ques-
tion asked by Lewis Richardson:* How long is the western coast-
line of Britain?
Following the coastline on a map using an odometer, a device F I G U R E 11 A
shown in Figure 11, Richardson found that the length l of the curvemeter or odometer
coastline depends on the scale s (say 1 : 10 000 or 1 : 500 000) of
the map used:
l = l 0 s 0.25 (2)

(Richardson found other numbers for other coasts.) The number l 0 is the length at scale
1 : 1. The main result is that the larger the map, the longer the coastline. What would hap-
pen if the scale of the map were increased even beyond the size of the original? The length
would increase beyond all bounds. Can a coastline really have infinite length? Yes, it can.
In fact, mathematicians have described many such curves; they are called fractals. An
infinite number of them exist, and Figure 12 shows one example.** Can you construct
Challenge 42 e another?
Length has other strange properties. The Italian mathematician Giuseppe Vitali was
Copyright © Christoph Schiller November 1997–May 2006

the first to discover that it is possible to cut a line segment of length 1 into pieces that
can be reassembled – merely by shifting them in the direction of the segment – into a

* Lewis Fray Richardson (1881–1953), English physicist and psychologist.

** Most of these curves are self-similar, i.e. they follow scaling laws similar to the above-mentioned. The term
‘fractal’ is due to the Polish mathematician Benoît Mandelbrot and refers to a strange property: in a certain
sense, they have a non-integral number D of dimensions, despite being one-dimensional by construction.
Mandelbrot saw that the non-integer dimension was related to the exponent e of Richardson by D = 1 + e,
thus giving D = 1.25 in the example above.
Coastlines and other fractals are beautifully presented in Heinz-Otto Peitgen, Hartmu t Jür-
gens & Dietmar Saupe, Fractals for the Classroom, Springer Verlag, 1992, pp. 232–245. It is also available
in several other languages.

Dvipsbugw
 54 i galilean motion • 2. galilean physics – motion in everyday life

n=1 n=2 n=3 n=∞

F I G U R E 12 A fractal: a self-similar curve of infinite length (far right), and its construction

line segment of length 2. Are you able to find such a division using the hint that it is only
Challenge 43 d possible using infinitely many pieces?
To sum up, length is well defined for lines that are straight or nicely curved, but not Dvipsbugw
for intricate lines, or for lines made of infinitely many pieces. We therefore avoid fractals
and other strangely shaped curves in the following, and we take special care when we
talk about infinitely small segments. These are the central assumptions in the first two
parts of this adventure, and we should never forget them. We will come back to these
assumptions in the third part.
In fact, all these problems pale when compared with the following problem. Com-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

monly, area and volume are defined using length. You think that it is easy? You’re wrong,
as well as being a victim of prejudices spread by schools around the world. To define area
and volume with precision, their definitions must have two properties: the values must
be additive, i.e. for finite and infinite sets of objects, the total area and volume have to be
the sum of the areas and volumes of each element of the set; and they must be rigid, i.e. if
one cuts an area or a volume into pieces and then rearranges the pieces, the value remains
the same. Do such concepts exist?
For areas in a plane, one proceeds in the following standard way: one defines the area
A of a rectangle of sides a and b as A = ab; since any polygon can be rearranged into
Challenge 44 n a rectangle with a finite number of straight cuts, one can then define an area value for
all polygons. Subsequently, one can define area for nicely curved shapes as the limit of
Page 178 the sum of infinitely many polygons. This method is called integration; it is introduced in
detail in the section on physical action.
However, integration does not allow us to define area for arbitrarily bounded regions.
Challenge 45 n (Can you imagine such a region?) For a complete definition, more sophisticated tools are
needed. They were discovered in 1923 by the famous mathematician Stefan Banach.* He
proved that one can indeed define an area for any set of points whatsoever, even if the bor-
der is not nicely curved but extremely complicated, such as the fractal curve previously
mentioned. Today this generalized concept of area, technically a ‘finitely additive isomet-
rically invariant measure,’ is called a Banach measure in his honour. Mathematicians sum
Copyright © Christoph Schiller November 1997–May 2006

up this discussion by saying that since in two dimensions there is a Banach measure, there
is a way to define the concept of area – an additive and rigid measure – for any set of points
whatsoever.**
What is the situation in three dimensions, i.e. for volume? We can start in the same
way as for area, by defining the volume V of a rectangular polyhedron with sides a, b,

* Stefan Banach (Krakow, 1892–Lvov, 1945), important Polish mathematician.

** Actually, this is true only for sets on the plane. For curved surfaces, such as the surface of a sphere, there
are complications that will not be discussed here. In addition, the problems mentioned in the definition of
length of fractals also reappear for area if the surface to be measured is not flat but full of hills and valleys.
A typical example is the area of the human lung: depending on the level of details examined, one finds area
values from a few up to over a hundred square metres.

Dvipsbugw
 galilean physics – motion in everyday life 55

c as V = abc. But then we encounter a first problem: a general polyhedron cannot be

cut into a cube by straight cuts! The limitation was discovered in 1900 and 1902 by Max
Dehn.* He found that the possibility depends on the values of the edge angles, or dihedral
angles, as the mathematicians call them. If one ascribes to every edge of a general poly-
hedron a number given by its length l times a special function д(α) of its dihedral angle
α, then Dehn found that the sum of all the numbers for all the edges of a solid does not
change under dissection, provided that the function fulfils д(α + β) = д(α) + д(β) and
д(π) = 0. An example of such a strange function д is the one assigning the value 0 to
any rational multiple of π and the value 1 to a basis set of irrational multiples of π. The Dvipsbugw
values for all other dihedral angles of the polyhedron can then be constructed by com-
bination of rational multiples of these basis angles. Using this function, you may then
Challenge 46 n deduce for yourself that a cube cannot be dissected into a regular tetrahedron because
their respective Dehn invariants are different.**
Despite the problems with Dehn invariants, one
can define a rigid and additive concept of volume for

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

polyhedra, since for all polyhedra and, in general, for
all ‘nicely curved’ shapes, one can again use integra-
tion for the definition of their volume.
Now let us consider general shapes and general
cuts in three dimensions, not just the ‘nice’ ones dihedral
angle
mentioned so far. We then stumble on the famous
Banach–Tarski theorem (or paradox). In 1924, Stefan
Ref. 34 Banach and Alfred Tarski*** proved that it is pos-
sible to cut one sphere into five pieces that can be
F I G U R E 13 A polyhedron with one of
recombined to give two spheres, each the size of the its dihedral angles (© Luca Gastaldi)
original. This counter-intuitive result is the Banach–
Tarski theorem. Even worse, another version of the theorem states: take any two sets not
extending to infinity and containing a solid sphere each; then it is always possible to dis-
sect one into the other with a finite number of cuts. In particular it is possible to dissect
a pea into the Earth, or vice versa. Size does not count!**** Volume is thus not a useful
concept at all.
The Banach–Tarski theorem raises two questions: first, can the result be applied to gold
or bread? That would solve many problems. Second, can it be applied to empty space?
Challenge 47 n In other words, are matter and empty space continuous? Both topics will be explored
later in our walk; each issue will have its own, special consequences. For the moment, we
Copyright © Christoph Schiller November 1997–May 2006

eliminate this troubling issue by restricting our interest to smoothly curved shapes (and
cutting knives). With this restriction, volumes of matter and of empty space do behave

* Max Dehn (1878–1952), German mathematician, student of David Hilbert.

** This is also told in the beautiful book by M. Aigler & G.M. Z iegler, Proofs from the Book, Springer
Verlag, 1999. The title is due to the famous habit of the great mathematician Paul Erdös to imagine that all
beautiful mathematical proofs can be assembled in the ‘book of proofs’.
*** Alfred Tarski (b. 1902 Warsaw, d. 1983 Berkeley), Polish mathematician.
**** The proof of the result does not need much mathematics; it is explained beautifully by Ian Stewart in
Paradox of the spheres, New Scientist, 14 January 1995, pp. 28–31. The Banach–Tarski paradox also exists in
four dimensions, as it does in any higher dimension. More mathematical detail can be found in the beautiful
Ref. 35 book by Steve Wagon.

Dvipsbugw
 56 i galilean motion • 2. galilean physics – motion in everyday life

nicely: they are additive and rigid, and show no paradoxes. Indeed, the cuts required for
the Banach–Tarski paradox are not smooth; it is not possible to perform them with an
everyday knife, as they require (infinitely many) infinitely sharp bends performed with
an infinitely sharp knife. Such a knife does not exist. Nevertheless, we keep in the back of
our mind that the size of an object or of a piece of empty space is a tricky quantity – and
that we need to be careful whenever we talk about it.

What is straight?
Dvipsbugw
When you see a solid object with a straight edge, it is a 99 %-safe bet that it is man-made.*
The contrast between the objects seen in a city – buildings, furniture, cars, electricity poles,
boxes, books – and the objects seen in a forest – trees, plants, stones, clouds – is evident:
in the forest nothing is straight or flat, in the city most objects are. How is it possible for
humans to produce straight objects while there are none to be found in nature?
Any forest teaches us the origin of straightness; it presents tall tree trunks and rays of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

daylight entering from above through the leaves. For this reason we call a line straight if
it touches either a plumb-line or a light ray along its whole length. In fact, the two defini-
Challenge 48 n tions are equivalent. Can you confirm this? Can you find another definition? Obviously,
we call a surface flat if for any chosen orientation and position it touches a plumb-line or
a light ray along its whole extension.
In summary, the concept of straightness – and thus also of flatness – is defined with
the help of bodies or radiation. In fact, all spatial concepts, like all temporal concepts,
require motion for their definition.

A hollow Earth?
Space and straightness pose subtle challenges. Some strange people maintain that all hu-
mans live on the inside of a sphere; they (usually) call this the hollow Earth theory. They
claim that the Moon, the Sun and the stars are all near the centre of the hollow sphere.
They also explain that light follows curved paths in the sky and that when conventional
physicists talk about a distance r from the centre of the Earth, the real hollow Earth dis-
Challenge 49 n tance is r he = R Earth
2
~r. Can you show that this model is wrong? Roman Sexl** used to
ask this question to his students and fellow physicists. The answer is simple: if you think
you have an argument to show that this view is wrong, you are mistaken! There is no way
of showing that such a view is wrong. It is possible to explain the horizon, the appearance
Challenge 50 e of day and night, as well as the satellite photographs of the round Earth, such as Figure 14.
Copyright © Christoph Schiller November 1997–May 2006

To explain what happened during a flight to the Moon is also fun. A coherent hollow
Earth view is fully equivalent to the usual picture of an infinitely extended space. We will
Page 495 come back to this problem in the section on general relativity.

* The most common counter-examples are numerous crystalline minerals, where the straightness is related
to the atomic structure. Another famous exception is the well-known Irish geological formation called the
Giant’s Causeway. Other candidates that might come to mind, such as certain bacteria which have (almost)
Ref. 36 square or (almost) triangular shapes are not counter-examples, as the shapes are only approximate.
** Roman Sexl, (1939–1986), important Austrian physicist, author of several influential textbooks on grav-
itation and relativity.

Dvipsbugw
 galilean physics – motion in everyday life 57

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 14 A photograph of the Earth – seen from the direction of the Sun

Curiosities and fun challenges about everyday space and time

Space and time lead to many thought-provoking questions.
**
How does one measure the speed of a gun bullet with a stop watch, in a space of 1 m3 ,
Challenge 51 n without electronics? Hint: the same method can also be used to measure the speed of
light.
**
Imagine a black spot on a white surface. What is the colour of the line separating the spot
Copyright © Christoph Schiller November 1997–May 2006

Challenge 52 n from the background? This question is often called Peirce’s puzzle.

**
Also bread is an (approximate) irregular fractal. The fractal dimension of bread is around
Challenge 53 n 2.7. Try to measure it.
**
Motoring poses many mathematical problems. A central one is the following parking
issue: what is the shortest distance d from the car in front necessary to leave a parking
Challenge 54 n spot without using reverse gear? (Assume that you know the geometry of your car, as
shown in Figure 16, and its smallest outer turning radius R, which is known for every

Dvipsbugw
 58 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 15 A model illustrating the hollow Earth theory, showing how day and night
appear (© Helmut Diehl)

d b

F I G U R E 16 Leaving a parking space

car.) Next question: what is the smallest gap required when you are allowed to manoeuvre
Copyright © Christoph Schiller November 1997–May 2006

Challenge 55 n back and forward as often as you like? Now a problem to which no solution seems to be
available in the literature: How does the gap depend on the number, n, of times you use
Challenge 56 n reverse gear? (The author offers 50 euro for the first well-explained solution sent to him.)
**
Challenge 57 n How often in 24 hours do the hour and minute hands of a clock lie on top of each other?
For clocks that also have a second hand, how often do all three hands lie on top of each
other?

**
How many times in twelve hours can the two hands of a clock be exchanged with the

Dvipsbugw
 galilean physics – motion in everyday life 59

TA B L E 9 The exponential notation: how to write small and large numbers

Number Exponential Number Exponential

n o tat i o n n o tat i o n

1 100
0.1 10−1 10 101
0.2 2 ë 10−1 20 2 ë 101
0.324 3.24 ë 10−1 32.4 3.24 ë 101
Dvipsbugw
0.01 10−2 100 102
0.001 10−3 1000 103
0.000 1 10−4 10 000 104
0.000 01 10−5 etc. 100 000 105 etc.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 58 n result that the new situation shows a valid time? What happens for clocks that also have
a third hand for seconds?
**
Challenge 59 n How many minutes does the Earth rotate in one minute?
**
What is the highest speed achieved by throwing (with and without a racket)? What was
Challenge 60 n the projectile used?
**
A rope is put around the Earth, on the Equator, as tightly as possible. The rope is
Challenge 61 n lengthened then by 1 m. Can a mouse slip under it?
**
Jack was rowing his boat on a river. When he was under a bridge, he dropped a ball into
the river. Jack continued to row in the same direction for 10 minutes after he dropped
the ball. He then turned around and rowed back. When he reached the ball, the ball had
Challenge 62 n floated 600 m from the bridge. How fast was the river flowing? Copyright © Christoph Schiller November 1997–May 2006

**
Adam and Bert are brothers. Adam is 18 years old. Bert is twice as old as at the time when
Adam was the age that Bert is now. How old is Bert?
**
Scientists use a special way to write large and small numbers, explained in Table 9.
Ref. 37 In 1996 the smallest experimentally probed distance was 10−19 m, achieved between
quarks at Fermilab. (To savour the distance value, write it down without the exponent.)
Challenge 63 n What does this measurement mean for the continuity of space?

**

Dvipsbugw
 60 i galilean motion • 2. galilean physics – motion in everyday life

‘Where am I?’ is a common question; ‘When am I?’ is never asked, not even in other
Challenge 64 n languages. Why?

**
Challenge 65 n Is there a smallest time interval in nature? A smallest distance?
**
Given that you know what straightness is, how would you characterize or define the
Challenge 66 n curvature of a curved line using numbers? And that of a surface? Dvipsbugw

**
Challenge 67 n What is the speed of your eyelid?
**
The surface area of the human body is about 200 m2 . Can you say where this large number

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 68 n comes from?

**
Fractals in three dimensions bear many surprises. Take a regular tetrahedron; then glue
on every one of its triangular faces a smaller regular tetrahedron, so that the surface of
the body is again made up of many equal regular triangles. Repeat the process, gluing
still smaller tetrahedrons to these new (more numerous) triangular surfaces. What is the
Challenge 69 n shape of the final fractal, after an infinite number of steps?

**
Zeno reflected on what happens to a moving object at a given instant of time. To discuss
with him, you decide to build the fastest possible shutter for a photographic camera that
you can imagine. You have all the money you want. What is the shortest shutter time you
Challenge 70 n would achieve?

**
Can you prove Pythagoras’ theorem by geometrical means alone, without using
Challenge 71 n coordinates? (There are more than 30 possibilities.)
**
Copyright © Christoph Schiller November 1997–May 2006

Challenge 72 n Why are most planets and moons (almost) spherical?

**
A rubber band connects the tips of the two hands of a clock. What is the path followed
Challenge 73 n by the mid-point of the band?
**
There are two important quantities connected to angles. As shown in Figure 17, what is
usually called a (plane) angle is defined as the ratio between the lengths of the arc and the
radius. A right angle is π~2 radian (or π~2 rad) or 90°.
The solid angle is the ratio between area and the square of the radius. An eighth of a

Dvipsbugw
 galilean physics – motion in everyday life 61

a
r A
α Ω
r
A
a Ω = −2
α=− r
r

Dvipsbugw
F I G U R E 17 The definition of plane and solid angles

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

sky yks

horizon noziroh

earth htrae

F I G U R E 18 How the apparent size of the Moon and the Sun changes

sphere is π~2 or steradian π~2 sr. As a result, a small solid angle shaped like a cone and
Challenge 74 n the angle of the cone tip are different. Can you find the relationship?
**
The definition of angle helps to determine the size of a firework display. Measure the time
T, in seconds, between the moment that you see the rocket explode in the sky and the
Copyright © Christoph Schiller November 1997–May 2006

moment you hear the explosion, measure the (plane) angle α of the ball with your hand.
The diameter D is
D  6 s~° T α . (3)

Challenge 75 e Why? For more about fireworks, see the http://cc.oulu.fi/~kempmp website. By the way,
the angular distance between the knuckles of an extended fist are about 3°, 2° and 3°, the
Challenge 76 n size of an extended hand 20°. Can you determine the other angles related to your hand?
**
Measuring angular size with the eye only is tricky. For example, can you say whether the
Challenge 77 e Moon is larger or smaller than the nail of your thumb at the end of your extended arm?

Dvipsbugw
 62 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw

F I G U R E 19 How the apparent size of the Moon changes during its

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

orbit (© Anthony Ayiomamitis)

F I G U R E 20 A
vernier/nonius/clavius

Angular size is not an intuitive quantity; it requires measurement instruments.

A famous example, shown in Figure 18, illustrates the difficulty of estimating angles.
Both the Sun and the Moon seem larger when they are on the horizon. In ancient times,
Ptolemy explained this illusion by an unconscious apparent distance change induced by
the human brain. In fact, the Moon is even further away from the observer when it is just
above the horizon, and thus its image is smaller than it was a few hours earlier, when it
Challenge 78 n was high in the sky. Can you confirm this?
In fact, the Moon’s size changes much more due to another effect: the orbit of the Moon
is elliptical. An example of this is shown in Figure 19.
**
Copyright © Christoph Schiller November 1997–May 2006

Cylinders can be used to roll a flat object over the floor; they keep the object plane always
at the same distance from the floor. What cross-sections other than circular allow you to
Challenge 79 n realize the same feat? How many examples can you find?
**
Galileo also made mistakes. In his famous book, the Dialogues, he says that the curve
formed by a thin chain hanging between two nails is a parabola, i.e. the curve defined
Challenge 80 d by y = x 2 . That is not correct. What is the correct curve? You can observe the shape
(approximately) in the shape of suspension bridges.

**

Dvipsbugw
 galilean physics – motion in everyday life 63

How does a vernier work? It is called nonius in other languages. The first name is derived
from a French military engineer* who did not invent it, the second is a play of words
on the Latinized name of the Portuguese inventor of a more elaborate device** and the
Latin word for ‘nine’. In fact, the device as we know it today – shown in Figure 20 – was
designed around 1600 by Christophonius Clavius,*** the same astronomer who made
the studies that formed the basis of the Gregorian calendar reform of 1582. Are you able
to design a vernier/nonius/clavius that, instead of increasing the precision tenfold, does
Challenge 81 n so by an arbitrary factor? Is there a limit to the attainable precision?
Dvipsbugw
**
Draw three circles, of different sizes, that touch each other. Now draw a fourth circle in
the space between, touching the outer three. What simple relation do the inverse radii of
Challenge 82 n the four circles obey?
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Take a tetrahedron OABC whose triangular sides OAB, OBC and OAC are rectangular in
O. In other words, OA, OB and OC are all perpendicular to each other. In the tetrahedron,
the areas of the triangles OAB, OBC and OAC are respectively 8, 4 and 1. What is the area
Challenge 83 n of triangle ABC?
**
With two rulers, you can add and subtract numbers by lying them side by side. Are you
Challenge 84 n able to design rulers that allow you to multiply and divide in the same manner? More
elaborate devices using this principle were called slide rules and were the precursors of
electronic calculators; they were in use all over the world until the 1970s.
**
How many days would a year have if the Earth turned the other way with the same rota-
Challenge 85 n tion frequency?

**
Challenge 86 n Where is the Sun in the spectacular situation shown in Figure 21?
**
Ref. 38 Could a two-dimensional universe exist? Alexander Dewdney described such a universe
Copyright © Christoph Schiller November 1997–May 2006

Challenge 87 d in a book. Can you explain why a two-dimensional universe is impossible?

* Pierre Vernier (1580–1637), French military officer interested in cartography.

** Pedro Nuñes or Peter Nonnius (1502–1578), Portuguese mathematician and cartographer.
*** Christophonius Clavius or Schlüssel (1537–1612), Bavarian astronomer.

Dvipsbugw
64 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw

F I G U R E 21 Anticrepuscular rays (© Peggy Peterson)

How to describe motion – kinematics

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

La filosofia è scritta in questo grandissimo libro

“ che continuamente ci sta aperto innanzi agli

occhi (io dico l’universo) ... Egli è scritto in
lingua matematica.*
Galileo Galilei, Il saggiatore VI.

Experiments show that the properties of Galilean time and space are extracted from the
environment by most higher animals and by young children. Later, when children learn
to speak, they put these experiences into concepts, as was just done above. With the help
”
of these concepts, grown-up children then say that motion is change of position with time.
This description is illustrated by rapidly flipping the lower left corners of this book, start-
ing at page 173. Each page simulates an instant of time, and the only change that takes
place during motion is in the position of the object, represented by the dark spot. The
other variations from one picture to the next, which are due to the imperfections of print-
ing techniques, can be taken to simulate the inevitable measurement errors.
It is evident that calling ‘motion’ the change of position with time is neither an explan-
ation nor a definition, since both the concepts of time and position are deduced from
motion itself. It is only a description of motion. Still, the description is useful, because it
allows for high precision, as we will find out by exploring gravitation and electrodyna-
mics. After all, precision is our guiding principle during this promenade. Therefore the
detailed description of changes in position has a special name: it is called kinematics.
Copyright © Christoph Schiller November 1997–May 2006

The set of all positions taken by an object over time forms a path or trajectory. The
origin of this concept is evident when one watches fireworks** or again the previously
mentioned flip movie in the lower left corners after page 173. With the description of space
and time by real numbers, a trajectory can be described by specifying its three coordinates
(x, y, z) – one for each dimension – as continuous functions of time t. (Functions are

* Science is written in this huge book that is continuously open before our eyes (I mean the universe) ... It is
written in mathematical language.
** On the world of fireworks, see the frequently asked questions list of the usenet group rec.pyrotechnics, or
search the web. A simple introduction is the article by J.A. Conkling, Pyrotechnics, Scientific American
pp. 66–73, July 1990.

Dvipsbugw
 how to describe motion – kinematics 65

collision
Dvipsbugw
F I G U R E 22 Two ways to test that the time of free fall does not depend on horizontal velocity

defined in detail on page 650.) This is usually written as x = x(t) = (x(t), y(t), z(t)).
For example, observation shows that the height z of any thrown or falling stone changes
as
z(t) = z 0 + v 0 (t − t 0 ) − 21 д(t − t o )2 (4)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where t 0 is the time the fall starts, z 0 is the initial height, v 0 is the initial velocity in the
vertical direction and д = 9.8 m~s2 is a constant that is found to be the same, within about
one part in 300, for all falling bodies on all points of the surface of the Earth. Where do the
Ref. 39 value 9.8 m~s2 and its slight variations come from? A preliminary answer will be given
shortly, but the complete elucidation will occupy us during the larger part of this hike.
Equation (4) allows us to determine the depth of a well, given the time a stone takes
Challenge 88 n to reach its bottom. The equation also»gives the speed v with which one hits the ground
after jumping from a tree, namely v = 2дh . A height of 3 m yields a velocity of 27 km~h.
The velocity is thus proportional only to the square root of the height. Does this mean
Challenge 89 n that one’s strong fear of falling results from an overestimation of its actual effects?
Galileo was the first to state an important result about free fall: the motions in the
horizontal and vertical directions are independent. He showed that the time it takes for a
cannon ball that is shot exactly horizontally to fall is independent of the strength of the
gunpowder, as shown in Figure 22. Many great thinkers did not agree with this statement
Ref. 40 even after his death: in 1658 the Academia del Cimento even organized an experiment
to check this assertion, by comparing the flying cannon ball with one that simply fell
Challenge 90 n vertically. Can you imagine how they checked the simultaneity? Figure 22 also shows
how you can check this at home. In this experiment, whatever the powder load of the
cannon, the two bodies will always collide, thus proving the assertion.
Copyright © Christoph Schiller November 1997–May 2006

In other words, a canon ball is not accelerated in the horizontal direction. Its horizontal
motion is simply unchanging. By extending the description of equation (4) with the two
expressions for the horizontal coordinates x and y, namely

x(t) = x 0 + v x0 (t − t 0 )
y(t) = y 0 + v y0 (t − t 0 ) , (5)

a complete description for the path followed by thrown stones results. A path of this shape
is called a parabola; it is shown in Figures 9, 22 and 23.*A parabolic shape is also used for
* Apart from the graphs shown in Figure 23, there is also the configuration space spanned by the coordinates

Dvipsbugw
 66 i galilean motion • 2. galilean physics – motion in everyday life

configuration space-time hodograph phase space

space diagrams graph

z z vz mv z

x t vx z
Dvipsbugw
x mv x

t x

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 23 Various types of graphs describing the same path of a thrown stone

F I G U R E 24 Three
superimposed images of a frass
pellet shot away by a caterpillar
(© Stanley Caveney)

Challenge 91 n light reflectors inside pocket lamps or car headlights. Can you show why?

Throwing and shooting

The kinematic description of motion is useful for answering a whole range of questions.
Copyright © Christoph Schiller November 1997–May 2006

**
Numerous species of moth and butterfly caterpillars shoot away their frass – to put it
Ref. 41 more crudely: their shit – so that its smell does not help predators to locate them. Stanley
Caveney and his team took photographs of this process. Figure 24 shows a caterpillar
(yellow) of the skipper Calpodes ethlius inside a rolled up green leaf caught in the act.
Given that the record distance observed is 1.5 m (though by another species, Epargyreus
Challenge 92 n clarus), what is the ejection speed? How do caterpillars achieve it?

of all particles of a system; only for a single particle it is equal to the real space. The phase space diagram is
also called state space diagram.

Dvipsbugw
 how to describe motion – kinematics 67

y
derivative:
dy/dt

slope: Δ y
Δ y/Δ t
Δ t
t

Dvipsbugw
F I G U R E 25 Derivatives

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
What is the horizontal distance one can reach with a stone, given the speed and the angle
Challenge 93 n from the horizontal at which it is thrown?
**
Challenge 94 n How can the speed of falling rain be measured using an umbrella? This method can also
Page 276 be used to measure the speed of light.

**
Challenge 95 n What is the maximum numbers of balls that could be juggled at the same time?
**
Finding an upper limit for the long jump is interesting. The running speed world
record in 1997 was 12 m~s  43 km~h by Ben Johnson, and the women’s record was
Ref. 42 11 m~s  40 km~h. In fact, long jumpers never run much faster than about 9.5 m~s. How
much extra jump distance could they achieve if they could run full speed? How could
Ref. 43 they achieve that? In addition, long jumpers take off at angles of about 20°, as they are
not able to achieve a higher angle at the speed they are running. How much would they
Challenge 96 n gain if they could achieve 45°? Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 97 n Is it true that rain drops would kill if it weren’t for the air resistance of the atmosphere?
What about ice?
**
Challenge 98 n Are bullets fired from a gun falling back after being fired into the air dangerous?

The last two issues arise because equation (4) does not hold in all cases. For example,
leaves or potato crisps do not follow it. As Galileo already knew, this is a consequence of
air resistance; we will discuss it shortly. In fact, even without air resistance, the path of a
Challenge 99 n stone is not always a parabola; can you find such a situation?

Dvipsbugw
 68 i galilean motion • 2. galilean physics – motion in everyday life

What is rest?
In the Galilean description of nature, motion and rest are opposites. In other words, a
body is at rest when its position, i.e. its coordinates, do not change with time. In other
words, (Galilean) rest is defined as

x(t) = const . (6)

Later we will see that this definition, contrary to first impressions, is not much use and Dvipsbugw
will have to be modified. The definition of rest implies that non-resting objects can be
distinguished by comparing the rapidity of their displacement. One thus can define the
velocity v of an object as the change of its position x with time t. This is usually written as

v=
dx
. (7)
dt

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In this expression, valid for each coordinate separately, d~dt means ‘change with time’;
one can thus say that velocity is the derivative of position with respect to time. The speed
v is the name given to the magnitude of the velocity v. Derivatives are written as fractions
in order to remind the reader that they are derived from the idea of slope. The expression

dy ∆y
is meant as an abbreviation of lim , (8)
dt ∆t 0 ∆t

a shorthand for saying that the derivative at a point is the limit of the slopes in the neigh-
Challenge 100 e bourhood of the point, as shown in Figure 25. This definition implies the working rules

d(y + z) dy dz d dy d2 y
= =c = =
d(c y) dy d(yz) dy dz
+ , , , z+y ,(9)
dt dt dt dt dt dt dt dt 2 dt dt dt
c being any number. This is all one ever needs to know about derivatives. The quantities
dt and dy, sometimes useful by themselves, are called differentials. These concepts are due
to Gottfried Wilhelm Leibniz.* Derivatives lie at the basis of all calculations based on the
continuity of space and time. Leibniz was the person who made it possible to describe
and use velocity in physical formulae and, in particular, to use the idea of velocity at a
given point in time or space for calculations.
Copyright © Christoph Schiller November 1997–May 2006

The definition of velocity assumes that it makes sense to take the limit ∆t 0. In other
words, it is assumed that infinitely small time intervals do exist in nature. The definition
of velocity with derivatives is possible only because both space and time are described by
sets which are continuous, or in mathematical language, connected and complete. In the
rest of our walk we shall not forget that from the beginning of classical physics, infinities
are present in its description of nature. The infinitely small is part of our definition of

* Gottfried Wilhelm Leibniz (b. 1646 Leipzig, d. 1716 Hannover), Saxon lawyer, physicist, mathematician,
philosopher, diplomat and historian. He was one of the great minds of mankind; he invented the differ-
ential calculus (before Newton) and published many successful books in the various fields he explored,
among them De arte combinatoria, Hypothesis physica nova, Discours de métaphysique, Nouveaux essais sur
l’entendement humain, the Théodicée and the Monadologia.

Dvipsbugw
 how to describe motion – kinematics 69

velocity. Indeed, differential calculus can be defined as the study of infinity and its uses.
We thus discover that the appearance of infinity does not automatically render a descrip-
tion impossible or imprecise. In order to remain precise, physicists use only the smallest
two of the various possible types of infinities. Their precise definition and an overview of
Page 648 other types are introduced in the intermezzo following this chapter.
The appearance of infinity in the usual description of motion
was first criticized in his famous ironical arguments by Zeno of
Ref. 44 Elea (around 445 bce), a disciple of Parmenides. In his so-called
third argument, Zeno explains that since at every instant a given Dvipsbugw
object occupies a part of space corresponding to its size, the notion
of velocity at a given instant makes no sense; he provokingly con-
cludes that therefore motion does not exist. Nowadays we would
not call this an argument against the existence of motion, but
against its usual description, in particular against the use of infin-
Challenge 101 e itely divisible space and time. (Do you agree?) Nevertheless, the Gottfried Leibniz

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

description criticized by Zeno actually works quite well in every-
day life. The reason is simple but deep: in daily life, changes are indeed continuous.
Large changes in nature are made up of many small changes. This property of nature is
not obvious. For example, we note that we have tacitly assumed that the path of an object
is not a fractal or some other badly behaved entity. In everyday life this is correct; in other
domains of nature it is not. The doubts of Zeno will be partly rehabilitated later in our
Page 1001 walk, and increasingly so the more we proceed. The rehabilitation is only partial, as the
solution will be different from that which he envisaged; on the other hand, the doubts
about the idea of ‘velocity at a point’ will turn out to be well-founded. For the moment
though, we have no choice: we continue with the basic assumption that in nature changes
happen smoothly.
Why is velocity necessary as a concept? Aiming for precision in the description of
motion, we need to find the complete list of aspects necessary to specify the state of an
object. The concept of velocity is obviously on this list. Continuing along the same lines,
we call acceleration a of a body the change of velocity v with time, or

dv d2 x
a= = . (10)
dt dt 2
Acceleration is what we feel when the Earth trembles, an aeroplane takes off, or a bicycle
Copyright © Christoph Schiller November 1997–May 2006

goes round a corner. More examples are given in Table 10. Like velocity, acceleration has
both a magnitude and a direction, properties indicated by the use of bold letters for their
abbreviations. *

* Such physical quantities are called vectors. In more precise, mathematical language, a vector is an element
of a set, called vector space, in which the following properties hold for all vectors a and b and for all numbers
c and d:
c(a + b) = ca + cb , (c + d)a = ca + da , (cd)a = c(da) and 1a = a . (11)
Another example of vector space is the set of all positions of an object. Does the set of all rotations form a
Challenge 102 n vector space? All vector spaces allow the definition of a unique null vector and of a single negative vector
for each vector in it.
Note that vectors do not have specified points at which they start: two arrows with same direction and

Dvipsbugw
 70 i galilean motion • 2. galilean physics – motion in everyday life

TA B L E 10 Some measured acceleration values

O b s e r va t i o n Accelera-
tion

What is the lowest you can find? Challenge 105 n

Acceleration of the galaxy M82 by its ejected jet 10 fm~s2
Acceleration of a young star by an ejected jet 10 pm~s2
Acceleration of the Sun in its orbit around the Milky Way 0.2 nm~s2 Dvipsbugw
Unexplained deceleration of the Pioneer satellites 0.8 nm~s2
Acceleration at Equator due to Earth’s rotation 0.34 mm~s2
Centrifugal acceleration due to the Earth’s rotation 33 mm~s2
Electron acceleration in household electricity wire due to alternating 50 mm~s2
current
Gravitational acceleration on the Moon 1.6 m~s2

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Gravitational acceleration on the Earth’s surface, depending on location 9.8  0.1 m~s2
Standard gravitational acceleration 9.806 65 m~s2
Highest acceleration for a car or motorbike with engine-driven wheels 15 m~s2
Gravitational acceleration on Jupiter’s surface 240 m~s2
Acceleration of cheetah 32 m~s2
Acceleration that triggers air bags in cars 360 m~s2
Fastest leg-powered acceleration (by the froghopper, Philaenus spumarius, 4 km~s2
an insect)
Tennis ball against wall 0.1 Mm~s2
Bullet acceleration in rifle 5 Mm~s2
Fastest centrifuges 0.1 Gm~s2
Acceleration of protons in large accelerator 90 Tm~s2
Acceleration of protons inside nucleus 1031 m~s2
Highest possible acceleration in nature 1052 m~s2

Higher derivatives than acceleration can also be defined in the same manner. They
Challenge 104 n add little to the description of nature, because as we will show shortly neither these nor
even acceleration itself are useful for the description of the state of motion of a system.
Copyright © Christoph Schiller November 1997–May 2006

length are the same vector, even if they start at different points in space.
In many vector spaces the concept of length (specifying the ‘magnitude’) can be introduced, usually via
an intermediate step. A vector space is called Euclidean if one can define for it a scalar product between two
vectors, a number ab satisfying

aa E 0 , ab = ba , (a + a′ )b = ab + a′ b , a(b + b′ ) = ab + ab′ and (ca)b = a(cb) = c(ab) . (12)

In Cartesian coordinate notation, the standard scalar product is given by ab = a x b x + a y b y + a z b z . Whenever

it vanishes the two vectors are orthogonal. The lengthºor norm of a vector can then be defined as the square
root of the scalar product of a vector with itself: a = aa .
The scalar product is also useful for specifying directions. Indeed, the scalar product between two vectors
Challenge 103 n encodes the angle between them. Can you deduce this important relation?

Dvipsbugw
 how to describe motion – kinematics 71

Objects and point particles

Wenn ich den Gegenstand kenne, so kenne ich

“ auch sämtliche Möglichkeiten seines

Vorkommens in Sachverhalten.*
Ludwig Wittgenstein, Tractatus, 2.0123

One aim of the study of motion is to find a complete and precise description of both states
and objects. With the help of the concept of space, the description of objects can be refined
considerably. In particular, one knows from experience that all objects seen in daily life
” Dvipsbugw
Challenge 106 e have an important property: they can be divided into parts. Often this observation is
expressed by saying that all objects, or bodies, have two properties. First, they are made
out of matter,** defined as that aspect of an object responsible for its impenetrability,
i.e. the property preventing two objects from being in the same place. Secondly, bodies
have a certain form or shape, defined as the precise way in which this impenetrability is
distributed in space.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In order to describe motion as accurately as possible, it is convenient to start with
those bodies that are as simple as possible. In general, the smaller a body, the simpler
it is. A body that is so small that its parts no longer need to be taken into account is
called a particle. (The older term corpuscle has fallen out of fashion.) Particles are thus
idealized small stones. The extreme case, a particle whose size is negligible compared with
the dimensions of its motion, so that its position is described completely by a single triplet
of coordinates, is called a point particle or a point mass. In equation (4), the stone was
assumed to be such a point particle.
Do point-like objects, i.e. objects smaller than anything one can measure, exist in daily
life? Yes and no. The most notable examples are the stars. At present, angular sizes as small
as 2 µrad can be measured, a limit given by the fluctuations of the air in the atmosphere.
In space, such as for the Hubble telescope orbiting the Earth, the angular limit is due to
the diameter of the telescope and is of the order of 10 nrad. Practically all stars seen from
Earth are smaller than that, and are thus effectively ‘point-like’, even when seen with the
most powerful telescopes.
As an exception to the general rule, the size of a few large and nearby stars, of red
giant type, can be measured with special instruments.*** Betelgeuse, the higher of the
two shoulders of Orion shown in Figure 26, Mira in Cetus, Antares in Scorpio, Aldebaran
in Taurus and Sirius in Canis Major are examples of stars whose size has been measured;
Ref. 46 they are all only a few light years from Earth. Of course, like the Sun, all other stars have
Copyright © Christoph Schiller November 1997–May 2006

Challenge 107 n a finite size, but one cannot prove this by measuring dimensions in photographs. (True?)

* If I know an object I also know all its possible occurrences in states of affairs.
Ref. 45 ** Matter is a word derived from the Latin ‘materia’, which originally meant ‘wood’ and was derived via
intermediate steps from ‘mater’, meaning ‘mother’.
*** The website http://www.astro.uiuc.edu/~kaler/sow/sowlist.html gives an introduction to the different
types of stars. The http://www.astro.wisc.edu/~dolan/constellations/constellations.html website provides de-
tailed and interesting information about constellations.
For an overview of the planets, see the beautiful book by K.R. L ang & C.A. Whitney, Vagabonds de
l’espace – Exploration et découverte dans le système solaire, Springer Verlag, 1993. The most beautiful pictures
of the stars can be found in D. Malin, A View of the Universe, Sky Publishing and Cambridge University
Press, 1993.

Dvipsbugw
 72 i galilean motion • 2. galilean physics – motion in everyday life

α γ
Betelgeuse
Bellatrix

ε δ Mintaka
ζ Alnilam
Alnitag

β
κ Dvipsbugw
Rigel
Saiph

F I G U R E 26 Orion (in natural colours) and Betelgeuse

The difference between ‘point-like’ and finite size sources can be seen with the naked
eye: at night, stars twinkle, but planets do not. (Check it!) This effect is due to the tur-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 108 e
bulence of air. Turbulence has an effect on the almost point-like stars because it deflects
light rays by small amounts. On the other hand, air turbulence is too weak to lead to
twinkling of sources of larger angular size, such as planets or artificial satellites,* because
the deflection is averaged out in this case.
An object is point-like for the naked eye if its angular size is smaller than about
Challenge 109 n 2 ′ = 0.6 mrad. Can you estimate the size of a ‘point-like’ dust particle? By the way, an
object is invisible to the naked eye if it is point-like and if its luminosity, i.e. the intensity
of the light from the object reaching the eye, is below some critical value. Can you esti-
mate whether there are any man-made objects visible from the Moon, or from the space
Challenge 110 n shuttle?
The above definition of ‘point-like’ in everyday life is obviously misleading. Do proper,
real point particles exist? In fact, is it at all possible to show that a particle has vanishing
size? This question will be central in the last two parts of our walk. In the same way, we
need to ask and check whether points in space do exist. Our walk will lead us to the
astonishing result that all the answers to these questions are negative. Can you imagine
Challenge 111 n why? Do not be disappointed if you find this issue difficult; many brilliant minds have
had the same problem.
However, many particles, such as electrons, quarks or photons are point-like for all
practical purposes. Once one knows how to describe the motion of point particles, one
can also describe the motion of extended bodies, rigid or deformable, by assuming that
Copyright © Christoph Schiller November 1997–May 2006

they are made of parts. This is the same approach as describing the motion of an animal as
a whole by combining the motion of its various body parts. The simplest description, the
continuum approximation, describes extended bodies as an infinite collection of point
particles. It allows us to understand and to predict the motion of milk and honey, the
motion of the air in hurricanes and of perfume in rooms. The motion of fire and all other
gaseous bodies, the bending of bamboo in the wind, the shape changes of chewing gum,
Ref. 47 and the growth of plants and animals can also be described in this way.
Page 704 A more precise description than the continuum approximation is given below. Nev-

* A satellite is an object circling a planet, like the Moon; an artificial satellite is a system put into orbit by
humans, like the Sputniks.

Dvipsbugw
 how to describe motion – kinematics 73

F I G U R E 27 How an object can rotate continuously without tangling up the connection to a second
object
Dvipsbugw

Figure still to be added

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 28 Legs and ‘wheels’ in living beings

ertheless, all observations so far have confirmed that the motion of large bodies can be
described to high precision as the result of the motion of their parts. This approach will
guide us through the first two parts of our mountain ascent. Only in the third part will
we discover that, at a fundamental scale, this decomposition is impossible.

Legs and wheels

The parts of a body determine its shape. Shape is an important aspect of bodies: among
other things, it tells us how to count them. In particular, living beings are always made of
a single body. This is not an empty statement: from this fact we can deduce that animals
cannot have wheels or propellers, but only legs, fins, or wings. Why?
Living beings have only one surface; simply put, they have only one piece of skin. Math-
Appendix D ematically speaking, animals are connected. This is often assumed to be obvious, and it
Ref. 48 is often mentioned that the blood supply, the nerves and the lymphatic connections to a
rotating part would get tangled up. However, this argument is not so simple, as Figure 27
shows. It shows that it is indeed possible to rotate a body continuously against a second
one, without tangling up the connections. Can you find an example for this kind of mo-
Copyright © Christoph Schiller November 1997–May 2006

Challenge 112 n tion in your own body? Are you able to see how many cables may be attached to the
Challenge 113 n rotating body of the figure without hindering the rotation?
Despite the possibility of animals having rotating parts, the method of Figure 27 still
Challenge 114 n cannot be used to make a practical wheel or propeller. Can you see why? Evolution had
no choice: it had to avoid animals with parts rotating around axles. That is the reason
that propellers and wheels do not exist in nature. Of course, this limitation does not rule
Ref. 49 out that living bodies move by rotation as a whole: tumbleweed, seeds from various trees,
some insects, certain other animals, children and dancers occasionally move by rolling
or rotating as a whole.
Single bodies, and thus all living beings, can only move through deformation of their
shape: therefore they are limited to walking, running, crawling or flapping wings or fins,

Dvipsbugw
 74 i galilean motion • 2. galilean physics – motion in everyday life

as shown in Figure 28. In contrast, systems of several bodies, such as bicycles, pedal boats
or other machines, can move without any change of shape of their components, thus
enabling the use of axles with wheels, propellers or other rotating devices.*
In summary, whenever we observe a construction in which some part is turning con-
tinuously (and without the ‘wiring’ of the figure) we know immediately that it is an arte-
fact: it is a machine, not a living being (but built by one). However, like so many state-
ments about living creatures, this one also has exceptions. The distinction between one
and two bodies is poorly defined if the whole system is made of only a few molecules. This
happens most clearly inside bacteria. Organisms such as Escherichia coli, the well-known Dvipsbugw
bacterium found in the human gut, or bacteria from the Salmonella family, all swim using
flagella. Flagella are thin filaments, similar to tiny hairs that stick out of the cell membrane.
In the 1970s it was shown that each flagellum, made of one or a few long molecules with
Page 972 a diameter of a few tens of nanometres, does in fact turn about its axis. A bacterium is
able to turn its flagella in both clockwise and anticlockwise directions, can achieve more
Ref. 50 than 1000 turns per second, and can turn all its flagella in perfect synchronization. (These

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

wheels are so tiny that they do not need a mechanical connection.) Therefore wheels ac-
tually do exist in living beings, albeit only tiny ones. But let us now continue with our
study of simple objects.

Objects and images

Walking through a forest we observe two rather different types of motion: the breeze
moves the leaves, and at the same time their shadows move on the ground. Shadows
Ref. 51 are a simple type of image. Both objects and images are able to move. Running tigers,
falling snowflakes, and material ejected by volcanoes are examples of motion, since they
all change position over time. For the same reason, the shadow following our body, the
beam of light circling the tower of a lighthouse on a misty night, and the rainbow that
constantly keeps the same apparent distance from the hiker are examples of motion.
Everybody who has ever seen an animated cartoon knows that images can move in
more surprising ways than objects. Images can change their size, shape and even colour,
a feat only few objects are able to perform.** Images can appear and disappear without
trace, multiply, interpenetrate, go backwards in time and defy gravity or any other force.

* Despite the disadvantage of not being able to use rotating parts and of being restricted to one piece only,
nature’s moving constructions, usually called animals, often outperform human built machines. As an ex-
ample, compare the size of the smallest flying systems built by evolution with those built by humans. (See,
Copyright © Christoph Schiller November 1997–May 2006

e.g. http://pixelito.reference.be.) There are two reasons for this discrepancy. First, nature’s systems have in-
tegrated repair and maintenance systems. Second, nature can build large structures inside containers with
small openings. In fact, nature is very good at what people do when they build sailing ships inside glass
Challenge 115 n bottles. The human body is full of such examples; can you name a few?
** Excluding very slow changes such as the change of colour of leaves in the Fall, in nature only certain
crystals, the octopus, the chameleon and a few other animals achieve this. Of man-made objects, television,
Challenge 116 n computer displays, heated objects and certain lasers can do it. Do you know more examples? An excellent
source of information on the topic of colour is the book by K. Nassau, The Physics and Chemistry of Colour
– the fifteen causes of colour, J. Wiley & Sons, 1983. In the popular science domain, the most beautiful book is
the classic work by the Flemish astronomer Marcel G.J. Minnaert, Light and Colour in the Outdoors,
Springer, 1993, an updated version based on his wonderful book series, De natuurkunde van ‘t vrije veld,
Ref. 52 Thieme & Cie, Zutphen. Reading it is a must for all natural scientists. On the web, there is also the – much
simpler – http://webexhibits.org/causesofcolour website.

Dvipsbugw
 objects and images 75

push

F I G U R E 29 In which direction
does the bicycle turn?

Images, even ordinary shadows, can move faster than light. Images can float in space Dvipsbugw
and keep the same distance from approaching objects. Objects can do almost none of
Ref. 53 this. In general, the ‘laws of cartoon physics’ are rather different from those in nature. In
fact, the motion of images does not seem to follow any rules, in contrast to the motion of
objects. On the other hand, both objects and images differ from their environment in that
they have boundaries defining their size and shape. We feel the need for precise criteria
allowing the two cases to be distinguished.
Making a clear distinction between images and objects is performed using the same

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

method that children or animals use when they stand in front of a mirror for the first
time: they try to touch what they see. Indeed, if we are able to touch what we see – or
more precisely, if we are able to move it – we call it an object, otherwise an image.* Images
Page 787 cannot be touched, but objects can. Images cannot hit each other, but objects can. And as
everybody knows, touching something means feeling that it resists movement. Certain
bodies, such as butterflies, pose little resistance and are moved with ease, others, such as
ships, resist more, and are moved with more difficulty. This resistance to motion – more
precisely, to change of motion – is called inertia, and the difficulty with which a body can
be moved is called its (inertial) mass. Images have neither inertia nor mass.
Summing up, for the description of motion we must distinguish bodies, which can be
touched and are impenetrable, from images, which cannot and are not. Everything visible
Challenge 117 n is either an object or an image; there is no third possibility. (Do you agree?) If the object
is so far away that it cannot be touched, such as a star or a comet, it can be difficult to
decide whether one is dealing with an image or an object; we will encounter this difficulty
Challenge 118 n repeatedly. For example, how would you show that comets are objects and not images?
In the same way that objects are made of matter, images are made of radiation. Images
Ref. 54 are the domain of shadow theatre, cinema, television, computer graphics, belief systems
and drug experts. Photographs, motion pictures, ghosts, angels, dreams and many hallu-
cinations are images (sometimes coupled with brain malfunction). To understand images,
we need to study radiation (plus the eye and the brain). However, due to the importance
Copyright © Christoph Schiller November 1997–May 2006

of objects – after all we are objects ourselves – we study the latter first.

Motion and contact

Democritus affirms that there is only one type

Ref. 55
“ of movement: That resulting from collision.
Aetius, Opinions.

”
* One could propose including the requirement that objects may be rotated; however, this requirement gives
difficulties in the case of atoms, as explained on page 751, and with elementary particles, so that rotation is
not made a separate requirement.

Dvipsbugw
76 i galilean motion • 2. galilean physics – motion in everyday life

v1 v2 Dvipsbugw

v'2
v'1

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 30 Collisions F I G U R E 31 The standard kilogram (© BIPM)
define mass

When a child rides a monocycle, she or he makes use of a general rule in our world:
one body acting on another puts it in motion. Indeed, in about six hours, anybody can
learn to ride and enjoy a monocycle. As in all of life’s pleasures, such as toys, animals,
women, machines, children, men, the sea, wind, cinema, juggling, rambling and loving,
Copyright © Christoph Schiller November 1997–May 2006

something pushes something else. Thus our first challenge is to describe this transfer of
motion in more precise terms.
Contact is not the only way to put something into motion; a counter-example is an
apple falling from a tree or one magnet pulling another. Non-contact influences are more
fascinating: nothing is hidden, but nevertheless something mysterious happens. Contact
motion seems easier to grasp, and that is why one usually starts with it. However, despite
this choice, non-contact forces are not easily avoided. Taking this choice one has a similar
experience to that of cyclists. (See Figure 29.) If you ride a bicycle at a sustained speed
and try to turn left by pushing the right-hand steering bar, you will turn right.* In other
words, despite our choice the rest of our walk will rapidly force us to study non-contact
interactions as well.

Dvipsbugw
 objects and images 77

What is mass?

“
∆ός µοι ποῦ στω καὶ κινῶ τὴν γῆν.
Da ubi consistam, et terram movebo.*
Archimedes

When we push something we are unfamiliar with, such as when we kick an object on
the street, we automatically pay attention to the same aspect that children explore when
they stand in front of a mirror for the first time, or when they see a red laser spot for
”
the first time. We check whether the unknown entity can be pushed and pay attention to Dvipsbugw
how the unknown object moves under our influence. The high precision version of the
experiment is shown in Figure 30. Repeating the experiment with various pairs of objects,
we find – as in everyday life – that a fixed quantity m i can be ascribed to every object i.
The more difficult it is to move an object, the higher the quantity; it is determined by the
relation
=−
m2 ∆v 1
(13)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

m1 ∆v 2

where ∆v is the velocity change produced by the collision. The number m i is called the
mass of the object i.
In order to have mass values that are common to everybody, the mass value for one par-
ticular, selected object has to be fixed in advance. This special object, shown in Figure 31
is called the standard kilogram and is kept with great care under vacuum in a glass con-
tainer in Sèvres near Paris. It is touched only once every few years because otherwise dust,
humidity, or scratches would change its mass. Through the standard kilogram the value
of the mass of every other object in the world is determined.
The mass thus measures the difficulty of getting something moving. High masses are
harder to move than low masses. Obviously, only objects have mass; images don’t. (By the
Ref. 45 way, the word ‘mass’ is derived, via Latin, from the Greek µαζα – bread – or the Hebrew
‘mazza’ – unleavened bread – quite a change in meaning.)
Experiments with everyday life objects also show that throughout any collision, the
sum of all masses is conserved:
Q m i = const . (14)
i

The principle of conservation of mass was first stated by Antoine-Laurent Lavoisier.**

Conservation of mass implies that the mass of a composite system is the sum of the mass
Copyright © Christoph Schiller November 1997–May 2006

of the components. In short, Galilean mass is a measure for the quantity of matter.
* This surprising effect obviously works only above a certain minimal speed. Can you determine what this
Challenge 119 n speed is? Be careful! Too strong a push will make you fall.
* ‘Give me a place to stand, and I’ll move the Earth.’ Archimedes (c. 283–212), Greek scientist and engineer.
Ref. 56 This phrase was attributed to him by Pappus. Already Archimedes knew that the distinction used by lawyers
between movable and immovable property made no sense.
** Antoine-Laurent Lavoisier (1743–1794), French chemist and a genius. Lavoisier was the first to under-
stand that combustion is a reaction with oxygen; he discovered the components of water and introduced
mass measurements into chemistry. When he was (unjustly) sentenced to the guillotine during the French
revolution, he decided to use the experience for a scientific experiment; he decided to blink his eyes as fre-
quently as possible after his head was cut off, in order to show others how long it takes to lose consciousness.
Lavoisier managed to blink eleven times.

Dvipsbugw
78 i galilean motion • 2. galilean physics – motion in everyday life

F I G U R E 32 Is this dangerous? Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Antoine Lavoisier
The definition of mass can also be given in another way. We can
ascribe a number m i to every object i such that for collisions free of
outside interference the following sum is unchanged throughout the
collision:
Q m i vi = const . (15)
i

The product of the velocity vi and the mass m i is called the mo-
mentum of the body. The sum, or total momentum of the system, is
the same before and after the collision; it is a conserved quantity. Mo-
Copyright © Christoph Schiller November 1997–May 2006

mentum conservation defines mass. The two conservation principles Christiaan Huygens
(14) and (15) were first stated in this way by the important Dutch physicist Christiaan
Huygens.* Some typical momentum values are given in Table 11.
Momentum conservation implies that when a moving sphere hits a resting one of the
same mass, a simple rule determines the angle between the directions the two spheres take

* Christiaan Huygens (b. 1629 ’s Gravenhage, d. 1695 Hofwyck) was one of the main physicists and math-
ematicians of his time. Huygens clarified the concepts of mechanics; he also was one of the first to show that
light is a wave. He wrote influential books on probability theory, clock mechanisms, optics and astronomy.
Among other achievements, Huygens showed that the Orion Nebula consists of stars, discovered Titan, the
moon of Saturn, and showed that the rings of Saturn consist of rock. (This is in contrast to Saturn itself,
whose density is lower than that of water.)

Dvipsbugw
 objects and images 79

TA B L E 11 Some measured momentum values

O b s e r va t i o n Momentum

Green photon momentum 2 ë 10−28 Ns

Average momentum of oxygen molecule in air 10−26 Ns
X-ray photon momentum 10−23 Ns
γ photon momentum 10−17 Ns
Highest particle momentum in accelerators 1 f Ns
Planck momentum 6.5 Ns Dvipsbugw
Fast billiard ball 3 Ns
Flying rifle bullet 10 Ns
Box punch 15 to 50 Ns
Comfortably walking human 80 Ns
Car on highway 40 kNs
Impact of meteorite with 2 km diameter 100 TNs

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Momentum of a galaxy in galaxy collision up to 1046 Ns

Challenge 120 n after the collision. Can you find this rule? It is particularly useful when playing billiards.
We will find out later that it is not valid in special relativity.
Another consequence is shown in Figure 32: a man lying on a bed of nails with two
large blocks of concrete on his stomach. Another man is hitting the concrete with a heavy
sledgehammer. As the impact is mostly absorbed by the concrete, there is no pain and no
Challenge 121 n danger – unless the concrete is missed. Why?
The above definition of mass has been generalized by the physicist and philosopher
Ernst Mach* in such a way that it is valid even if the two objects interact without contact,
as long as they do so along the line connecting their positions. The mass ratio between
two bodies is defined as a negative inverse acceleration ratio, thus as

=− ,
m2 a1
(16)
m1 a2

where a is the acceleration of each body during the interaction. This definition has been
studied in much detail in the physics community, mainly in the nineteenth century. A
Copyright © Christoph Schiller November 1997–May 2006

few points sum up the results:

— The definition of mass implies the conservation of momentum P mv. Momentum con-
servation is not a separate principle. Conservation of momentum cannot be checked
experimentally, because mass is defined in such a way that the principle holds.

* Ernst Mach (1838 Chrlice–1916 Vaterstetten), Austrian physicist and philosopher. The mach unit for aero-
plane speed as a multiple of the speed of sound in air (about 0.3 km~s) is named after him. He developed the
so-called Mach–Zehnder interferometer; he also studied the basis of mechanics. His thoughts about mass
and inertia influenced the development of general relativity, and led to Mach’s principle, which will appear
later on. He was also proud to be the last scientist denying – humorously, and against all evidence – the
existence of atoms.

Dvipsbugw
 80 i galilean motion • 2. galilean physics – motion in everyday life

— The definition of mass implies the equality of the products m 1 a 1 and −m 2 a 2 . Such
products are called forces. The equality of acting and reacting forces is not a separate
principle; mass is defined in such a way that the principle holds.
— The definition of mass is independent of whether contact is involved or not, and
whether the origin of the accelerations is due to electricity, gravitation, or other in-
teractions.* Since the interaction does not enter the definition of mass, mass values
defined with the help of the electric, nuclear or gravitational interaction all agree, as
long as momentum is conserved. All known interactions conserve momentum. For
some unfortunate historical reasons, the mass value measured with the electric or Dvipsbugw
nuclear interactions is called the ‘inertial’ mass and the mass measured using grav-
ity is called the ‘gravitational’ mass. As it turns out, this artificial distinction has no
real meaning; this becomes especially clear when one takes an observation point that
is far away from all the bodies concerned.
— The definition of mass is valid only for observers at rest or in inertial motion. More
about this issue later.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

By measuring the masses of bodies around us, as given in Table 12, we can explore
the science and art of experiments. We also discover the main properties of mass. It is
additive in everyday life, as the mass of two bodies combined is equal to the sum of the
two separate masses. Furthermore, mass is continuous; it can seemingly take any positive
value. Finally, mass is conserved; the mass of a system, defined as the sum of the mass of
all constituents, does not change over time if the system is kept isolated from the rest of
the world. Mass is not only conserved in collisions but also during melting, evaporation,
digestion and all other processes.
Later we will find that in the case of mass all these properties, summarized in Table 13,
are only approximate. Precise experiments show that none of them are correct.** For the
moment we continue with the present, Galilean concept of mass, as we have not yet a
better one at our disposal.
In a famous experiment in the sixteenth century, for several weeks Santorio Santorio
(Sanctorius) (1561–1636), friend of Galileo, lived with all his food and drink supply, and
also his toilet, on a large balance. He wanted to test mass conservation. How did the
Challenge 123 n measured weight change with time?
The definition of mass through momentum conservation implies that when an object
falls, the Earth is accelerated upwards by a tiny amount. If one could measure this tiny
amount, one could determine the mass of the Earth. Unfortunately, this measurement is
Copyright © Christoph Schiller November 1997–May 2006

Challenge 124 n impossible. Can you find a better way to determine the mass of the Earth?
Summarizing Table 13, the mass of a body is thus most precisely described by a positive
real number, often abbreviated m or M. This is a direct consequence of the impenetrability
of matter. Indeed, a negative (inertial) mass would mean that such a body would move in
the opposite direction of any applied force or acceleration. Such a body could not be kept

* As mentioned above, only central forces obey the relation (16) used to define mass. Central forces act
Page 90 between the centre of mass of bodies. We give a precise definition later. However, since all fundamental forces
are central, this is not a restriction. There seems to be one notable exception: magnetism. Is the definition
Challenge 122 n of mass valid in this case?
** In particular, in order to define mass we must be able to distinguish bodies. This seems a trivial require-
ment, but we discover that this is not always possible in nature.

Dvipsbugw
 objects and images 81

TA B L E 12 Some measured mass values

O b s e r va t i o n Mass

Mass increase due to absorption of one green photon 3.7 ë 10−36 kg

Lightest known object: electron 9.109 381 88(72) ë 10−31 kg
Atom of argon 39.962 383 123(3) u = 66.359 1(1) yg
Lightest object ever weighed (a gold particle) 0.39 ag
Human at early age (fertilized egg) 10−8 g Dvipsbugw
Water adsorbed on to a kilogram metal weight 10−5 g
Planck mass 2.2 ë 10−5 g
Fingerprint 10−4 g
Typical ant 10−4 g
Water droplet 1 mg
Honey bee 0.1 g

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Heaviest living things, such as the fungus Armillaria 106 kg
ostoyae or a large Sequoia Sequoiadendron giganteum
Largest ocean-going ship 400 ë 106 kg
Largest object moved by man (Troll gas rig) 687.5 ë 106 kg
Large antarctic iceberg 1015 kg
Water on Earth 1021 kg
Solar mass 2.0 ë 1030 kg
Our galaxy 1041 kg
Total mass visible in the universe 1054 kg

TA B L E 13 Properties of Galilean mass

Masses Physical M at h e m at i c a l Definition

propert y name

Can be distinguished distinguishability element of set Page 646

Can be ordered sequence order Page 1195
Can be compared measurability metricity Page 1205
Can change gradually continuity completeness Page 1214
Copyright © Christoph Schiller November 1997–May 2006

Can be added quantity of matter additivity Page 69

Beat any limit infinity unboundedness, openness Page 647
Do not change conservation invariance m = const
Do not disappear impenetrability positivity mE0

in a box; it would break through any wall trying to stop it. Strangely enough, negative mass
bodies would still fall downwards in the field of a large positive mass (though more slowly
Challenge 125 e than an equivalent positive mass). Are you able to confirm this? However, a small positive
mass object would float away from a large negative-mass body, as you can easily deduce
by comparing the various accelerations involved. A positive and a negative mass of the

Dvipsbugw
 82 i galilean motion • 2. galilean physics – motion in everyday life

same value would stay at constant distance and spontaneously accelerate away along the
Challenge 126 e line connecting the two masses. Note that both energy and momentum are conserved in
all these situations.* Negative-mass bodies have never been observed. Antimatter, which
Page 315, page 759 will be discussed later, also has positive mass.

Is motion eternal?
Every body continues in the state of rest or of

“ uniform motion in a straight line except in so

far as it doesn’t.
Arthur Eddington**
Dvipsbugw

The product p = mv of mass and velocity is called the momentum of a particle; it describes
the tendency of an object to keep moving during collisions. The larger it is, the harder it
is to stop the object. Like velocity, momentum has a direction and a magnitude: it is a
”
vector. In French, momentum is called ‘quantity of motion’, a more appropriate term. In

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the old days, the term ‘motion’ was used instead of ‘momentum’, for example by Newton.
Relation (15), the conservation of momentum, therefore expresses the conservation of
motion during interactions.
Momentum and energy are extensive quantities. That means that it can be said of both
that they flow from one body to the other, and that they can be accumulated in bodies,
in the same way that water flows and can be accumulated in containers. Imagining mo-
mentum as something that can be exchanged between bodies in collisions is always useful
when thinking about the description of moving objects.
Momentum is conserved. That explains the limitations you might experience when
being on a perfectly frictionless surface, such as ice or a polished, oil covered marble: you
cannot propel yourself forward by patting your own back. (Have you ever tried to put
a cat on such a marble surface? It is not even able to stand on its four legs. Neither are
Challenge 129 n humans. Can you imagine why?) Momentum conservation also answers the puzzles of
Figure 33.
The conservation of momentum and mass also means that teleportation (‘beam me
Challenge 130 n up’) is impossible in nature. Can you explain this to a non-physicist?
Momentum conservation implies that momentum can be imagined to be like an invis-
ible fluid. In an interaction, the invisible fluid is transferred from one object to another.
However, the sum is always constant.
Momentum conservation implies that motion never stops; it is only exchanged. On
Copyright © Christoph Schiller November 1997–May 2006

the other hand, motion often ‘disappears’ in our environment, as in the case of a stone
dropped to the ground, or of a ball left rolling on grass. Moreover, in daily life we often ob-
serve the creation of motion, such as every time we open a hand. How do these examples
fit with the conservation of momentum?
* For more curiosities, see R.H. Price, Negative mass can be positively amusing, American Journal of Phys-
ics 61, pp. 216–217, 1993. Negative mass particles in a box would heat up a box made of positive mass while
Page 83 traversing its walls, and accelerating, i.e. losing energy, at the same time. They would allow one to build a
Challenge 127 e perpetuum mobile of the second kind, i.e. a device circumventing the second principle of thermodynamics.
Moreover, such a system would have no thermodynamic equilibrium, because its energy could decrease
forever. The more one thinks about negative mass, the more one finds strange properties contradicting ob-
Challenge 128 n servations. By the way, what is the range of possible mass values for tachyons?
** Arthur Eddington (1882–1944), British astrophysicist.

Dvipsbugw
 objects and images 83

It turns out that the answer lies in the

microscopic aspects of these systems. A
muscle only transforms one type of mo-
tion, namely that of the electrons in cer-
tain chemical compounds* into another,
the motion of the fingers. The working of
muscles is similar to that of a car engine wine
transforming the motion of electrons in
the fuel into motion of the wheels. Both Dvipsbugw
systems need fuel and get warm in the pro- cork
cess.
We must also study the microscopic be-
haviour when a ball rolls on grass until
it stops. The disappearance of motion is wine
called friction. Studying the situation care-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fully, one finds that the grass and the ball
heat up a little during this process. Dur- stone
ing friction, visible motion is transformed
into heat. Later, when we discover the F I G U R E 33 What happens?
structure of matter, it will become clear
that heat is the disorganized motion of the microscopic constituents of every material.
When these constituents all move in the same direction, the object as a whole moves;
when they oscillate randomly, the object is at rest, but is warm. Heat is a form of motion.
Friction thus only seems to be disappearance of motion; in fact it is a transformation of
ordered into unordered motion.
Despite momentum conservation, macroscopic perpetual motion does not exist, since
friction cannot be completely eliminated.** Motion is eternal only at the microscopic
scale. In other words, the disappearance and also the spontaneous appearance of motion
in everyday life is an illusion due to the limitations of our senses. For example, the mo-
tion proper of every living being exists before its birth, and stays after its death. The same
happens with its energy. This result is probably the closest one can get to the idea of ever-

Ref. 57 * Usually adenosine triphosphate (ATP), the fuel of most processes in animals.
** Some funny examples of past attempts to built a perpetual motion machine are described in Stanislav
Michel, Perpetuum mobile, VDI Verlag, 1976. Interestingly, the idea of eternal motion came to Europe
Copyright © Christoph Schiller November 1997–May 2006

from India, via the Islamic world, around the year 1200, and became popular as it opposed the then standard
view that all motion on Earth disappears over time. See also the http://www.geocities.com/mercutio78_99/
pmm.html and the http://www.lhup.edu/~dsimanek/museum/unwork.htm websites. The conceptual mis-
take made by eccentrics and used by crooks is always the same: the hope of overcoming friction. (In fact,
this applied only to the perpetual motion machines of the second kind; those of the first kind – which are
even more in contrast with observation – even try to generate energy from nothing.)
If the machine is well constructed, i.e. with little friction, it can take the little energy it needs for the
sustenance of its motion from very subtle environmental effects. For example, in the Victoria and Albert
Ref. 58 Museum in London one can admire a beautiful clock powered by the variations of air pressure over time.
Low friction means that motion takes a long time to stop. One immediately thinks of the motion of the
Challenge 131 n planets. In fact, there is friction between the Earth and the Sun. (Can you guess one of the mechanisms?)
But the value is so small that the Earth has already circled around the Sun for thousands of millions of years,
and will do so for quite some time more.

Dvipsbugw
 84 i galilean motion • 2. galilean physics – motion in everyday life

lasting life from evidence collected by observation. It is perhaps less than a coincidence
that energy used to be called vis viva, or ‘living force’, by Leibniz and many others.
Since motion is conserved, it has no origin. Therefore, at this stage of our walk we
cannot answer the fundamental questions: Why does motion exist? What is its origin?
The end of our adventure is nowhere near.

More on conservation – energy

When collisions are studied in detail, a second conserved quantity turns up. Experiments Dvipsbugw
show that in the case of perfect, or elastic collisions – collisions without friction – the
following quantity, called the kinetic energy T of the system, is also conserved:

T = Q 21 m i v 2i = Q 21 m i v 2i = const . (17)
i i

Kinetic energy thus depends on the mass and on the square of the speed v of a body. Kin-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

etic energy is the ability that a body has to induce change in bodies it hits. The full name
‘kinetic energy’ was introduced by Gustave-Gaspard Coriolis.* Coriolis also introduced
Challenge 132 n the factor 1/2, in order that the relation dT~dv = p would be obeyed. (Why?) Energy is a
word taken from ancient Greek; originally it was used to describe character, and meant
‘intellectual or moral vigour’. It was taken into physics by Thomas Young (1773–1829) in
1807 because its literal meaning is ‘force within’. (The letters E, W, A and several others are
also used to denote energy.) Another, equivalent definition of energy will become clear
later: energy is what can be transformed into heat.
(Physical) energy is the measure of the ability to generate motion. A body has a lot
of energy if it has the ability to move many other bodies. Energy is a number; it has no
direction. The total momentum of two equal masses moving with opposite velocities is
zero; their total energy increases with velocity. Energy thus also measures motion, but
in a different way than momentum. Energy measures motion in a more global way. An
equivalent definition is the following. Energy is the ability to perform work. Here, the
physical concept of work is just the precise version of what is meant by work in everyday
life.**
Do not be surprised if you do not grasp the difference between momentum and energy
straight away: physicists took about two centuries to figure it out. For some time they even
insisted on using the same word for both, and often they didn’t know which situation
required which concept. So you are allowed to take a few minutes to get used to it.
Copyright © Christoph Schiller November 1997–May 2006

Both energy and momentum measure how systems change. Momentum tells how sys-
tems change over distance, energy measures how systems change over time. Momentum
is needed to compare motion here and there. Energy is needed to compare motion now
and later. Some measured energy values are given in Table 14.
One way to express the difference between energy and momentum is to think about
the following challenges. Is it more difficult to stop a running man with mass º m and speed
Challenge 133 e v, or one with mass m~2 and speed 2v, or one with mass m~2 and speed 2 v? You may
want to ask a rugby-playing friend for confirmation.

* Gustave-Gaspard Coriolis (b. 1792 Paris, d. 1843 Paris), French engineer and mathematician.
** (Physical) work is the product of force and distance in direction of the force.

Dvipsbugw
 objects and images 85

TA B L E 14 Some measured energy values

O b s e r va t i o n Energy

Average kinetic energy of oxygen molecule in air 6 ë 10−21 J

Green photon energy 5.6 ë 10−20 J
X-ray photon energy 10−15 J
γ photon energy 10−12 J
Highest particle energy in accelerators 10−7 J
Comfortably walking human 20 J Dvipsbugw
Flying arrow 50 J
Right hook in boxing 50 J
Energy in torch battery 1 kJ
Flying rifle bullet 10 kJ
Apple digestion 0.2 MJ
Car on highway 1 MJ

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Highest laser pulse energy 1.8 MJ
Lightning flash up to 1 GJ
Planck energy 2.0 GJ
Small nuclear bomb (20 ktonne) 84 TJ
Earthquake of magnitude 7 2 PJ
Largest nuclear bomb (50 Mtonne) 210 PJ
Impact of meteorite with 2 km diameter 1 EJ
Yearly machine energy use 420 EJ
Rotation energy of Earth 2 ë 1029 J
Supernova explosion 1044 J
Gamma ray burst up to 1047 J
Energy content E = mc 2 of Sun’s mass 1.8 ë 1047 J
Energy content of Galaxy’s central black hole 4 ë 1053 J

Another distinction is illustrated by athletics: the real long jump

world record, almost 10 m, is still kept by an athlete who in the early
twentieth century ran with two weights in his hands, and then threw
Copyright © Christoph Schiller November 1997–May 2006

the weights behind him at the moment he took off. Can you explain
Challenge 134 n the feat?
When a car travelling at 100 m~s runs head-on into a parked car
Challenge 135 n of the same kind and make, which car receives the greatest damage?
What changes if the parked car has its brakes on?
To get a better feeling for energy, here is an additional approach. Robert Mayer
The world consumption of energy by human machines (coming from
solar, geothermal, biomass, wind, nuclear, hydro, gas, oil, coal, or animal sources) in the
Ref. 59 year 2000 was about 420 EJ,* for a world population of about 6000 million people. To see

Page 1154 * For the explanation of the abbreviation E, see Appendix B.

Dvipsbugw
 86 i galilean motion • 2. galilean physics – motion in everyday life

what this energy consumption means, we translate it into a personal power consumption;
we get about 2.2 kW. The watt W is the unit of power, and is simply defined as 1 W = 1 J~s,
reflecting the definition of (physical) power as energy used per unit time. As a working
person can produce mechanical work of about 100 W, the average human energy con-
sumption corresponds to about 22 humans working 24 hours a day. (See Table 15 for
some power values found in nature.) In particular, if we look at the energy consumption
in First World countries, the average inhabitant there has machines working for them
Challenge 136 n equivalent to several hundred ‘servants’. Can you point out some of these machines?
Kinetic energy is thus not conserved in everyday life. For example, in non-elastic colli- Dvipsbugw
sions, such as that of a piece of chewing gum hitting a wall, kinetic energy is lost. Friction
destroys kinetic energy. At the same time, friction produces heat. It was one of the im-
portant conceptual discoveries of physics that total energy is conserved if one includes
the discovery that heat is a form of energy. Friction is thus in fact a process transforming
kinetic energy, i.e. the energy connected with the motion of a body, into heat. On a mi-
croscopic scale, energy is conserved.* Indeed, without energy conservation, the concept

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of time would not be definable. We will show this connection shortly.

Is velocity absolute? – The theory of everyday relativity

Why don’t we feel all the motions of the Earth? The two parts of the answer were already
given in 1632. First of all, as Galileo explained, we do not feel the accelerations of the
Earth because the effects they produce are too small to be detected by our senses. Indeed,
many of the mentioned accelerations do induce measurable effects in high-precision ex-
Page 393 periments, e.g. in atomic clocks.
But the second point made by Galileo is equally important. We do not feel translational,
unaccelerated motions because this is impossible in principle. We cannot feel that we are
moving! Galileo discussed the issue by comparing the observations of two observers: one
on the ground and another on the most modern means of transportation of the time, a
ship. Galileo asked whether a man on the ground and a man in a ship moving at constant
speed experience (or ‘feel’) anything different. Einstein used observers in trains. Later it
Challenge 137 e became fashionable to use travellers in rockets. (What will come next?) Galileo explained
that only relative velocities between bodies produce effects, not the absolute values of the
velocities. For the senses, there is no difference between constant, undisturbed motion,
however rapid it may be, and rest. This is now called Galileo’s principle of relativity. In
everyday life we feel motion only if the means of transportation trembles (thus if it ac-
Copyright © Christoph Schiller November 1997–May 2006

celerates), or if we move against the air. Therefore Galileo concludes that two observers
in straight and undisturbed motion against each other cannot say who is ‘really’ moving.
Whatever their relative speed, neither of them ‘feels’ in motion.**

* In fact, the conservation of energy was stated in its full generality in public only in 1842, by Julius Robert
Mayer. He was a medical doctor by training, and the journal Annalen der Physik refused to publish his paper,
as it supposedly contained ‘fundamental errors’. What the editors called errors were in fact mostly – but not
only – contradictions of their prejudices. Later on, Helmholtz, Kelvin, Joule and many others acknowledged
Mayer’s genius. However, the first to have stated energy conservation in its modern form was the French
physicist Sadi Carnot (1796–1832) in 1820. To him the issue was so clear that he did not publish the result.
In fact he went on and discovered the second ‘law’ of thermodynamics. Today, energy conservation, also
called the first ‘law’ of thermodynamics, is one of the pillars of physics, as it is valid in all its domains.
** In 1632, in his Dialogo, Galileo writes: ‘Shut yourself up with some friend in the main cabin below decks

Dvipsbugw
objects and images 87

TA B L E 15 Some measured power values

O b s e r va t i o n Power

Power of flagellar motor in bacterium 0.1 pW

Incandescent light bulb light output 1 to 5 W
Incandescent light bulb electricity consumption 25 to 100 W
A human, during one work shift of eight hours 100 W
One horse, for one shift of eight hours 300 W Dvipsbugw
Eddy Merckx, the great bicycle athlete, during one hour 500 W
Official horse power power unit 735 W
Large motorbike 100 kW
Electrical power station output 0.1 to 6 GW
World’s electrical power production in 2000 450 GW
Power used by the geodynamo 200 to 500 GW

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Input on Earth surface: Sun’s irradiation of Earth Ref. 60 0.17 EW
Input on Earth surface: thermal energy from inside of the Earth 32 TW
Input on Earth surface: power from tides (i.e. from Earth’s rotation) 3 TW
Input on Earth surface: power generated by man from fossil fuels 8 to 11 TW
Lost from Earth surface: power stored by plants’ photosynthesis 40 TW
World’s record laser power 1 PW
Output of Earth surface: sunlight reflected into space 0.06 EW
Output of Earth surface: power radiated into space at 287 K 0.11 EW
Sun’s output 384.6 YW
Maximum power in nature, c 5 ~4G 9.1 ë 1051 W

on some large ship, and have with you there some flies, butterflies, and other small flying animals. Have
a large bowl of water with some fish in it; hang up a bottle that empties drop by drop into a wide vessel
beneath it. With the ship standing still, observe carefully how the little animals fly with equal speed to all
sides of the cabin. The fish swim indifferently in all directions; the drops fall into the vessel beneath; and,
in throwing something to your friend, you need throw it no more strongly in one direction than another,
the distances being equal: jumping with your feet together, you pass equal spaces in every direction. When
you have observed all these things carefully (though there is no doubt that when the ship is standing still
Copyright © Christoph Schiller November 1997–May 2006

everything must happen in this way), have the ship proceed with any speed you like, so long as the motion is
uniform and not fluctuating this way and that, you will discover not the least change in all the effects named,
nor could you tell from any of them whether the ship was moving or standing still. In jumping, you will pass
on the floor the same spaces as before, nor will you make larger jumps toward the stern than toward the prow
even though the ship is moving quite rapidly, despite the fact that during the time you are in the air the floor
under you will be going in a direction opposite to your jump. In throwing something to your companion,
you will need no more force to get it to him whether he is in the direction of the bow or the stern, with
yourself situated opposite. The droplets will fall as before into the vessel beneath without dropping toward
the stern, although while the drops are in the air the ship runs many spans. The fish in their water will swim
toward the front of their bowl with no more effort than toward the back, and will go with equal ease to bait
placed anywhere around the edges of the bowl. Finally the butterflies and flies will continue their flights
indifferently toward every side, nor will it ever happen that they are concentrated toward the stern, as if
tired out from keeping up with the course of the ship, from which they will have been separated during long

Dvipsbugw
 88 i galilean motion • 2. galilean physics – motion in everyday life

Rest is relative. Or more clearly: rest is an observer-dependent concept. This result

of Galilean physics is so important that Poincaré introduced the expression ‘theory of
relativity’ and Einstein repeated the principle explicitly when he published his famous
theory of special relativity. However, these names are awkward. Galilean physics is also
a theory of relativity! The relativity of rest is common to all of physics; it is an essential
aspect of motion.
Undisturbed or uniform motion has no observable effect; only change of motion does.
As a result, every physicist can deduce something simple about the following statement
by Wittgenstein: Dvipsbugw

Daß die Sonne morgen aufgehen wird, ist eine Hypothese; und das heißt:
wir wissen nicht, ob sie aufgehen wird.*

The statement is wrong. Can you explain why Wittgenstein erred here, despite his strong
Challenge 138 n desire not to?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Rotation
Rotation keeps us alive. Without the change of day and night, we would be either fried or
frozen to death, depending on our location on our planet. A short summary of rotation
is thus appropriate. We saw before that a body is described by its reluctance to move;
similarly, a body also has a reluctance to turn. This quantity is called its moment of inertia
and is often abbreviated Θ. The speed or rate of rotation is described by angular velocity,
usually abbreviated ω. A few values found in nature are given in Table 16.
The observables that describe rotation are similar to those describing linear motion,
as shown in Table 17. Like mass, the moment of inertia is defined in such a way that the
sum of angular momenta L – the product of moment of inertia and angular velocity – is
conserved in systems that do not interact with the outside world:

Q Θ i ω i = Q L i = const . (18)
i i

In the same way that linear momentum conservation defines mass, angular momentum
conservation defines the moment of inertia.
The moment of inertia can be related to the mass and shape of a body. If the body is
imagined to consist of small parts or mass elements, the resulting expression is
Copyright © Christoph Schiller November 1997–May 2006

Θ = Q m n r n2 , (19)
n

intervals by keeping themselves in the air. And if smoke is made by burning some incense, it will be seen
going up in the form of a little cloud, remaining still and moving no more toward one side than the other.
The cause of all these correspondences of effects is the fact that the ship’s motion is common to all the things
contained in it, and to the air also. That is why I said you should be below decks; for if this took place above
in the open air, which would not follow the course of the ship, more or less noticeable differences would be
seen in some of the effects noted.’
* ‘It is a hypothesis that the Sun will rise tomorrow; and this means that we do not know whether it will rise.’
This well-known statement is found in Ludwig Wittgenstein, Tractatus, 6.36311.

Dvipsbugw
 objects and images 89

TA B L E 16 Some measured rotation frequencies

O b s e r va t i o n Angular velocity
ω = 2π~T

Galactic rotation 2π ë 0.14 ë 10−15 ~ s = 2π ~220 ë 106 a

Average Sun rotation around its axis 2π ë3.8 ë 10−7 ~ s = 2π ~ 30 d
Typical lighthouse 2π ë 0.08~ s
Pirouetting ballet dancer 2π ë 3~ s
Ship’s diesel engine 2π ë 5~ s Dvipsbugw
Helicopter motor 2π ë 5.3~ s
Washing machine up to 2π ë 20~ s
Bacterial flagella 2π ë 100~ s
Racing car engine up to 2π ë 600~ s
Fastest turbine built 2π ë 103 ~ s
Fastest pulsars (rotating stars) up to at least 2π ë 716~ s

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ultracentrifuge A 2π ë 2 ë 103 ~ s
Dental drill up to 2π ë 13 ë 103 ~ s
Proton rotation 2π ë 1020 ~ s
Highest possible, Planck angular velocity 2πë 1035 ~ s

TA B L E 17 Correspondence between linear and rotational motion

Q ua n t i t y Linear motion R o tat i o n

State time t time t

position x angle φ
momentum p = mv angular momentum L = Θω
energy mv 2 ~2 energy Θω 2 ~2
Motion velocity v angular velocity ω
acceleration a angular acceleration α
Reluctance to move mass m moment of inertia Θ
Motion change force ma torque Θα
Copyright © Christoph Schiller November 1997–May 2006

where r n is the distance from the mass element m n to the axis of rotation. Can you con-
Challenge 139 e firm the expression? Therefore, the moment of inertia of a body depends on the chosen
Challenge 140 n axis of rotation. Can you confirm that this is so for a brick?
Obviously, the value of the moment of inertia also depends on the location of the axis
used for its definition. For each axis direction, one distinguishes an intrinsic moment of
inertia, when the axis passes through the centre of mass of the body, from an extrinsic
moment of inertia, when it does not.* In the same way, one distinguishes intrinsic and
* Extrinsic and intrinsic moment of inertia are related by
2
Θ ext = Θ int + md , (20)
where d is the distance between the centre of mass and the axis of extrinsic rotation. This relation is called

Dvipsbugw
 90 i galilean motion • 2. galilean physics – motion in everyday life

middle finger: "r x p"

L fingers in
rotation
sense;
thumb
r shows
index: "p" angular
A momentum
thumb: "r"
p Dvipsbugw

F I G U R E 34 Angular momentum and the two versions of the right-hand rule

extrinsic angular momenta. (By the way, the centre of mass of a body is that imaginary
point which moves straight during vertical fall, even if the body is rotating. Can you find
a way to determine its location for a specific body?)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 142 n
Every object that has an orientation also has an intrinsic angular momentum. (What
Challenge 143 n about a sphere?) Therefore, point particles do not have intrinsic angular momenta – at
least in first approximation. (This conclusion will change in quantum theory.) The ex-
trinsic angular momentum L of a point particle is given by

L=rp= L=rp=
2A(T)m 2A(T)m
so that (21)
T T

where p is the momentum of the particle, A(T) is the surface swept by the position vector
r of the particle during time T.* The angular momentum thus points along the rotation
axis, following the right-hand rule, as shown in Figure 34.
We then define a corresponding rotational energy as

L2
E rot = 21 Θ ω 2 = . (23)
2Θ
The expression is similar to the expression for the kinetic energy of a particle. Can you

Challenge 141 n Steiner’s parallel axis theorem. Are you able to deduce it?
* For the curious, the result of the cross product or vector product ab between two vectors a and b is defined
Copyright © Christoph Schiller November 1997–May 2006

as that vector that is orthogonal to both, whose orientation is given by the right-hand rule, and whose length
is given by ab sin ∢(a, b), i.e. by the surface area of the parallelogram spanned by the two vectors. From the
Challenge 144 e definition you can show that the vector product has the properties

a  b = −b  a , a  (b + c) = a  b + a  c , λa  b = λ(a  b) = a  λb , a  a = 0 ,
a(b  c) = b(c  a) = c(a  b) , a  (b  c) = b(ac) − c(ab) ,
(a  b)(c  d) = a(b  (c  d)) = (ac)(bd) − (bc)(ad) ,
(a  b)  (c  d) = c((a  b)d) − d((a  b)c) , a  (b  c) + b  (c  a) + c  (a  b) = 0 . (22)

Page 1203 The vector product exists (almost) only in three-dimensional vector spaces. (See Appendix D.) The cross
product vanishes if and only if the vectors are parallel. The parallelepiped spanned by three vectors a, b and
Challenge 145 e c has the volume V = c(a  b). The pyramid or tetrahedron formed by the three vectors has one sixth of
that volume.

Dvipsbugw
 objects and images 91

frictionless
axis

Dvipsbugw
F I G U R E 35 How a snake turns itself around its axis F I G U R E 36 Can the ape reach
the banana?

guess how much larger the rotational energy of the Earth is compared with the yearly

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 146 n electricity usage of humanity? In fact, if you can find a way to harness this energy, you
will become famous.
As in the case of linear motion, rotational energy and angular momentum are not
always conserved in the macroscopic world: rotational energy can change due to friction,
and angular momentum can change due to external forces (torques). However, for closed
systems both quantities always conserved on the microscopic scale.
On a frictionless surface, as approximated by smooth ice or by a marble floor covered
by a layer of oil, it is impossible to move forward. In order to move, we need to push
against something. Is this also the case for rotation?
Surprisingly, it is possible to turn even without pushing against something. You can
check this on a well-oiled rotating office chair: simply rotate an arm above the head. After
each turn of the hand, the orientation of the chair has changed by a small amount. Indeed,
conservation of angular momentum and of rotational energy do not prevent bodies from
changing their orientation. Cats learn this in their youth. After they have learned the
trick, if they are dropped legs up, they can turn themselves in such a way that they always
Ref. 61 land feet first. Snakes also know how to rotate themselves, as Figure 35 shows. During
the Olympic Games one can watch board divers and gymnasts perform similar tricks.
Challenge 147 d Rotation is thus different from translation in this aspect. (Why?)
Angular momentum is conserved. This statement is valid for any axis, provided that
friction plays no role. external forces (torques) play no role. To make the point, Jean-Marc
Copyright © Christoph Schiller November 1997–May 2006

Ref. 3 Lévy-Leblond poses the problem of Figure 36. Can the ape reach the banana without
leaving the plate, assuming that the plate on which the ape rests can turn around the axis
Challenge 148 n without friction?

Rolling wheels
Rotation is an interesting phenomenon in many ways. A rolling wheel does not turn
around its axis, but around its point of contact. Let us show this.
A wheel of radius R is rolling if the speed of the axis v axis is related to the angular
velocity ω by
ω=
v axis
. (24)
R

Dvipsbugw
 92 i galilean motion • 2. galilean physics – motion in everyday life

ωr ω d = vp
P ωR

r ω R = vaxis

d
R

C
Dvipsbugw
F I G U R E 37 The velocities and unit F I G U R E 38 A simulated
vectors for a rolling wheel photograph of a rolling wheel with
spokes

For any point P on the wheel, with distance r from the axis, the velocity v P is the sum

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of the motion of the axis and the motion around the axis. Figure 37 shows that v P is
orthogonal to d, the distance between the point P and the contact point of the wheel. The
Challenge 149 e figure also shows that the length ratio between v P and d is the same as between v axis and
R. As a result, we can write
vP = ω  d , (25)

which shows that a rolling wheel does indeed rotate about its point of contact with the
ground.
Surprisingly, when a wheel rolls, some points on it move towards the wheel’s axis, some
stay at a fixed distance and others move away from it. Can you determine where these
Challenge 150 n various points are located? Together, they lead to an interesting pattern when a rolling
Ref. 62 wheel with spokes, such as a bicycle wheel, is photographed.
Ref. 63 With these results you can tackle the following beautiful challenge. When a turning
bicycle wheel is put on a slippery surface, it will slip for a while and then end up rolling.
Challenge 151 d How does the final speed depend on the initial speed and on the friction?

How do we walk?
Golf is a good walk spoiled.

“
Why do we move our arms when walking or running? To conserve energy. In fact, when
a body movement is performed with as little energy as possible, it is natural and graceful.
Mark Twain

”
Copyright © Christoph Schiller November 1997–May 2006

(This can indeed be taken as the actual definition of grace. The connection is common
Ref. 17 knowledge in the world of dance; it is also a central aspect of the methods used by actors
to learn how to move their bodies as beautifully as possible.)
To convince yourself about the energy savings, try walking or running with your arms
fixed or moving in the opposite direction to usual: the effort required is considerably
higher. In fact, when a leg is moved, it produces a torque around the body axis which
has to be counterbalanced. The method using the least energy is the swinging of arms.
Since the arms are lighter than the legs, they must move further from the axis of the body,
to compensate for the momentum; evolution has therefore moved the attachment of the

Dvipsbugw
 objects and images 93

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 39 The measured motion of a walking human (© Ray McCoy)

rotating Moon sky

Earth or and
Sun stars

F I G U R E 40 The parallaxis – not drawn to scale

arms, the shoulders, farther apart than those of the legs, the hips. Animals on two legs but
Copyright © Christoph Schiller November 1997–May 2006

no arms, such as penguins or pigeons, have more difficulty walking; they have to move
their whole torso with every step.
Which muscles do most of the work when walking, the motion that experts call gait? In
Ref. 64 1980, Serge Gracovetsky found that in human gait most power comes from the muscles
along the spine, not from the legs. (Indeed, people without legs are also able to walk.)
When you take a step, the lumbar muscles straighten the spine; this automatically makes it
turn a bit to one side, so that the knee of the leg on that side automatically comes forward.
When the foot is moved, the lumbar muscles can relax, and then straighten again for the
next step. In fact, one can experience the increase in tension in the back muscles when
Challenge 152 e walking without moving the arms, thus confirming where the human engine is located.
Human legs differ from those of apes in a fundamental aspect: humans are able to run.

Dvipsbugw
 94 i galilean motion • 2. galilean physics – motion in everyday life

In fact the whole human body has been optimized for running, an ability that no other
primate has. The human body has shed most of its hair to achieve better cooling, has
evolved the ability to run while keeping the head stable, has evolved the right length of
arms for proper balance when running, and even has a special ligament in the back that
works as a shock absorber while running. In other words, running is the most human of
all forms of motion.

Is the Earth rotating?

Dvipsbugw
Eppur si muove!

“
The search for answers to this question gives a beautiful cross section of the history of
Anonymous*

”
classical physics. Around the year 265 bce, the Greek thinker Aristarchos of Samos main-
Ref. 65 tained that the Earth rotates. He had measured the parallax of the Moon (today known to
be up to 0.95°) and of the Sun (today known to be 8.8 ′ ).** The parallax is an interesting

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

effect; it is the angle describing the difference between the directions of a body in the sky
when seen by an observer on the surface of the Earth and when seen by a hypothetical
observer at the Earth’s centre. (See Figure 40.) Aristarchos noticed that the Moon and the
Sun wobble across the sky, and this wobble has a period of 24 hours. He concluded that
the Earth rotates.
Page 276 Measurements of the aberration of light
also show the rotation of the Earth; it can
sphere 5 km
be detected with a telescope while looking at
the stars. The aberration is a change of the ex-
pected light direction, which we will discuss Earth
shortly. At the Equator, Earth rotation adds 5 km
5 km
an angular deviation of 0.32 ′ , changing sign Equator
every 12 hours, to the aberration due to the
motion of the Earth around the Sun, about
20.5 ′ . In modern times, astronomers have
found a number of additional proofs, but 5 km
none is accessible to the man on the street. F I G U R E 41 Earth’s deviation from spherical
Furthermore, the measurements showing shape due to its rotation
that the Earth is not a sphere, but is flattened
at the poles, confirmed the rotation of the Earth. Again, however, this eighteenth century
Copyright © Christoph Schiller November 1997–May 2006

measurement by Maupertuis*** is not accessible to everyday observation.

Then, in the years 1790 to 1792 in Bologna, Giovanni Battista Guglielmini (1763–1817)
finally succeeded in measuring what Galileo and Newton had predicted to be the simplest
proof for the Earth’s rotation. On the Earth, objects do not fall vertically, but are slightly
* ‘And yet she moves’ is the sentence falsely attributed to Galileo about the Earth. It is true, however, that at
his trial he was forced to publicly retract the idea of a moving Earth to save his life (see the footnote on page
220).
** For the definition of angles see page 61 and for the definition of angle units see Appendix B.
*** Pierre Louis Moreau de Maupertuis (1698–1759), French physicist and mathematician. He was one of
the key figures in the quest for the principle of least action, which he named in this way. He was also founding
president of the Berlin Academy of Sciences.

Dvipsbugw
 objects and images 95

v = ω (R+h)
h N h

v=ω R
ϕ

N
Equator
E

Dvipsbugw
S

F I G U R E 42 The deviations of free fall towards the east and

towards the Equator due to the rotation of the Earth

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

deviated to the east. This deviation appears because an object keeps the larger hori-
zontal velocity it had at the height from which it started falling, as shown in Figure 42.
Guglielmini’s result was the first non-astronomical proof of the Earth’s rotation. The ex-
periments were repeated in 1802 by Johann Friedrich Benzenberg (1777–1846). Using
metal balls which he dropped from the Michaelis tower in Hamburg – a height of 76 m
– Benzenberg found that the deviation to the east was 9.6 mm. Can you confirm that the
value measured by Benzenberg almost agrees with the assumption that the Earth turns
Challenge 153 ny once every 24 hours? (There is also a much smaller deviation towards the Equator, not
measured by Guglielmini, Benzenberg or anybody after them up to this day; however,
it completes the list of effects on free fall by the rotation of the Earth.) Both deviations
are easily understood if we remember that falling objects describe an ellipse around the
centre of the rotating Earth. The elliptical shape shows that the path of a thrown stone
does not lie on a plane for an observer standing on Earth; for such an observer, the exact
path thus cannot be drawn on a piece of paper.
In 1835, the engineer and mathematician Gustave-Gaspard Coriolis (1792–1843), the
Frenchman who also introduced the modern concepts of ‘work’ and of ‘kinetic energy’,
found a closely related effect that nobody had as yet noticed in everyday life. An object
travelling in a rotating background does not move in a straight line. If the rotation is
anticlockwise, as is the case for the Earth on the northern hemisphere, the velocity of an
object is slightly turned to the right, while its magnitude stays constant. This so-called
Copyright © Christoph Schiller November 1997–May 2006

Coriolis acceleration (or Coriolis force) is due to the change of distance to the rotation
Challenge 154 ny axis. Can you deduce the analytical expression for it, namely aC = 2ω  v?
The Coriolis acceleration determines the handedness of many large-scale phenomena
with a spiral shape, such as the directions of cyclones and anticyclones in meteorology,
the general wind patterns on Earth and the deflection of ocean currents and tides. Most
beautifully, the Coriolis acceleration explains why icebergs do not follow the direction
Ref. 66 of the wind as they drift away from the polar caps. The Coriolis acceleration also plays
a role in the flight of cannon balls (that was the original interest of Coriolis), in satellite
Ref. 67 launches, in the motion of sunspots and even in the motion of electrons in molecules. All
these phenomena are of opposite sign on the northern and southern hemispheres and
thus prove the rotation of the Earth. (In the First World War, many naval guns missed

Dvipsbugw
 96 i galilean motion • 2. galilean physics – motion in everyday life

Earth's
centre
ϕ
Eq
u ato
r
ψ1
ψ2 Dvipsbugw

F I G U R E 43 The turning motion of a pendulum showing the

rotation of the Earth

their targets in the southern hemisphere because the engineers had compensated them

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

for the Coriolis effect in the northern hemisphere.)
Ref. 68 Only in 1962, after several earlier attempts by other researchers, Asher Shapiro was the
first to verify that the Coriolis effect has a tiny influence on the direction of the vortex
formed by the water flowing out of a bathtub. Instead of a normal bathtub, he had to use a
carefully designed experimental set-up because, contrary to an often-heard assertion, no
such effect can be seen in a real bathtub. He succeeded only by carefully eliminating all
disturbances from the system; for example, he waited 24 hours after the filling of the reser-
voir (and never actually stepped in or out of it!) in order to avoid any left-over motion of
water that would disturb the effect, and built a carefully designed, completely rotationally-
symmetric opening mechanism. Others have repeated the experiment in the southern
Ref. 68 hemisphere, confirming the result. In other words, the handedness of usual bathtub vor-
tices is not caused by the rotation of the Earth, but results from the way the water starts
to flow out. But let us go on with the story about the Earth’s rotation.
Finally, in 1851, the French physician-turned-physicist Jean Bernard Léon Foucault
(b. 1819 Paris, d. 1868 Paris) performed an experiment that removed all doubts and
rendered him world-famous practically overnight. He suspended a 67 m long pendulum*
in the Panthéon in Paris and showed the astonished public that the direction of its swing
changed over time, rotating slowly. To anybody with a few minutes of patience to watch
the change of direction, the experiment proved that the Earth rotates. If the Earth did not
rotate, the swing of the pendulum would always continue in the same direction. On a ro-
tating Earth, the direction changes to the left, as shown in Figure 43, unless the pendulum
Copyright © Christoph Schiller November 1997–May 2006

is located at the Equator.** The time over which the orientation of the swing performs
a full turn – the precession time – can be calculated. Study a pendulum swinging in the
Challenge 156 d North–South direction and you will find that the precession time TFoucault is given by

Challenge 155 d * Why was such a long pendulum necessary? Understanding the reasons allows one to repeat the experiment
Ref. 69 at home, using a pendulum as short as 70 cm, with the help of a few tricks.
** The discovery also shows how precision and genius go together. In fact, the first person to observe the
effect was Vincenzo Viviani, a student of Galilei, as early as 1661! Indeed, Foucault had read about Viviani’s
work in the publications of the Academia dei Lincei. But it took Foucault’s genius to connect the effect to
the rotation of the Earth; nobody had done so before him.

Dvipsbugw
 objects and images 97

r
E

m m W
Dvipsbugw
S

F I G U R E 45 Showing the rotation of the F I G U R E 46 Demonstrating the rotation

Earth through the rotation of an axis of the Earth with water

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

TFoucault =
24 h
(26)
sin φ

where φ is the latitude of the location of the pendulum, e.g. 0° at the Equator and 90° at
the North Pole. This formula is one of the most beautiful results of Galilean kinematics.*
Foucault was also the inventor and namer of the gyroscope. He
built the device, shown in Figure 44, in 1852, one year after his pen-
dulum. With it, he again demonstrated the rotation of the Earth.
Once a gyroscope rotates, the axis stays fixed in space – but only
when seen from distant stars or galaxies. (This is not the same as
Challenge 157 n talking about absolute space. Why?) For an observer on Earth, the
axis direction changes regularly with a period of 24 hours. Gyro-
scopes are now routinely used in ships and in aeroplanes to give
the direction of north, because they are more precise and more re-
liable than magnetic compasses. In the most modern versions, one
F I G U R E 44 The
uses laser light running in circles instead of rotating masses.**
gyroscope
In 1909, Roland von Eőtvős measured a simple effect: due to the
rotation of the Earth, the weight of an object depends on the direction in which it moves.
As a result, a balance in rotation around the vertical axis does not stay perfectly horizontal:
Copyright © Christoph Schiller November 1997–May 2006

Challenge 159 n the balance starts to oscillate slightly. Can you explain the origin of the effect?
Ref. 70 In 1910, John Hagen published the results of an even simpler experiment, proposed
by Louis Poinsot in 1851. Two masses are put on a horizontal bar that can turn around a
vertical axis, a so-called isotomeograph. If the two masses are slowly moved towards the
support, as shown in Figure 45, and if the friction is kept low enough, the bar rotates.
Obviously, this would not happen if the Earth were not rotating. Can you explain the
Challenge 160 n observation? This little-known effect is also useful for winning bets between physicists.

* The calculation of the period of Foucault’s pendulum assumed that the precession rate is constant during
a rotation. This is only an approximation (though usually a good one).
Challenge 158 n ** Can you guess how rotation is detected in this case?

Dvipsbugw
 98 i galilean motion • 2. galilean physics – motion in everyday life

In 1913, Arthur Compton showed that a closed tube filled with water and some small
Ref. 82 floating particles (or bubbles) can be used to show the rotation of the Earth. The device
is called a Compton tube or Compton wheel. Compton showed that when a horizontal
tube filled with water is rotated by 180°, something happens that allows one to prove that
the Earth rotates. The experiment, shown in Figure 46, even allows measurement of the
Challenge 161 d latitude of the point where the experiment is made. Can you guess what happens?
In 1925, Albert Michelson* and his collaborators in Illinois constructed a vacuum in-
terferometer with the incredible perimeter of 1.9 km. Interferometers produce bright and
Page 728 dark fringes of light; the position of the fringes depends on the speed at which the inter- Dvipsbugw
ferometers rotates. The fringe shift is due to an effect first measured in 1913 by the French
physicist Georges Sagnac: the rotation of a complete ring interferometer with angular
Challenge 162 n frequency (vector) Ω produces a fringe shift of angular phase ∆φ given by

∆φ =
8π Ω A
(27)
cλ

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where A is the area (vector) enclosed by the two interfering light rays, λ their wavelength
and c the speed of light. The effect is now called the Sagnac effect, even though it had
Ref. 71 already been predicted 20 years earlier by Oliver Lodge.** Michelson and his team found
a fringe shift with a period of 24 hours and of exactly the magnitude predicted by the
rotation of the Earth. Modern high precision versions use ring lasers with areas of only
a few square metres, but which are able to measure variations of the rotation rates of the
Earth of less than one part per million. Indeed, over the course of a year the length of a
day varies irregularly by a few milliseconds, mostly due to influences from the Sun or the
Ref. 72 Moon, due to weather changes and due to hot magma flows deep inside the Earth.*** But
also earthquakes, the el Ninño effect in the climate and the filling of large water dams have
effects on the rotation of the Earth. All these effects can be studied with such precision
interferometers; these can also be used for research into the motion of the soil due to
lunar tides or earthquakes, and for checks on the theory of relativity.
In summary, observations show that the Earth surface rotates at 463 m~s at the Equator,
a larger value than that of the speed of sound in air – about 340 m~s in usual conditions
– and that we are in fact whirling through the universe.

How does the Earth rotate?

Is the rotation of the Earth constant over geological time scales? That is a hard question.
Copyright © Christoph Schiller November 1997–May 2006

If you find a method leading to an answer, publish it! (The same is true for the question
Ref. 73 whether the length of the year is constant.) Only a few methods are known, as we will
find out shortly.

* Albert Abraham Michelson (b. 1852 Strelno, d. 1931 Pasadena) Prussian–Polish–US-American physicist,
obsessed by the precise measurement of the speed of light, received the Nobel Prize for Physics in 1907.
** Oliver Lodge (1851–1940) was a British physicist who studied electromagnetic waves and tried to com-
municate with the dead. A strange but influential figure, his ideas are often cited when fun needs to be made
of physicists; for example, he was one of those (rare) physicists who believed that at the end of the nineteenth
century physics was complete.
*** The growth of leaves on trees and the consequent change in the Earth’s moment of inertia, already studied
in 1916 by Harold Jeffreys, is too small to be seen, so far.

Dvipsbugw
 objects and images 99

nutation period
year 15000: is 18.6 years year 2000:
North pole is North pole is
Vega in Polaris in the
Lyra Ursa minor
Dvipsbugw
precession

N
Moon’s
path
Moon

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

equatorial
bulge
Eq equatorial
uat bulge
or

th
Earth’s pa

F I G U R E 47 The precession and the nutation of the Earth’s axis

The rotation of the Earth is not even constant during a human lifespan. It varies by a
few parts in 108 . In particular, on a ‘secular’ time scale, the length of the day increases by
about 1 to 2 ms per century, mainly because of the friction by the Moon and the melting
of the polar ice caps. This was deduced by studying historical astronomical observations
Ref. 74 of the ancient Babylonian and Arab astronomers. Additional ‘decadic’ changes have an
amplitude of 4 or 5 ms and are due to the motion of the liquid part of the Earth’s core.
The seasonal and biannual changes of the length of the day – with an amplitude of
Copyright © Christoph Schiller November 1997–May 2006

0.4 ms over six months, another 0.5 ms over the year, and 0.08 ms over 24 to 26 months –
are mainly due to the effects of the atmosphere. In the 1950s the availability of precision
measurements showed that there is even a 14 and 28 day period with an amplitude of
0.2 ms, due to the Moon. In the 1970s, when wind oscillations with a length scale of about
50 days were discovered, they were also found to alter the length of the day, with an
amplitude of about 0.25 ms. However, these last variations are quite irregular.
But why does the Earth rotate at all? The rotation derives from the rotating gas cloud
at the origin of the solar system. This connection explains that the Sun and all planets,
except one, turn around themselves in the same direction, and that they also all turn
Ref. 75 around the Sun in that same direction. But the complete story is outside the scope of this
text.

Dvipsbugw
 100 i galilean motion • 2. galilean physics – motion in everyday life

The rotation around its axis is not the only motion of the Earth; it performs other
motions as well. This was already known long ago. In 128 bce, the Greek astronomer
Hipparchos discovered what is today called the (equinoctial) precession. He compared a
measurement he made himself with another made 169 years before. Hipparchos found
that the Earth’s axis points to different stars at different times. He concluded that the sky
was moving. Today we prefer to say that the axis of the Earth is moving. During a period
of 25 800 years the axis draws a cone with an opening angle of 23.5°. This motion, shown
in Figure 47, is generated by the tidal forces of the Moon and the Sun on the equatorial
bulge of the Earth that results form its flattening. The Sun and the Moon try to align the Dvipsbugw
axis of the Earth at right angles to the Earth’s path; this torque leads to the precession of
the Earth’s axis. (The same effect appears for any spinning top or in the experiment with
the suspended wheel shown on page 174.)
In addition, the axis of the Earth is not even fixed relative to the Earth’s surface. In 1884,
by measuring the exact angle above the horizon of the celestial North Pole, Friedrich Küst-
ner (1856–1936) found that the axis of the Earth moves with respect to the Earth’s crust, as

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Bessel had suggested 40 years earlier. As a consequence of Küstner’s discovery, the Inter-
national Latitude Service was created. The polar motion Küstner discovered turned out to
consist of three components: a small linear drift – not yet understood – a yearly elliptical
motion due to seasonal changes of the air and water masses, and a circular motion* with
a period of about 1.2 years due to fluctuations in the pressure at the bottom of the oceans.
In practice, the North Pole moves with an amplitude of 15 m around an average central
Ref. 76 position.
In 1912, the German meteorologist and geophysicist Alfred Wegener (1880–1930) dis-
covered an even larger effect. After studying the shapes of the continental shelves and the
geological layers on both sides of the Atlantic, he conjectured that the continents move,
and that they are all fragments of a single continent that broke up 200 million years ago.**
Even though at first derided across the world, Wegener’s discoveries were correct. Mod-
ern satellite measurements, shown in Figure 48, confirm this model. For example, the
American continent moves away from the European continent by about 10 mm every
year. There are also speculations that this velocity may have been much higher at certain
periods in the past. The way to check this is to look at the magnetization of sedimental
rocks. At present, this is still a hot topic of research. Following the modern version of the
model, called plate tectonics, the continents (with a density of 2.7 ë 103 kg~m3 ) float on the
Page 105, page 600 fluid mantle of the Earth (with a density of 3.1 ë 103 kg~m3 ) like pieces of cork on water,
Ref. 77 and the convection inside the mantle provides the driving mechanism for the motion.
Copyright © Christoph Schiller November 1997–May 2006

Does the Earth move?

The centre of the Earth is not at rest in the universe. In the third century bce Aristarchos
of Samos maintained that the Earth turns around the Sun. However, a fundamental diffi-
* The circular motion, a wobble, was predicted by the great Swiss mathematician Leonhard Euler (1707–
1783). Using this prediction and Küstner’s data, in 1891 Seth Chandler claimed to be the discoverer of the
circular component.
** In this old continent, called Gondwanaland, there was a huge river that flowed westwards from the Chad
to Guayaquil in Ecuador. After the continent split up, this river still flowed to the west. When the Andes
appeared, the water was blocked, and many millions of years later, it flowed back. Today, the river still flows
eastwards and is called the Amazonas.

Dvipsbugw
 objects and images 101

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 48 The continental plates are the objects of
tectonic motion

culty of the heliocentric system is that the stars look the same all year long. How can this
be, if the Earth travels around the Sun? The distance between the Earth and the Sun has
been known since the seventeenth century, but it was only in 1837 that Friedrich Wilhelm
Bessel* became the first to observe the parallax of a star. This was a result of extremely
careful measurements and complex calculations: he discovered the Bessel functions in
order to realize it. He was able to find a star, 61 Cygni, whose apparent position changed
with the month of the year. Seen over the whole year, the star describes a small ellipse in
the sky, with an opening of 0.588 ′′ (this is the modern value). After carefully eliminating
all other possible explanations, he deduced that the change of position was due to the
motion of the Earth around the Sun, and from the size of the ellipse he determined the
Challenge 163 n distance to the star to be 105 Pm, or 11.1 light years.
Bessel had thus managed for the first time to measure the distance of a star. By doing
so he also proved that the Earth is not fixed with respect to the stars in the sky and that the
Earth indeed revolves around the Sun. The motion itself was not a surprise. It confirmed
the result of the mentioned aberration of light, discovered in 1728 by James Bradley** and
Copyright © Christoph Schiller November 1997–May 2006

Page 276 to be discussed shortly; the Earth moves around the Sun.
With the improvement of telescopes, other motions of the Earth were discovered. In
1748, James Bradley announced that there is a small regular change of the precession,
which he called nutation, with a period of 18.6 years and an angular amplitude of 19.2 ′′ .

* Friedrich Wilhelm Bessel (1784–1846), Westphalian astronomer who left a successful business career to
dedicate his life to the stars, and became the foremost astronomer of his time.
** James Bradley, (1693–1762), English astronomer. He was one of the first astronomers to understand the
value of precise measurement, and thoroughly modernized Greenwich. He discovered the aberration of light,
a discovery that showed that the Earth moves and also allowed him to measure the speed of light; he also
discovered the nutation of the Earth.

Dvipsbugw
 102 i galilean motion • 2. galilean physics – motion in everyday life

precession ellipticity change

rotation axis

Earth
Sun

Sun

rotation tilt change

axis Dvipsbugw

Earth perihelion shift

Sun

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Earth
orbital inclination change
P
Sun
P
Sun

F I G U R E 49 Changes in the Earth’s motion around the Sun

Nutation occurs because the plane of the Moon’s orbit around the Earth is not exactly the
same as the plane of the Earth’s orbit around the Sun. Are you able to confirm that this
Challenge 164 ny situation produces nutation?
Astronomers also discovered that the 23.5° tilt – or obliquity – of the Earth’s axis, the
angle between its intrinsic and its orbital angular momentum, actually changes from 22.1°
to 24.5° with a period of 41 000 years. This motion is due to the attraction of the Sun and
the deviations of the Earth from a spherical shape. During the Second World War, in
1941, the Serbian astronomer Milutin Milankovitch (1879–1958) retreated into solitude
and studied the consequences. In his studies he realized that this 41 000 year period of
the tilt, together with an average period of 22 000 years due to precession,* gives rise to
the more than 20 ice ages in the last 2 million years. This happens through stronger or
Copyright © Christoph Schiller November 1997–May 2006

weaker irradiation of the poles by the Sun. The changing amounts of melted ice then lead
to changes in average temperature. The last ice age had is peak about 20 000 years ago and
ended around 11 800 years ago; the next is still far away. A spectacular confirmation of the
ice age cycles, in addition to the many geological proofs, came through measurements of
oxygen isotope ratios in sea sediments, which allow the average temperature over the past
Ref. 78 million years to be tracked.
The Earth’s orbit also changes its eccentricity with time, from completely circular to
slightly oval and back. However, this happens in very complex ways, not with periodic

* In fact, the 25 800 year precession leads to three insolation periods, of 23 700, 22 400 and 19 000 years, due
to the interaction between precession and perihelion shift.

Dvipsbugw
 objects and images 103

120 000 al = 1.2 Zm

our galaxy

orbit of our local star system

Dvipsbugw

500 al = 5 Em Sun's path

50 000 al = 500 Em

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 50 The motion of the Sun around the galaxy

regularity, and is due to the influence of the large planets of the solar system on the Earth’s
orbit. The typical time scale is 100 000 to 125 000 years.
In addition, the Earth’s orbit changes in inclination with respect to the orbits of the
other planets; this seems to happen regularly every 100 000 years. In this period the in-
clination changes from +2.5° to −2.5° and back.
Even the direction in which the ellipse points changes with time. This so-called peri-
helion shift is due in large part to the influence of the other planets; a small remaining
part will be important in the chapter on general relativity. It was the first piece of data
confirming the theory.
Obviously, the length of the year also changes with time. The measured variations
are of the order of a few parts in 1011 or about 1 ms per year. However, knowledge of
these changes and of their origins is much less detailed than for the changes in the Earth’s
rotation.
The next step is to ask whether the Sun itself moves. Indeed it does. Locally, it moves
with a speed of 19.4 km~s towards the constellation of Hercules. This was shown by Wil-
liam Herschel in 1783. But globally, the motion is even more interesting. The diameter of
the galaxy is at least 100 000 light years, and we are located 26 000 light years from the
Copyright © Christoph Schiller November 1997–May 2006

centre. (This has been known since 1918; the centre of the galaxy is located in the direc-
tion of Sagittarius.) At our position, the galaxy is 1 300 light years thick; presently, we are
Ref. 79 68 light years ‘above’ the centre plane. The Sun, and with it the solar system, takes about
225 million years to turn once around the galactic centre, its orbital velocity being around
220 km~s. It seems that the Sun will continue moving away from the galaxy plane until it
is about 250 light years above the plane, and then move back, as shown in Figure 50. The
oscillation period is estimated to be around 60 million years, and has been suggested as
the mechanism for the mass extinctions of animal life on Earth, possibly because some
gas cloud may be encountered on the way. The issue is still a hot topic of research.
We turn around the galaxy centre because the formation of galaxies, like that of solar
systems, always happens in a whirl. By the way, can you confirm from your own observa-

Dvipsbugw
 104 i galilean motion • 2. galilean physics – motion in everyday life

Challenge 165 n tion that our galaxy itself rotates?

Finally, we can ask whether the galaxy itself moves. Its motion can indeed be observed
because it is possible to give a value for the motion of the Sun through the universe, de-
fining it as the motion against the background radiation. This value has been measured
Ref. 80 to be 370 km~s. (The velocity of the Earth through the background radiation of course
depends on the season.) This value is a combination of the motion of the Sun around
the galaxy centre and of the motion of the galaxy itself. This latter motion is due to the
gravitational attraction of the other, nearby galaxies in our local group of galaxies.*
In summary, the Earth really moves, and it does so in rather complex ways. As Henri Dvipsbugw
Poincaré would say, if we are in a given spot today, say the Panthéon in Paris, and come
back to the same spot tomorrow at the the same time, we are in fact 31 million kilometres
away. This state of affairs would make time travel extremely difficult even if it were pos-
sible (which it is not); whenever you went back to the past, you would have to get to the
old spot exactly!

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Is rotation relative?
When we turn rapidly, our arms lift. Why does this happen? How can our body detect
whether we are rotating or not? There are two possible answers. The first approach, pro-
moted by Newton, is to say that there is an absolute space; whenever we rotate against this
space, the system reacts. The other answer is to note that whenever the arms lift, the stars
also rotate, and in exactly the same manner. In other words, our body detects rotation
because we move against the average mass distribution in space.
The most cited discussion of this question is due to Newton. Instead of arms, he ex-
plored the water in a rotating bucket. As usual for philosophical issues, Newton’s answer
was guided by the mysticism triggered by his father’s early death. Newton saw absolute
space as a religious concept and was not even able to conceive an alternative. Newton
thus sees rotation as an absolute concept. Most modern scientist have fewer problems
and more common sense than Newton; as a result, today’s consensus is that rotation ef-
fects are due to the mass distribution in the universe: rotation is relative. However, we
have to be honest; the question cannot be settled by Galilean physics. We will need gen-
eral relativity.

Curiosities and fun challenges about everyday motion

It is a mathematical fact that the casting of this

“
Copyright © Christoph Schiller November 1997–May 2006

pebble from my hand alters the centre of gravity

of the universe.
Thomas Carlyle,** Sartor Resartus III.

Here are a few facts to ponder about motion.

**
”
A car at a certain speed uses 7 litres of gasoline per 100 km. What is the combined air and

* This is roughly the end of the ladder. Note that the expansion of the universe, to be studied later, produces
no motion.
Challenge 166 n ** Thomas Carlyle (1797–1881). Do you agree with the quotation?

Dvipsbugw
 objects and images 105

F I G U R E 51 Is it safe to let the cork

go? Dvipsbugw

Challenge 167 n rolling resistance? (Assume that the engine has an efficiency of 25%.)
**
When travelling in the train, you can test Galileo’s statement about everyday relativity o

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fmotion. Close your eyes and ask somebody to turn you around many times: are you able
Challenge 168 e to say in which direction the train is running?
**
A train starts to travel at a constant speed of 10 m~s between two cities A and B, 36 km
apart. The train will take one hour for the journey. At the same time as the train, a fast
dove starts to fly from A to B, at 20 m~s. Being faster than the train, the dove arrives at
B first. The dove then flies back towards A; when it meets the train, it backturns again to
city B. It goes on flying back and forward until the train reaches B. What distance did the
Challenge 169 e dove cover?
**
A good bathroom scales, used to determine the weight of objects, does not show a con-
Challenge 170 n stant weight when you step on it and stay motionless. Why not?

**
A cork is attached to a thin string a metre long. The string is passed over a long rod
held horizontally, and a wine glass is attached at the other end. If you let go the cork in
Challenge 171 n Figure 51, nothing breaks. Why not? And what happens? Copyright © Christoph Schiller November 1997–May 2006

**
In 1901, Duncan MacDougalls, a medical doctor, measured the weight of dying people, in
the hope to see whether death leads to a mass change. He found a sudden change of about
20 g at the moment of death, with large variations from person to person. He attributed
Challenge 172 n it to the soul. Is this explanation satisfactory? (If you know a better one, publish it!)
**
The Earth’s crust is less dense (2.7 kg~l) than the Earth’s mantle (3.1 kg~l) and floats on it.
As a result, the lighter crust below a mountain ridge must be much deeper than below a
plain. If a mountain rises 1 km above the plain, how much deeper must the crust be below
Challenge 173 n it? The simple block model shown in Figure 52 works fairly well; first, it explains why, near

Dvipsbugw
 106 i galilean motion • 2. galilean physics – motion in everyday life

mountain
h
plain ocean
continental crust

d
mantle
magma

Dvipsbugw

F I G U R E 52 A simple model for continents and mountains

mountains, measurements of the deviation of free fall from the vertical line lead to so
much lower values than those expected without a deep crust. Later, sound measurements

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

have confirmed directly that the continental crust is indeed thicker beneath mountains.

**
Balance a pencil vertically (tip upwards!) on a piece of paper near the edge of a table. How
Challenge 174 e can you pull out the paper without letting the pencil fall?
**
Take a pile of coins. One can push out the coins, starting with the one at the bottom,
by shooting another coin over the table surface. The method also helps to visualize two-
Challenge 175 e dimensional momentum conservation.
**
In early 2004, two men and a woman earned £ 1.2 million in a single evening in a Lon-
don casino. They did so by applying the formulae of Galilean mechanics. They used the
method pioneered by various physicists in the 1950s who built various small computers
that could predict the outcome of a roulette ball from the initial velocity imparted by the
Ref. 81 croupier. In the case in Britain, the group added a laser scanner to a smart phone that
measured the path of a roulette ball and predicted the numbers where it would arrive.
In this way, they increased the odds from 1 in 37 to about 1 in 6. After six months of
investigations, Scotland Yard ruled that they could keep the money they won.
Copyright © Christoph Schiller November 1997–May 2006

**
Is a return flight by plane – from a point A to B and back to A – faster if the wind blows
Challenge 176 e or not?
**
The toy of Figure 53 shows interesting behaviour: when a number of spheres are lifted
and dropped to hit the resting ones, the same number of spheres detach on the other side,
whereas the previously dropped spheres remain motionless. At first sight, all this seems
to follow from energy and momentum conservation. However, energy and momentum
conservation provide only two equations, which are insufficient to explain or determine

Dvipsbugw
 objects and images 107

before the hit observed after the hit

F I G U R E 53 A well-known toy

Dvipsbugw
before the hit observed after the hit
V=0 v V‘ v’ 0

2L,2M L, M

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 54 An elastic collision that seems not to obey energy conservation

wall

ladder

F I G U R E 55 The centre of F I G U R E 56
mass defines stability How does the
ladder fall?

the behaviour of five spheres. Why then do the spheres behave in this way? And why do
Challenge 177 d they all swing in phase when a longer time has passed?
**
A surprising effect is used in home tools such as hammer drills. We remember that when a
Copyright © Christoph Schiller November 1997–May 2006

small ball elastically hits a large one at rest, both balls move after the hit, and the small one
Ref. 83 obviously moves faster than the large one. Despite this result, when a short cylinder hits a
long one of the same diameter and material, but with a length that is some integer multiple
of that of the short one, something strange happens. After the hit, the small cylinder
remains almost at rest, whereas the large one moves, as shown in Figure 54. Even though
the collision is elastic, conservation of energy seems not to hold in this case. (In fact this
is the reason that demonstrations of elastic collisions in schools are always performed
Challenge 178 d with spheres.) What happens to the energy?

**
Does a wall get a stronger jolt when it is hit by a ball rebounding from it or when it is hit

Dvipsbugw
 108 i galilean motion • 2. galilean physics – motion in everyday life

Challenge 179 n by a ball that remains stuck to it?

**
Housewives know how to extract a cork of a wine bottle using a cloth. Can you imagine
Challenge 180 n how? They also know how to extract the cork with the cloth if the cork has fallen inside
the bottle. How?
**
The sliding ladder problem, shown schematically in Figure 56, asks for the detailed mo- Dvipsbugw
tion of the ladder over time. The problem is more difficult than it looks, even if friction
Challenge 181 ny is not taken into account. Can you say whether the lower end always touches the floor?
**
Ref. 84 A common fly on the stern of a 30 000 ton ship of 100 m length tilts it by less than the
diameter of an atom. Today, distances that small are easily measured. Can you think of at

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 182 n least two methods, one of which should not cost more than 2000 euro?

**
The level of acceleration a human can survive depends on the duration over which one
is subjected to it. For a tenth of a second, 30 д = 300 m~s2 , as generated by an ejector
seat in an aeroplane, is acceptable. (It seems that the record acceleration a human has
survived is about 80 д = 800 m~s2 .) But as a rule of thumb it is said that accelerations of
15 д = 150 m~s2 or more are fatal.

**
The highest microscopic accelerations are observed in particle collisions, where one gets
values up to 1035 m~s2 . The highest macroscopic accelerations are probably found in the
collapsing interiors of supernovae, exploding stars which can be so bright as to be visible
in the sky even during the daytime. A candidate on Earth is the interior of collapsing
bubbles in liquids, a process called cavitation. Cavitation often produces light, an effect
discovered by Frenzel and Schulte in 1934 and called sonoluminescence. (See Figure 57.)
It appears most prominently when air bubbles in water are expanded and contracted by
underwater loudspeakers at around 30 kHz and allows precise measurements of bubble
motion. At a certain threshold intensity, the bubble radius changes at 1500 m~s in as little
Ref. 85 as a few µm, giving an acceleration of several 1011 m~s2 .
Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 183 n If a gun located at the Equator shoots a bullet vertically, where does the bullet fall?
**
Challenge 184 n Why are most rocket launch sites as near as possible to the Equator?
**
Would travelling through interplanetary space be healthy? People often fantasize about
long trips through the cosmos. Experiments have shown that on trips of long dura-
tion, cosmic radiation, bone weakening and muscle degeneration are the biggest dangers.

Dvipsbugw
 objects and images 109

no figure yet

Dvipsbugw
F I G U R E 57 Observation of sonoluminescence and a
diagram of the experimental set-up

Many medical experts question the viability of space travel lasting longer than a couple of
years. Other dangers are rapid sunburn, at least near the Sun, and exposure to the vacuum.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 86 So far only one man has experienced vacuum without protection. He lost consciousness
after 14 seconds, but survived unharmed.
**
Challenge 185 n How does the kinetic energy of a rifle bullet compare to that of a running man?
**
Challenge 186 n In which direction does a flame lean if it burns inside a jar on a rotating turntable?

**
A ping-pong ball is attached by a string to a stone, and the whole is put under water in
a jar. The set-up is shown in Figure 58. Now the jar is accelerated horizontally. In which
Challenge 187 n direction does the ball move? What do you deduce for a jar at rest?
**
Challenge 188 n What happens to the size of an egg when one places it in a jar of vinegar for a few days?
**
Does centrifugal acceleration exist? Most university students go through the shock of
meeting a teacher who says that it doesn’t because it is a ‘fictitious’ quantity, in the face
Copyright © Christoph Schiller November 1997–May 2006

of what one experiences every day in a car when driving around a bend. Simply ask the
teacher who denies it to define ‘existence’. (The definition physicists usually use is given
Page 677 in the Intermezzo following this chapter.) Then check whether the definition applies to
Challenge 189 n the term and make up your own mind.
**
Rotation holds a surprise for anybody who studies it carefully. Angular momentum is
a quantity with a magnitude and a direction. However, it is not a vector, as any mirror
shows. The angular momentum of a body circling in a plane parallel to a mirror behaves
in a different way from a usual arrow: its mirror image is not reflected if it points towards
Challenge 190 e the mirror! You can easily check this for yourself. For this reason, angular momentum

Dvipsbugw
 110 i galilean motion • 2. galilean physics – motion in everyday life

ping-pong ball
string
stone

F I G U R E 58 How does the ball F I G U R E 59 The famous

move when the jar is Celtic stone and a
accelerated? version made with a
spoon Dvipsbugw

is called a pseudovector. The fact has no important consequences in classical physics; but
we have to keep it in mind for later occasions.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
What is the best way to transport a number of full coffee or tea cups while at the same
Challenge 191 n time avoiding spilling any precious liquid?
**
The Moon recedes from the Earth by 3.8 cm a year, due to friction. Can you find the
Challenge 192 n mechanism responsible?

**
What is the amplitude of a pendulum oscillating in such a way that the absolute value of
Challenge 193 ny its acceleration at the lowest point and at the return point are equal?
**
Can you confirm that the value of the acceleration of a drop of water falling through
Challenge 194 ny vapour is д~7?
**
What are earthquakes? Earthquakes are large examples of the same process that make a
door squeak. The continental plates correspond to the metal surfaces in the joints of the
door.
Copyright © Christoph Schiller November 1997–May 2006

Earthquakes can be described as energy sources. The Richter scale is a direct measure
of this energy. The Richter magnitude M s of an earthquake, a pure number, is defined
from its energy E in joule via

log(E~1 J) − 4.8
Ms = . (28)
1.5
The strange numbers in the expression have been chosen to put the earthquake values as
near as possible to the older, qualitative Mercalli scale (now called EMS98) that classifies
the intensity of earthquakes. However, this is not fully possible; the most sensitive instru-
ments today detect earthquakes with magnitudes of −3. The highest value every meas-

Dvipsbugw
 objects and images 111

ured was a Richter magnitude of 10, in Chile in 1960. Magnitudes above 12 are probably
Challenge 195 n impossible. (Can you show why?)

**
Figure 59 shows the so-called Celtic wiggle stone, a stone that starts rotating on a plane
Ref. 83 surface when it is put into oscillation. The size can vary between a few centimetres and a
few metres. By simply bending a spoon one can realize a primitive form of this strange
device, if the bend is not completely symmetrical. The rotation is always in the same
direction. If the stone is put into rotation in the wrong direction, after a while it stops and Dvipsbugw
Challenge 196 d starts rotating in the other sense! Can you explain the effect?
**
What is the motion of the point on the surface of the Earth that has Sun in its zenith (i.e.,
Challenge 197 ny vertically above it), when seen on a map of the Earth during one day, and day after day?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
The moment of inertia of a body depends on the shape of the body; usually, angular mo-
mentum and the angular velocity do not point in the same direction. Can you confirm
Challenge 198 n this with an example?
**
Can it happen that a satellite dish for geostationary TV satellites focuses the sunshine
Challenge 199 n onto the receiver?

**
Why is it difficult to fire a rocket from an aeroplane in the direction opposite to the motion
Challenge 200 n of the plane?
**
You have two hollow spheres: they have the same weight, the same size and are painted in
the same colour. One is made of copper, the other of aluminium. Obviously, they fall with
Challenge 201 ny the same speed and acceleration. What happens if they both roll down a tilted plane?
**
An ape hangs on a rope. The rope hangs over a wheel and is attached to a mass of equal
Copyright © Christoph Schiller November 1997–May 2006

weight hanging down on the other side, as shown in Figure 60. The rope and the wheel
Challenge 202 n are massless and frictionless. What happens when the ape climbs the rope?

**
Challenge 203 ny What is the shape of a rope when rope jumping?
**
Challenge 204 n How can you determine the speed of a rifle bullet with only a scale and a metre stick?

**
Why does a gun make a hole in a door but cannot push it open, in exact contrast to what

Dvipsbugw
 112 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw

F I G U R E 60 How

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

does the ape
move?

Challenge 205 e a finger can do?

**
Challenge 206 n Can a water skier move with a higher speed than the boat pulling him?

**
Take two cans of the same size and weight, one full of ravioli and one full of peas. Which
Challenge 207 e one rolls faster on an inclined plane?
**
Challenge 208 n What is the moment of inertia of a homogeneous sphere?

**
The moment of inertia is determined by the values of its three principal axes. These are
all equal for a sphere and for a cube. Does it mean that it is impossible to distinguish a
Challenge 209 n sphere from a cube by their inertial behaviour?
Copyright © Christoph Schiller November 1997–May 2006

**
You might know the ‘Dynabee’, a hand-held gyroscopic device that can be accelerated to
Challenge 210 d high speed by proper movements of the hand. How does it work?
**
It is possible to make a spinning top with a metal paperclip. It is even possible to make
Challenge 211 n one of those tops with turn on their head when spinning. Can you find out how?

**
Is it true that the Moon in the first quarter in the northern hemisphere looks like the

Dvipsbugw
 objects and images 113

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 61 A long exposure of the stars at night - over
the Mauna Kea telescope in Hawaii (Gemini)

Challenge 212 n Moon in the last quarter in the southern hemisphere?

**
An impressive confirmation that the Earth is round can be seen at sunset, if one turns,
against usual habits, one’s back on the Sun. On the eastern sky one can see the impressive
rise of the Earth’s shadow. (In fact, more precise investigations show that it is not the
shadow of the Earth alone, but the shadow of its ionosphere.) One can admire a vast
shadow rising over the whole horizon, clearly having the shape of a segment of a huge
circle. Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 213 n How would Figure 61 look if taken at the Equator?

**
Since the Earth is round, there are many ways to drive from one point on the Earth to
another along a circle segment. This has interesting consequences for volley balls and for
girl-watching. Take a volleyball and look at its air inlet. If you want to move the inlet to a
different position with a simple rotation, you can choose the rotation axis in may different
Challenge 214 e ways. Can you confirm this? In other words, when we look in a given direction and then
want to look in another, the eye can accomplish this change in different ways. The option
chosen by the human eye had already been studied by medical scientists in the eighteenth

Dvipsbugw
 114 i galilean motion • 2. galilean physics – motion in everyday life

century. It is called Listing’s ‘law’.* It states that all axes that nature chooses lie in one plane.
Challenge 215 n Can you imagine its position in space? Men have a real interest that this mechanism is
strictly followed; if not, looking at girls on the beach could cause the muscles moving the
eyes to get knotted up.

Legs or wheels? – Again

The acceleration and deceleration of standard wheel-driven cars is never much greater
than about 1 д = 9.8 m~s2 , the acceleration due to gravity on our planet. Higher acceler- Dvipsbugw
ations are achieved by motorbikes and racing cars through the use of suspensions that
divert weight to the axes and by the use of spoilers, so that the car is pushed downwards
with more than its own weight. Modern spoilers are so efficient in pushing a car towards
the track that racing cars could race on the roof of a tunnel without falling down.
Through the use of special tyres these downwards forces are transformed into grip;
modern racing tyres allow forward, backward and sideways accelerations (necessary for

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

speed increase, for braking and for turning corners) of about 1.1 to 1.3 times the load. En-
gineers once believed that a factor 1 was the theoretical limit and this limit is still some-
times found in textbooks; but advances in tyre technology, mostly by making clever use
of interlocking between the tyre and the road surface as in a gear mechanism, have al-
lowed engineers to achieve these higher values. The highest accelerations, around 4 д,
are achieved when part of the tyre melts and glues to the surface. Special tyres designed
to make this happen are used for dragsters, but high performance radio-controlled model
cars also achieve such values.
How do all these efforts compare to using legs? High jump athletes can achieve peak
accelerations of about 2 to 4 д, cheetahs over 3 д, bushbabies up to 13 д, locusts about
Ref. 87 18 д, and fleas have been measured to accelerate about 135 д. The maximum accelera-
tion known for animals is that of click beetles, a small insect able to accelerate at over
2000 m~s2 = 200 д, about the same as an airgun pellet when fired. Legs are thus definit-
ively more efficient accelerating devices than wheels – a cheetah can easily beat any car
or motorbike – and evolution developed legs, instead of wheels, to improve the chances
of an animal in danger getting to safety. In short, legs outperform wheels.
Challenge 216 n There are other reasons for using legs instead of wheels. (Can you name some?) For
example, legs, unlike wheels, allow walking on water. Most famous for this ability is the
basilisk,** a lizard living in Central America. This reptile is about 50 cm long and has a
mass of about 90 g. It looks like a miniature Tyrannosaurus rex and is able to run over
Copyright © Christoph Schiller November 1997–May 2006

water surfaces on its hind legs. The motion has been studied in detail with high-speed
cameras and by measurements using aluminium models of the animal’s feet. The experi-
ments show that the feet slapping on the water provides only 25 % of the force necessary
Ref. 88 to run above water; the other 75 % is provided by a pocket of compressed air that the
basilisks create between their feet and the water once the feet are inside the water. In fact,

* If you are interested in learning in more detail how nature and the eye cope with the complexities of three di-
mensions, see the http://schorlab.berkeley.edu/vilis/whatisLL.htm and http://www.med.uwo.ca/physiology/
courses/LLConsequencesWeb/ListingsLaw/perceptual2.htm websites.
** In the Middle Ages, the term ‘basilisk’ referred to a mythical monster supposed to appear shortly before
the end of the world. Today, it is a small reptile in the Americas.

Dvipsbugw
 objects and images 115

Dvipsbugw

F I G U R E 62 A basilisk lizard (Basiliscus basiliscus) running on water,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

showing how the propulsing leg pushes into the water (© TERRA)

basilisks mainly walk on air.* It was calculated that humans are also able to walk on water,
provided their feet hit the water with a speed of 100 km~h using the simultaneous physical
power of 15 sprinters. Quite a feat for all those who ever did so.
There is a second method of walking and running on water;
this second method even allows its users to remain immobile
on top of the water surface. This is what water striders, insects
of the family Gerridae with a overall length of up to 15 mm, are
able to do (together with several species of spiders). Like all in-
sects, the water strider has six legs (spiders have eight). The wa-
ter strider uses the back and front legs to hover over the surface,
helped by thousands of tiny hairs attached to its body. The hairs,
F I G U R E 63 A water strider
together with the surface tension of water, prevent the strider (© Charles Lewallen)
from getting wet. If you put shampoo into the water, the wa-
ter strider sinks and can no longer move. The water strider uses its large middle legs as
oars to advance over the surface, reaching speeds of up to 1 m~s doing so. In short, water
striders actually row over water.
Legs pose many interesting problems. Engineers know that a staircase is comfortable
to walk only if for each step the depth l plus twice the height h is a constant: l + 2h =
Copyright © Christoph Schiller November 1997–May 2006

Challenge 217 n 0.63  0.02 m. This is the so-called staircase formula. Why does it hold?
Challenge 218 n All animals have an even number of legs. Do you know an exception? Why not? In
fact, one can argue that no animal has less than four legs. Why is this the case?
On the other hand, all animals with two legs have the legs side by side, whereas systems
Challenge 219 e with wheels have them one behind the other. Why is this not the other way round?
But let us continue with the study of motion transmitted over distance, without the
use of any contact at all.

Ref. 89 * Both effects used by basilisks are also found in fast canoeing.

Dvipsbugw
 116 i galilean motion • 2. galilean physics – motion in everyday life

Dynamics due to gravitation

Caddi come corpo morto cade.

“ Dante, Inferno, c. V, v. 142.*

The first and main contact-free method to generate motion we discover in our environ- ”
ment is height. Waterfalls, snow, rain and falling apples all rely on it. It was one of the
fundamental discoveries of physics that height has this property because there is an in-
teraction between every body and the Earth. Gravitation produces an acceleration along Dvipsbugw
the line connecting the centres of gravity of the body and the Earth. Note that in order to
make this statement, it is necessary to realize that the Earth is a body in the same way as
a stone or the Moon, that this body is finite and that therefore it has a centre and a mass.
Today, these statements are common knowledge, but they are by no means evident from
everyday personal experience.**
How does gravitation change when two bodies are far apart? The experts on distant ob-
jects are the astronomers. Over the years they have performed numerous measurements

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of the movements of the Moon and the planets. The most industrious of all was Tycho
Brahe,*** who organized an industrial-scale search for astronomical facts sponsored by
his king. His measurements were the basis for the research of his young assistant, the Swa-
bian astronomer Johannes Kepler**** who found the first precise description of planetary
Page 677 motion. In 1684, all observations of planets and stones were condensed into an astonish-
ingly simple result by the English physicist Robert Hooke:***** every body of mass M
attracts any other body towards its centre with an acceleration whose magnitude a is
given by
a=G 2
M
(29)
r

* ‘I fell like dead bodies fall.’ Dante Alighieri (1265, Firenze–1321, Ravenna), the powerful Italian poet.
** In several myths about the creation or the organization of the world, such as the biblical one or the Indian
one, the Earth is not an object, but an imprecisely defined entity, such as an island floating or surrounded by
Challenge 220 n water with unclear boundaries and unclear method of suspension. Are you able to convince a friend that the
Earth is round and not flat? Can you find another argument apart from the roundness of the Earth’s shadow
when it is visible on the Moon?
Ref. 90 A famous crook, Robert Peary, claimed to have reached the North Pole in 1909. (In fact, Roald Amundsen
reached the both the South and the North Pole first.) Peary claimed to have taken a picture there, but that
Challenge 221 n picture, which went round the world, turned out to be the proof that he had not been there. Can you imagine
how?
Copyright © Christoph Schiller November 1997–May 2006

By the way, if the Earth is round, the top of two buildings is further apart than their base. Can this effect
Challenge 222 n be measured?
*** Tycho Brahe (1546–1601), famous Danish astronomer, builder of Uraniaborg, the astronomical castle.
He consumed almost 10 % of the Danish gross national product for his research, which produced the first
star catalogue and the first precise position measurements of planets.
**** Johannes Kepler (1571 Weil der Stadt–1630 Regensburg); after helping his mother successfully defend
herself in a trial where she was accused of witchcraft, he studied Protestant theology and became a teacher
of mathematics, astronomy and rhetoric. His first book on astronomy made him famous, and he became
assistant to Tycho Brahe and then, at his teacher’s death, the Imperial Mathematician. He was the first to
use mathematics in the description of astronomical observations, and introduced the concept and field of
‘celestial physics’.
***** Robert Hooke, (1635–1703), important English physicist and secretary of the Royal Society. He also
wrote the Micrographia, a beautifully illustrated exploration of the world of the very small.

Dvipsbugw
 dynamics due to gravitation 117

where r is the centre-to-centre distance of the two bodies. This is called the universal
‘law’ of gravitation, or universal gravity, because it is valid in general. The proportionality
constant G is called the gravitational constant; it is one of the fundamental constants of
nature, like the speed of light or the quantum of action. More about it will be said shortly.
The effect of gravity thus decreases with increasing distance; gravity depends on the in-
verse distance squared of the bodies under consideration. If bodies are small compared
with the distance r, or if they are spherical, expression (29) is correct as it stands; for
non-spherical shapes the acceleration has to be calculated separately for each part of the
bodies and then added together. Dvipsbugw
This inverse square dependence is often called Newton’s ‘law’ of gravitation, because
the English physicist Isaac Newton proved more elegantly than Hooke that it agreed with
all astronomical and terrestrial observations. Above all, however, he organized a better
Ref. 91 public relations campaign, in which he falsely claimed to be the originator of the idea.
Newton published a simple proof showing that this description of astronomical mo-
tion also gives the correct description for stones thrown through the air, down here on

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

‘father Earth’. To achieve this, he compared the acceleration a m of the Moon with that of
stones д. For the ratio between these two accelerations, the inverse square relation pre-
dicts a value a m ~д = R 2 ~d m
2
, where R is the radius of the Earth and d m the distance of
the Moon. The Moon’s distance can be measured by triangulation, comparing the pos-
ition of the Moon against the starry background from two different points on Earth.*
The result is d m ~R = 60  3, depending on the orbital position of the Moon, so that an
average ratio a m ~д = 3.6 ë 103 is predicted from universal gravity. But both accelerations
can also be measured directly. At the surface of the Earth, stones are subject to an accele-
ration due to gravitation with magnitude д = 9.8 m~s2 , as determined by measuring the
time that stones need to fall a given distance. For the Moon, the definition of acceleration,
a = dv~dt, in the case of circular motion – roughly correct here – gives a m = d m (2π~T)2 ,
where T = 2.4 Ms is the time the Moon takes for one orbit around the Earth.** The meas-
urement of the radius of the Earth*** yields R = 6.4 Mm, so that the average Earth–Moon

* The first precise – but not the first – measurement was achieved in 1752 by the French astronomers Lalande
and La Caille, who simultaneously measured the position of the Moon seen from Berlin and from Le Cap.
** This is deduced easily by noting that for an object in circular motion, the magnitude v of the velocity
Challenge 223 n v = dx~dt is given as v = 2πr~T. The drawing of the vector v over time, the so-called hodograph, shows that
it behaves exactly like the position of the object. Therefore the magnitude a of the acceleration a = dv~dt is
given by the corresponding expression, namely a = 2πv~T.
Copyright © Christoph Schiller November 1997–May 2006

*** This is the hardest quantity to measure oneself. The most surprising way to determine the Earth’s size is
Ref. 92 the following: watch a sunset in the garden of a house, with a stopwatch in hand. When the last ray of the
Sun disappears, start the stopwatch and run upstairs. There, the Sun is still visible; stop the stopwatch when
the Sun disappears again and note the time t. Measure the height distance h of the two eye positions where
Challenge 224 ny the Sun was observed. The Earth’s radius R is then given by R = k h~t 2 , with k = 378 ë 106 s2 .
Ref. 93 There is also a simple way to measure the distance to the Moon, once the size of the Earth is known. Take
a photograph of the Moon when it is high in the sky, and call θ its zenith angle, i.e. its angle from the vertical.
Make another photograph of the Moon a few hours later, when it is just above the horizon. On this picture,
Page 62 unlike a common optical illusion, the Moon is smaller, because it is further away. With a sketch the reason
for this becomes immediately clear. If q is the ratio of the two angular diameters, the Earth–Moon distance
Challenge 225 ny d m is given by the relation d m2
= R 2 + [2Rq cos θ~(1 − q 2 )]2 . Enjoy finding its derivation from the sketch.
Another possibility is to determine the size of the Moon by comparing it with the size of the shadow of
the Earth during an eclipse. The distance to the Moon is then computed from its angular size, about 0.5°.

Dvipsbugw
 118 i galilean motion • 2. galilean physics – motion in everyday life

Moon

Earth

figure to be inserted Dvipsbugw

F I G U R E 64 A physicist’s and an artist’s view of the fall of the Moon: a diagram by Christiaan Huygens
(not to scale) and a marble statue by Auguste Rodin

distance is d m = 0.38 Gm. One thus has a m ~д = 3.6 ë 103 , in agreement with the above pre-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

diction. With this famous ‘Moon calculation’ we have thus shown that the inverse square
property of gravitation indeed describes both the motion of the Moon and that of stones.
Challenge 226 n You might want to deduce the value of GM.
From the observation that on the Earth all motion eventually comes to rest, whereas
in the sky all motion is eternal, Aristotle and many others had concluded that motion in
the sublunar world has different properties from motion in the translunar world. Several
thinkers had criticized this distinction, notably the French philosopher and rector of the
Ref. 94 University of Paris, Jean Buridan.* The Moon calculation was the most important result
showing this distinction to be wrong. This is the reason for calling the expression (29)
the universal ‘law’ of gravitation.
This result allows us to answer another old question. Why does the Moon not fall
from the sky? Well, the preceding discussion showed that fall is motion due to gravitation.
Therefore the Moon actually is falling, with the peculiarity that instead of falling towards
the Earth, it is continuously falling around it. Figure 64 illustrates the idea. The Moon is
continuously missing the Earth.**
Universal gravity also explains why the Earth and most planets are (almost) spherical.
Since gravity increases with decreasing distance, a liquid body in space will always try to
form a spherical shape. Seen on a large scale, the Earth is indeed liquid. We also know
that the Earth is cooling down – that is how the crust and the continents formed. The
Copyright © Christoph Schiller November 1997–May 2006

sphericity of smaller solid objects encountered in space, such as the Moon, thus means
that they used to be liquid in older times.

* Jean Buridan (c. 1295 to c. 1366) was also one of the first modern thinkers to speculate on a rotation of
the Earth about an axis.
** Another way to put it is to use the answer of the Dutch physicist Christiaan Huygens (1629–1695): the
Moon does not fall from the sky because of the centrifugal acceleration. As explained on page 109, this
explanation is nowadays out of favour at most universities.
Ref. 95 There is a beautiful problem connected to the left part of the figure: Which points on the surface of the
Challenge 227 d Earth can be hit by shooting from a mountain? And which points can be hit by shooting horizontally?

Dvipsbugw
 dynamics due to gravitation 119

Properties of gravitation
Gravitation implies that the path of a stone is not a parabola, as stated earlier, but actually
an ellipse around the centre of the Earth. This happens for exactly the same reason that
Challenge 228 ny the planets move in ellipses around the Sun. Are you able to confirm this statement?
Universal gravitation allows us to solve a mystery. The puzzling acceleration value
д = 9.8 m~s2 we encountered in equation (4) is thus due to the relation

д = GM Earth ~R Earth
2
. (30) Dvipsbugw

The equation can be deduced from equation (29) by taking the Earth to be spherical. The
everyday acceleration of gravity д thus results from the size of the Earth, its mass, and the
universal constant of gravitation G. Obviously, the value for д is almost constant on the
surface of the Earth because the Earth is almost a sphere. Expression (30) also explains
why д gets smaller as one rises above the Earth, and the deviations of the shape of the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Earth from sphericity explain why д is different at the poles and higher on a plateau.
Challenge 229 n (What would it be on the Moon? On Mars? On Jupiter?)
By the way, it is possible to devise a simple machine, other than a yo-yo, that slows
down the effective acceleration of gravity by a known amount, so that one can measure
Challenge 230 n its value more easily. Can you imagine it?
Note that 9.8 is roughly π 2 . This is not a coincidence: the metre has been chosen in
such a way to make this correct. The period T of a swinging pendulum, i.e. a back and
Challenge 231 n forward swing, is given by* ¾
T = 2π
l
, (31)
д

where l is the length of the pendulum, and д is the gravitational acceleration. (The pen-
dulum is assumed to be made of a mass attached to a string of negligible mass.) The
oscillation time of a pendulum depends only on the length of the string and the planet it
is located on. If the metre had been defined such that T~2 = 1 s, the value of the normal
Challenge 233 e acceleration д would have been exactly π 2 m~s2 . This was the first proposal for the defin-
ition of the metre; it was made in 1673 by Huygens and repeated in 1790 by Talleyrand,
but was rejected by the conference that defined the metre because variations in the value
of д with geographical position and temperature-induced variations of the length of a
pendulum induce errors that are too large to yield a definition of useful precision.
Copyright © Christoph Schiller November 1997–May 2006

Finally, the proposal was made to define the metre as 1~40 000 000 of the circum-
ference of the Earth through the poles, a so-called meridian. This proposal was almost
identical to – but much more precise than – the pendulum proposal. The meridian defin-
ition of the metre was then adopted by the French national assembly on 26 March 1791,
* Formula (31) is noteworthy mainly for all that is missing. The period of a pendulum does not depend on
the mass of the swinging body. In addition, the period of a pendulum does not depend on the amplitude.
(This is true as long as the oscillation angle is smaller than about 15°.) Galileo discovered this as a student,
when observing a chandelier hanging on a long rope in the dome of Pisa. Using his heartbeat as a clock he
found that even though the amplitude of the swing got smaller and smaller, the time for the swing stayed
the same.
A leg also moves like a pendulum, when one walks normally. Why then do taller people tend to walk
Challenge 232 n faster?

Dvipsbugw
 120 i galilean motion • 2. galilean physics – motion in everyday life

with the statement that ‘a meridian passes under the feet of every human being, and all
meridians are equal’. (Nevertheless, the distance from Equator to the poles is not exactly
Ref. 96 10 Mm; that is a strange story. One of the two geographers who determined the size of
the first metre stick was dishonest. The data he gave for his measurements – the general
method of which is shown in Figure 65 – was fabricated. Thus the first official metre stick
in Paris was shorter than it should be.)
But we can still ask: Why does the Earth have the mass
and size it has? And why does G have the value it has? The
first question asks for a history of the solar system; it is still Dvipsbugw
unanswered and is topic of research. The second question
is addressed in Appendix B.
If all objects attract each other, it should also be the case
for objects in everyday life. Gravity must also work side-
ways. This is indeed the case, even though the effects are
so small that they were measured only long after universal

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

gravity had predicted them. Measuring this effect allows
the gravitational constant G to be determined.
Note that measuring the gravitational constant G is also
the only way to determine the mass of the Earth. The
first to do so, in 1798, was the English physicist Henry
Cavendish; he used the machine, ideas and method of
John Michell who died when attempting the experiment.
Michell and Cavendish* called the aim and result of their
experiments ‘weighing the Earth’. Are you able to imagine
Challenge 234 n how they did it? The value found in modern experiments
is
F I G U R E 65 The measurements
G = 6.7 ë 10−11
Nm ~kg = 6.7 ë 10
2 2 −11
m ~kg s .
3 2
(32) that lead to the definition of the
metre (© Ken Alder)

Cavendish’s experiment was thus the first to confirm that

gravity also works sideways.
For example, two average people 1 m apart feel an acceleration towards each other
Challenge 235 n that is less than that exerted by a common fly when landing on the skin. Therefore we
usually do not notice the attraction to other people. When we notice it, it is much stronger
than that. This simple calculation thus proves that gravitation cannot be the true cause
Copyright © Christoph Schiller November 1997–May 2006

of people falling in love, and that sexual attraction is not of gravitational origin, but of
a different source. The basis for this other interaction, love, will be studied later in our
walk: it is called electromagnetism.
But gravity has more interesting properties to offer. The effects of gravitation can also
be described by another observable, namely the (gravitational) potential φ. We then have

* Henry Cavendish (1731–1810) was one of the great geniuses of physics; rich and solitary, he found many
rules of nature, but never published them. Had he done so, his name would be much more well known.
John Michell (1724–1793) was church minister, geologist and amateur astronomer.

Dvipsbugw
 dynamics due to gravitation 121

the simple relation that the acceleration is given by the gradient of the potential

a = −∇φ or a = −grad φ . (33)

The gradient is just a learned term for ‘slope along the steepest direction’. It is defined
for any point on a slope, is large for a steep one and small for a shallow one and it
points in the direction of steepest ascent, as shown in Figure 66. The gradient is ab-
breviated ∇, pronounced ‘nabla’ and is mathematically defined as the vector ∇φ =
(∂φ~∂x, ∂φ~∂y, ∂φ~∂z) = grad φ. The minus sign in (33) is introduced by convention, Dvipsbugw
in order to have higher potential values at larger heights.* For a point-like or a spherical
body of mass M, the potential φ is

φ = −G
M
. (34)
r

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

A potential considerably simplifies the description of motion, since a potential is additive:
given the potential of a point particle, one can calculate the potential and then the motion
around any other irregularly shaped object.**
The potential φ is an interesting quantity; with a single number at every position in
space we can describe the vector aspects of gravitational acceleration. It automatically
gives that gravity in New Zealand acts in the opposite direction to gravity in Paris. In
addition, the potential suggests the introduction of the so-called potential energy U by
setting
U = mφ (36)

and thus allowing us to determine the change of kinetic energy T of a body falling from
a point 1 to a point 2 via

T1 − T2 = U 2 − U 1 m 1 v1 2 − m 2 v2 2 = mφ 2 − mφ 1 .
1 1
or (37)
2 2
* In two or more dimensions slopes are written ∂φ~∂z – where ∂ is still pronounced ‘d’ – because in those
cases the expression dφ~dz has a slightly different meaning. The details lie outside the scope of this walk.
** Alternatively, for a general, extended body, the potential is found by requiring that the divergence of its
gradient is given by the mass (or charge) density times some proportionality constant. More precisely, one
has
Copyright © Christoph Schiller November 1997–May 2006

∆φ = 4πGρ (35)
where ρ = ρ(x, t) is the mass volume density of the body and the operator ∆, pronounced ‘delta’, is defined
as ∆ f = ∇∇ f = ∂ 2 f ~∂x 2 + ∂ 2 f ~∂y 2 + ∂ 2 f ~∂z 2 . Equation (35) is called the Poisson equation for the potential
φ. It is named after Siméon-Denis Poisson (1781–1840), eminent French mathematician and physicist. The
positions at which ρ is not zero are called the sources of the potential. The so-called source term ∆φ of
a function is a measure for how much the function φ(x) at a point x differs from the average value in a
Challenge 236 ny region around that point. (Can you show this, by showing that ∆φ  φ̄ − φ(x)?) In other words, the Poisson
equation (35) implies that the actual value of the potential at a point is the same as the average value around
that point minus the mass density multiplied by 4πG. In particular, in the case of empty space the potential
at a point is equal to the average of the potential around that point.
Often the concept of gravitational field is introduced, defined as g = −∇φ. We avoid this in our walk,
because we will discover that, following the theory of relativity, gravity is not due to a field at all; in fact even
the concept of gravitational potential turns out to be only an approximation.

Dvipsbugw
 122 i galilean motion • 2. galilean physics – motion in everyday life

In other words, the total energy, defined as the sum of kinetic and potential energy, is
conserved in motion due to gravity. This is a characteristic property of gravitation. Not
all accelerations can be derived from a potential; systems with this property are called
conservative. The accelerations due to friction are not conservative, but those due to elec-
tromagnetism are.
Interestingly, the number of dimensions of space
d is coded into the potential of a spherical mass: its f (x,y)
Challenge 237 n dependence on the radius r is in fact 1~r d−2 . The ex-
ponent d − 2 has been checked experimentally to Dvipsbugw
high precision; no deviation of d from 3 has ever
Ref. 99 been found.
The concept of potential helps in understanding y
x
the shape of the Earth. Since most of the Earth is
Ref. 100 still liquid when seen on a large scale, its surface is grad f
always horizontal with respect to the direction de- F I G U R E 66 The potential and the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

termined by the combination of the accelerations gradient
of gravity and rotation. In short, the Earth is not a
sphere. It is not an ellipsoid either. The mathematical shape defined by the equilibrium
Ref. 101 requirement is called a geoid. The geoid shape differs from a suitably chosen ellipsoid
Challenge 238 ny by at most 50 m. Can you describe the geoid mathematically? The geoid is an excellent
approximation to the actual shape of the Earth; sea level differs from it by less than 20
metres. The differences can be measured with satellite radar and are of great interest to
geologists and geographers. For example, it turns out that the South Pole is nearer to the
equatorial plane than the North Pole by about 30 m. This is probably due to the large land
masses in the northern hemisphere.
Above we saw how the inertia of matter,
Page 94 through the so-called ‘centrifugal force’,
increases the radius of the Earth at the
Equator. In other words, the Earth is
flattened at the poles. The Equator has
a radius a of 6.38 Mm, whereas the dis-
tance b from the poles to the centre of
the Earth is 6.36 Mm. The precise flatten-
ing (a − b)~a has the value 1~298.3 =
Appendix B 0.0034. As a result, the top of Mount Chim-
Copyright © Christoph Schiller November 1997–May 2006

borazo in Ecuador, even though its height

is only 6267 m above sea level, is about
20 km farther away from the centre of the
Earth than the top of Mount Sagarmatha*
in Nepal, whose height above sea level is
8850 m. The top of Mount Chimborazo is
in fact the point on the surface most dis- F I G U R E 67 The shape of the Earth, with
exaggerated height scale (©
tant from the centre of the Earth. GeoForschungsZentrum Potsdam)
As a consequence, if the Earth stopped

* Mount Sagarmatha is sometimes also called Mount Everest.

Dvipsbugw
 dynamics due to gravitation 123

rotating (but kept its shape), the water of the oceans would flow north; all of Europe would
be under water, except for the few mountains of the Alps that are higher than about 4 km.
The northern parts of Europe would be covered by between 6 km and 10 km of water.
Mount Sagarmatha would be over 11 km above sea level. If one takes into account the
resulting change of shape of the Earth, the numbers come out smaller. In addition, the
change in shape would produce extremely strong earthquakes and storms. As long as
there are none of these effects, we can be sure that the Sun will indeed rise tomorrow,
Page 88 despite what some philosophers might pretend.
Dvipsbugw
Dynamics – how do things move in various dimensions?
Let us give a short summary. If a body can move only along a (possibly curved) line, the
concepts of kinetic and potential energy are sufficient to determine the way it moves. In
short, motion in one dimension follows directly from energy conservation.
If more than two spatial dimensions are involved, energy conservation is insufficient

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

to determine how a body moves. If a body can move in two dimensions, and if the forces
involved are internal (which is always the case in theory, but not in practice), the con-
servation of angular momentum can be used. The full motion in two dimensions thus
follows from energy and angular momentum conservation. For example, all properties
of free fall follow from energy and angular momentum conservation. (Are you able to
Challenge 239 n show this?)
In the case of motion in three dimensions, a more general rule for determining motion
is necessary. It turns out that all motion follows from a simple principle: the time average
of the difference between kinetic and potential energy must be as small as possible. This
Page 176 is called the least action principle. We will explain the details of this calculation method
later.
For simple gravitational motions, motion is two-dimensional. Most three-
dimensional problems are outside the scope of this text; in fact, some of these problems
are still subjects of research. In this adventure, we will explore three-dimensional motion
only for selected cases that provide important insights.

Gravitation in the sky

The expression for the acceleration due to gravity a = GM~r 2 also describes the motion
of all the planets around the Sun. Anyone can check that the planets always stay within
the zodiac, a narrow stripe across the sky. The centre line of the zodiac gives the path of
Copyright © Christoph Schiller November 1997–May 2006

the Sun and is called the ecliptic, since the Moon must be located on it to produce an
eclipse. But the detailed motion of the planets is not easy to describe.* A few generations
before Hooke, the Swabian astronomer Johannes Kepler had deduced several ‘laws’ in his
painstaking research about the movements of the planets in the zodiac. The three main
ones are as follows:
1. Planets move on ellipses with the Sun located at one focus (1609).
2. Planets sweep out equal areas in equal times (1609).

* The apparent height of the ecliptic changes with the time of the year and is the reason for the changing
seasons. Therefore seasons are a gravitational effect as well.

Dvipsbugw
 124 i galilean motion • 2. galilean physics – motion in everyday life

3. All planets have the same ratio T 2 ~d 3 between the orbit duration T and the semima-
jor axis d (1619).
The main results are given in Figure 68. The sheer
work required to deduce the three ‘laws’ was enorm-
ous. Kepler had no calculating machine available, not
even a slide rule. The calculation technology he used d
was the recently discovered logarithms. Anyone who Sun
d
has used tables of logarithms to perform calculations Dvipsbugw
can get a feeling for the amount of work behind these planet
three discoveries.
The second ‘law’ about equal swept areas implies
that planets move faster when they are near the Sun. F I G U R E 68 The motion of a planet
around the Sun, showing its
It is a way to state the conservation of angular mo- semimajor axis d, which is also the
mentum. But now comes the central point. The huge spatial average of its distance from
volume of work by Brahe and Kepler can be summar- the Sun

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ized in the expression a = GM~r 2 . Can you confirm
Challenge 240 ny that all three laws follow from Hooke’s expression of universal gravity? Publishing this
result was the main achievement of Newton. Try to repeat his achievement; it will show
you not only the difficulties, but also the possibilities of physics, and the joy that puzzles
give.
Newton solved the puzzle with geometric drawing. Newton was not able to write down,
Ref. 24 let alone handle, differential equations at the time he published his results on gravitation.
In fact, it is well known that Newton’s notation and calculation methods were poor. (Much
poorer than yours!) The English mathematician Godfrey Hardy* used to say that the
insistence on using Newton’s integral and differential notation, rather than the earlier
and better method, still common today, due to his rival Leibniz – threw back English
mathematics by 100 years.
Kepler, Hooke and Newton became famous because they brought order to the descrip-
tion of planetary motion. This achievement, though of small practical significance, was
widely publicized because of the age-old prejudices linked with astrology.
However, there is more to gravitation. Universal gravity explains the motion and shape
of the Milky Way and of the other galaxies, the motion of many weather phenomena and
Challenge 241 n explains why the Earth has an atmosphere but the Moon does not. (Can you explain this?)
In fact, universal gravity explains much more about the Moon.
Copyright © Christoph Schiller November 1997–May 2006

The Moon
How long is a day on the Moon? The answer is roughly 14 Earth-days. That is the time
that it takes for the Moon to see the Sun again in the same position.
One often hears that the Moon always shows the same side to the Earth. But this is
wrong. As one can check with the naked eye, a given feature in the centre of the face of
the Moon at full Moon is not at the centre one week later. The various motions leading

* Godfrey Harold Hardy (1877–1947) was an important English number theorist, and the author of the
well-known A Mathematician’s Apology. He also ‘discovered’ the famous Indian mathematician Srinivasa
Ramanujan, bringing him to Britain.

Dvipsbugw
dynamics due to gravitation 125

Dvipsbugw

Loading movie file.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 69 The change of the moon during the month, showing its libration (© Martin
Elsässer)

to this change are called librations; they are shown in the movie in Figure 69.* The mo-
tions appear mainly because the Moon does not describe a circular, but an elliptical orbit
around the Earth and because the axis of the Moon is slightly inclined, compared with
that of its rotation around the Earth. As a result, only around 45 % of the Moon’s surface
is permanently hidden from Earth.
The first photographs of the hidden area were taken in the 1960s by a Soviet artifi-
cial satellite; modern satellites provided exact maps, as shown in Figure 70. The hidden
surface is much more irregular than the visible one, as the hidden side is the one that
Copyright © Christoph Schiller November 1997–May 2006

intercepts most asteroids attracted by the Earth. Thus the gravitation of the Moon helps
to deflect asteroids from the Earth. The number of animal life extinctions is thus reduced
to a small, but not negligible number. In other words, the gravitational attraction of the
Moon has saved the human race from extinction many times over.**
The trips to the Moon in the 1970s also showed that the Moon originated from the
Earth itself: long ago, an object hit the Earth almost tangentially and threw a sizeable

* The movie is in DivX 5 AVI format and requires a software plug-in in Acrobat Reader that can play it.
** The web pages http://cfa-www.harvard.edu/iau/lists/Closest.html and InnerPlot.html give an impression
of the number of objects that almost hit the Earth every year. Without the Moon, we would have many
additional catastrophes.

Dvipsbugw
 126 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw

F I G U R E 70 Maps (not photographs) of the near side (left) and far side (right) of
the moon, showing how often the latter saved the Earth from meteorite impacts
(courtesy USGS)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fraction of material up into the sky. This is the only mechanism able to explain the large
Ref. 102 size of the Moon, its low iron content, as well as its general material composition.
Ref. 103 The Moon is receding from the Earth at 3.8 cm a year. This result confirms the old
deduction that the tides slow down the Earth’s rotation. Can you imagine how this meas-
Challenge 242 n urement was performed? Since the Moon slows down the Earth, the Earth also changes
shape due to this effect. (Remember that the shape of the Earth depends on its speed of
rotation.) These changes in shape influence the tectonic activity of the Earth, and maybe
also the drift of the continents.
The Moon has many effects on animal life. A famous example is the midge Clunio,
Ref. 104 which lives on coasts with pronounced tides. Clunio spends between six and twelve weeks
as a larva, then hatches and lives for only one or two hours as an adult flying insect, during
which time it reproduces. The midges will only reproduce if they hatch during the low
tide phase of a spring tide. Spring tides are the especially strong tides during the full and
new moons, when the solar and lunar effects combine, and occur only every 14.8 days. In
1995, Dietrich Neumann showed that the larvae have two built-in clocks, a circadian and
a circalunar one, which together control the hatching to precisely those few hours when
the insect can reproduce. He also showed that the circalunar clock is synchronized by the
brightness of the Moon at night. In other words, the larvae monitor the Moon at night
Copyright © Christoph Schiller November 1997–May 2006

and then decide when to hatch: they are the smallest known astronomers.
If insects can have circalunar cycles, it should come as no surprise that women also
Ref. 105 have such a cycle. However, in this case the origin of the cycle length is still unknown.
The Moon also helps to stabilize the tilt of the Earth’s axis, keeping it more or less
fixed relative to the plane of motion around the Sun. Without the Moon, the axis would
change its direction irregularly, we would not have a regular day and night rhythm, we
would have extremely large climate changes, and the evolution of life would have been
Ref. 106 impossible. Without the Moon, the Earth would also rotate much faster and we would
Ref. 107 have much less clement weather. The Moon’s main remaining effect on the Earth, the
Page 100 precession of its axis, is responsible for the ice ages.
Furthermore, the Moon shields the Earth from cosmic radiation by greatly increasing

Dvipsbugw
 dynamics due to gravitation 127

the Earth’s magnetic field. In other words, the Moon is of central importance for the
evolution of life. Understanding how often Earth-sized planets have Moon-sized satellites
Ref. 108 is thus important for the estimation of the probability that life exists on other planets. So
far, it seems that large satellites are rare; there are only four known moons that are larger
than that of the Earth, but they circle much larger planets, namely Jupiter and Saturn.
Indeed, the formation of satellites is still an area of research. But let us return to the effects
of gravitation in the sky.

Orbits Dvipsbugw

The path of a body orbiting another under the influence of gravity is an ellipse with the
central body at one focus. A circular orbit is also possible, a circle being a special case
of an ellipse. Single encounters of two objects can also be parabolas or hyperbolas, as
shown in Figure 71. Circles, ellipses, parabolas and hyperbolas are collectively known as
conic sections. Indeed each of these curves can be produced by cutting a cone with a knife.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 243 e Are you able to confirm this?
If orbits are mostly ellipses, it follows that comets
return. The English astronomer Edmund Halley
hyperbola
(1656–1742) was the first to draw this conclusion and
to predict the return of a comet. It arrived at the pre-
dicted date in 1756, and is now named after him. The
period of Halley’s comet is between 74 and 80 years; parabola
the first recorded sighting was 22 centuries ago, and
it has been seen at every one of its 30 passages since,
the last time in 1986. mass
Depending on the initial energy and the initial an- ellipse
circle
gular momentum of the body with respect to the cent-
ral planet, there are two additional possibilities: para-
bolic paths and hyperbolic paths. Can you determ-
ine the conditions of the energy and the angular mo-
Challenge 244 ny mentum needed for these paths to appear?
In practice, parabolic paths do not exist in nature.
(Though some comets seem to approach this case
when moving around the Sun; almost all comets fol- F I G U R E 71 The possible orbits due to
low elliptical paths). Hyperbolic paths do exist; artifi- universal gravity
Copyright © Christoph Schiller November 1997–May 2006

cial satellites follow them when they are shot towards

a planet, usually with the aim of changing the direction of the satellite’s journey across
the solar system.
Why does the inverse square law lead to conic sections? First, for two bodies, the total
angular momentum L is a constant:

L = mr 2 φ̇ (38)

Dvipsbugw
 128 i galilean motion • 2. galilean physics – motion in everyday life

and therefore the motion lies in a plane. Also the energy E is a constant

E = m( )2 + m(r )2 − G
1 dr 1 dφ mM
. (39)
2 dt 2 dt r
Challenge 245 ny Together, the two equations imply that

L2
r=
1
¼ . (40) Dvipsbugw
Gm M 1 + 1 + 2E L 2
2
cos φ
G 2 m3 M2

Now, any curve defined by the general expression

r= r=
C C
or (41)
1 + e cos φ 1 − e cos φ

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is an ellipse for 0 < e < 1, a parabola for e = 1 and a hyperbola for e A 1, one focus being
at the origin. The quantity e, called the eccentricity, describes how squeezed the curve is.
In other words, a body in orbit around a central mass follows a conic section.
In all orbits, also the heavy mass moves. In fact, both bodies orbit around the common
centre of mass. Both bodies follow the same type of curve (ellipsis, parabola or hyperbola),
Challenge 246 e but the dimensions of the two curves differ.
If more than two objects move under mutual gravitation, many additional possibilities
for motions appear. The classification and the motions are quite complex. In fact, this so-
called many-body problem is still a topic of research, and the results are mathematically
fascinating. Let us look at a few examples.
When several planets circle a star, they also attract each other. Planets thus do not
move in perfect ellipses. The largest deviation is a perihelion shift, as shown in Figure 49.
It is observed for Mercury and a few other planets, including the Earth. Other deviations
from elliptical paths appear during a single orbit. In 1846, the observed deviations of the
motion of the planet Uranus from the path predicted by universal gravity were used to pre-
dict the existence of another planet, Neptune, which was discovered shortly afterwards.
Page 80 We have seen that mass is always positive and that gravitation is thus always attractive;
there is no antigravity. Can gravity be used for levitation nevertheless, maybe using more
than two bodies? Yes; there are two examples.* The first are the geostationary satellites,
which are used for easy transmission of television and other signals from and towards
Copyright © Christoph Schiller November 1997–May 2006

Earth.
The Lagrangian libration points are the second example. Named after their discoverer,
these are points in space near a two-body system, such as Moon–Earth or Earth–Sun,
in which small objects have a stable equilibrium position. A general overview is given in
Challenge 247 ny Figure 72. Can you find their precise position, remembering to take rotation into account?
There are three additional Lagrangian points on the Earth–Moon line. How many of them
Challenge 248 ny are stable?
There are thousands of asteroids, called Trojan asteroids, at and around the Lagrangian
points of the Sun–Jupiter system. In 1990, a Trojan asteroid for the Mars–Sun system was

* Levitation is discussed in detail on page 602.

Dvipsbugw
 dynamics due to gravitation 129

discovered. Finally, in 1997, an ‘almost Trojan’ asteroid was found that follows the Earth
on its way around the Sun (it is only transitionary and follows a somewhat more complex
Ref. 109 orbit). This ‘second companion’ of the Earth has a diameter of 5 km. Similarly, on the
main Lagrangian points of the Earth–Moon system a high concentration of dust has been
observed.
To sum up, the single equation a = −GMr~r 3 cor-
rectly describes a large number of phenomena in the
sky. The first person to make clear that this expression
describes everything happening in the sky was Pierre Si- Dvipsbugw
planet (or Sun)
mon Laplace* in his famous treatise Traité de méca-
nique céleste. When Napoleon told him that he found L5
π/3
no mention about the creator in the book, Laplace gave
a famous, one sentence summary of his book: Je n’ai pas
π/3
eu besoin de cette hypothèse. ’I had no need for this hy- π/3 π/3
pothesis.’ In particular, Laplace studied the stability of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

L4
the solar system, the eccentricity of the lunar orbit, and moon (or planet)
the eccentricities of the planetary orbits, always getting
F I G U R E 72 The two stable
full agreement between calculation and measurement. Lagrangian points
These results are quite a feat for the simple expres-
sion of universal gravitation; they also explain why it is called ‘universal’. But how pre-
cise is the formula? Since astronomy allows the most precise measurements of gravita-
tional motion, it also provides the most stringent tests. In 1849, Urbain Le Verrier con-
cluded after intensive study that there was only one known example of a discrepancy
between observation and universal gravity, namely one observation for the planet Mer-
cury. (Nowadays a few more are known.) The point of least distance to the Sun of the
orbit of planet Mercury, its perihelion, changes at a rate that is slightly less than that pre-
Ref. 110 dicted: he found a tiny difference, around 38 ′′ per century. (This was corrected to 43 ′′ per
century in 1882 by Simon Newcomb.) Le Verrier thought that the difference was due to
a planet between Mercury and the Sun, Vulcan, which he chased for many years without
success. The study of motion had to wait for Albert Einstein to find the correct explana-
tion of the difference.

Tides
Why do physics texts always talk about tides? Because, as general relativity will show, tides
Copyright © Christoph Schiller November 1997–May 2006

prove that space is curved! It is thus useful to study them in a bit more detail. Gravitation
explains the sea tides as results of the attraction of the ocean water by the Moon and the
Sun. Tides are interesting; even though the amplitude of the tides is only about 0.5 m
on the open sea, it can be up to 20 m at special places near the coast. Can you imagine
Challenge 249 n why? The soil is also lifted and lowered by the Sun and the Moon, by about 0.3 m, as satel-
Ref. 39 lite measurements show. Even the atmosphere is subject to tides, and the corresponding
Ref. 111 pressure variations can be filtered out from the weather pressure measurements.

* Pierre Simon Laplace (b. 1749 Beaumont-en-Auge, d. 1827 Paris), important French mathematician. His
treatise appeared in five volumes between 1798 and 1825. He was the first to propose that the solar system
was formed from a rotating gas cloud, and one of the first people to imagine and explore black holes.

Dvipsbugw
130 i galilean motion • 2. galilean physics – motion in everyday life

before
Sun

t = t1 :
deformed

after
t=0:
spherical Dvipsbugw

F I G U R E 73 Tidal deformations due F I G U R E 74 The origin of tides

to gravity

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Tides appear for any extended body moving in the gravitational field of another. To
understand the origin of tides, picture a body in orbit, like the Earth, and imagine its
components, such as the segments of Figure 73, as being held together by springs. Uni-
versal gravity implies that orbits are slower the more distant they are from a central body.
As a result, the segment on the outside of the orbit would like to be slower than the central
one; but it is pulled by the rest of the body through the springs. In contrast, the inside seg-
ment would like to orbit more rapidly but is retained by the others. Being slowed down,
the inside segments want to fall towards the Sun. In sum, both segments feel a pull away
from the centre of the body, limited by the springs that stop the deformation. Therefore,
extended bodies are deformed in the direction of the field inhomogeneity.
For example, as a result of tidal forces, the Moon always has (roughly) the same face
to the Earth. In addition, its radius in direction of the Earth is larger by about 5 m than
the radius perpendicular to it. If the inner springs are too weak, the body is torn into
pieces; in this way a ring of fragments can form, such as the asteroid ring between Mars
and Jupiter or the rings around Saturn.
Copyright © Christoph Schiller November 1997–May 2006

Let us return to the Earth. If a body is surrounded by water, it will form bulges in the
direction of the applied gravitational field. In order to measure and compare the strength
of the tides from the Sun and the Moon, we reduce tidal effects to their bare minimum.
As shown in Figure 74, we can study the deformation of a body due to gravity by study-
ing the deformation of four pieces. We can study it in free fall, because orbital motion
and free fall are equivalent. Now, gravity makes some of the pieces approach and others
diverge, depending on their relative positions. The figure makes clear that the strength of
the deformation – water has no built-in springs – depends on the change of gravitational
acceleration with distance; in other words, the relative acceleration that leads to the tides
is proportional to the derivative of the gravitational acceleration.
Using the numbers from Appendix B, the gravitational accelerations from the Sun and

Dvipsbugw
 dynamics due to gravitation 131

the Moon measured on Earth are

a Sun = = 5.9 mm~s2

GM Sun
2
d Sun

= = 0.033 mm~s2
GM Moon
a Moon 2
(42)
d Moon

and thus the attraction from the Moon is about 178 times weaker than that from the Sun.
Dvipsbugw
When two nearby bodies fall near a large mass, the relative acceleration is proportional
to their distance, and follows da = da~dr dr. The proportionality factor da~dr = ∇a,
called the tidal acceleration (gradient), is the true measure of tidal effects. Near a large
Challenge 250 e spherical mass M, it is given by
=− 3
da 2GM
(43)
dr r

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

which yields the values

=− 3 = −0.8 ë 10−13 ~s2

da Sun 2GM Sun
dr d Sun

=− 3 = −1.7 ë 10−13 ~s2 .

da Moon 2GM Moon
(44)
dr d Moon

In other words, despite the much weaker pull of the Moon, its tides are predicted to be
over twice as strong as the tides from the Sun; this is indeed observed. When Sun, Moon
and Earth are aligned, the two tides add up; these so-called spring tides are especially
strong and happen every 14.8 days, at full and new moon.
Tides also produce friction. The friction leads to a slowing of the Earth’s rotation.
Nowadays, the slowdown can be measured by precise clocks (even though short time
Ref. 72 variations due to other effects, such as the weather, are often larger). The results fit well
Page 451 with fossil results showing that 400 million years ago, in the Devonian period, a year had
400 days, and a day about 22 hours. It is also estimated that 900 million years ago, each
of the 481 days of a year were 18.2 hours long. The friction at the basis of this slowdown
also results in an increase in the distance of the Moon from the Earth by about 3.8 cm
Challenge 251 n per year. Are you able to explain why?
As mentioned above, the tidal motion of the soil is also responsible for the trigger-
Copyright © Christoph Schiller November 1997–May 2006

ing of earthquakes. Thus the MoonMoon, dangers of can have also dangerous effects on
Earth. The most fascinating example of tidal effects is seen on Jupiter’s satellite Io. Its
tides are so strong that they induce intense volcanic activity, as shown in Figure 75, with
eruption plumes as high as 500 km. If tides are even stronger, they can destroy the body
altogether, as happened to the body between Mars and Jupiter that formed the planetoids,
or (possibly) to the moons that led to Saturn’s rings.
In summary, tides are due to relative accelerations of nearby mass points. This has an
Page 377 important consequence. In the chapter on general relativity we will find that time multi-
plied by the speed of light plays the same role as length. Time then becomes an additional
dimension, as shown in Figure 76. Using this similarity, two free particles moving in the

Dvipsbugw
 132 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 75 A spectacular result of tides:
volcanism on Io (NASA)

x
t1

α
b

t2 M
t
F I G U R E 76 Particles falling side F I G U R E 77 Masses bend light
by side approach over time

same direction correspond to parallel lines in space-time. Two particles falling side-by-
side also correspond to parallel lines. Tides show that such particles approach each other.
Page 418 In other words, tides imply that parallel lines approach each other. But parallel lines can
Copyright © Christoph Schiller November 1997–May 2006

approach each other only if space-time is curved. In short, tides imply curved space-time
and space. This simple reasoning could have been performed in the eighteenth century;
however, it took another 200 years and Albert Einstein’s genius to uncover it.

Can light fall?

Die Maxime, jederzeit selbst zu denken, ist die

“ Aufklärung.
Immanuel Kant*

”
* The maxim to think at all times for oneself is the enlightenment.

Dvipsbugw
 dynamics due to gravitation 133

Towards the end of the seventeenth century people discovered that light has a finite velo-
Page 275 city – a story which we will tell in detail later. An entity that moves with infinite velocity
cannot be affected by gravity, as there is no time to produce an effect. An entity with a
finite speed, however, should feel gravity and thus fall.
Does its speed increase when light reaches the surface of the Earth? For almost three
centuries people had no means of detecting any such effect; so the question was not in-
vestigated. Then, in 1801, the Prussian astronomer Johann Soldner (1776–1833) was the
Ref. 112 first to put the question in a different way. Being an astronomer, he was used to measur- Dvipsbugw
ing stars and their observation angles. He realized that light passing near a massive body
would be deflected due to gravity.
Soldner studied a body on a hyperbolic path, moving with velocity c past a spherical
mass M at distance b (measured from the centre), as shown in Figure 77. Soldner deduced
Challenge 252 ny the deflection angle
α univ. grav. =
2 GM
.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(45)
b c2
One sees that the angle is largest when the motion is just grazing the mass M. For light
deflected by the mass of the Sun, the angle turns out to be at most a tiny 0.88 ′′ = 4.3 µrad.
In Soldner’s time, this angle was too small to be measured. Thus the issue was forgotten.
Had it been pursued, general relativity would have begun as an experimental science, and
Page 409 not as the theoretical effort of Albert Einstein! Why? The value just calculated is different
from the measured value. The first measurement took place in 1919;* it found the correct
dependence on the distance, but found a deflection of up to 1.75 ′′ , exactly double that of
expression (45). The reason is not easy to find; in fact, it is due to the curvature of space,
as we will see. In summary, light can fall, but the issue hides some surprises.

What is mass? – Again

Mass describes how an object interacts with others. In our walk, we have encountered
two of its aspects. Inertial mass is the property that keeps objects moving and that offers
resistance to a change in their motion. Gravitational mass is the property responsible for
the acceleration of bodies nearby (the active aspect) or of being accelerated by objects
nearby (the passive aspect). For example, the active aspect of the mass of the Earth de-
termines the surface acceleration of bodies; the passive aspect of the bodies allows us
to weigh them in order to measure their mass using distances only, e.g. on a scale or a
Copyright © Christoph Schiller November 1997–May 2006

balance. The gravitational mass is the basis of weight, the difficulty of lifting things.**
Is the gravitational mass of a body equal to its inertial mass? A rough answer is given
by the experience that an object that is difficult to move is also difficult to lift. The simplest
experiment is to take two bodies of different masses and let them fall. If the acceleration
is the same for all bodies, inertial mass is equal to (passive) gravitational mass, because
in the relation ma = ∇(GMm~r) the left-hand m is actually the inertial mass, and the
right-hand m is actually the gravitational mass.

Challenge 253 ny * By the way, how would you measure the deflection of light near the bright Sun?
Challenge 254 ny ** What are the values shown by a balance for a person of 85 kg juggling three balls of 0.3 kg each?

Dvipsbugw
 134 i galilean motion • 2. galilean physics – motion in everyday life

But in the seventeenth century Galileo had made widely known an even older argu-
ment showing without a single experiment that the acceleration is indeed the same for all
bodies. If larger masses fell more rapidly than smaller ones, then the following paradox
would appear. Any body can be seen as being composed of a large fragment attached to a
small fragment. If small bodies really fell less rapidly, the small fragment would slow the
large fragment down, so that the complete body would have to fall less rapidly than the
larger fragment (or break into pieces). At the same time, the body being larger than its
fragment, it should fall more rapidly than that fragment. This is obviously impossible: all
masses must fall with the same acceleration. Dvipsbugw
Many accurate experiments have been performed since Galileo’s original discussion.
In all of them the independence of the acceleration of free fall from mass and material
Ref. 113 composition has been confirmed with the precision they allowed. In other words, as far
as we can tell, gravitational mass and inertial mass are identical. What is the origin of this
mysterious equality?
This so-called ‘mystery’ is a typical example of disinformation, now common across

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 77 the whole world of physics education. Let us go back to the definition of mass as a negat-
ive inverse acceleration ratio. We mentioned that the physical origins of the accelerations
do not play a role in the definition because the origin does not appear in the expression.
In other words, the value of the mass is by definition independent of the interaction. That
means in particular that inertial mass, based on electromagnetic interaction, and gravit-
ational mass are identical by definition.
We also note that we have never defined a separate concept of ‘passive gravitational
mass’. The mass being accelerated by gravitation is the inertial mass. Worse, there is no
Challenge 255 ny way to define a ‘passive gravitational mass’. Try it! All methods, such as weighing an ob-
ject, cannot be distinguished from those that determine inertial mass from its reaction to
acceleration. Indeed, all methods of measuring mass use non-gravitational mechanisms.
Scales are a good example.
If the ‘passive gravitational mass’ were different from the inertial mass, we would have
strange consequences. For those bodies for which it were different we would get into
trouble with energy conservation. Also assuming that ‘active gravitational mass’ differs
from inertial mass gets us into trouble.
Another way of looking at the issue is as follows. How could ‘gravitational mass’ dif-
fer from inertial mass? Would the difference depend on relative velocity, time, position,
composition or on mass itself? Each of these possibilities contradicts either energy or
momentum conservation.
Copyright © Christoph Schiller November 1997–May 2006

No wonder that all measurements confirm the equality of all mass types. The issue is
Page 395 usually resurrected in general relativity, with no new results. ‘Both’ masses remain equal;
mass is a unique property of bodies. Another issue remains, though. What is the origin of
mass? Why does it exist? This simple but deep question cannot be answered by classical
physics. We will need some patience to find out.

Dvipsbugw
 dynamics due to gravitation 135

Dvipsbugw

F I G U R E 78 Brooms fall more

rapidly than stones (© Luca
Gastaldi)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Curiosities and fun challenges about gravitation
Fallen ist weder gefährlich noch eine Schande;

“ **
Liegen bleiben ist beides.*
Konrad Adenauer

”
Gravity on the Moon is only one sixth of that on the Earth. Why does this imply that
it is difficult to walk quickly and to run on the Moon (as can be seen in the TV images
recorded there)?

**
The inverse square expression of universal gravity has a limitation: it does not allow one to
make sensible statements about the matter in the universe. Universal gravity predicts that
a homogeneous mass distribution is unstable; indeed, an inhomogeneous distribution
is observed. However, universal gravity does not predict the average mass density, the
darkness at night, the observed speeds of the distant galaxies, etc. In fact, not a single
property of the universe is predicted. To do this, we need general relativity.

**
Imagine that you have twelve coins of identical appearance, of which one is a forgery. The
forged one has a different mass from the eleven genuine ones. How can you decide which
Copyright © Christoph Schiller November 1997–May 2006

is the forged one and whether it is lighter or heavier, using a simple balance only three
Challenge 256 e times?

**
For a physicist, antigravity is repulsive gravity; it does not exist in nature. Nevertheless,
the term ‘antigravity’ is used incorrectly by many people, as a short search on the inter-
net shows. Some people call any effect that overcomes gravity, ‘antigravity’. However, this
definition implies that tables and chairs are antigravity devices. Following the definition,
* ‘Falling is neither dangerous nor a shame; to keep lying is both.’ Konrad Adenauer (b. 1876 Köln, d.
1967 Rhöndorf), German chancellor.

Dvipsbugw
 136 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw
1000 km

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 79 The starting F I G U R E 80 An honest balance?
situation for a bungee
jumper

most of the wood, steel and concrete producers are in the antigravity business. The inter-
net definition makes absolutely no sense.
**
Do all objects on Earth fall with the same acceleration of 9.8 m~s2 , assuming that air
resistance can be neglected? No; every housekeeper knows that. You can check this by
yourself. A broom angled at around 35° hits the floor before a stone, as the sounds of
Challenge 257 n impact confirm. Are you able to explain why?
**
Also Bungee jumpers are accelerated more strongly than д. For a rubber of mass m and
a jumper of mass M, the maximum acceleration a is

a = д ‹1 + (4 + ) .
1m m
(46)
8M M
Copyright © Christoph Schiller November 1997–May 2006

Challenge 258 n Can you deduce the relation from Figure 79?

**
Challenge 259 n Guess: What is the weight of a ball of cork with a radius of 1 m?
**
Challenge 260 n Guess: A heap of 1000 1 mm diameter iron balls is collected. What is its mass?

**
How can you use your observations made during your travels to show that the Earth is

Dvipsbugw
 dynamics due to gravitation 137

Challenge 261 n not flat?

**
Is the acceleration due to gravity constant? Not really. Every day, it is estimated that 108 kg
of material fall onto the Earth in the form of meteorites.
**
Both the Earth and the Moon attract bodies. The centre of mass of the Earth–Moon sys-
tem is 4800 km away from the centre of the Earth, quite near its surface. Why do bodies Dvipsbugw
Challenge 262 n on Earth still fall towards the centre of the Earth?
**
Does every spherical body fall with the same acceleration? No. If the weight of the object
is comparable to that of the Earth, the distance decreases in a different way. Can you con-
Challenge 263 ny firm this statement? What then is wrong about Galileo’s argument about the constancy

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of acceleration of free fall?

**
It is easy to lift a mass of a kilogram from the floor on a table. Twenty kilograms is harder.
Challenge 264 n A thousand is impossible. However, 6 ë 1024 kg is easy. Why?
**
The ratio of the strengths of the tides of Moon and Sun is roughly 7  3. Is it true that this
Challenge 265 ny is also the ratio between the mass densities of the two bodies?
**
The friction between the Earth and the Moon slows down the rotation of both. The Moon
stopped rotating millions of years ago, and the Earth is on its way to doing so as well.
When the Earth stops rotating, the Moon will stop moving away from Earth. How far will
Challenge 266 ny the Moon be from the Earth at that time? Afterwards however, even further in the future,
the Moon will move back towards the Earth, due to the friction between the Earth–Moon
system and the Sun. Even though this effect would only take place if the Sun burned for
Challenge 267 n ever, which is known to be false, can you explain it?
**
Copyright © Christoph Schiller November 1997–May 2006

When you run towards the east, you lose weight. There are two different reasons for this:
the ‘centrifugal’ acceleration increases so that the force with which you are pulled down
diminishes, and the Coriolis force appears, with a similar result. Can you estimate the
Challenge 268 ny size of the two effects?
**
What is the time ratio between a stone falling through a distance l and a pendulum
Challenge 269 n swinging though half a circle of radius l? (This problem is due to Galileo.) How many
digits of the number π can one expect to determine in this way?

**

Dvipsbugw
 138 i galilean motion • 2. galilean physics – motion in everyday life

Earth Moon Earth

Moon
Dvipsbugw

Sun

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 81 Which of the two Moon paths is correct?

Why can a spacecraft accelerate through the slingshot effect when going round a planet,
Challenge 270 n despite momentum conservation? It is speculated that the same effect is also the reason
for the few exceptionally fast stars that are observed in the galaxy. For example, the star
Ref. 97 HE0457-5439 moves with 720 km~s, which is much higher than the 100 to 200 km~s of
most stars in the Milky way. It seems that the role of the accelerating centre was taken by
a black hole.
**
Ref. 98 The orbit of a planet around the Sun has many interesting properties. What is the hodo-
Challenge 271 n graph of the orbit? What is the hodograph for parabolic and hyperbolic orbits?

**
A simple, but difficult question: if all bodies attract each other, why don’t or didn’t all stars
Challenge 272 n fall towards each other? Copyright © Christoph Schiller November 1997–May 2006

**
The acceleration д due to gravity at a depth of 3000 km is 10.05 m~s2 , over 2 % more than
Ref. 114 at the surface of the Earth. How is this possible? Also, on the Tibetan plateau, д is higher
than the sea level value of 9.81 m~s2 , even though the plateau is more distant from the
Challenge 273 n centre of the Earth than sea level is. How is this possible?
**
When the Moon circles the Sun, does its path have sections concave towards the Sun, as
Challenge 274 n shown at the right of Figure 81, or not, as shown on the left? (Independent of this issue,
both paths in the diagram disguise that fact that the Moon path does not lie in the same
plane as the path of the Earth around the Sun.)

Dvipsbugw
 dynamics due to gravitation 139

**
You can prove that objects attract each other (and that they are not only attracted by the
Earth) with a simple experiment that anybody can perform at home, as described on the
http://www.fourmilab.ch/gravitation/foobar/ website.
**
It is instructive to calculate the escape velocity of the Earth, i.e. that velocity with which
a body must be thrown so that it never falls back. It turns out to be 11 km~s. What is
the escape velocity for the solar system? By the way, the escape velocity of our galaxy is Dvipsbugw
129 km~s. What would happen if a planet or a system were so heavy that its escape velocity
Challenge 275 n would be larger than the speed of light?
**
For bodies of irregular shape, the centre of gravity of a body is not the same as the centre
Challenge 276 n of mass. Are you able to confirm this? (Hint: Find and use the simplest example possible.)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Can gravity produce repulsion? What happens to a small test body on the inside of a large
Challenge 277 ny C-shaped mass? Is it pushed towards the centre of mass?
**
Ref. 115 The shape of the Earth is not a sphere. As a consequence, a plumb line usually does not
Challenge 278 ny point to the centre of the Earth. What is the largest deviation in degrees?

**
Challenge 279 n What is the largest asteroid one can escape from by jumping?
**
If you look at the sky every day at 6 a.m., the Sun’s position varies during the year. The
result of photographing the Sun on the same film is shown in Figure 82. The curve, called
the analemma, is due to the inclination of the Earth’s axis, as well as the elliptical shape
of the path around the Sun. The shape of the analemma is also built into high quality
sundials. The top and the (hidden) bottom points correspond to the solstices.
**
Copyright © Christoph Schiller November 1997–May 2006

The constellation in which the Sun stands at noon (at the centre of the time zone) is sup-
posedly called the ‘zodiacal sign’ of that day. Astrologers say there are twelve of them,
namely Aries, Taurus, Gemini, Cancer, Leo, Virgo, Libra, Scorpius, Sagittarius, Capri-
cornus, Aquarius and Pisces and that each takes (quite precisely) a twelfth of a year or
a twelfth of the ecliptic. Any check with a calendar shows that at present, the midday
Sun is never in the zodiacal sign during the days usually connected to it. The relation has
Page 102 shifted by about a month since it was defined, due to the precession of the Earth’s axis. A
check with a map of the star sky shows that the twelve constellations do not have the same
length and that on the ecliptic there are fourteen of them, not twelve. There is Ophiuchus,
the snake constellation, between Scorpius and Sagittarius, and Cetus, the whale, between
Aquarius and Pisces. In fact, not a single astronomical statement about zodiacal signs is

Dvipsbugw
 140 i galilean motion • 2. galilean physics – motion in everyday life

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 82 The analemma over Delphi, between
January and December 2002 (© Anthony
Ayiomamitis)

dm
r

m
R

dM

F I G U R E 83 The vanishing of
gravitational force inside a
spherical shell of matter
Copyright © Christoph Schiller November 1997–May 2006

Ref. 116 correct. To put it clearly, astrology, in contrast to its name, is not about stars. (In some
languages, the term for ‘crook’ is derived from the word ‘astrologer’.)
**
Ref. 117 The gravitational acceleration for a particle inside a spherical shell is zero. The vanishing
of gravity in this case is independent of the particle shape and its position, and independ-
Challenge 280 n ent of the thickness of the shell.Can you find the argument using Figure 83? This works
only because of the 1~r 2 dependence of gravity. Can you show that the result does not
hold for non-spherical shells? Note that the vanishing of gravity inside a spherical shell

Dvipsbugw
 dynamics due to gravitation 141

usually does not hold if other matter is found outside the shell. How could one eliminate
Challenge 281 n the effects of outside matter?

**
For a long time, it was thought that there is no additional planet in our solar system
Ref. 118 outside Neptune and Pluto, because their orbits show no disturbances from another body.
Today, the view has changed. It is known that there are only eight planets: Pluto is not a
planet, but the first of a set of smaller objects beyond them, in the so-called Kuiper belt
and Oort cloud. (Astronomers have also agreed to continue to call Pluto a ‘planet’ despite Dvipsbugw
this evidence, to avoid debates.) Kuiper belt objects are regularly discovered. In 2003, an
object, called Sedna, was found that is almost as large as Pluto but three times farther
Ref. 119 from the Sun.
**
In astronomy new examples of motion are regularly discovered even in the present cen-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tury. Sometimes there are also false alarms. One example was the alleged fall of mini
comets on the Earth. They were supposedly made of a few dozen kilograms of ice and hit-
Ref. 120 ting the Earth every few seconds. It is now known not to happen. On the other hand, it is
known that many tons of asteroids fall on the Earth every day, in the form of tiny particles.
Incidentally, discovering objects hitting the Earth is not at all easy. Astronomers like to
point out that an asteroid as large as the one that led to the extinction of the dinosaurs
could hit the Earth without any astronomer noticing in advance, if the direction is slightly
unusual, such as from the south, where few telescopes are located.

**
Universal gravity allows only elliptical, parabolic or hyperbolic orbits. It is impossible
for a small object approaching a large one to be captured. At least, that is what we have
learned so far. Nevertheless, all astronomy books tell stories of capture in our solar system;
Challenge 282 n for example, several outer satellites of Saturn have been captured. How is this possible?

**
How would a tunnel have to be shaped in order that a stone would fall through it without
touching the walls? (Assume constant density.) If the Earth did not rotate, the tunnel
would be a straight line through its centre, and the stone would fall down and up again,
in a oscillating motion. For a rotating Earth, the problem is much more difficult. What is
Copyright © Christoph Schiller November 1997–May 2006

Challenge 283 n the shape when the tunnel starts at the Equator?

**
The International Space Station circles the Earth every 90 minutes at an altitude of about
380 km. You can see where it is from the website http://www.heavens-above.com. By the
way, whenever it is just above the horizon, the station is the third brightest object in the
Challenge 284 e night sky, superseded only by the Moon and Venus. Have a look at it.
**
Is it true that the centre of mass of the solar system, its barycentre, is always inside the
Challenge 285 n Sun? Even though a star or the Sun move very little when planets move around them,

Dvipsbugw
 142 i galilean motion • 2. galilean physics – motion in everyday life

this motion can be detected with precision measurements making use of the Doppler
Page 284 effect for light or radio waves. Jupiter, for example, produces a speed change of 13 m~s
in the Sun, the Earth 1 m~s. The first planets outside the solar system, around the pulsar
PSR1257+12 and the star Pegasi 51, was discovered in this way, in 1992 and 1995. In the
meantime, over 150 planets have been discovered with this method. So far, the smallest
planet discovered has 7 times the mass of the Earth.
**
Not all points on the Earth receive the same number of daylight hours during a year. The Dvipsbugw
Challenge 286 d effects are difficult to spot, though. Can you find one?
**
Can the phase of the Moon have a measurable effect on the human body? What about
Challenge 287 n the tidal effects of the Moon?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
There is an important difference between the heliocentric system and the old idea that
all planets turn around the Earth. The heliocentric system states that certain planets,
such as Mars and Venus, can be between the Earth and the Sun at certain times, and
behind the Sun at other times. In contrast, the geocentric system states that they are al-
ways in between. Why did such an important difference not immediately invalidate the
Challenge 288 n geocentric system?

**
The strangest reformulation of the description of motion given by ma = ∇U is the almost
Ref. 121 absurd looking equation
∇v = dv~ds (47)

where s is the motion path length. It is called the ray form of Newton’s equation of motion.
Challenge 289 n Can you find an example of its application?

**
Seen from Neptune, the size of the Sun is the same as that of Jupiter seen from the Earth
Challenge 290 n at the time of its closest approach. True? Copyright © Christoph Schiller November 1997–May 2006

**
What is gravity? This is not a simple question. In 1690, Nicolas Fatio de Duillier and in
Ref. 122 1747, Georges-Louis Lesage proposed an explanation for the 1~r 2 dependence. Lesage ar-
gued that the world is full of small particles – he called them ‘corpuscules ultra-mondains’
– flying around randomly and hitting all objects. Single objects do not feel the hits, since
they are hit continuously and randomly from all directions. But when two objects are
near to each other, they produce shadows for part of the flux to the other body, resulting
Challenge 291 ny in an attraction. Can you show that such an attraction has a 1~r 2 dependence?
However, Lesage’s proposal has a number of problems. The argument only works if the
collisions are inelastic. (Why?) However, that would mean that all bodies would heat up
Ref. 3 with time, as Jean-Marc Lévy-Leblond explains.

Dvipsbugw
 dynamics due to gravitation 143

There are two additional problems with the idea of Lesage. First, a moving body in free
space would be hit by more or faster particles in the front than in the back; as a result,
the body should be decelerated. Second, gravity would depend on size, but in a strange
way. In particular, three bodies lying on a line should not produce shadows, as no such
shadows are observed; but the naive model predicts such shadows.
Despite all the criticisms, this famous idea has regularly resurfaced in physics ever
since, even though such particles have never been found. Only in the third part of our
mountain ascent will we settle the issue.
Dvipsbugw
**
Challenge 292 ny For which bodies does gravity decrease as you approach them?
**
Could one put a satellite into orbit using a cannon? Does the answer depend on the dir-
Challenge 293 ny ection in which one shoots?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Two computer users share experiences. ‘I threw my Pentium III and Pentium IV out of
the window.’ ‘And?’ ‘The Pentium III was faster.’
**
Challenge 294 n How often does the Earth rise and fall when seen from the Moon? Does the Earth show
phases?

**
Challenge 295 ny What is the weight of the Moon? How does it compare with the weight of the Alps?
**
Owing to the slightly flattened shape of the Earth, the source of the Mississippi is about
20 km nearer to the centre of the Earth than its mouth; the water effectively runs uphill.
Challenge 296 n How can this be?
**
If a star is made of high density material, the speed of a planet orbiting near to it could
Challenge 297 n be greater than the speed of light. How does nature avoid this strange possibility?
Copyright © Christoph Schiller November 1997–May 2006

**
What will happen to the solar system in the future? This question is surprisingly hard
to answer. The main expert of this topic, French planetary scientist Jacques Laskar, simu-
Ref. 123 lated a few hundred million years of evolution using computer-aided calculus. He found
Page 259 that the planetary orbits are stable, but that there is clear evidence of chaos in the evolu-
tion of the solar system, at a small level. The various planets influence each other in subtle
and still poorly understood ways. Effects in the past are also being studied, such as the
energy change of Jupiter due to its ejection of smaller asteroids from the solar system, or
energy gains of Neptune. There is still a lot of research to be done in this field.

Dvipsbugw
 144 i galilean motion • 2. galilean physics – motion in everyday life

TA B L E 18 An unexplained property of nature: planet

distances and the values resulting from the Titius–Bode
rule

Planet n predicted measured

d i s ta n c e i n AU

Mercury −ª 0.4 0.4

Venus 0 0.7 0.7
Earth 1 1.0 1.0 Dvipsbugw
Mars 2 1.6 1.5
Planetoids 3 2.8 2.2 to 3.2
Jupiter 4 5.2 5.2
Saturn 5 10.0 9.5
Uranus 6 19.6 19.2
Neptune 7 38.8 30.1

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Pluto 8 77.2 39.5

**
One of the open problems of the solar system is the description of planet distances dis-
covered in 1766 by Johann Daniel Titius (1729–1796) and publicized by Johann Elert
Bode (1747–1826). Titius discovered that planetary distances d from the Sun can be ap-
proximated by
d = a + 2n b with a = 0.4 AU , b = 0.3 AU (48)

where distances are measured in astronomical units and n is the number of the planet.
The resulting approximation is compared with observations in Table 18.
Interestingly, the last three planets, as well as the planetoids, were discovered after
Bode’s and Titius’ deaths; the rule had successfully predicted Uranus’ distance, as well as
that of the planetoids. Despite these successes – and the failure for the last two planets –
nobody has yet found a model for the formation of the planets that explains Titius’ rule.
The large satellites of Jupiter and of Uranus have regular spacing, but not according to the
Titius–Bode rule.
Explaining or disproving the rule is one of the challenges that remains in classical
Copyright © Christoph Schiller November 1997–May 2006

Ref. 124 mechanics. Some researchers maintain that the rule is a consequence of scale invariance,
Ref. 125 others maintain that it is a accident or even a red herring. The last interpretation is also
suggested by the non-Titius–Bode behaviour of practically all extrasolar planets. The is-
sue is not closed.
**
Around 3000 years ago, the Babylonians had measured the orbital times of the seven
celestial bodies. Ordered from longest to shortest, they wrote them down in Table 19.
The Babylonians also introduced the week and the division of the day into 24 hours.
The Babylonians dedicated every one of the 168 hours of the week to a celestial body, fol-
lowing the order of the Table. They also dedicated the whole day to that celestial body

Dvipsbugw
 dynamics due to gravitation 145

TA B L E 19 The orbital
periods known to the
Babylonians

Body Period

Saturn 29 a
Jupiter 12 a
Mars 687 d
Sun 365 d Dvipsbugw
Venus 224 d
Mercury 88 d
Moon 29 d

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 84 A solar eclipse (11 August 1999, photographed from the Russian Mir
station) Copyright © Christoph Schiller November 1997–May 2006

that corresponds to the first hour of that day. The first day of the week was dedicated to
Challenge 298 e Saturn; the present ordering of the other days of the week then follows from Table 19. This
Ref. 126 story was told by Cassius Dio (c. 160 to c. 230). Towards the end of Antiquity, the order-
ing was taken up by the Roman empire. In Germanic languages, including English, the
Latin names of the celestial bodies were replaced by the corresponding Germanic gods.
The order Saturday, Sunday, Monday, Tuesday, Wednesday, Thursday and Friday is thus
a consequence of both the astronomical measurements and the astrological superstitions
of the ancients.

**

Dvipsbugw
 146 i galilean motion • 2. galilean physics – motion in everyday life

In 1722, the great mathematician Leonhard Euler made a mistake in his calculation that
led him to conclude that if a tunnel were built from one pole of the Earth to the other,
a stone falling into it would arrive at the Earth’s centre and then immediately turn and
go back up. Voltaire made fun of this conclusion for many years. Can you correct Euler
and show that the real motion is an oscillation from one pole to the other, and can you
Challenge 299 n calculate the time a pole-to-pole fall would take (assuming homogeneous density)?
What would be the oscillation time for an arbitrary straight surface-to-surface tunnel
Challenge 300 n of length l, thus not going from pole to pole?
Dvipsbugw
**
Figure 84 shows a photograph of the 1999 solar eclipse taken by the Russian space station
Mir. It clearly shows that a global view of a phenomenon can be quite different from a local
Challenge 301 n one. What is the speed of the shadow?
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In 2005, satellite measurements have shown that the water in the Amazon river presses
down the land up to 75 mm more in the season when it is full of water than in the season
Ref. 127 when it is almost empty.

What is cl assical mechanics?

All types of motion that can be described when the mass of a body is its only permanent
property form what is called mechanics. The same name is also given to the experts study-
ing the field. We can think of mechanics as the athletic part of physics;* both in athletics
and in mechanics only lengths, times and masses are measured.
More specifically, our topic of investigation so far is called classical mechanics, to dis-
tinguish it from quantum mechanics. The main difference is that in classical physics ar-
bitrary small values are assumed to exist, whereas this is not the case in quantum physics.
The use of real numbers for observable quantities is thus central to classical physics.
Classical mechanics is often also called Galilean physics or Newtonian physics. The basis
of classical mechanics, the description of motion using only space and time, is called kin-
ematics. An example is the description of free fall by z(t) = z 0 + v 0 (t − t 0 ) − 21 д(t − t 0 )2 .
The other, main part of classical mechanics is the description of motion as a consequence
of interactions between bodies; it is called dynamics. An example of dynamics is the for-
mula of universal gravity.
Copyright © Christoph Schiller November 1997–May 2006

The distinction between kinematics and dynamics can also be made in relativity, ther-
modynamics and electrodynamics. Even though we have not explored these fields of en-
quiry yet, we know that there is more to the world than gravity. A simple observation
makes the point: friction. Friction cannot be due to gravity, because friction is not ob-
served in the skies, where motion follows gravity rules only.** Moreover, on Earth, fric-

* This is in contrast to the actual origin of the term ‘mechanics’, which means ‘machine science’. It derives
from the Greek µηκανή, which means ‘machine’ and even lies at the origin of the English word ‘machine’
itself. Sometimes the term ‘mechanics’ is used for the study of motion of solid bodies only, excluding, e.g.,
hydrodynamics. This use fell out of favour in physics in the last century.
** This is not completely correct: in the 1980s, the first case of gravitational friction was discovered: the
Page 402 emission of gravity waves. We discuss it in detail later.

Dvipsbugw
 what is cl assical mechanics? 147

TA B L E 20 Some measured force values

O b s e r va t i o n Force

Value measured in a magnetic resonance force microscope 820 zN

Maximum force exerted by teeth 1.6 kN
Typical force exerted by sledgehammer 2 kN
Force exerted by quadricpes up to 3 kN
Force sustained by 1 cm2 of a good adhesive up to 10 kN Dvipsbugw
Force needed to tear a good rope used in rock climbing 30 kN
Maximum force measurable in nature 3.0 ë 1043 N

Challenge 302 e tion is independent of gravity, as you might want to check. There must be another inter-
action responsible for friction. We shall study it shortly. But one issue merits a discussion

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

right away.

Should one use force?

The direct use of force is such a poor solution to

“ any problem, it is generally employed only by

small children and large nations.
David Friedman

Everybody has to take a stand on this question, even students of physics. Indeed, many
types of forces are used and observed in daily life. One speaks of muscular, gravitational,
psychic, sexual, satanic, supernatural, social, political, economic and many others. Physi-
”
cists see things in a simpler way. They call the different types of forces observed between
objects interactions. The study of the details of all these interactions will show that, in
everyday life, they are of electrical origin.
For physicists, all change is due to motion. The term force then also takes on a more
restrictive definition. (Physical) force is defined as the change of momentum, i.e. as

F=
dp
. (49)
dt
Force is the change or flow of motion. If a force acts on a body, momentum flows into
Copyright © Christoph Schiller November 1997–May 2006

it. Indeed, momentum can be imagined to be some invisible and intangible liquid. Force
measures how much of this liquid flows from one body to another per unit time.
Using the Galilean definition of linear momentum p = mv, we can rewrite the defini-
tion of force (for constant mass) as
F = ma , (50)

where F = F(t, x) is the force acting on an object of mass m and where a = a(t, x) =
dv~dt = d2 x~dt 2 is the acceleration of the same object, that is to say its change of velocity.*

* This equation was first written down by the Swiss mathematician and physicist Leonhard Euler (1707–
1783) in 1747, over 70 years after Newton’s first law and 20 years after the death of Newton, to whom it is

Dvipsbugw
 148 i galilean motion • 2. galilean physics – motion in everyday life

The expression states in precise terms that force is what changes the velocity of masses. The
quantity is called ‘force’ because it corresponds in many, but not all aspects to muscular
force. For example, the more force is used, the further a stone can be thrown.
However, whenever the concept of force is used, it should be remembered that physical
force is different from everyday force or everyday effort. Effort is probably best approxim-
ated by the concept of (physical) power, usually abbreviated P, and defined (for constant
force) as
P= =Fëv
dW
(51) Dvipsbugw
dt
in which (physical) work W is defined as W = F ë s. Physical work is a form of energy, as
you might want to check. Note that a man who walks carrying a heavy rucksack is hardly
Challenge 303 n doing any work; why then does he get tired? Work, as a form of energy, has to be taken
into account when the conservation of energy is checked.
Challenge 304 n With the definition of work just given you can solve the following puzzles. What hap-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pens to the electricity consumption of an escalator if you walk on it instead of standing
Challenge 305 d still? What is the effect of the definition of power for the salary of scientists?
When students in exams say that the force acting on a thrown stone is least at the
Ref. 128 highest point of the trajectory, it is customary to say that they are using an incorrect
view, namely the so-called Aristotelian view, in which force is proportional to velocity.
Sometimes it is even said that they are using a different concept of state of motion. Critics
then add, with a tone of superiority, how wrong all this is. This is a typical example of
intellectual disinformation. Every student knows from riding a bicycle, from throwing
a stone or from pulling an object that increased effort results in increased speed. The
student is right; those theoreticians who deduce that the student has a mistaken concept
of force are wrong. In fact, instead of the physical concept of force, the student is just
using the everyday version, namely effort. Indeed, the effort exerted by gravity on a flying
stone is least at the highest point of the trajectory. Understanding the difference between
physical force and everyday effort is the main hurdle in learning mechanics.*
Often the flow of momentum, equation (49), is not recognized as the definition of
force. This is mainly due to an everyday observation: there seem to be forces without
any associated acceleration or change in momentum, such as in a string under tension
or in water at high pressure. When one pushes against a tree, there is no motion, yet a
force is applied. If force is momentum flow, where does the momentum go? It flows into
the slight deformations of the arms and the tree. In fact, when one starts pushing and
Copyright © Christoph Schiller November 1997–May 2006

thus deforming, the associated momentum change of the molecules, the atoms, or the
electrons of the two bodies can be observed. After the deformation is established, and
looking at even higher magnification, one can indeed find that a continuous and equal
flow of momentum is going on in both directions. The nature of this flow will be clarified

usually and falsely ascribed; it was Euler, not Newton, who first understood that this definition of force is
useful in every case of motion, whatever the appearance, be it for point particles or extended objects, and be
Ref. 24 it rigid, deformable or fluid bodies. Surprisingly and in contrast to frequently-made statements, equation
(50) is even correct in relativity, as shown on page 323.
* This stepping stone is so high that many professional physicists do not really take it themselves; this is
confirmed by the innumerable comments in papers that state that physical force is defined using mass, and, at
the same time, that mass is defined using force (the latter part of the sentence being a fundamental mistake).

Dvipsbugw
 what is cl assical mechanics? 149

in the part on quantum theory.

As force is net momentum flow, it is only needed as a separate concept in everyday life,
where it is useful in situations where net momentum flows are less than the total flows.
At the microscopic level, momentum alone suffices for the description of motion. For
example, the concept of weight describes the flow of momentum due to gravity. Thus we
will hardly ever use the term ‘weight’ in the microscopic part of our adventure.
Through its definition, the concept of force is distinguished clearly from ‘mass’, ‘mo-
mentum’, ‘energy’ and ‘power’. But where do forces originate? In other words, which ef-
fects in nature have the capacity to accelerate bodies by pumping momentum into ob- Dvipsbugw
jects? Table 21 gives an overview.
Every example of motion, from the one that lets us choose the direction of our gaze
to the one that carries a butterfly through the landscape, can be put into one of the two
left-most columns of Table 21. Physically, the two columns are separated by the following
criterion: in the first class, the acceleration of a body can be in a different direction from
its velocity. The second class of examples produces only accelerations that are exactly

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

opposed to the velocity of the moving body, as seen from the frame of reference of the
braking medium. Such a resisting force is called friction, drag or a damping. All examples
Challenge 307 e in the second class are types of friction. Just check.
Friction can be so strong that all motion of a body against its environment is made
impossible. This type of friction, called static friction or sticking friction, is common and
important: without it, turning the wheels of bicycles, trains or cars would have no effect.
Not a single screw would stay tightened. We could neither run nor walk in a forest, as
the soil would be more slippery than polished ice. In fact not only our own motion, but
all voluntary motion of living beings is based on friction. The same is the case for self-
moving machines. Without static friction, the propellers in ships, aeroplanes and heli-
copters would not have any effect and the wings of aeroplanes would produce no lift to
Challenge 308 n keep them in the air. (Why?) In short, static friction is required whenever we want to
move relative to our environment.
Once an object moves through its environment, it is hindered by another type of fric-
tion; it is called dynamic friction and acts between bodies in relative motion. Without it,
Ref. 129 falling bodies would always rebound to the same height, without ever coming to a stop;
neither parachutes nor brakes would work; worse, we would have no memory, as we will
see later.*
As the motion examples in the second column of Table 21 include friction, in those
examples macroscopic energy is not conserved: the systems are dissipative. In the first
Copyright © Christoph Schiller November 1997–May 2006

column, macroscopic energy is constant: the systems are conservative.

The first two columns can also be distinguished using a more abstract, mathematical
criterion: on the left are accelerations that can be derived from a potential, on the right,
decelerations that can not. As in the case of gravitation, the description of any kind of
motion is much simplified by the use of a potential: at every position in space, one needs
only the single value of the potential to calculate the trajectory of an object, instead of the
three values of the acceleration or the force. Moreover, the magnitude of the velocity of

* Recent research suggest that maybe in certain crystalline systems, such as tungsten bodies on silicon, under
ideal conditions gliding friction can be extremely small and possibly even vanish in certain directions of
Ref. 130 motion. This so-called superlubrication is presently a topic of research.

Dvipsbugw
150 i galilean motion • 2. galilean physics – motion in everyday life

TA B L E 21 Selected processes and devices changing the motion of bodies

S i t uat i o n s t h at c a n S i t uat i o n s t h at Motors and actu-

l e a d t o a c c e l e r at i o n only l ea d to de - at o r s
c e l e r at i o n
piezoelectricity
quartz under applied voltage thermoluminescence walking piezo tripod Dvipsbugw
gravitation
falling emission of gravity waves pulley
collisions
satellite in planet encounter car crash rocket motor
growth of mountains meteorite crash swimming of larvae

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

magnetic effects
compass needle near magnet electromagnetic braking electromagnetic gun
magnetostriction transformer losses linear motor
current in wire near magnet electric heating galvanometer
electric effects
rubbed comb near hair friction between solids electrostatic motor
bombs fire muscles, sperm flagella
television tube electron microscope Brownian motor
light
levitating objects by light light bath stopping atoms (true) light mill
solar sail for satellites light pressure inside stars solar cell
elasticity
bow and arrow trouser suspenders ultrasound motor
bent trees standing up again pillow, air bag bimorphs
osmosis
water rising in trees salt conservation of food osmotic pendulum
electro-osmosis tunable X-ray screening
heat & pressure
Copyright © Christoph Schiller November 1997–May 2006

freezing champagne bottle surfboard water resistance hydraulic engines

tea kettle quicksand steam engine
barometer parachute air gun, sail
earthquakes sliding resistance seismometer
attraction of passing trains shock absorbers water turbine
nuclei
radioactivity plunging into the Sun supernova explosion
biology
bamboo growth find example! Challenge 306 ny molecular motors

Dvipsbugw
 what is cl assical mechanics? 151

an object at any point can be calculated directly from energy conservation.

The processes from the second column cannot
be described by a potential. These are the cases ideal shape, cw = 0.0168
where we necessarily have to use force if we want
to describe the motion of the system. For example,
the force F due to wind resistance of a body is
roughly given by
typical passenger aeroplane, cw = 0.03
Dvipsbugw
F = c w ρAv 2
1
(52)
2
where A is the area of its cross-section and v its
velocity relative to the air, ρ is the density of air; typical sports car, c = 0.44
w
the drag coefficient c w is a pure number that de-
pends on the shape of the moving object. (A few ex-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

amples are given in Figure 85.) You may check that
Challenge 309 ny aerodynamic resistance cannot be derived from a
dolphin
potential.*
The drag coefficient c w is found experiment-
ally to be always larger than 0.0168, which corres-
F I G U R E 85 Shapes and air/water
ponds to the optimally streamlined tear shape. An resistance
aerodynamic car has a value of 0.25 to 0.3; but
many sports cars share with vans values of 0.44
and higher.**
Ref. 132 Wind resistance is also of importance to humans, in particular in athletics. It is estim-
ated that 100 m sprinters spend between 3 % and 6 % of their power overcoming drag.
Challenge 311 ny This leads to varying sprint times t w when wind of speed w is involved, related by the
expression
wt w 2
= 1.03 − 0.03 ‹1 −  ,
t0
(53)
tw 100

where the more conservative estimate of 3 % is used. An opposing wind speed of −2 m~s
gives an increase in time of 0.13 s, enough to change a potential world record into an
‘only’ excellent result. (Are you able to deduce the c w value for running humans from the
Challenge 312 ny formula?)
Copyright © Christoph Schiller November 1997–May 2006

Likewise, parachuting exists due to wind resistance. Can you determine how the speed
Challenge 313 n of a falling body changes with time, assuming constant shape and drag coefficient?
* Such a statement about friction is correct only in three dimensions, as is the case in nature; in the case of
Challenge 310 n a single dimension, a potential can always be found.
** Calculating drag coefficients in computers, given the shape of the body and the properties of the fluid, is
one of the most difficult tasks of science; the problem is still not fully solved.
The topic of aerodynamic shapes is even more interesting for fluid bodies. They are kept together by
surface tension. For example, surface tension keeps the hairs of a wet brush together. Surface tension also
determines the shape of rain drops. Experiments show that it is spherical for drops smaller than 2 mm
diameter, and that larger rain drops are lens shaped, with the flat part towards the bottom. The usual tear
Ref. 131 shape is not encountered in nature; something vaguely similar to it appears during drop detachment, but
never during drop fall.

Dvipsbugw
 152 i galilean motion • 2. galilean physics – motion in everyday life

In contrast, static friction has different properties. It is proportional to the force press-
Ref. 133 ing the two bodies together. Why? Studying the situation in more detail, sticking friction
is found to be proportional to the actual contact area. It turns out that putting two solids
into contact is rather like turning Switzerland upside down and putting it onto Austria;
the area of contact is much smaller than that estimated macroscopically. The important
point is that the area of actual contact is proportional to the normal force. The study of
what happens in that contact area is still a topic of research; researchers are investigating
the issues using instruments such as atomic force microscopes, lateral force microscopes
and triboscopes. These efforts resulted in computer hard discs which last longer, as the Dvipsbugw
friction between disc and the reading head is a central quantity in determining the life-
time.
All forms of friction are accompanied by an increase in the temperature of the mov-
ing body. The reason became clear after the discovery of atoms. Friction is not observed
in few – e.g. 2, 3, or 4 – particle systems. Friction only appears in systems with many
particles, usually millions or more. Such systems are called dissipative. Both the tem-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

perature changes and friction itself are due to the motion of large numbers of micro-
scopic particles against each other. This motion is not included in the Galilean descrip-
tion. When it is included, friction and energy loss disappear, and potentials can then be
used throughout. Positive accelerations – of microscopic magnitude – then also appear,
and motion is found to be conserved. As a result, all motion is conservative on a micro-
scopic scale. Therefore, on a microscopic scale it is possible to describe all motion without
the concept of force.* The moral of the story is that one should use force only in one situ-
ation: in the case of friction, and only when one does not want to go into the microscopic
details.**
Et qu’avons-nous besoin de ce moteur, quand

“ l’étude réfléchie de la nature nous prouve que le

mouvement perpétuel est la première de ses
lois ?***
Donatien de Sade Justine, ou les malheurs de la
vertu.

Complete states – initial conditions

Quid sit futurum cras, fuge quaerere ...****
”
“ Horace, Odi, lib. I, ode 9, v. 13.

”
* The first scientist who eliminated force from the description of nature was Heinrich Rudolf Hertz (b. 1857
Hamburg, d. 1894 Bonn), the famous discoverer of electromagnetic waves, in his textbook on mechanics,
Copyright © Christoph Schiller November 1997–May 2006

Die Prinzipien der Mechanik, Barth, 1894, republished by Wissenschaftliche Buchgesellschaft, Darmstadt,
1963. His idea was strongly criticized at that time; only a generation later, when quantum mechanics quietly
got rid of the concept for good, did the idea become commonly accepted. (Many have speculated about
the role Hertz would have played in the development of quantum mechanics and general relativity, had he
not died so young.) In his book, Hertz also formulated the principle of the straightest path: particles follow
geodesics. This same description is one of the pillars of general relativity, as we will see later on.
** In the case of human relations the evaluation should be somewhat more discerning, as the research by
Ref. 134 James Gilligan shows.
*** ‘And whatfor do we need this motor, when the reasoned study of nature proves to us that perpetual
motion is the first of its laws?’
Ref. 56 **** ‘What future will be tomorrow, never ask ...’ Horace is Quintus Horatius Flaccus (65–8 bce ), the great
Roman poet.

Dvipsbugw
 what is cl assical mechanics? 153

We often describe the motion of a body by specifying the time dependence of its position,
for example as

x(t) = x0 + v0 (t − t 0 ) + 21 a0 (t − t 0 )2 + 61 j0 (t − t 0 )3 + ... . (54)

The quantities with an index 0, such as the starting position x0 , the starting velocity v0 ,
etc., are called initial conditions. Initial conditions are necessary for any description of mo-
tion. Different physical systems have different initial conditions. Initial conditions thus Dvipsbugw
specify the individuality of a given system. Initial conditions also allow us to distinguish
the present situation of a system from that at any previous time: initial conditions specify
the changing aspects of a system. In other words, they summarize the past of a system.
Page 39 Initial conditions are thus precisely the properties we have been seeking for a descrip-
tion of the state of a system. To find a complete description of states we thus need only a
complete description of initial conditions. It turns out that for gravitation, as for all other

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

microscopic interactions, there is no need for initial acceleration a0 , initial jerk j0 , or
higher-order initial quantities. In nature, acceleration and jerk depend only on the prop-
erties of objects and their environment; they do not depend on the past. For example,
the expression a = GM~r 2 , giving the acceleration of a small body near a large one, does
not depend on the past, but only on the environment. The same happens for the other
fundamental interactions, as we will find out shortly.
Page 71 The complete state of a moving mass point is thus described by specifying its position
and its momentum at all instants of time. Thus we have achieved a complete description
of the intrinsic properties of point objects, namely by their mass, and of their states of mo-
tion, namely by their momentum, energy, position and time. For extended rigid objects
we also need orientation, angular velocity and angular momentum. Can you specify the
Challenge 314 ny necessary quantities in the case of extended elastic bodies or fluids?
The set of all possible states of a system is given a special name: it is called the phase
space. We will use the concept repeatedly. Like any space, it has a number of dimensions.
Challenge 315 n Can you specify it for a system consisting of N point particles?
However, there are situations in nature where the motion of an object depends on
characteristics other than its mass; motion can depend on its colour (can you find an
Challenge 316 n example?), on its temperature, and on a few other properties that we will soon discover.
Challenge 317 n Can you give an example of an intrinsic property that we have so far missed? And for
each intrinsic property there are state variables to discover. These new properties are the
Copyright © Christoph Schiller November 1997–May 2006

basis of the field of physical enquiry beyond mechanics. We must therefore conclude that
as yet we do not have a complete description of motion.
It is interesting to recall an older challenge and ask again: does the universe have initial
Challenge 318 n conditions? Does it have a phase space? As a hint, recall that when a stone is thrown, the
initial conditions summarize the effects of the thrower, his history, the way he got there
etc.; in other words, initial conditions summarize the effects that the environment had
during the history of a system.

An optimist is somebody who thinks that the

“ future is uncertain.

”
Dvipsbugw
 154 i galilean motion • 2. galilean physics – motion in everyday life

Do surprises exist? Is the future determined?

Die Ereignisse der Zukunft können wir nicht aus

“ den gegenwärtigen erschließen. Der Glaube an

den Kausalnexus ist ein Aberglaube.*
Ludwig Wittgenstein, Tractatus, 5.1361

“
Freedom is the recognition of necessity.
Friedrich Engels (1820–1895) ”
If, after climbing a tree, we jump down, we cannot halt the jump in the middle of the
trajectory; once the jump has begun, it is unavoidable and determined, like all passive
” Dvipsbugw

motion. However, when we begin to move an arm, we can stop or change its motion
from a hit to a caress. Voluntary motion does not seem unavoidable or predetermined.
Challenge 319 e Which of these two cases is the general one?
Let us start with the example that we can describe most precisely so far: the fall of a
body. Once the potential φ acting on a particle is given and taken into account, using

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

a(x) = −∇φ = −GMr~r 3 , (55)

and the state at a given time is given by initial conditions such as

x(t 0 ) = x 0 and v(t 0 ) = v 0 , (56)

we then can determine the motion in advance. The complete trajectory x(t) can be cal-
culated with these two pieces of information. Owing to this possibility, an equation such
as (55) is called an evolution equation for the motion of the object. (Note that the term
‘evolution’ has different meanings in physics and in biology.) An evolution equation al-
ways expresses the observation that not all types of change are observed in nature, but
only certain specific cases. Not all imaginable sequences of events are observed, but only
a limited number of them. In particular, equation (55) expresses that from one instant to
the next, objects change their motion based on the potential acting on them. Thus, given
an evolution equation and initial state, the whole motion of a system is uniquely fixed;
this property of motion is often called determinism. Since this term is often used with
different meanings, let us distinguish it carefully from several similar concepts, to avoid
misunderstandings.
Motion can be deterministic and at the same time still be unpredictable. The latter
Copyright © Christoph Schiller November 1997–May 2006

property can have four origins: an impracticably large number of particles involved, the
complexity of the evolution equations, insufficient information about initial conditions,
and strange shapes of space-time. The weather is an example where the first three condi-
tions are fulfilled at the same time.** Nevertheless, its motion is still deterministic. Near
black holes all four cases apply together. We will discuss black holes in the section on
general relativity. Nevertheless, near black holes, motion is still deterministic.
Motion can be both deterministic and time random, i.e. with different outcomes in

* We cannot infer the events of the future from those of the present. Superstition is nothing but belief in the
causal nexus.
** For a beautiful view of clouds, see the http://www.goes.noass.gov website.

Dvipsbugw
 what is cl assical mechanics? 155

similar experiments. A roulette ball’s motion is deterministic, but it is also random.* As

we will see later, quantum-mechanical situations fall into this category, as do all examples
of irreversible motion, such as a drop of ink spreading out in clear water. In all such
cases the randomness and the irreproducibility are only apparent; they disappear when
the description of states and initial conditions in the microscopic domain are included.
In short, determinism does not contradict (macroscopic) irreversibility. However, on the
microscopic scale, deterministic motion is always reversible.
A final concept to be distinguished from determinism is acausality. Causality is the
requirement that a cause must precede the effect. This is trivial in Galilean physics, but Dvipsbugw
becomes of importance in special relativity, where causality implies that the speed of light
is a limit for the spreading of effects. Indeed, it seems impossible to have deterministic
motion (of matter and energy) which is acausal, i.e. faster than light. Can you confirm
Challenge 320 n this? This topic will be looked at more deeply in the section on special relativity.
Saying that motion is ‘deterministic’ means that it is fixed in the future and also in
the past. It is sometimes stated that predictions of future observations are the crucial test

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

for a successful description of nature. Owing to our often impressive ability to influence
the future, this is not necessarily a good test. Any theory must, first of all, describe past
observations correctly. It is our lack of freedom to change the past that results in our lack
of choice in the description of nature that is so central to physics. In this sense, the term
‘initial condition’ is an unfortunate choice, because it automatically leads us to search for
the initial condition of the universe and to look there for answers to questions that can be
answered without that knowledge. The central ingredient of a deterministic description is
that all motion can be reduced to an evolution equation plus one specific state. This state
can be either initial, intermediate, or final. Deterministic motion is uniquely specified
into the past and into the future.
To get a clear concept of determinism, it is useful to remind ourselves why the concept
of ‘time’ is introduced in our description of the world. We introduce time because we ob-
serve first that we are able to define sequences in observations, and second, that unrestric-
ted change is impossible. This is in contrast to films, where one person can walk through
a door and exit into another continent or another century. In nature we do not observe
metamorphoses, such as people changing into toasters or dogs into toothbrushes. We are
able to introduce ‘time’ only because the sequential changes we observe are extremely re-
Challenge 321 n stricted. If nature were not reproducible, time could not be used. In short, determinism
expresses the observation that sequential changes are restricted to a single possibility.
Since determinism is connected to the use of the concept of time, new questions arise
Copyright © Christoph Schiller November 1997–May 2006

whenever the concept of time changes, as happens in special relativity, in general relativity
and in theoretical high energy physics. There is a lot of fun ahead.
In summary, every description of nature that uses the concept of time, such as that of
everyday life, that of classical physics and that of quantum mechanics, is intrinsically and
inescapably deterministic, since it connects observations of the past and the future, elim-
inating alternatives. In short, the use of time implies determinism, and vice versa. When

* Mathematicians have developed a large number of tests to determine whether a collection of numbers may
be called random; roulette results pass all these tests – in honest casinos only, however. Such tests typically
check the equal distribution of numbers, of pairs of numbers, of triples of numbers, etc. Other tests are the
Ref. 135 χ 2 test, the Monte Carlo test(s), and the gorilla test.

Dvipsbugw
 156 i galilean motion • 2. galilean physics – motion in everyday life

drawing metaphysical conclusions, as is so popular nowadays when discussing quantum

Page 831 theory, one should never forget this connection. Whoever uses clocks but denies determ-
inism is nurturing a split personality!*
The idea that motion is determined often produces fear, because we are taught to asso-
ciate determinism with lack of freedom. On the other hand, we do experience freedom
in our actions and call it free will. We know that it is necessary for our creativity and for
our happiness. Therefore it seems that determinism is opposed to happiness.
But what precisely is free will? Much ink has been consumed trying to find a precise
definition. One can try to define free will as the arbitrariness of the choice of initial con- Dvipsbugw
ditions. However, initial conditions must themselves result from the evolution equations,
so that there is in fact no freedom in their choice. One can try to define free will from the
idea of unpredictability, or from similar properties, such as uncomputability. But these
definitions face the same simple problem: whatever the definition, there is no way to prove
experimentally that an action was performed freely. The possible definitions are useless.
In short, free will cannot be observed. (Psychologists also have a lot of their own data to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

support this, but that is another topic.)
No process that is gradual – in contrast to sudden – can be due to free will; gradual
processes are described by time and are deterministic. In this sense, the question about
free will becomes one about the existence of sudden changes in nature. This will be a re-
curring topic in the rest of this walk. Does nature have the ability to surprise? In everyday
life, nature does not. Sudden changes are not observed. Of course, we still have to invest-
igate this question in other domains, in the very small and in the very large. Indeed, we
will change our opinion several times. The lack of surprises in everyday life is built deep
into our body: the concept of curiosity is based on the idea that everything discovered is
useful afterwards. If nature continually surprised us, curiosity would make no sense.
Another observation contradicts the existence of surprises: in the beginning of our
walk we defined time using the continuity of motion; later on we expressed this by saying
that time is a consequence of the conservation of energy. Conservation is the opposite of
surprise. By the way, a challenge remains: can you show that time would not be definable
Challenge 322 n even if surprises existed only rarely?
In summary, so far we have no evidence that surprises exist in nature. Time exists be-
cause nature is deterministic. Free will cannot be defined with the precision required by
physics. Given that there are no sudden changes, there is only one consistent definition of
free will: it is a feeling, in particular of independence of others, of independence from fear
Ref. 136 and of accepting the consequences of one’s actions. Free will is a feeling of satisfaction.
Copyright © Christoph Schiller November 1997–May 2006

This solves the apparent paradox; free will, being a feeling, exists as a human experience,
even though all objects move without any possibility of choice. There is no contradic-
tion.**

* That can be a lot of fun though.

** That free will is a feeling can also be confirmed by careful introspection. The idea of free will always
appears after an action has been started. It is a beautiful experiment to sit down in a quiet environment, with
the intention to make, within an unspecified number of minutes, a small gesture, such as closing a hand. If
Challenge 323 e you carefully observe, in all detail, what happens inside yourself around the very moment of decision, you
find either a mechanism that led to the decision, or a diffuse, unclear mist. You never find free will. Such an
experiment is a beautiful way to experience deeply the wonders of the self. Experiences of this kind might
also be one of the origins of human spirituality, as they show the connection everybody has with the rest of

Dvipsbugw
 what is cl assical mechanics? 157

Ref. 137 Even if human action is determined, it is still authentic. So why is determinism so
frightening? That is a question everybody has to ask themself. What difference does de-
Challenge 324 e terminism imply for your life, for the actions, the choices, the responsibilities and the
pleasures you encounter?* If you conclude that being determined is different from being
free, you should change your life! Fear of determinism usually stems from refusal to take
the world the way it is. Paradoxically, it is precisely he who insists on the existence of free
will who is running away from responsibility.
You do have the ability to surprise yourself.

A strange summary about motion

“ Richard Bandler and John Grinder

”
Dvipsbugw

Darum kann es in der Logik auch nie

“ Überraschungen geben.**
Ludwig Wittgenstein, Tractatus, 6.1251

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Classical mechanics describes nature in a rather simple way. Objects are permanent and
massive entities localized in space-time. States are changing properties of objects, de-
scribed by position in space and instant in time, by energy and momentum, and by their
rotational equivalents. Time is the relation between events measured by a clock. Clocks
are devices in undisturbed motion whose position can be observed. Space and position
is the relation between objects measured by a metre stick. Metre sticks are devices whose
shape is subdivided by some marks, fixed in an invariant and observable manner. Motion
is change of position with time (times mass); it is determined, does not show surprises,
is conserved (even in death), and is due to gravitation and other interactions.
Even though this description works rather well, it contains a circular definition. Can
Challenge 326 n you spot it? Each of the two central concepts of motion is defined with the help of the
other. Physicists worked for about 200 years on classical mechanics without noticing or
wanting to notice the situation. Even thinkers with an interest in discrediting science
Challenge 327 n did not point it out. Can an exact science be based on a circular definition? Obviously,
physics has done quite well so far. Some even say the situation is unavoidable in principle.
Despite these opinions, undoing this logical loop is one of the aims of the rest of our walk.
To achieve it, we need to increase substantially the level of precision in our description
of motion.
Whenever precision is increased, imagination is restricted. We will discover that many
types of motion that seem possible are not. Motion is limited. Nature limits speed, size,
Copyright © Christoph Schiller November 1997–May 2006

acceleration, mass, force, power and many other quantities. Continue reading only if you
are prepared to exchange fantasy for precision. It will be no loss, as you will gain some-
thing else: the workings of nature will fascinate you.

nature.
Challenge 325 n * If nature’s ‘laws’ are deterministic, are they in contrast with moral or ethical ‘laws’? Can people still be held
responsible for their actions?
** Hence there can never be surprises in logic.

Dvipsbugw
 158 i galilean motion

Biblio graphy
Aiunt enim multum legendum esse, non multa.

“ Plinius, Epistulae.*

1 For a history of science in antiquity, see Lucio Russo, La rivoluzione dimenticata, Fel-
”
trinelli 1996, also available in several other languages. Cited on page 28.
2 An overview of motion illusions can be found on the excellent website http://wwww. Dvipsbugw
mittelbach.de/ot. Cited on page 30.
3 A beautiful book explaining physics and its many applications in nature and technology
vividly and thoroughly is Paul G. Hewitt, John Suchocki & Leslie A. Hewitt,
Conceptual Physical Science, Bejamin/Cummings, 1999.
A book famous for its passion for curiosity is R ichard P. Feynman, R.B. Leighton
& M. Sands, The Feynman Lectures on Physics, Addison Wesley, 1977.
A lot can be learned about motion from quiz books. One of the best is the well-structured

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

collection of beautiful problems that require no mathematics, written by Jean-Marc
Lév y-Leblond, La physique en questions – mécanique, Vuibert, 1998.
Another excellent quiz collection is Yakov Perelman, Oh, la physique, Dunod, 2000,
a translation from the Russian original.
A good problem book is W.G. R ees, Physics by Example: 200 Problems and Solutions,
Cambridge University Press, 1994.
A good history of physical ideas is given in the excellent text by David Park, The How
and the Why, Princeton University Press, 1988.
An excellent introduction into physics, is Robert Pohl, Pohl’s Einführung in die Physik,
Klaus Lüders & Robert O. Pohl editors, Springer, 2004, in two volumes with CDs. It is a new
edition of a book that is over 70 years old; but the didactic quality, in particular of the exper-
imental side of physics, is unsurpassed. Cited on pages 29, 91, 142, and 217.
4 A well-known principle in the social sciences states that, given a question, for every possible
answer, however weird it may seem, there is somebody – and often a whole group – who
holds it as his opinion. One just has to go through literature (or the internet) to confirm
this.
About group behaviour in general, see R. Axelrod, The Evolution of Cooperation,
Harper Collins, 1984. The propagation and acceptance of ideas, such as those of physics,
are also an example of human cooperation, with all its potential dangers and weaknesses.
Cited on page 30.
5 All the known texts by Parmenides and Heraclitos can be found in Jean-Paul Dumont,
Copyright © Christoph Schiller November 1997–May 2006

Les écoles présocratiques, Folio-Gallimard, 1988. Views about the non-existence of motion
have also been put forward by much more modern and much more contemptible authors,
Page 678 such as in 1710 by Berkeley. Cited on page 30.
6 An example of people worried by Zeno is given by William McL aughlin, Resolving
Zeno’s paradoxes, Scientific American pp. 66–71, November 1994. The actual argument was
not about a hand slapping a face, but about an arrow hitting the target. See also Ref. 44.
Cited on page 30.

* ‘Read much, but not anything.’ Ep. 7, 9, 15. Gaius Plinius Secundus (b. 23/4 Novum Comum, d.
79 Vesuvius eruption), Roman writer, especially famous for his large, mainly scientific work Historia nat-
uralis, which has been translated and read for almost 2000 years.

Dvipsbugw
bibliography 159

7 The full text of La Beauté and the other poems from Les fleurs du mal, one of the finest books
of poetry ever written, can be found at the http://hypermedia.univ-paris8.fr/bibliotheque/
Baudelaire/Spleen.html website. Cited on page 32.
8 The most famous text is Jearl Walker, The Flying Circus of Physics, Wiley, 1975. For
more interesting physical effects in everyday life, see Erwein Fl achsel, Hundertfünfzig
Physikrätsel, Ernst Klett Verlag, 1985. The book also covers several clock puzzles, in puzzle
numbers 126 to 128. Cited on page 32.
9 A concise and informative introduction into the history of classical physics is given in the
first chapter of the book by F.K. R ichtmeyer, E.H. Kennard & J.N. Cooper, Intro- Dvipsbugw
duction to Modern Physics, McGraw–Hill, 1969. Cited on page 33.
10 A good overview over the arguments used to prove the existence of god from motion is given
by Michael Buckley, Motion and Motion’s God, Princeton University Press, 1971. The
intensity of the battles waged around these failed attempts is one of the tragicomic chapters
of history. Cited on page 33.
11 Thomas Aquinas, Summa Theologiae or Summa Theologica, 1265–1273, online in Latin

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

at http://www.newadvent.org/summa, in English on several other servers. Cited on page
33.
12 For an exploration of ‘inner’ motions, see the beautiful text by R ichard Schwartz, In-
ternal Family Systems Therapy, The Guilford Press, 1995. Cited on page 33.
13 See e.g. the fascinating text by David G. Chandler, The Campaigns of Napoleon - The
Mind and Method of History’s Greatest Soldier, Macmillan, 1966. Cited on page 33.
14 R ichard Marcus, American Roulette, St Martin’s Press, 2003, a thriller and a true story.
Cited on page 33.
15 A good and funny book on behaviour change is the well-known text by R. Bandler, Using
Your Brain for a Change, Real People Press, 1985. See also R ichard Bandler & John
Grinder, Frogs into princes – Neuro Linguistic Programming, Eden Grove Editions, 1990.
Cited on pages 34 and 41.
16 A beautiful book about the mechanisms of human growth from the original cell to full size
is Lewis Wolpert, The Triumph of the Embryo, Oxford University Press, 1991. Cited on
page 34.
17 On the topic of grace and poise, see e.g. the numerous books on the Alexander technique,
such as M. Gelb, Body Learning – An Introduction to the Alexander Technique, Aurum
Press, 1981, and R ichard Brennan, Introduction to the Alexander Technique, Little
Brown and Company, 1996. Among others, the idea of the Alexander technique is to re-
turn to the situation that the muscle groups for sustention and those for motion are used
Copyright © Christoph Schiller November 1997–May 2006

only for their respective function, and not vice versa. Any unnecessary muscle tension, such
as neck stiffness, is a waste of energy due to the use of sustention muscles for movement and
of motion muscles for sustention. The technique teaches the way to return to the natural use
of muscles.
Motion of animals was discussed extensively already in the seventeenth century by
G. B orelli, De motu animalium, 1680. An example of a more modern approach is J.J.
Collins & I. Stewart, Hexapodal gaits and coupled nonlinear oscillator models, Biolo-
gical Cybernetics 68, pp. 287–298, 1993. See also I. Stewart & M. Golubitsky, Fearful
Symmetry, Blackwell, 1992. Cited on pages 35 and 92.
18 The results on the development of children mentioned here and in the following have been
drawn mainly from the studies initiated by Jean Piaget; for more details on child develop-
ment, see the Intermezzo following this chapter, on page 634. At http://www.piaget.org you

Dvipsbugw
160 i galilean motion

can find the website maintained by the Jean Piaget Society. Cited on pages 36, 45, and 46.
19 The reptilian brain (eat? flee? ignore?), also called the R-complex, includes the brain stem,
the cerebellum, the basal ganglia and the thalamus; the old mammalian (emotions) brain,
also called the limbic system, contains the amygdala, the hypothalamus and the hippocam-
pus; the human (and primate) (rational) brain, called the neocortex, consists of the famous
grey matter. For images of the brain, see the atlas by John Nolte, The Human Brain: An
Introduction to its Functional Anatomy, Mosby, fourth edition, 1999. Cited on page 37.
20 The lower left corner movie can be reproduced on a computer after typing the following
lines in the Mathematica software package: Cited on page 37. Dvipsbugw
« Graphics‘Animation‘
Nxpixels=72; Nypixels=54; Nframes=Nxpixels 4/3;
Nxwind=Round[Nxpixels/4]; Nywind=Round[Nypixels/3];
front=Table[Round[Random[]],{y,1,Nypixels},{x,1,Nxpixels}];
back =Table[Round[Random[]],{y,1,Nypixels},{x,1,Nxpixels}];
frame=Table[front,{nf,1,Nframes}];

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Do[ If[ x>n-Nxwind && x<n && y>Nywind && y<2Nywind,
frame[[n,y,x]]=back[[y,x-n]] ],
{x,1,Nxpixels}, {y,1,Nypixels}, {n,1,Nframes}];
film=Table[ListDensityPlot[frame[[nf]], Mesh-> False,
Frame-> False, AspectRatio-> N[Nypixels/Nxpixels],
DisplayFunction-> Identity], {nf,1,Nframes}]
ShowAnimation[film]
But our motion detection system is much more powerful than the example shown in the
lower left corners. The following, different movie makes the point.
« Graphics‘Animation‘
Nxpixels=72; Nypixels=54; Nframes=Nxpixels 4/3;
Nxwind=Round[Nxpixels/4]; Nywind=Round[Nypixels/3];
front=Table[Round[Random[]],{y,1,Nypixels},{x,1,Nxpixels}];
back =Table[Round[Random[]],{y,1,Nypixels},{x,1,Nxpixels}];
frame=Table[front,{nf,1,Nframes}];
Do[ If[ x>n-Nxwind && x<n && y>Nywind && y<2Nywind,
frame[[n,y,x]]=back[[y,x]] ],
{x,1,Nxpixels}, {y,1,Nypixels}, {n,1,Nframes}];
film=Table[ListDensityPlot[frame[[nf]], Mesh-> False,
Frame-> False, AspectRatio-> N[Nypixels/Nxpixels],
DisplayFunction-> Identity], {nf,1,Nframes}]
Copyright © Christoph Schiller November 1997–May 2006

ShowAnimation[film]
Similar experiments, e.g. using randomly changing random patterns, show that the eye per-
ceives motion even in cases where all Fourier components of the image are practically zero;
such image motion is called drift-balanced or non-Fourier motion. Several examples are
presented in J. Z anker, Modelling human motion perception I: Classical stimuli, Natur-
wissenschaften 81, pp. 156–163, 1994, and J. Z anker, Modelling human motion perception
II: Beyond Fourier motion stimuli, Naturwissenschaften 81, pp. 200–209, 1994.
21 An introduction into perception research is E. Bruce Goldstein, Perception,
Books/Cole, 5th edition, 1998. Cited on pages 33 and 37.
22 All fragments from Heraclitus are from John Mansley Robinson, An Introduction to
Early Greek Philosophy, Houghton Muffin 1968, chapter 5. Cited on page 38.

Dvipsbugw
 bibliography 161

23 An introduction to Newton the alchemist are the two books by Betty Jo Teeter Dobbs,
The Foundations of Newton’s Alchemy, Cambridge University Press, 1983, and The Janus Face
of Genius, Cambridge University Press, 1992. Newton is found to be a sort of highly intellec-
tual magician, desperately looking for examples of processes where gods interact with the
material world. An intense but tragic tale. A good overview is provided by R.G. Keesing,
Essay Review: Newton’s Alchemy, Contemporary Physics 36, pp. 117–119, 1995.
Newton’s infantile theology, typical for god seekers who grew up without a father, can be
found in the many books summarizing the letter exchanges between Clarke, his secretary,
and Leibniz, Newton’s rival for fame. Cited on page 42. Dvipsbugw
24 An introduction to the story of classical mechanics, which also destroys a few of the myths
surrounding it – such as the idea that Newton could solve differential equations or that he
introduced the expression F = ma – is given by Clifford A. Truesdell, Essays in the
History of Mechanics, Springer, 1968. Cited on pages 42, 124, and 148.
25 C. Liu, Z. Du tton, C.H. Behroozi & L. Vestergaard Hau, Observation of coher-
ent optical information storage in an atomic medium using halted light pulses, Nature 409,
pp. 490–493, 2001. There is also a comment on the paper by E.A. Cornell, Stopping light

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

in its track, 409, pp. 461–462, 2001. However, despite the claim, the light pulses of course
have not been halted. Can you give at least two reasons without even reading the paper, and
Challenge 328 n maybe a third after reading it?
The work was an improvement on the previous experiment where a group velocity of
light of 17 m~s had been achieved, in an ultracold gas of sodium atoms, at nanokelvin tem-
peratures. This was reported by L. Vestergaard Hau, S.E. Harris, Z. Du tton &
C.H. Behroozi, Light speed reduction to 17 meters per second in an ultracold atomic gas,
Nature 397, pp. 594–598, 1999. Cited on pages 44 and 343.
26 R ainer Flindt, Biologie in Zahlen – Eine Datensammlung in Tabellen mit über 10.000
Einzelwerten, Spektrum Akademischer Verlag, 2000. Cited on page 44.
27 Two jets with that speed have been observed by I.F. Mirabel & L.F. Rodríguez, A
superluminal source in the Galaxy, Nature 371, pp. 46–48, 1994, as well as the comments
on p. 18. Cited on page 44.
28 A beautiful introduction to the effects of the slowest motions is Detlev Busche, Jürgen
Kempf & Ingrid Stengl, Landschaftsformen der Erde – Bildatlas der Geomorphologie,
Primus Verlag, 2005. Cited on page 45.
29 An introduction to the sense of time as a result of clocks in the brain is found in R.B. Ivry
& R. Spencer, The neural representation of time, Current Opinion in Neurobiology 14,
pp. 225–232, 2004. The chemical clocks in our body are described in John D. Palmer, The
Living Clock, Oxford University Press, 2002, or in A. Ahlgren & F. Halberg, Cycles of
Nature: An Introduction to Biological Rhythms, National Science Teachers Association, 1990.
Copyright © Christoph Schiller November 1997–May 2006

See also the http://www.msi.umn.edu/~halberg/introd/ website. Cited on page 47.

30 This has been shown among others by the work of Anna Wierzbicka mentioned in more de-
tail in the Intermezzo following this chapter, on page 643. The passionate best seller by the
Chomskian author Steven Pinker, The Language Instinct – How the Mind Creates Lan-
guage, Harper Perennial, 1994, also discusses issues related to this matter, refuting amongst
others on page 63 the often repeated false statement that the Hopi language is an exception.
Cited on page 47.
31 Aristotle rejects the idea of the flow of time in chapter IV of his Physics. See the full text on
the http://classics.mit.edu/Aristotle/physics.4.iv.html website. Cited on page 49.
32 Perhaps the most informative of the books about the ‘arrow of time’ is Hans Dieter Zeh,
The Physical Basis of the Direction of Time, Springer Verlag, 4th edition, 2001. It is still the

Dvipsbugw
162 i galilean motion

best book on the topic. Most other texts – have a look on the internet – lack clarity of ideas.
A typical conference proceeding is J.J. Halliwell, J. Pérez–Mercader & Woj-
ciech H. Zurek, Physical Origins of Time Asymmetry, Cambridge University Press, 1994.
Cited on page 49.
33 On the issue of absolute and relative motion there are many books about few issues. Ex-
amples are John Barbour, Absolute or Relative Motion? Vol. 1: A Study from the Machian
Point of View of the Discovery and the Structure of Spacetime Theories, Cambridge Univer-
sity Press, 1989, John Barbour, Absolute or Relative Motion? Vol. 2: The Deep Structure
of General Relativity, Oxford University Press, 2005, or John Earman, World Enough and Dvipsbugw
Spacetime: Absolute vs Relational Theories of Spacetime, MIT Press, 1989. Cited on page 53.
34 R. Dougherty & M. Foreman, Banach–Tarski decompositions using sets with the
property of Baire, Journal of the American Mathematical Society 7, pp. 75–124, 1994. See
also Al an L.T. Paterson, Amenability, American Mathematical Society, 1998, and
Robert M. French, The Banach–Tarski theorem, The Mathematical Intelligencer 10,
pp. 21–28, 1998. Finally, there are the books by Bernard R. Gelbaum & John M.H.
Olmsted, counter-examples in Analysis, Holden–Day, 1964, and their Theorems and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

counter-examples in Mathematics, Springer, 1993. Cited on page 55.
35 The beautiful but not easy text is Steve Wagon, The Banach Tarski Paradox, Cambridge
University Press, 1993. Cited on pages 55 and 1235.
36 About the shapes of salt water bacteria, see the corresponding section in the interesting book
by Bernard Dixon, Power Unseen – How Microbes Rule the World, W.H. Freeman, 1994.
The book has about 80 sections, in which as many microorganisms are vividly presented.
Cited on page 56.
37 The smallest distances are probed in particle accelerators; the distance can be determined
from the energy of the particle beam. In 1996, the value of 10−19 m (for the upper limit of the
size of quarks) was taken from the experiments described in F. Abe & al., Measurement of
dijet angular distributions by the collider detector at Fermilab, Physical Review Letters 77,
pp. 5336–5341, 1996. Cited on page 59.
38 Alexander K. Dewdney, The Planiverse – Computer Contact with a Two-dimensional
World, Poseidon Books/Simon & Schuster, 1984. Several other fiction authors had explored
the option of a two-dimensional universe before, always answering, incorrectly, in the af-
firmative. Cited on page 63.
39 There is a whole story behind the variations of д. It can be found in Chuji Tsuboi, Grav-
ity, Allen & Unwin, 1979, or in Wolf gang Torge, Gravimetry, de Gruyter, 1989, or in
Mil an Burša & Karel Pěč, The Gravity Field and the Dynamics of the Earth, Springer,
1993. The variation of the height of the soil by around 0.3 m due to the Moon is one of the
interesting effects found by these investigations. Cited on pages 65 and 129.
Copyright © Christoph Schiller November 1997–May 2006

40 Andrea Frova, La fisica sotto il naso – 44 pezzi facili, Biblioteca Universale Rizzoli, Mil-
ano, 2001. Cited on page 65.
41 The study of shooting faeces (i.e., shit) and its mechanisms is a part of modern biology. The
reason that caterpillars do this was determined by M. Weiss, Good housekeeping: why do
shelter-dwelling caterpillars fling their frass?, Ecology Letters 6, pp. 361–370, 2003, who also
gives the present record of 1.5 m for the 24 mg pellets of Epargyreus clarus. The picture of
the flying frass is from S. Caveney, H. McLean & D. Surry, Faecal firing in a skipper
caterpillar is pressure-driven, The Journal of Experimental Biology 201, pp. 121–133, 1998.
Cited on page 66.
42 This was discussed in the Frankfurter Allgemeine Zeitung, 2nd of August, 1997, at the time
of the world athletics championship. The values are for the fastest part of the race of a 100 m

Dvipsbugw
bibliography 163

sprinter; the exact values cited were called the running speed world records in 1997, and were
given as 12.048 m~s = 43.372 km~h by Ben Johnson for men, and 10.99 m~s = 39.56 km~h
for women. Cited on page 67.
43 Long jump data and literature can be found in three articles all entitled Is a good long
jumper a good high jumper?, in the American Journal of Physics 69, pp. 104–105, 2001. In
particular, world class long jumpers run at 9.35  0.15 m~s, with vertical take-off speeds
of 3.35  0.15 m~s, giving take-off angles of about (only) 20°. A new technique for achiev-
ing higher take-off angles would allow the world long jump record to increase dramatically.
Cited on page 67. Dvipsbugw
44 The arguments of Zeno can be found in Aristotle, Physics, VI, 9. It can be found trans-
lated in almost any language. The http://classics.mit.edu/Aristotle/physics.6.vi.html website
provides an online version in English. Cited on pages 69 and 158.
45 Etymology can be a fascinating topic, e.g. when research discovers the origin of the Ger-
man word ‘Weib’ (‘woman’, related to English ‘wife’). It was discovered, via a few texts in
Tocharian – an extinct Indo-European language from a region inside modern China – to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

mean originally ‘shame’. It was used for the female genital region in an expression meaning
‘place of shame’. With time, this expression became to mean ‘woman’ in general, while being
shortened to the second term only. This connection was discovered by the German linguist
Klaus T. Schmidt; it explains in particular why the word is not feminine but neutral, i.e. why
it uses the article ‘das’ instead of ‘die’. Julia Simon, private communication.
Etymology can also be simple and plain fun, for example when one discovers in the Ox-
ford English Dictionary that ‘testimony’ and ‘testicle’ have the same origin; indeed in Latin
the same word ‘testis’ was used for both concepts. Cited on pages 71 and 77.
46 An overview of the latest developments is given by J.T. Armstrong, D.J. Hunter, K.J.
Johnston & D. Mozurkewich, Stellar optical interferometry in the 1990s, Physics
Today pp. 42–49, May 1995. More than 100 stellar diameters were known already in 1995.
Several dedicated powerful instruments are being planned. Cited on page 71.
47 A good biology textbook on growth is Arthur F. Hopper & Nathan H. Hart, Found-
ations of Animal Deveopment, Oxford University Press, 2006. Cited on page 72.
48 This is discussed for example in C.L. Stong, The amateur scientist – how to supply electric
power to something which is turning, Scientific American pp. 120–125, December 1975. It
also discusses how to make a still picture of something rotating simply by using a few prisms,
the so-called Dove prisms. Other examples of attaching something to a rotating body are
given by E. R ieflin, Some mechanisms related to Dirac’s strings, American Journal of
Physics 47, pp. 379–381, 1979. Cited on page 73.
Copyright © Christoph Schiller November 1997–May 2006

49 James A. Young, Tumbleweed, Scientific American 264, pp. 82–87, March 1991. The
tumbleweed is in fact quite rare, except in in Hollywood westerns, where all directors feel
obliged to give it a special appearance. Cited on page 73.
50 The first experiments to prove the rotation of the flagella were by M. Silverman & M.I.
Simon, Flagellar rotation and the mechanism of bacterial motility, Nature 249, pp. 73–74,
1974. For some pretty pictures of the molecules involved, see K. Namba, A biological mo-
lecular machine: bacterial flagellar motor and filament, Wear 168, pp. 189–193, 1993. The
present record speed of rotation, 1700 rotations per second, is reported by Y. Magar-
iyama, S. Sugiyama, K. Muramoto, Y. Maekawa, I. Kawagishi, Y. Imae & S.
Kud o, Very fast flagellar rotation, Nature 371, p. 752, 1994.
More on bacteria can be learned from David Dusenbery, Life at a Small Scale, Sci-
entific American Library, 1996. Cited on page 74.

Dvipsbugw
164 i galilean motion

51 On shadows, see the agreeable popular text by Roberto Casati, Alla scoperta dell’ombra
– Da Platone a Galileo la storia di un enigma che ha affascinato le grandi menti dell’umanità,
Oscar Mondadori, 2000, and his websites located at http://www.shadowmill.com and http://
roberto.casati.free.fr/casati/roberto.htm. Cited on page 74.
52 There is also the beautiful book by Penelope Farrant, Colour in Nature, Blandford,
1997. Cited on page 74.
53 The laws of cartoon physics can easily be found using any search engine on the internet.
Cited on page 75.
Dvipsbugw
54 For the curious, an overview of the illusions used in the cinema and in television, which
lead to some of the strange behaviour of images mentioned above, is given in Bernard
Wilkie, The Technique of Special Effects in Television, Focal Press, 1993, and his other books,
or in the Cinefex magazine. Cited on page 75.
55 Aetius, Opinions, I, XXIII, 3. See Jean-Paul Dumont, Les écoles présocratiques, Folio
Essais, Gallimard, p. 426, 1991. Cited on page 75.
56 Giuseppe Fumagalli, Chi l’ha detto?, Hoepli, 1983. Cited on pages 77 and 152.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

57 For the role and chemistry of adenosine triphosphate (ATP) in cells and in living beings, see
any chemistry book, or search the internet. The uncovering of the mechanisms around ATP
has led to Nobel Prizes in chemistry in 1978 and in 1997. Cited on page 83.
58 A picture of this unique clock can be found in the article by A. Garrett, Perpetual motion
– a delicious delirium, Physics World pp. 23–26, December 1990. Cited on page 83.
59 A Shell study estimated the world’s total energy consumption in 2000 to be 500 EJ. The
US Department of Energy estimated it to be around 416 EJ. We took the lower value here.
A discussion and a breakdown into electricity usage (14 EJ) and other energy forms, with
variations per country, can be found in S. Benka, The energy challenge, Physics Today
55, pp. 38–39, April 2002, and in E.J. Monitz & M.A. Kenderdine, Meeting energy
challenges: technology and policy, Physics Today 55, pp. 40–46, April 2002. Cited on page
85.
60 For an overview, see the paper by J.F. Mulligan & H.G. Hertz, An unpublished lecture
by Heinrich Hertz: ‘On the energy balance of the Earth’, American Journal of Physics 65,
pp. 36–45, 1997. Cited on page 87.
61 For a beautiful photograph of this feline feat, see the cover of the journal and the article of
J. Darius, A tale of a falling cat, Nature 308, p. 109, 1984. Cited on page 91.
62 Natthi L. Sharma, A new observation about rolling motion, European Journal of Physics
17, pp. 353–356, 1996. Cited on page 92.
63 C. Singh, When physical intuition fails, American Journal of Physics 70, pp. 1103–1109, 2002.
Copyright © Christoph Schiller November 1997–May 2006

Cited on page 92.

64 Serge Gracovetsky, The Spinal Engine, Springer Verlag, 1990. It is now also knon that
human gait is chaotic. This is explained by M. Perc, The dynamics of human gait, European
Journal of Physics 26, pp. 525–534, 2005. Cited on page 93.
65 Thomas Heath, Aristarchus of Samos – the Ancient Copernicus, Dover, 1981, reprinted
from the original 1913 edition. Aristarchos’ treaty is given in Greek and English. Aristarchos
was the first proposer of the heliocentric system. Aristarchos had measured the length of the
day (in fact, by determining the number of days per year) to the astonishing precision of
less than one second. This excellent book also gives an overview of Greek astronomy before
Aristarchos, explained in detail for each Greek thinker. Aristarchos’ text is also reprinted
in Aristarchos, On the sizes and the distances of the Sun and the Moon, c. 280 bce in

Dvipsbugw
bibliography 165

Michael J. Crowe, Theories of the World From Antiquity to the Copernican Revolution,
Dover, 1990, especially on pp. 27–29. Cited on page 94.
66 The influence of the Coriolis effect on icebergs was studied most thoroughly by the Swedish
physicist turned oceanographer Walfrid Ekman (1874–1954); the topic was suggested by
the great explorer Fridtjof Nansen, who also made the first observations. In his honour, one
speaks of the Ekman layer, Ekman transport and Ekman spirals. Any text on oceanography
or physical geography will give more details about them. Cited on page 95.
67 An overview of the effects of the Coriolis acceleration a = −2ω  v in the rotating frame is
given by Edward A. Desloge, Classical Mechanics, Volume 1, John Wiley & Sons, 1982. Dvipsbugw
Even the so-called Gulf Stream, the current of warm water flowing from the Caribbean to
the North Sea, is influenced by it. Cited on page 95.
68 The original publication is by A.H. Shapiro, Bath-tub vortex, Nature 196, pp. 1080–1081,
1962. He also produced two movies of the experiment. The experiment has been repeated
many times in the northern and in the southern hemisphere, where the water drains clock-
wise; the first southern hemisphere test was L.M. Trefethen & al., The bath-tub vortex
in the southern hemisphere, Nature 201, pp. 1084–1085, 1965. A complete literature list is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

found in the letters to the editor of the American Journal of Physics 62, p. 1063, 1994. Cited
on page 96.
69 The tricks are explained by H. R ichard Crane, Short Foucault pendulum: a way to
eliminate the precession due to ellipticity, American Journal of Physics 49, pp. 1004–1006,
1981, and particularly in H. R ichard Crane, Foucault pendulum wall clock, American
Journal of Physics 63, pp. 33–39, 1993. The Foucault pendulum was also the topic of the thesis
of Heike Kamerling Onnes, Nieuwe bewijzen der aswenteling der aarde, Universiteit
Groningen, 1879. Cited on page 96.
70 The reference is J.G. Hagen, La rotation de la terre : ses preuves mécaniques anciennes et
nouvelles, Sp. Astr. Vaticana Second. App. Rome, 1910. His other experiment is published as
J.G. Hagen, How Atwood’s machine shows the rotation of the Earth even quantitatively,
International Congress of Mathematics, Aug. 1912. Cited on page 97.
71 R. Anderson, H.R. Bilger & G.E. Stedman, The Sagnac-effect: a century of Earth-
rotated interferometers, American Journal of Physics 62, pp. 975–985, 1994.
See also the clear and extensive paper by G.E. Stedman, Ring laser tests of funda-
mental physics and geophysics, Reports on Progress in Physics 60, pp. 615–688, 1997. Cited
on page 98.
72 About the length of the day, see the http://maia.usno.navy.mil website, or the books by
K. L ambeck, The Earth’s Variable Rotation: Geophysical Causes and Consequences, Cam-
bridge University Press, 1980, and by W.H. Munk & G.J.F. MacDonald, The Rotation
Copyright © Christoph Schiller November 1997–May 2006

of the Earth, Cambridge University Press, 1960. Cited on pages 98 and 131.
73 One example of data is by C.P. Sonett, E.P. Kvale, A. Z akharian, M.A. Chan &
T.M. Demko, Late proterozoic and paleozoic tides, retreat of the moon, and rotation of
the Earth, Science 273, pp. 100–104, 5 July 1996. They deduce from tidal sediment analysis
that days were only 18 to 19 hours long in the Proterozoic, i.e. 900 million years ago; they
assume that the year was 31 million seconds long from then to today. Another determination
was by G.E. Williams, Precambrian tidal and glacial clastic deposits: implications for
precambrian Earth–Moon dynamics and palaeoclimate, Sedimentary Geology 120, pp. 55–
74, 1998. Using a geological formation called tidal rhythmites, he deduced that about 600
million years ago there were 13 months per year and a day had 22 hours. Cited on page 98.
74 The story of this combination of history and astronomy is told in R ichard Stephenson,
Historical Eclispes and Earth’s Rotation, Cambridge University Press, 1996. Cited on page 99.

Dvipsbugw
166 i galilean motion

75 On the rotation and history of the solar system, see S. Brush, Theories of the origin of the
solar system 1956–1985, Reviews of Modern Physics 62, pp. 43–112, 1990. Cited on page 99.
76 The website http://maia.usno.navy.mil shows the motion of the Earth’s axis over the last ten
years. The International Latitude Service founded by Küstner is now part of the International
Earth Rotation Service; more information can be found on the http://www.iers.org website.
The latest idea is that two-thirds of the circular component of the polar motion, which in
the USA is called ‘Chandler wobble’ after the person who attributed to himself the discovery
by Küstner, is due to fluctuations of the ocean pressure at the bottom of the oceans and one-
third is due to pressure changes in the atmosphere of the Earth. This is explained by R.S. Dvipsbugw
Gross, The excitation of the Chandler wobble, Geophysical Physics Letters 27, pp. 2329–
2332, 2000. Cited on page 100.
77 For more information about Alfred Wegener, see the (simple) text by Kl aus Rohrbach,
Alfred Wegener – Erforscher der wandernden Kontinente, Verlag Freies Geistesleben, 1993;
about plate tectonics, see the http://www.scotese.com website. About earthquakes, see the
http://www.geo.ed.ac.uk/quakexe/quakes and the http://www.iris.edu/seismon website. See
the http://vulcan.wr.usgs.gov and the http://www.dartmouth.edu/~volcano/ websites for in-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

formation about volcanoes. Cited on page 100.
78 J.D. Hays, J. Imbrie & N.J. Shackleton, Variations in the Earth’s orbit: pacemaker of
the ice ages, Science 194, pp. 1121–1132, 1976. They found the result by literally digging in the
mud that covers the ocean floor in certain places. Note that the web is full of information
on the ice ages. Just look up ‘Milankovitch’ in a search engine. Cited on page 102.
79 Roberta Humphreys & Jeffrey L arsen, The sun’s distance above the galactic plane,
Astronomical Journal 110, pp. 2183–2188, November 1995. Cited on page 103.
80 C.L. Bennet, M.S. Turner & M. White, The cosmic rosetta stone, Physics Today 50,
pp. 32–38, November 1997. Cited on page 104.
81 A good roulette prediction story from the 1970s is told by Thomas A. Bass, The Eu-
daemonic Pie also published under the title The Newtonian Casino, Backinprint, 2000. An
overview up to 1998 is given in the paper Edward O. Thorp, The invention of the first
wearable computer, Proceedings of the Second International Symposium on Wearable Com-
puters (ISWC 1998), 19-20 October 1998, Pittsburgh, Pennsylvania, USA (IEEE Computer Soci-
ety), pp. 4–8, 1998, downloadable at http://csdl.computer.org/comp/proceedings/iswc/1998/
9074/00/9074toc.htm. Cited on pages 106 and 806.
82 The original papers are A.H. Compton, A laboratory method of demonstrating the Earth’s
rotation, Science 37, pp. 803–806, 1913, A.H. Compton, Watching the Earth revolve, Sci-
entific American Supplement no. 2047, pp. 196–197, 1915, and A.H. Compton, A determina-
tion of latitude, azimuth and the length of the day independent of astronomical observations,
Copyright © Christoph Schiller November 1997–May 2006

Physical Review (second series) 5, pp. 109–117, 1915. Cited on page 98.
83 This and many other physics surprises are described in the beautiful lecture script by Josef
Zweck, Physik im Alltag, the notes of his lectures held in 1999/2000 at the Universität Re-
gensburg. Cited on pages 107 and 111.
84 The equilibrium of ships, so important in car ferries, is an interesting part of shipbuilding;
an introduction was already given by Leonhard Euler, Scientia navalis, 1749. Cited on
page 108.
85 K.R. Weninger, B.P. Barber & S.J. Pu tterman, Pulsed Mie scattering measure-
ments of the collapse of a sonoluminescing bubble, Physical Review Letters 78, pp. 1799–
1802, 1997. Cited on page 108.
86 On http://www.sff.net/people/geoffrey.landis/vacuum.html you can read a description of

Dvipsbugw
bibliography 167

what happened. See also the http://www.sff.net/people/geoffrey.landis/ebullism.html and

http://imagine.gsfc.nasa.gov/docs/ask_astro/answers/970603.html websites. They all give
details on the effects of vacuum on humans. Cited on page 109.
87 R. McN. Alexander, Leg design and jumping technique for humans, other vertebrates
and insects, Philosophical Transactions of the Royal Society in London B 347, pp. 235–249,
1995. Cited on page 114.
88 J. W. Gl asheen & T. A. McMahon, A hydrodynamic model of locomotion in the
basilisk lizard, Nature 380, pp. 340–342, For pictures, see also New Scientist, p. 18, 30 March
1996, or Scientific American, pp. 48–49, September 1997, or the website by the author at Dvipsbugw
http://rjf2.biol.berkeley.edu/Full_Lab/FL_Personnel/J_Glasheen/J_Glasheen.html.
Several shore birds also have the ability to run over water, using the same mechanism.
Cited on page 114.
89 A. Fernandez–Nieves & F.J. de l as Nieves, About the propulsion system of a kayak
and of Basiliscus basiliscus, European Journal of Physics 19, pp. 425–429, 1998. Cited on page
115.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

90 The material on the shadow discussion is from the book by Robert M. Pryce, Cook and
Peary, Stackpole Books, 1997. See also Wally Herbert, The Noose of Laurels, Doubleday
1989. The sad story of Robert Peary is also told in the centenary number of National Geo-
graphic, September 1988. Since the National Geographic Society had financed Peary in his
attempt and had supported him until the US Congress had declared him the first man at the
Pole, the (partial) retraction is noteworthy. (The magazine then changed its mind again later
on, to sell more copies.) By the way, the photographs of Cook, who claimed to have been
at the North Pole even before Peary, have the same problem with the shadow length. Both
men have a history of cheating about their ‘exploits’. As a result, the first man at the North
Pole was probably Roald Amundsen, who arrived there a few years later, and who was also
the first man at the South Pole. Cited on page 116.
91 The story is told in M. Nauenberg, Hooke, orbital motion, and Newton’s Principia, Amer-
ican Journal of Physics 62, 1994, pp. 331–350. Cited on page 117.
92 More details are given by D. R awlins, in Doubling your sunsets or how anyone can meas-
ure the Earth’s size with wristwatch and meter stick, American Journal of Physics 47, 1979,
pp. 126–128. Another simple measurement of the Earth radius, using only a sextant, is given
by R. O’Keefe & B. Ghavimi–Al agha, in The World Trade Centre and the distance
to the world’s centre, American Journal of Physics 60, pp. 183–185, 1992. Cited on page 117.
93 More details on astronomical distance measurements can be found in the beautiful little
book by A. van Helden, Measuring the Universe, University of Chicago Press, 1985, and in
Nigel Henbest & Heather Cooper, The Guide to the Galaxy, Cambridge University
Copyright © Christoph Schiller November 1997–May 2006

Press, 1994. Cited on page 117.

94 A lot of details can be found in M. Jammer, Concepts of Mass in Classical and Modern Phys-
ics, reprinted by Dover, 1997, and in Concepts of Force, a Study in the Foundations of Mechan-
ics, Harvard University Press, 1957. These eclectic and thoroughly researched texts provide
numerous details and explain various philosophical viewpoints, but lack clear statements
and conclusions on the accurate description of nature; thus are not of help on fundamental
issues.
Jean Buridan (c. 1295 to c. 1366) criticizes the distinction of sublunar and translunar mo-
tion in his book De Caelo, one of his numerous works. Cited on page 118.
95 D. Topper & D.E. Vincent, An analysis of Newton’s projectile diagram, European
Journal of Physics 20, pp. 59–66, 1999. Cited on page 118.

Dvipsbugw
168 i galilean motion

96 The absurd story of the metre is told in the historical novel by Ken Alder, The Measure of
All Things : The Seven-Year Odyssey and Hidden Error that Transformed the World, The Free
Press, 2003. Cited on page 120.
97 H. Edelmann, R. Napiwotzki, U. Heber, N. Christlieb & D. R eimers, HE
0437-5439: an unbound hyper-velocity B-type star, The Astrophysical Journal 634, pp. L181–
L184, 2005. Cited on page 138.
98 This is explained for example by D.K. Firpić & I.V. Aniçin, The planets, after all, may
run only in perfect circles – but in the velocity space!, European Journal of Physics 14, pp. 255–
258, 1993. Cited on pages 138 and 1248. Dvipsbugw
99 About the measurement of spatial dimensions via gravity – and the failure to find any hint
for a number different from three – see the review by E.G. Adelberger, B.R. Heckel
& A.E. Nelson, Tests of the gravitational inverse-square law, Annual Review of Nuclear
and Particle Science 53, pp. 77–121, 2003, also http://www.arxiv.org/abs/hep-ph/0307284, or
the review by J.A. Hewett & M. Spiropulu, Particle physics probes of extra spacetime
dimensions, Annual Review of Nuclear and Particle Science 52, pp. 397–424, 2002, http://

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

www.arxiv.org/abs/hep-ph/0205106. Cited on page 122.
100 There are many books explaining the origin of the precise shape of the Earth, such as the
pocket book S. Anders, Weil die Erde rotiert, Verlag Harri Deutsch, 1985. Cited on page
122.
101 The shape of the Earth is described most precisely with the World Geodetic System. For
an extensive presentation, see the http://www.eurocontrol.be/projects/eatchip/wgs84/start.
html website. See also the website of the International Earth Rotation Service at http://hpiers.
obspm.fr. Cited on page 122.
102 W.K. Hartman, R.J. Phillips & G.J. Taylor, editors, Origin of the Moon, Lunar and
Planetary Institute, 1986. Cited on page 126.
103 If you want to read about the motion of the Moon in all its fascinating details, have a look
at Martin C. Gu tzwiller, Moon–Earth–Sun: the oldest three body problem, Reviews
of Modern Physics 70, pp. 589–639, 1998. Cited on page 126.
104 Dietrich Neumann, Physiologische Uhren von Insekten – Zur Ökophysiologie lun-
arperiodisch kontrollierter Fortpflanzungszeiten, Naturwissenschaften 82, pp. 310–320, 1995.
Cited on page 126.
105 The origin of the duration of the menstrual cycle is not yet settled; however, there are ex-
planations on how it becomes synchronized with other cycles. For a general explanation see
Arkady Pikovsky, Michael Rosenblum & Jürgen Kurths, Synchronization: A
Universal Concept in Nonlinear Science, Cambridge University Press, 2002. Cited on page
Copyright © Christoph Schiller November 1997–May 2006

126.
106 J. L askkar, F. Jou tel & P. Robu tel, Stability of the Earth’s obliquity by the moon,
Nature 361, pp. 615–617, 1993. However, the question is not completely settled, and other
opinions exist. Cited on page 126.
107 Neil F. Comins, What if the Moon Did not Exist? – Voyages to Earths that Might Have
Been, Harper Collins, 1993. Cited on page 126.
108 See for example the discussion by J.J. Lissauer, It is not easy to make the moon, Nature
389, pp. 327–328, 1997. Cited on page 127.
109 Paul A. Wiegert, Kimmo A. Innanen & Seppo Mikkol a, An asteroidal compan-
ion to the Earth, Nature 387, pp. 685–686, 12 June 1997, together with the comment on
pp. 651–652. Details on the orbit and on the fact that Lagrangian points do not always form

Dvipsbugw
bibliography 169

equilateral triangles can be found in F. Namouni, A.A. Christou & C.D. Murray, Co-
orbital dynamics at large eccentricity and inclination, Physical Review Letters 83, pp. 2506–
2509, 1999. Cited on page 129.
110 Simon Newcomb, Astronomical Papers of the American Ephemeris 1, p. 472, 1882. Cited
on page 129.
111 A beautiful introduction is the classic G. Falk & W. Ruppel, Mechanik, Relativität, Grav-
itation – ein Lehrbuch, Springer Verlag, Dritte Auflage, 1983. Cited on page 129.
112 J. Soldner, Berliner Astronomisches Jahrbuch auf das Jahr 1804, 1801, p. 161. Cited on page
Dvipsbugw
133.
113 The equality was first tested with precision by Rol and von Eőtvős, Annalen der Physik
& Chemie 59, p. 354, 1896, and by R. von Eőtvős, V. Pekár & E. Fekete, Beiträge
zum Gesetz der Proportionalität von Trägheit und Gravität, Annalen der Physik 4, Leipzig
68, pp. 11–66, 1922. He found agreement to 5 parts in 109 . More experiments were performed
by P.G. Roll, R. Krotkow & R.H. Dicke, The equivalence of inertial and passive grav-
itational mass, Annals of Physics (NY) 26, pp. 442–517, 1964, one of the most interesting and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

entertaining research articles in experimental physics, and by V.B. Braginsky & V.I.
Panov, Soviet Physics – JETP 34, pp. 463–466, 1971. Modern results, with errors less than
one part in 1012 , are by Y. Su & al., New tests of the universality of free fall, Physical Review
D50, pp. 3614–3636, 1994. Several experiments have been proposed to test the equality in
space to less than one part in 1016 . Cited on page 134.
114 See L. Hod ges, Gravitational field strength inside the Earth, American Journal of Physics
59, pp. 954–956, 1991. Cited on page 138.
115 P. Mohazzabi & M.C. James, Plumb line and the shape of the Earth, American Journal
of Physics 68, pp. 1038–1041, 2000. Cited on page 139.
116 From Neil de Gasse Tyson, The Universe Down to Earth, Columbia University Press,
1994. Cited on page 140.
117 This is a small example from the beautiful text by Mark P. Silverman, And Yet It Moves:
Strange Systems and Subtle Questions in Physics, Cambridge University Press, 1993. It is a
treasure chest for anybody interested in the details of physics. Cited on page 140.
118 G.D. Quinl an, Planet X: a myth exposed, Nature 363, pp. 18–19, 1993. Cited on page 141.
119 See http://en.wikipedia.org/wiki/90377_Sedna. Cited on page 141.
120 See Robert Matthews, Not a snowball’s chance ..., New Scientist 12 July 1997, pp. 24–27.
The original claim is by Louis A. Frank, J.B. Sigwarth & J.D. Craven, On the in-
flux of small comets into the Earth’s upper atmosphere, parts I and II, Geophysical Research
Letters 13, pp. 303–306, pp. 307–310, 1986. The latest observations have disproved the claim.
Copyright © Christoph Schiller November 1997–May 2006

Cited on page 141.

121 The ray form is beautifully explained by J. Evans, The ray form of Newton’s law of motion,
American Journal of Physics 61, pp. 347–350, 1993. Cited on page 142.
122 G.-L. Lesage, Lucrèce Newtonien, Nouveaux mémoires de l’Académie Royale des Sciences
et Belles Lettres pp. 404–431, 1747, or http://www3.bbaw.de/bibliothek/digital/struktur/
03-nouv/1782/jpg-0600/00000495.htm. Cited on page 142.
123 J. L askar, A numerical experiment on the chaotic behaviour of the solar system, Nature
338, pp. 237–238, 1989, and J. L askar, The chaotic motion of the solar system - A numer-
ical estimate of the size of the chaotic zones, Icarus 88, pp. 266–291, 1990. The work by Laskar
was later expanded by Jack Wisdom, using specially built computers, following only the plan-
ets, without taking into account the smaller objects. For more details, see G.J. Sussman

Dvipsbugw
170 i galilean motion

& J. Wisd om, Chaotic Evolution of the Solar System, Science 257, pp. 56–62, 1992. Today,
such calculations can be performed on your home PC with computer code freely available
on the internet. Cited on page 143.
124 B. Dubrulle & F. Graner, Titius-Bode laws in the solar system. 1: Scale invariance ex-
plains everything, Astronomy and Astrophysics 282, pp. 262–268, 1994, and Titius-Bode laws
in the solar system. 2: Build your own law from disk models, Astronomy and Astrophysics
282, pp. 269–-276, 1994. Cited on page 144.
125 M. Lecar, Bode’s Law, Nature 242, pp. 318–319, 1973, and M. Henon, A comment on
“The resonant structure of the solar system” by A.M. Molchanov, Icarus 11, pp. 93–94, 1969. Dvipsbugw
Cited on page 144.
126 Cassius Dio, Historia Romana, c. 220, book 37, 18. For an English translation, see the
site http://penelope.uchicago.edu/Thayer/E/Roman/Texts/Cassius_Dio/37*.html. Cited on
page 145.
127 M. Bevis, D. Alsd orf, E. Kendrick, L.P. Fortes, B. Forsberg, R. Malley & J.
Becker, Seasonal fluctuations in the mass of the Amazon River system and Earth’s elastic

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

response, Geophysical Research Letters 32, p. L16308, 2005. Cited on page 146.
128 D. Hestenes, M. Wells & G. Swackhamer, Force concept inventory, Physics Teacher
30, pp. 141–158, 1982. The authors developed tests to check the understanding of the concept
of physical force in students; the work has attracted a lot of attention in the field of physics
teaching. Cited on page 148.
129 For a general overview on friction, from physics to economics, architecture and organiza-
tional theory, see N. Åkerman, editor, The Necessity of Friction – Nineteen Essays on a
Vital Force, Springer Verlag, 1993. Cited on page 149.
130 See M. Hirano, K. Shinjo, R. Kanecko & Y. Murata, Observation of superlubricity
by scanning tunneling microscopy, Physical Review Letters 78, pp. 1448–1451, 1997. See also
the discussion of their results by Serge Fayeulle, Superlubricity: when friction stops,
Physics World pp. 29–30, May 1997. Cited on page 149.
131 C. Donald Ahrens, Meteorology Today: An Introduction to the Weather, Climate, and
the Environment, West Publishing Company, 1991. Cited on page 151.
132 This topic is discussed with lucidity by J.R. Mureika, What really are the best 100 m per-
formances?, Athletics: Canada’s National Track and Field Running Magazine, July 1997. It can
also be found as http://www.arxiv.org/abs/physics/9705004, together with other papers on
similar topics by the same author. Cited on page 151.
133 F.P. B owden & D. Tabor, The Friction and Lubrication of Solids, Oxford University Press,
Part I, revised edition, 1954, and part II, 1964. Cited on page 152.
Copyright © Christoph Schiller November 1997–May 2006

134 A powerful book on human violence is James Gilligan, Violence – Our Deadly Epidemic
and its Causes, Grosset/Putnam, 1992. Cited on page 152.
135 The main tests of randomness of number series – among them the gorilla test – can be found
in the authoritative paper by G. Marsaglia & W.W. Tsang, Some difficult-to-pass tests
of randomness, Journal of Statistical Software 7, p. 8, 2002. It can also be downloaded from
the http://www.jstatsoft.org/v07/i03/tuftests.pdf website. Cited on page 155.
136 For one aspect of the issue, see for example the captivating book by Bert Hellinger,
Zweierlei Glück, Carl Auer Systeme Verlag, 1997. The author explains how to live serenely
and with the highest possible responsibility for one’s actions, by reducing entanglements
with the destiny of others. He describes a simple powerful technique to realise this goal.
A completely different viewpoint is given by Nobel Peace Price winner Aung
San Suu Kyi, Freedom from Fear, Penguin, 1991. Cited on page 156.

Dvipsbugw
bibliography 171

137 Henrik Walter, Neurophilosophie der Willensfreiheit, Mentis Verlag, Paderborn 1999.
Also available in English translation. Cited on page 157.

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 global descriptions of motion – the simplicity of complexity 173

F I G U R E 86 What shape of rail allows F I G U R E 87 Can motion be described in a

the black stone to glide most rapidly manner common to all observers?
from point A to the lower point B? Dvipsbugw

3. global descriptions of motion – the simplicit y

of complexit y

“
Πλεῖν ἀνάγκε, ζῆν οὐκ ἀνάγκη.*

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Pompeius

”
A ll over the Earth – even in Australia – people observe that stones fall ‘down’. This
ncient observation led to the discovery of the universal law of gravity. To find it,
ll that was necessary was to look for a description of gravity that was valid globally.
The only additional observation that needs to be recognized in order to deduce the result
a = GM~r 2 is the variation of gravity with height.
In short, thinking globally helps us to make our description of motion more precise.
How can we describe motion as globally as possible? It turns out that there are six ap-
proaches to this question, each of which will be helpful on our way to the top of Motion
Mountain. We will start with an overview, and then explore the details of each approach.
— The first global approach to motion arises from a limitation of what we have learned
so far. When we predict the motion of a particle from its current acceleration, we are
using the most local description of motion possible. For example, whenever we use
an evolution equation we use the acceleration of a particle at a certain place and time
to determine its position and motion just after that moment and in the immediate
neighbourhood of that place.
Evolution equations thus have a mental ’horizon’ of radius zero.
The opposite approach is illustrated in the famous problem of Figure 86. The chal-
Copyright © Christoph Schiller November 1997–May 2006

lenge is to find the path that allows the fastest possible gliding motion from a high
Challenge 329 d point to a distant low point. To solve this we need to consider the motion as a whole,
for all times and positions. The global approach required by questions such as this one
will lead us to a description of motion which is simple, precise and fascinating: the
so-called principle of cosmic laziness, also known as the principle of least action.
— The second global approach to motion emerges when we compare the various descrip-
tions of the same system produced by different observers. For example, the observa-
tions by somebody falling from a cliff, a passenger in a roller coaster, and an observer

* Navigare necesse, vivere non necesse. ‘To navigate is necessary, to live is not.’ Gnaeus Pompeius Magnus
Ref. 138 (106–48 bce ), as cited by Plutarchus (c. 45 to c. 125).

Dvipsbugw
 174 i galilean motion • 3. the global simplicity of complexity

bicycle rope
wheel
a b b

F C P

a b b

F I G U R E 88 What happens F I G U R E 89 How to draw a straight line

when one rope is cut? with a compass: fix point F, put a pencil
into joint P and move C with a compass Dvipsbugw
along a circle

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 90 A south-pointing carriage

on the ground will usually differ. The relationships between these observations lead
us to a global description, valid for everybody. This approach leads us to the theory of
relativity.
— The third global approach to motion is to exploring the motion of extended and ri-
gid bodies, rather than mass points. The counter-intuitive result of the experiment in
Figure 88 shows why this is worthwhile.
In order to design machines, it is essential to understand how a group of rigid bod-
ies interact with one another. As an example, the mechanism in Figure 89 connects
the motion of points C and P. It implicitly defines a circle such that one always has the
relation r C = 1~r P between the distances of C and P from its centre. Can you find that
Challenge 330 ny circle?
Ref. 139 Another famous challenge is to devise a wooden carriage, with gearwheels that
connect the wheels to an arrow in such a way that whatever path the carriage takes,
the arrow always points south (see Figure 90). The solution to this is useful in helping
Copyright © Christoph Schiller November 1997–May 2006

Challenge 331 d
us to understand general relativity, as we will see.
Another interesting example of rigid motion is the way that human movements,
such as the general motions of an arm, are composed from a small number of basic
Ref. 140 motions. All these examples are from the fascinating field of engineering; unfortu-
nately, we will have little time to explore this topic in our hike.
— The fourth global approach to motion is the description of non-rigid extended bodies.
For example, fluid mechanics studies the flow of fluids (like honey, water or air) around
solid bodies (like spoons, ships, sails or wings). Fluid mechanics thus seeks to explain
how insects, birds and aeroplanes fly,* why sailboats can sail against the wind, what

* The mechanisms of insect flight are still a subject of active research. Traditionally, fluid dynamics has

Dvipsbugw
 the global simplicity of complexity 175

? or ?

F I G U R E 91 How and where does a falling brick chimney break?

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 92 Why do hot-air balloons F I G U R E 93 What
stay inflated? How can you measure determines the
the weight of a bicycle rider using number of petals in
only a ruler? a daisy?

Ref. 141 happens when a hard-boiled egg is made to spin on a thin layer of water, or how a
Challenge 332 n bottle full of wine can be emptied in the fastest way possible.
As well as fluids, we can study the behaviour of deformable solids. This area of re-
search is called continuum mechanics. It deals with deformations and oscillations of
extended structures. It seeks to explain, for example, why bells are made in particular
Challenge 333 n shapes; how large bodies – such as falling chimneys – break when under stress; and
how cats can turn themselves the right way up as they fall. During the course of our
journey we will repeatedly encounter issues from this field, which impinges even upon
general relativity and the world of elementary particles.
— The fifth global approach to motion is the study of the motion of huge numbers of
particles. This is called statistical mechanics. The concepts needed to describe gases,
such as temperature and pressure (see Figure 92), will be our first steps towards the
understanding of black holes.
— The sixth global approach to motion involves all of the above-mentioned view-
Copyright © Christoph Schiller November 1997–May 2006

points at the same time. Such an approach is needed to understand everyday ex-
perience, and life itself. Why does a flower form a specific number of petals? How
does an embryo differentiate in the womb? What makes our hearts beat? How do
mountains ridges and cloud patterns emerge? How do stars and galaxies evolve?
How are sea waves formed by the wind?

concentrated on large systems, like boats, ships and aeroplanes. Indeed, the smallest human-made object
that can fly in a controlled way – say, a radio-controlled plane or helicopter – is much larger and heavier
than many flying objects that evolution has engineered. It turns out that controlling the flight of small things
requires more knowledge and more tricks than controlling the flight of large things. There is more about this
topic on page 969.

Dvipsbugw
 176 i galilean motion • 3. the global simplicity of complexity

All these are examples of self-organization; life scientists simply speak of growth.
Whatever we call these processes, they are characterized by the spontaneous appear-
ance of patterns, shapes and cycles. Such processes are a common research theme
across many disciplines, including biology, chemistry, medicine, geology and engin-
eering.
We will now give a short introduction to these six global approaches to motion. We will
begin with the first approach, namely, the global description of moving point-like ob-
jects. The beautiful method described below was the result of several centuries of collect- Dvipsbugw
ive effort, and is the highlight of mechanics. It also provides the basis for all the further
descriptions of motion that we will meet later on.

Measuring change with action

Motion can be described by numbers. For a single particle, the relations between the spa-
tial and temporal coordinates describe the motion. The realization that expressions like
(x(t), y(t), z(t)) could be used to describe the path of a moving particle was a milestone

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

in the development of modern physics.
We can go further. Motion is a type of change. And this change can itself be usefully
described by numbers. In fact, change can be measured by a single number. This realiza-
tion was the next important milestone. Physicists took almost two centuries of attempts
to uncover the way to describe change. As a result, the quantity that measures change
has a strange name: it is called (physical) action.* To remember the connection of ’action’
with change, just think about a Hollywood movie: a lot of action means a large amount
of change.
Imagine taking two snapshots of a system at different times. How could you define the
amount of change that occurred in between? When do things change a lot, and when do
they change only a little? First of all, a system with a lot of motion shows a lot of change. So
it makes sense that the action of a system composed of independent subsystems should
be the sum of the actions of these subsystems.
Secondly, change often – but not always – builds up over time; in other cases, recent
change can compensate for previous change. Change can thus increase or decrease with
time.
Thirdly, for a system in which motion is stored, transformed or shifted from one sub-
system to another, the change is smaller than for a system where this is not the case.
Copyright © Christoph Schiller November 1997–May 2006

* Note that this ‘action’ is not the same as the ‘action’ appearing in statements such as ‘every action has an
equal and opposite reaction’. This last usage, coined by Newton, has not stuck; therefore the term has been
recycled. After Newton, the term ‘action’ was first used with an intermediate meaning, before it was finally
given the modern meaning used here. This last meaning is the only meaning used in this text.
Another term that has been recycled is the ‘principle of least action’. In old books it used to have a different
meaning from the one in this chapter. Nowadays, it refers to what used to be called Hamilton’s principle in
the Anglo-Saxon world, even though it is (mostly) due to others, especially Leibniz. The old names and
meanings are falling into disuse and are not continued here.
Behind these shifts in terminology is the story of an intense two-centuries-long attempt to describe mo-
tion with so-called extremal or variational principles: the objective was to complete and improve the work
initiated by Leibniz. These principles are only of historical interest today, because all are special cases of the
Ref. 142 principle of least action described here.

Dvipsbugw
measuring change with action 177

TA B L E 22 Some action values for changes either observed or imagined

Change A p p r ox i m at e a c t i o n
va l u e

Smallest measurable change 0.5 ë 10−34 Js

Exposure of photographic film 1.1 ë 10−34 Js to 10−9 Js
Wing beat of a fruit fly c. 1 pJs
Flower opening in the morning c. 1 nJs Dvipsbugw
Getting a red face c. 10 mJs
Held versus dropped glass 0.8 Js
Tree bent by the wind from one side to the other 500 Js
Making a white rabbit vanish by ‘real’ magic 100 PJs
Hiding a white rabbit c. 0.1 Js
Maximum brain change in a minute c. 5 Js

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Levitating yourself within a minute by 1 m c. 40 kJs
Car crash c. 2 kJs
Birth c. 2 kJs
Change due to a human life c. 1 EJs
Driving car stops within the blink of an eye 20 kJs
Large earthquake c. 1 PJs
Driving car disappears within the blink of an eye 1 ZJs
Sunrise c. 0.1 ZJs
Gamma ray burster before and after explosion c. 1046 Js
Universe after one second has elapsed undefined and undefinable

L
L(t) = T − U

average L

integral
∫ L(t)dt
Copyright © Christoph Schiller November 1997–May 2006

t
∆t t m
ti tf
elapsed time

F I G U R E 94 Defining a total effect as an

accumulation (addition, or integral) of small effects
over time

Dvipsbugw
 178 i galilean motion • 3. the global simplicity of complexity

The mentioned properties imply that the natural measure of

change is the average difference between kinetic and potential
energy multiplied by the elapsed time. This quantity has all the
right properties: it is (usually) the sum of the corresponding
quantities for all subsystems if these are independent; it gener-
ally increases with time (unless the evolution compensates for
something that happened earlier); and it decreases if the sys-
Challenge 334 e tem transforms motion into potential energy.
Thus the (physical) action S, measuring the change in a sys- Dvipsbugw
tem, is defined as
Joseph Lagrange
S = L ë (t f − t i ) = T − U ë (t f − t i ) = (T − U) dt =
tf tf
∫ti
∫
ti
L dt ,
(57)
Page 120 where T is the kinetic energy, U the potential energy we already know, L is the differ-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ence between these, and the overbar indicates a time average. The quantity L is called the
Lagrangian (function) of the system,* describes what is being added over time, whenever
things change. The sign ∫ is a stretched ‘S’, for ‘sum’, and is pronounced ‘integral of ’. In
intuitive terms it designates the operation (called integration) of adding up the values of
a varying quantity in infinitesimal time steps dt. The initial and the final times are writ-
ten below and above the integration sign, respectively. Figure 94 illustrates the idea: the
integral is simply the size of the dark area below the curve L(t).
Challenge 335 e Mathematically, the integral of the curve L(t) is defined as

f
L(t) dt = lim Q L(t m )∆t = L ë (t f − t i ) .
tf
∫ti ∆t 0 m=i
(58)

In other words, the integral is the limit, as the time slices get smaller, of the sum of the
areas of the individual rectangular strips that approximate the function.** Since the P
sign also means a sum, and since an infinitesimal ∆t is written dt, we can understand the
notation used for integration. Integration is a sum over slices. The notation was developed
by Gottfried Leibniz to make exactly this point. Physically speaking, the integral of the
Lagrangian measures the effect that L builds up over time. Indeed, action is called ‘effect’
in some languages, such as German.
In short, then, action is the integral of the Lagrangian over time. The unit of action,
Copyright © Christoph Schiller November 1997–May 2006

and thus of physical change, is the unit of energy (the Joule), times the unit of time (the
second). Thus change is measured in Js. A large value means a big change. Table 22 shows
some approximate values of actions.

* It is named after Giuseppe Lodovico Lagrangia (b. 1736 Torino, d. 1813 Paris), better known as
Joseph Louis Lagrange. He was the most important mathematician of his time; he started his career in Turin,
then worked for 20 years in Berlin, and finally for 26 years in Paris. Among other things he worked on num-
ber theory and analytical mechanics, where he developed most of the mathematical tools used nowadays for
calculations in classical mechanics and classical gravitation. He applied them successfully to many motions
in the solar system.
** For more details on integration see Appendix D.

Dvipsbugw
 measuring change with action 179

F I G U R E 95 The
minimum of a curve has Dvipsbugw
vanishing slope

To understand the definition of action in more detail, we will start with the simplest
case: a system for which the potential energy is zero, such as a particle moving freely. Ob-
viously, a large kinetic energy means a lot of change. If we observe the particle at two in-
stants, the more distant they are the larger the change. Furthermore, the observed change

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is larger if the particle moves more rapidly, as its kinetic energy is larger. This is not sur-
prising.
Next, we explore a single particle moving in a potential. For example, a falling stone
loses potential energy in exchange for a gain in kinetic energy. The more energy is ex-
changed, the more change there is. Hence the minus sign in the definition of L. If we
explore a particle that is first thrown up in the air and then falls, the curve for L(t) first
is below the times axis, then above. We note that the definition of integration makes us
count the grey surface below the time axis negatively. Change can thus be negative, and
be compensated by subsequent change, as expected.
To measure change for a system made of several independent components, we simply
add all the kinetic energies and subtract all the potential energies. This technique allows
us to define actions for gases, liquids and solid matter. Even if the components interact,
we still get a sensible result. In short, action is an additive quantity.
Physical action thus measures, in a single number, the change observed in a system
between two instants of time. The observation may be anything at all: an explosion, a
caress or a colour change. We will discover later that this idea is also applicable in relativity
and quantum theory. Any change going on in any system of nature can be measured with
a single number. Copyright © Christoph Schiller November 1997–May 2006

The principle of least action

We now have a precise measure of change, which, as it turns out, allows a simple and
powerful description of motion. In nature, the change happening between two instants is
always the smallest possible. In nature, action is minimal.* Of all possible motions, nature
always chooses for which the change is minimal. Let us study a few examples.
In the simple case of a free particle, when no potentials are involved, the principle of
minimal action implies that the particle moves in a straight line with constant velocity.
Challenge 336 e All other paths would lead to larger actions. Can you verify this?

* In fact, in some pathological situations the action is maximal, so that the snobbish form of the principle is
that the action is ‘stationary,’ or an ‘extremum,’ meaning minimal or maximal. The condition of vanishing

Dvipsbugw
 180 i galilean motion • 3. the global simplicity of complexity

When gravity is present, a thrown stone flies along a parabola (or more precisely, along
an ellipse) because any other path, say one in which the stone makes a loop in the air,
Challenge 337 e would imply a larger action. Again you might want to verify this for yourself.
All observations support this simple and basic statement: things always move in a way
that produces the smallest possible value for the action. This statement applies to the full
path and to any of its segments. Betrand Russell called it the ‘law of cosmic laziness’.
It is customary to express the idea of minimal change in a different way. The action
varies when the path is varied. The actual path is the one with the smallest action. You will
recall from school that at a minimum the derivative of a quantity vanishes: a minimum Dvipsbugw
has a horizontal slope. In the present case, we do not vary a quantity, but a complete path;
hence we do not speak of a derivative or slope, but of a variation. It is customary to write
the variation of action as δS. The principle of least action thus states:

U The actual trajectory between specified end points satisfies δS = 0. (59)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Mathematicians call this a variational principle. Note that the end points have to be spe-
cified: we have to compare motions with the same initial and final situations.
Before discussing the principle further, we can check that it is equivalent to the evolu-
tion equation.* To do this, we can use a standard procedure, part of the so-called calculus

variation, given below, encompasses both cases.

* For those interested, here are a few comments on the equivalence of Lagrangians and evolution equations.
Page 149 First of all, Lagrangians do not exist for non-conservative, or dissipative systems. We saw that there is no
potential for any motion involving friction (and more than one dimension); therefore there is no action in
these cases. One approach to overcome this limitation is to use a generalized formulation of the principle
of least action. Whenever there is no potential, we can express the work variation δW between different
trajectories x i as
δW = Q m i ẍ i δx i . (60)
i
Motion is then described in the following way:

(δT + δW)dt = 0 provided δx(t i ) = δx(t f ) = 0 .

tf
U The actual trajectory satifies ∫
ti
(61)

The quantity being varied has no name; it represents a generalized notion of change. You might want to
Challenge 338 ny check that it leads to the correct evolution equations. Thus, although proper Lagrangian descriptions exist
only for conservative systems, for dissipative systems the principle can be generalized and remains useful.
Many physicists will prefer another approach. What a mathematician calls a generalization is a special
case for a physicist: the principle (61) hides the fact that all friction results from the usual principle of min-
Copyright © Christoph Schiller November 1997–May 2006

imal action, if we include the complete microscopic details. There is no friction in the microscopic domain.
Friction is an approximate, macroscopic concept.
Nevertheless, more mathematical viewpoints are useful. For example, they lead to interesting limitations
for the use of Lagrangians. These limitations, which apply only if the world is viewed as purely classical –
which it isn’t – were discovered about a hundred years ago. In those times computers where not available,
and the exploration of new calculation techniques was important. Here is a summary.
The coordinates used in connection with Lagrangians are not necessarily the Cartesian ones. Generalized
coordinates are especially useful when there are constraints on the motion. This is the case for a pendulum,
where the weight always has to be at the same distance from the suspension, or for an ice skater, where the
Ref. 143 skate has to move in the direction in which it is pointing. Generalized coordinates may even be mixtures of
positions and momenta. They can be divided into a few general types.
Generalized coordinates are called holonomic–scleronomic if they are related to Cartesian coordinates in
a fixed way, independently of time: physical systems described by such coordinates include the pendulum

Dvipsbugw
 measuring change with action 181

of variations. The condition δS = 0 implies that the action, i.e. the area under the curve
in Figure 94, is a minimum. A little bit of thinking shows that if the Lagrangian is of the
Challenge 339 ny form L(x n , v n ) = T(v n ) − U(x n ), then

Œ ‘=
d ∂T ∂U
(62)
dt ∂v n ∂x n

where n counts all coordinates of all particles.* For a single particle, these Lagrange’s equa-
Dvipsbugw
Challenge 340 e tions of motion reduce to
ma = ∇U . (64)

This is the evolution equation: it says that the force on a particle is the gradient of the
potential energy U. The principle of least action thus implies the equation of motion.
Challenge 341 n (Can you show the converse?)
In other words, all systems evolve in such a way that the change is as small as possible.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Nature is economical. Nature is thus the opposite of a Hollywood thriller, in which the
action is maximized; nature is more like a wise old man who keeps his actions to a min-
imum.
The principle of minimal action also states that the actual trajectory is the one for
which the average of the Lagrangian over the whole trajectory is minimal (see Figure 94).
Nature is a Dr. Dolittle. Can you verify this? This viewpoint allows one to deduce Lag-
Challenge 342 ny range’s equations (62) directly.
The principle of least action distinguishes the actual trajectory from all other imagin-
able ones. This observation lead Leibniz to his famous interpretation that the actual world

and a particle in a potential. Coordinates are called holonomic–rheonomic if the dependence involves time.
An example of a rheonomic systems would be a pendulum whose length depends on time. The two terms
Page 242 rheonomic and scleronomic are due to Ludwig Boltzmann. These two cases, which concern systems that are
only described by their geometry, are grouped together as holonomic systems. The term is due to Heinrich
Page 559 Hertz.
The more general situation is called anholonomic, or nonholonomic. Lagrangians work well only for holo-
nomic systems. Unfortunately, the meaning of the term ‘nonholonomic’ has changed. Nowadays, the term
is also used for certain rheonomic systems. The modern use calls nonholonomic any system which involves
velocities. Therefore, an ice skater or a rolling disc is often called a nonholonomic system. Care is thus ne-
cessary to decide what is meant by nonholonomic in any particular context.
Even though the use of Lagrangians, and of action, has its limitations, these need not bother us at micro-
Copyright © Christoph Schiller November 1997–May 2006

scopic level, since microscopic systems are always conservative, holonomic and scleronomic. At the funda-
mental level, evolution equations and Lagrangians are indeed equivalent.
* The most general form for a Lagrangian L(q n , q̇ n , t), using generalized holonomic coordinates q n , leads
to Lagrange equations of the form
Œ ‘=
d ∂L ∂L
. (63)
dt ∂ q̇ n ∂q n
In order to deduce these equations, we also need the relation δ q̇ = d~dt(δq). This relation is valid only for
holonomic coordinates introduced in the previous footnote and explains their importance.
It should also be noted that the Lagrangian for a moving system is not unique; however, the study of how
Ref. 144 the various Lagrangians for a given moving system are related is not part of this walk.
By the way, the letter q for position and p for momentum were introduced in physics by the mathem-
atician Carl Jacobi (b. 1804 Potsdam, d. 1851 Berlin).

Dvipsbugw
 182 i galilean motion • 3. the global simplicity of complexity

is the ‘best of all possible worlds.’* We may dismiss this as metaphysical speculation, but
we should still be able to feel the fascination of the issue. Leibniz was so excited about the
principle of least action because it was the first time that actual observations were distin-
guished from all other imaginable possibilities. For the first time, the search for reasons
why things are the way they are became a part of physical investigation. Could the world
be different from what it is? In the principle of least action, we have a hint of a negative
Challenge 343 n answer. (What do you think?) The final answer will emerge only in the last part of our
adventure.
As a way to describe motion, the Lagrangian has several advantages over the evolu- Dvipsbugw
tion equation. First of all, the Lagrangian is usually more compact than writing the corres-
ponding evolution equations. For example, only one Lagrangian is needed for one system,
however many particles it includes. One makes fewer mistakes, especially sign mistakes,
as one rapidly learns when performing calculations. Just try to write down the evolution
equations for a chain of masses connected by springs; then compare the effort with a deriv-
Challenge 344 ny ation using a Lagrangian. (The system behaves like a chain of atoms.) We will encounter

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

another example shortly: David Hilbert took only a few weeks to deduce the equations
of motion of general relativity using a Lagrangian, whereas Albert Einstein had worked
for ten years searching for them directly.
In addition, the description with a Lagrangian is valid with any set of coordinates de-
scribing the objects of investigation. The coordinates do not have to be Cartesian; they
can be chosen as one prefers: cylindrical, spherical, hyperbolic, etc. These so-called gen-
eralized coordinates allow one to rapidly calculate the behaviour of many mechanical sys-
tems that are in practice too complicated to be described with Cartesian coordinates. For
example, for programming the motion of robot arms, the angles of the joints provide a
clearer description than Cartesian coordinates of the ends of the arms. Angles are non-
Cartesian coordinates. They simplify calculations considerably: the task of finding the
most economical way to move the hand of a robot from one point to another can be
solved much more easily with angular variables.
More importantly, the Lagrangian allows one to quickly deduce the essential proper-
ties of a system, namely, its symmetries and its conserved quantities. We will develop this
Page 195 important idea shortly, and use it regularly throughout our walk.
Finally, the Lagrangian formulation can be generalized to encompass all types of inter-
actions. Since the concepts of kinetic and potential energy are general, the principle of
least action can be used in electricity, magnetism and optics as well as mechanics. The
principle of least action is central to general relativity and to quantum theory, and allows
Copyright © Christoph Schiller November 1997–May 2006

one to easily relate both fields to classical mechanics.

As the principle of least action became well known, people applied it to an ever-increa-
Ref. 142 sing number of problems. Today, Lagrangians are used in everything from the study of ele-
mentary particle collisions to the programming of robot motion in artificial intelligence.
However, we should not forget that despite its remarkable simplicity and usefulness, the
Lagrangian formulation is equivalent to the evolution equations. It is neither more gen-
Challenge 345 n eral nor more specific. In particular, it is not an explanation for any type of motion, but
only a view of it. In fact, the search of a new physical ‘law’ of motion is just the search

* This idea was ridiculed by the French philosopher Voltaire (1694–1778) in his lucid writings, notably in
the brilliant book Candide, written in 1759, and still widely available.

Dvipsbugw
 measuring change with action 183

for a new Lagrangian. This makes sense, as the description of nature always requires the
description of change. Change in nature is always described by actions and Lagrangians.
The principle of least action states that the action is minimal when the end point of
Ref. 145 the motion, and in particular the time between them, are fixed. It is less well known that
the reciprocal principle also holds: if the action is kept fixed, the elapsed time is maximal.
Challenge 346 ny Can you show this?
Even though the principle of least action is not an explanation of motion, it somehow
calls for one. We need some patience, though. Why nature follows the principle of least
action, and how it does so, will become clear when we explore quantum theory. Dvipsbugw
Never confuse movement with action.
Ref. 146

Why is motion so often bounded?

“ Ernest Hemingway

”
The optimist thinks this is the best of all

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

possible worlds, and the pessimist knows it.
Robert Oppenheimer

Ref. 147
Looking around ourselves on Earth and in the sky, we find that matter is not evenly dis-
tributed. Matter tends to be near other matter: it is lumped together in aggregates. Some
major examples of aggregates are given in Figure 96 and Table 23. In the mass–size dia-
”
gram of Figure 96, both scales are logarithmic. One notes three straight lines: a line m  l
extending from the Planck mass* upwards, via black holes, to the universe itself; a line
m  1~l extending from the Planck mass downwards, to the lightest possible aggregate;
and the usual matter line with m  l 3 , extending from atoms upwards, via the Earth and
the Sun. The first of the lines, the black hole limit, is explained by general relativity; the
last two, the aggregate limit and the common matter line, by quantum theory.**
The aggregates outside the common matter line also show that the stronger the inter-
action that keeps the components together, the smaller the aggregate. But why is matter
mainly found in lumps?
First of all, aggregates form because of the existence of attractive interactions between
objects. Secondly, they form because of friction: when two components approach, an ag-
gregate can only be formed if the released energy can be changed into heat. Thirdly, ag-
gregates have a finite size because of repulsive effects that prevent the components from
collapsing completely. Together, these three factors ensure that bound motion is much
more common than unbound, ‘free’ motion.
Copyright © Christoph Schiller November 1997–May 2006

Only three types of attraction lead to aggregates: gravity, the attraction of electric
charges, and the strong nuclear interaction. Similarly, only three types of repulsion are
observed: rotation, pressure, and the Pauli exclusion principle (which we will encounter
Page 785 later on). Of the nine possible combinations of attraction and repulsion, not all appear in
Challenge 347 n nature. Can you find out which ones are missing from Figure 96 and Table 23, and why?
Together, attraction, friction and repulsion imply that change and action are minim-
ized when objects come and stay together. The principle of least action thus implies the
¼
Page 1157 * The Planck mass is given by m Pl = ħc~G = 21.767(16) µg.
** Figure 96 suggests that domains beyond physics exist; we will discover later on that this is not the case, as
mass and size are not definable in those domains.

Dvipsbugw
 184 i galilean motion • 3. the global simplicity of complexity

universe
mass
[kg]

galaxy
1040 black
holes star cluster
Sun

it
lim
Earth
Dvipsbugw

e
neutron

ol
1020

kh
bl :
e ce
star

th ien
ac
nd sc
beyond science: beyond Planck length limit
mountain

yo nd

beyond science: undefined

e
be eyo

in
b

ter l
mat
human
100

mon
Planck mass

com

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

cell

heavy DNA
10-20 nucleus
uranium
muon hydrogen
proton
electron
Aggregates
m

10-40 neutrino
icr
os
co
pi
ca
gg

Elementary
re
ga

particles lightest
te
lim

10-60 imaginable
it

aggregate

10-40 10-20 100 1020 size [m]

F I G U R E 96 Aggregates in nature

stability of aggregates. By the way, formation history also explains why so many aggreg-
Challenge 348 ny ates rotate. Can you tell why?
Copyright © Christoph Schiller November 1997–May 2006

But why does friction exist at all? And why do attractive and repulsive interactions
exist? And why is it – as it would appear from the above – that in some distant past matter
was not found in lumps? In order to answer these questions, we must first study another
global property of motion: symmetry.

TA B L E 23 Some major aggregates observed in nature

A g g r e g at e Size O b s. C o n st i t u e n t s
(diameter) num.

gravitationally bound aggregates

Dvipsbugw
measuring change with action 185

A g g r e g at e Size O b s. C o n st i t u e n t s
(diameter) num.

matter across universe c. 100 Ym 1 superclusters of galaxies, hydrogen

andhelium atoms
quasar 1012 to 1014 m 20 ë 106 baryons and leptons
supercluster of galaxies c. 3 Ym 107 galaxy groups and clusters
galaxy cluster c. 60 Zm 25 ë 109 10 to 50 galaxies
galaxy group or cluster c. 240 Zm 50 to over 2000 galaxies
Dvipsbugw
our local galaxy group 50 Zm 1 c. 40 galaxies
general galaxy 0.5 to 2 Zm 3.5 ë 1012 1010 to 3 ë 1011 stars, dust and gas
clouds, probably solar systems
our galaxy 1.0(0.1) Zm 1 1011 stars, dust and gas clouds, solar
systems
interstellar clouds up to 15 Em Q 105 hydrogen, ice and dust

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

solar system a unknown A 100 star, planets
our solar system 30 Pm 1 Sun, planets (Pluto’s orbit’s diameter:
11.8 Tm), moons, planetoids, comets,
asteroids, dust, gas
Oort cloud 6 to 30 Pm 1 comets, dust
Kuiper belt 60 Tm 1 planetoids, comets, dust
star b 10 km to 100 Gm 10221 ionized gas: protons, neutrons,
electrons, neutrinos, photons
our star 1.39 Gm
planet a (Jupiter, Earth) 143 Mm, 12.8 Mm 9+c. 100 solids, liquids, gases; in particular,
heavy atoms
planetoids (Varuna, etc) 50 to 1 000 km c. 10 solids
(est. 109 )
moons 10 to 1 000 km c. 50 solids
neutron stars 10 km c. 1000 mainly neutrons
electromagnetically bound aggregates c
asteroids, mountains d 1 m to 930 km A26 000 (109 estimated) solids, usually
monolithic
comets 10 cm to 50 km A 106 ice and dust
Copyright © Christoph Schiller November 1997–May 2006

planetoids, solids, liquids, 1 nm to A 100 km n.a. molecules, atoms

gases, cheese
animals, plants, kefir 5 µm to 1 km 10262 organs, cells
brain 0.15 m 1010 neurons and other cell types
cells: 10311 organelles, membranes, molecules
smallest (nanobacteria) c. 5 µm molecules
amoeba 600 µm molecules
largest (whale nerve, c. 30 m molecules
single-celled plants)
molecules: c. 10782 atoms

Dvipsbugw
 186 i galilean motion • 3. the global simplicity of complexity

A g g r e g at e Size O b s. C o n st i t u e n t s
(diameter) num.

H2 c. 50 pm 10722 atoms
DNA (human) 2 m (total per cell) 1021 atoms
atoms, ions 30 pm to 300 pm 10802 electrons and nuclei
aggregates bound by the weak interaction c
none Dvipsbugw
c
aggregates bound by the strong interaction
nucleus A 10−15 m 10792 nucleons
nucleon (proton, neutron) c. 10−15 m 10802 quarks
mesons c. 10−15 m n.a. quarks
neutron stars: see above

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

a. Only in 1994 was the first evidence found for objects circling stars other than our Sun; of over 100 extrasolar
planets found so far, most are found around F, G and K stars, including neutron stars. For example, three
Ref. 148 objects circle the pulsar PSR 1257+12, and a matter ring circles the star β Pictoris. The objects seem to be dark
stars, brown dwarfs or large gas planets like Jupiter. Due to the limitations of observation systems, none of
the systems found so far form solar systems of the type we live in. In fact, only a few Earth-like planets have
been found so far.
b. The Sun is among the brightest 7 % of stars. Of all stars, 80 %, are red M dwarfs, 8 % are orange K dwarfs,
and 5 % are white D dwarfs: these are all faint. Almost all stars visible in the night sky belong to the bright
7 %. Some of these are from the rare blue O class or blue B class (such as Spica, Regulus and Riga); 0.7 %
consist of the bright, white A class (such as Sirius, Vega and Altair); 2 % are of the yellow–white F class (such
as Canopus, Procyon and Polaris); 3.5 % are of the yellow G class (like Alpha Centauri, Capella or the Sun).
Exceptions include the few visible K giants, such as Arcturus and Aldebaran, and the rare M supergiants,
Page 466 such as Betelgeuse and Antares. More on stars later on.
c. For more details on microscopic aggregates, see the table of composites in Appendix C.
d. It is estimated that there are about 109 asteroids (or planetoids) larger than 1 km and about 1020 that are
Ref. 149 heavier than 100 kg. By the way, no asteroids between Mercury and the Sun – the hypothetical Vulcanoids –
have been found so far.

Curiosities and fun challenges about L agrangians

Lagrangians and variational principles form a fascinating topic, which has charmed phys-
Copyright © Christoph Schiller November 1997–May 2006

icists for the last four centuries.

**
When Lagrange published his book Mécanique analytique, in 1788, it formed one of the
high points in the history of mechanics. He was proud of having written a systematic
exposition of mechanics without a single figure. Obviously the book was difficult to read
and was not a sales success. Therefore his methods took another generation to come into
general use.

**
Given that action is the basic quantity describing motion, we can define energy as action

Dvipsbugw
 measuring change with action 187

per unit time, and momentum as action per unit distance. The energy of a system thus
describes how much it changes over time, and the momentum how much it changes over
Challenge 349 n distance. What are angular momentum and rotational energy?
**
‘In nature, effects of telekinesis or prayer are impossible, as in most cases the change inside
the brain is much smaller than the change claimed in the outside world.’ Is this argument
Challenge 350 n correct?
Dvipsbugw
**
In Galilean physics, the Lagrangian is the difference between kinetic and potential energy.
Later on, this definition will be generalized in a way that sharpens our understanding of
this distinction: the Lagrangian becomes the difference between a term for free particles
and a term due to their interactions. In other words, particle motion is a continuous
compromise between what the particle would do if it were free and what other particles

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

want it to do. In this respect, particles behave a lot like humans beings.

**
Challenge 351 ny Explain: why is T + U constant, whereas T − U is minimal?
**
In nature, the sum T + U of kinetic and potential energy is constant during motion (for
closed systems), whereas the average of the difference T −U is minimal. Is it possible to de-
duce, by combining these two facts, that systems tend to a state with minimum potential
Challenge 352 ny energy?
**
There is a principle of least effort describing the growth of trees. When a tree – a mono-
podal phanerophyte – grows and produces leaves, between 40% and 60% of the mass it
consists of, namely the water and the minerals, has to be lifted upwards from the ground.*
Therefore, a tree gets as many branches as high up in the air as possible using the smallest
amount of energy. This is the reason why not all leaves are at the very top of a tree. Can
Challenge 353 ny you deduce more details about trees from this principle?
**
Copyright © Christoph Schiller November 1997–May 2006

Another minimization principle can be used to understand the construction of animal

Ref. 150 bodies, especially their size and the proportions of their inner structures. For example,
the heart pulse and breathing frequency both vary with animal mass m as m−1~4 , and
the dissipated power varies as m 3~4 . It turns out that such exponents result from three
properties of living beings. First, they transport energy and material through the organ-
ism via a branched network of vessels: a few large ones, and increasingly many smaller
ones. Secondly, the vessels all have the same minimum size. And thirdly, the networks
are optimized in order to minimize the energy needed for transport. Together, these rela-
tions explain many additional scaling rules; they might also explain why animal lifespan

* The rest of the mass comes form the CO2 in the air.

Dvipsbugw
 188 i galilean motion • 3. the global simplicity of complexity

α air

water

Dvipsbugw
F I G U R E 97 Refraction of
light is due to travel-time
optimization

scales as m−1~4 , or why most mammals have roughly the same number of heart beats in

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

a lifetime.
A competing explanation, using a different minimization principle, states that quarter
powers arise in any network built in order that the flow arrives to the destination by the
Ref. 151 most direct path.
**
The minimization principle for the motion of light is even more beautiful: light always
takes the path that requires the shortest travel time. It was known long ago that this idea
describes exactly how light changes direction when it moves from air to water. In water,
light moves more slowly; the speed ratio between air and water is called the refractive in-
dex of water. The refractive index, usually abbreviated n, is material-dependent. The value
for water is about 1.3. This speed ratio, together with the minimum-time principle, leads
to the ‘law’ of refraction, a simple relation between the sines of the two angles. Can you
Challenge 354 n deduce it? (In fact, the exact definition of the refractive index is with respect to vacuum,
Challenge 355 n not to air. But the difference is negligible: can you imagine why?)
For diamond, the refractive index is 2.4. The high value is one reason for the sparkle
Challenge 356 n of diamonds cut with the 57-face brilliant cut. Can you think of some other reasons?
**
Can you confirm that each of these minimization principles is a special case of the prin-
Copyright © Christoph Schiller November 1997–May 2006

Challenge 357 n ciple of least action? In fact, this is the case for all known minimization principles in
nature. Each of them, like the principle of least action, is a principle of least change.

**
In Galilean physics, the value of the action depends on the speed of the observer, but not
on his position or orientation. But the action, when properly defined, should not depend
on the observer. All observers should agree on the value of the observed change. Only
special relativity will fulfil the requirement that action be independent of the observer’s
Challenge 358 n speed. How will the relativistic action be defined?

**

Dvipsbugw
 motion and symmetry 189

Dvipsbugw

F I G U R E 98 Forget-me-not, also called Myosotis

(Boraginaceae) (© Markku Savela)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Measuring all the change that is going on in the universe presupposes that the universe
Challenge 359 n is a physical system. Is this the case?

Motion and symmetry

The second way to describe motion globally is to describe it in such a way that all observ-
ers agree. An object under observation is called symmetric if it looks the same when seen
from different points of view. For example, a forget-me-not flower, shown in Figure 98, is
symmetrical because it looks the same after turning around it by 72 degrees; many fruit
tree flowers have the same symmetry. One also says that under change of viewpoint the
flower has an invariant property, namely its shape. If many such viewpoints are possible,
one talks about a high symmetry, otherwise a low symmetry. For example, a four-leaf
clover has a higher symmetry than a usual, three-leaf one. Different points of view imply
different observers; in physics, the viewpoints are often called frames of reference and are
described mathematically by coordinate systems.
High symmetry means many agreeing observers. At first sight, not many objects or
observations in nature seem to be symmetrical. But this is a mistake. On the contrary,
we can deduce that nature as a whole is symmetric from the simple fact that we have the
Copyright © Christoph Schiller November 1997–May 2006

Challenge 360 n ability to talk about it! Moreover, the symmetry of nature is considerably higher than that
of a forget-me-not. We will discover that this high symmetry is at the basis of the famous
expression E 0 = mc 2 .

Why can we think and talk?

The hidden harmony is stronger than the

Ref. 152
“ apparent.
Heraclitos of Ephesos, about 500 bce

Why can we understand somebody when he is talking about the world, even though we ”
are not in his shoes? We can for two reasons: because most things look similar from differ-

Dvipsbugw
 190 i galilean motion • 3. the global simplicity of complexity

ent viewpoints, and because most of us have already had similar experiences beforehand.
‘Similar’ means that what we and what others observe somehow correspond. In other
words, many aspects of observations do not depend on viewpoint. For example, the num-
ber of petals of a flower has the same value for all observers. We can therefore say that this
quantity has the highest possible symmetry. We will see below that mass is another such
example. Observables with the highest possible symmetry are called scalars in physics.
Other aspects change from observer to observer. For example, the apparent size varies
with the distance of observation. However, the actual size is observer-independent. In
general terms, any type of viewpoint-independence is a form of symmetry, and the obser- Dvipsbugw
vation that two people looking at the same thing from different viewpoints can under-
stand each other proves that nature is symmetric. We start to explore the details of this
symmetry in this section and we will continue during most of the rest of our hike.
In the world around us, we note another general property: not only does the same
phenomenon look similar to different observers, but different phenomena look similar
to the same observer. For example, we know that if fire burns the finger in the kitchen, it

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

will do so outside the house as well, and also in other places and at other times. Nature
shows reproducibility. Nature shows no surprises. In fact, our memory and our thinking
Challenge 361 n are only possible because of this basic property of nature. (Can you confirm this?) As
we will see, reproducibility leads to additional strong restrictions on the description of
nature.
Without viewpoint-independence and reproducibility, talking to others or to one-
self would be impossible. Even more importantly, we will discover that viewpoint-
independence and reproducibility do more than determine the possibility of talking to
each other: they also fix the content of what we can say to each other. In other words, we
will see that our description of nature follows logically, almost without choice, from the
simple fact that we can talk about nature to our friends.

Viewpoints
Tolerance ... is the suspicion that the other

“ might be right.
Kurt Tucholski (1890–1935), German writer
Tolerance – a strength one mainly wishes to
”
“ political opponents.
Wolfram Weidner (b. 1925) German journalist

When a young human starts to meet other people in childhood, it quickly finds out that ”
Copyright © Christoph Schiller November 1997–May 2006

certain experiences are shared, while others, such as dreams, are not. Learning to make
this distinction is one of the adventures of human life. In these pages, we concentrate
on a section of the first type of experiences: physical observations. However, even among
these, distinctions are to be made. In daily life we are used to assuming that weights,
volumes, lengths and time intervals are independent of the viewpoint of the observer.
We can talk about these observed quantities to anybody, and there are no disagreements
over their values, provided they have been measured correctly. However, other quantities
do depend on the observer. Imagine talking to a friend after he jumped from one of the
trees along our path, while he is still falling downwards. He will say that the forest floor
is approaching with high speed, whereas the observer below will maintain that the floor

Dvipsbugw
 motion and symmetry 191

is stationary. Obviously, the difference between the statements is due to their different
viewpoints. The velocity of an object (in this example that of the forest floor or of the
friend himself) is thus a less symmetric property than weight or size. Not all observers
agree on its value.
In the case of viewpoint-dependent observations, understanding is still possible with
the help of a little effort: each observer can imagine observing from the point of view of
the other, and check whether the imagined result agrees with the statement of the other.*
If the statement thus imagined and the actual statement of the other observer agree, the
observations are consistent, and the difference in statements is due only to the different Dvipsbugw
viewpoints; otherwise, the difference is fundamental, and they cannot agree or talk. Using
this approach, you can even argue whether human feelings, judgements, or tastes arise
Challenge 362 n from fundamental differences or not.
The distinction between viewpoint-independent (invariant) and viewpoint-
dependent quantities is an essential one. Invariant quantities, such as mass or shape,
describe intrinsic properties, and quantities depending on the observer make up the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

state of the system. Therefore, we must answer the following questions in order to find a
complete description of the state of a physical system:
— Which viewpoints are possible?
— How are descriptions transformed from one viewpoint to another?
— Which observables do these symmetries admit?
— What do these results tell us about motion?
In the discussion so far, we have studied viewpoints differing in location, in orientation,
in time and, most importantly, in motion. With respect to each other, observers can be at
rest, move with constant speed, or accelerate. These ‘concrete’ changes of viewpoint are
those we will study first. In this case the requirement of consistency of observations made
Page 86 by different observers is called the principle of relativity. The symmetries associated with
Page 200 this type of invariance are also called external symmetries. They are listed in Table 25.
A second class of fundamental changes of viewpoint concerns ‘abstract’ changes. View-
points can differ by the mathematical description used: such changes are called changes
of gauge. They will be introduced first in the section on electrodynamics. Again, it is re-
quired that all statements be consistent across different mathematical descriptions. This
requirement of consistency is called the principle of gauge invariance. The associated sym-
metries are called internal symmetries.
The third class of changes, whose importance may not be evident from everyday life,
Copyright © Christoph Schiller November 1997–May 2006

is that of the behaviour of a system under exchange of its parts. The associated invariance
is called permutation symmetry. It is a discrete symmetry, and we will encounter it in the
second part of our adventure.
The three consistency requirements described above are called ‘principles’ because
these basic statements are so strong that they almost completely determine the ‘laws’ of
physics, as we will see shortly. Later on we will discover that looking for a complete de-
scription of the state of objects will also yield a complete description of their intrinsic

* Humans develop the ability to imagine that others can be in situations different from their own at the
Ref. 153 age of about four years. Therefore, before the age of four, humans are unable to conceive special relativity;
afterwards, they can.

Dvipsbugw
 192 i galilean motion • 3. the global simplicity of complexity

properties. But enough of introduction: let us come to the heart of the topic.

Symmetries and groups

Since we are looking for a complete description of motion, we need to understand and de-
scribe the full set of symmetries of nature. A system is said to be symmetric or to possess
a symmetry if it appears identical when observed from different viewpoints. We also say
that the system possesses an invariance under change from one viewpoint to the other.
Viewpoint changes are called symmetry operations or transformations. A symmetry is thus Dvipsbugw
a transformation, or more generally, a set of transformations. However, it is more than
that: the successive application of two symmetry operations is another symmetry oper-
ation. To be more precise, a symmetry is a set G = ˜a, b, c, ... of elements, the trans-
formations, together with a binary operation X called concatenation or multiplication and
pronounced ‘after’ or ‘times’, in which the following properties hold for all elements a, b
and c:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

associativity, i.e. (a X b) X c = a X (b X c)
a neutral element e exists such that e X a = a X e = a
an inverse element a −1 exists such that a −1 X a = a X a −1 = e . (65)

Any set that fulfils these three defining properties, or axioms, is called a (mathematical)
group. Historically, the notion of group was the first example of a mathematical structure
which was defined in a completely abstract manner.* Can you give an example of a group
Challenge 363 n taken from daily life? Groups appear frequently in physics and mathematics, because
Ref. 154 symmetries are almost everywhere, as we will see.** Can you list the symmetry operations
Challenge 364 n of the pattern of Figure 99?

Representations
Looking at a symmetric and composed system such as the one shown in Figure 99, we
Challenge 365 e notice that each of its parts, for example each red patch, belongs to a set of similar objects,
usually called a multiplet. Taken as a whole, the multiplet has (at least) the symmetry
properties of the whole system. For some of the coloured patches in Figure 99 we need
four objects to make up a full multiplet, whereas for others we need two, or only one, as
in the case of the central star. In fact, in any symmetric system each part can be classified
Copyright © Christoph Schiller November 1997–May 2006

according to what type of multiplet it belongs to. Throughout our mountain ascent we will
perform the same classification with every part of nature, with ever-increasing precision.

* The term is due to Evariste Galois (1811–1832), the structure to Augustin-Louis Cauchy (1789–1857) and
the axiomatic definition to Arthur Cayley (1821–1895).
** In principle, mathematical groups need not be symmetry groups; but it can be proven that all groups can
be seen as transformation groups on some suitably defined mathematical space, so that in mathematics we
can use the terms ‘symmetry group’ and ‘group’ interchangeably.
A group is called Abelian if its concatenation operation is commutative, i.e. if a X b = b X a for all pairs of
elements a and b. In this case the concatenation is sometimes called addition. Do rotations form an abelian
group?
A subset G 1 ⊂ G of a group G can itself be a group; one then calls it a subgroup and often says sloppily
that G is larger than G 1 or that G is a higher symmetry group than G 1 .

Dvipsbugw
 motion and symmetry 193

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © 1990 Christoph Schiller

F I G U R E 99 A Hispano–Arabic ornament from the Governor’s Palace in Sevilla

A multiplet is a set of parts that transform into each other under all symmetry trans-
formations. Mathematicians often call abstract multiplets representations. By specifying
Copyright © Christoph Schiller November 1997–May 2006

to which multiplet a component belongs, we describe in which way the component is

part of the whole system. Let us see how this classification is achieved.
In mathematical language, symmetry transformations are often described by matrices.
For example, in the plane, a reflection along the first diagonal is represented by the matrix

D(refl) = Œ ‘ ,
0 1
(66)
1 0

since every point (x, y) becomes transformed to (y, x) when multiplied by the matrix
Challenge 366 e D(refl). Therefore, for a mathematician a representation of a symmetry group G is an
assignment of a matrix D(a) to each group element a such that the representation of

Dvipsbugw
 194 i galilean motion • 3. the global simplicity of complexity

the concatenation of two elements a and b is the product of the representations D of the
elements:
D(a X b) = D(a)D(b) . (67)

For example, the matrix of equation (66), together with the corresponding matrices for
all the other symmetry operations, have this property.*
For every symmetry group, the construction and classification of all possible represent-
ations is an important task. It corresponds to the classification of all possible multiplets a
symmetric system can be made of. In this way, understanding the classification of all mul- Dvipsbugw
tiplets and parts which can appear in Figure 99 will teach us how to classify all possible
parts of which an object or an example of motion can be composed!
A representation D is called unitary if all matrices D(a) are unitary.** Almost all rep-
resentations appearing in physics, with only a handful of exceptions, are unitary: this
term is the most restrictive, since it specifies that the corresponding transformations are
one-to-one and invertible, which means that one observer never sees more or less than

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

another. Obviously, if an observer can talk to a second one, the second one can also talk
to the first.
The final important property of a multiplet, or representation, concerns its structure.
If a multiplet can be seen as composed of sub-multiplets, it is called reducible, else irre-
ducible; the same is said about representations. The irreducible representations obviously
cannot be decomposed any further. For example, the symmetry group of Figure 99, com-
monly called D4 , has eight elements. It has the general, faithful, unitary and irreducible
Challenge 367 e matrix representation

* There are some obvious, but important, side conditions for a representation: the matrices D(a) must be
invertible, or non-singular, and the identity operation of G must be mapped to the unit matrix. In even more
compact language one says that a representation is a homomorphism from G into the group of non-singular
or invertible matrices. A matrix D is invertible if its determinant det D is not zero.
In general, if a mapping f from a group G to another G ′ satisfies

f (a XG b) = f (a) XG ′ f (b) , (68)

the mapping f is called an homomorphism. A homomorphism f that is one-to-one (injective) and onto
(surjective) is called a isomorphism. If a representation is also injective, it is called faithful, true or proper.
Copyright © Christoph Schiller November 1997–May 2006

In the same way as groups, more complex mathematical structures such as rings, fields and associative
algebras may also be represented by suitable classes of matrices. A representation of the field of complex
numbers is given in Appendix D.
** The transpose AT of a matrix A is defined element-by-element by (AT )ik = A ki . The complex conjugate A‡
of a matrix A is defined by (A‡ )ik = (A ik )‡ . The adjoint A† of a matrix A is defined by A† = (AT )‡ . A matrix
is called symmetric if AT = A, orthogonal if AT = A−1 , Hermitean or self-adjoint (the two are synonymous
in all physical applications) if A† = A (Hermitean matrices have real eigenvalues), and unitary if A† = A−1 .
Unitary matrices have eigenvalues of norm one. Multiplication by a unitary matrix is a one-to-one mapping;
since the time evolution of physical systems is a mapping from one time to another, evolution is always
described by a unitary matrix. A real matrix obeys A‡ = A, an antisymmetric or skew-symmetric matrix
is defined by AT = −A, an anti-Hermitean matrix by A† = −A and an anti-unitary matrix by A† = −A−1 .
All the mappings described by these special types of matrices are one-to-one. A matrix is singular, i.e. not
one-to-one, if det A = 0.

Dvipsbugw
 motion and symmetry 195

TA B L E 24 Correspondences between the symmetries of an ornament, a flower and nature as a whole

System H i s pa n o – A r - Flower Motion

abic
pat t e r n
Structure and set of ribbons and set of petals, stem motion path and
components patches observables
System pattern symmetry flower symmetry symmetry of Lagrangian Dvipsbugw
symmetry
Mathematical D4 C5 in Galilean relativity:
description of the position, orientation,
symmetry group instant and velocity changes
Invariants number of multiplet petal number number of coordinates,
elements magnitude of scalars,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

vectors and tensors
Representations multiplet types of multiplet types of tensors, including scalars
of the elements components and vectors
components
Most symmetric singlet part with circular scalar
representation symmetry
Simplest faithful quartet quintet vector
representation
Least symmetric quartet quintet no limit (tensor of infinite
representation rank)

Œ
cos nπ~2 − sin nπ~2
‘ n = 0..3, Œ
−1 0
‘,Œ ‘,Œ ‘,Œ
0 −1
‘.
1 0 0 1
sin nπ~2 cos nπ~2 0 1 0 −1 1 0 −1 0
(69)
The representation is an octet. The complete list of possible irreducible representations
Challenge 368 ny of the group D4 is given by singlets, doublets and quartets. Can you find them all? These
representations allow the classification of all the white and black ribbons that appear in
Copyright © Christoph Schiller November 1997–May 2006

the figure, as well as all the coloured patches. The most symmetric elements are singlets,
the least symmetric ones are members of the quartets. The complete system is always a
singlet as well.
With these concepts we are ready to talk about motion with improved precision.

Symmetries, motion and Galilean physics

Every day we experience that we are able to talk to each other about motion. It must
therefore be possible to find an invariant quantity describing it. We already know it: it is
the action. Lighting a match is a change. It is the same whether it is lit here or there, in
one direction or another, today or tomorrow. Indeed, the (Galilean) action is a number
whose value is the same for each observer at rest, independent of his orientation or the

Dvipsbugw
 196 i galilean motion • 3. the global simplicity of complexity

time at which he makes his observation.

In the case of the Arabic pattern of Figure 99, the symmetry allows us to deduce the
list of multiplets, or representations, that can be its building blocks. This approach must
be possible for motion as well. We deduced the classification of the ribbons in the Arabic
pattern into singlets, doublets, etc. from the various possible observation viewpoints. For
a moving system, the building blocks, corresponding to the ribbons, are the observables.
Since we observe that nature is symmetric under many different changes of viewpoint,
we can classify all observables. To do so, we need to take the list of all viewpoint trans-
formations and deduce the list of all their representations. Dvipsbugw
Our everyday life shows that the world stays unchanged after changes in position, ori-
entation and instant of observation. One also speaks of space translation invariance, rota-
tion invariance and time translation invariance. These transformations are different from
those of the Arabic pattern in two respects: they are continuous and they are unboun-
ded. As a result, their representations will generally be continuously variable and without
bounds: they will be quantities or magnitudes. In other words, observables will be con-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

structed with numbers. In this way we have deduced why numbers are necessary for any
description of motion.*
Since observers can differ in orientation, most representations will be objects possess-
ing a direction. To cut a long story short, the symmetry under change of observation
position, orientation or instant leads to the result that all observables are either ‘scalars’,
‘vectors’ or higher-order ‘tensors.’**
A scalar is an observable quantity which stays the same for all observers: it corresponds
to a singlet. Examples are the mass or the charge of an object, the distance between two
points, the distance of the horizon, and many others. Their possible values are (usually)
continuous, unbounded and without direction. Other examples of scalars are the poten-
tial at a point and the temperature at a point. Velocity is obviously not a scalar; nor is the
Challenge 370 n coordinate of a point. Can you find more examples and counter-examples?
Energy is a puzzling observable. It is a scalar if only changes of place, orientation and
instant of observation are considered. But energy is not a scalar if changes of observer
speed are included. Nobody ever searched for a generalization of energy that is a scalar
also for moving observers. Only Albert Einstein discovered it, completely by accident.
More about this issue shortly.
Any quantity which has a magnitude and a direction and which ‘stays the same’ with
respect to the environment when changing viewpoint is a vector. For example, the arrow
between two fixed points on the floor is a vector. Its length is the same for all observers;
Copyright © Christoph Schiller November 1997–May 2006

its direction changes from observer to observer, but not with respect to its environment.
On the other hand, the arrow between a tree and the place where a rainbow touches the
Earth is not a vector, since that place does not stay fixed with respect to the environment,
when the observer changes.
Mathematicians say that vectors are directed entities staying invariant under coordin-
ate transformations. Velocities of objects, accelerations and field strength are examples of
Challenge 371 e vectors. (Can you confirm this?) The magnitude of a vector is a scalar: it is the same for

* Only scalars, in contrast to vectors and higher-order tensors, may also be quantities which only take a
Challenge 369 e discrete set of values, such as +1 or −1 only. In short, only scalars may be discrete observables.
** Later on, spinors will be added to, and complete, this list.

Dvipsbugw
 motion and symmetry 197

any observer. By the way, a famous and baffling result of nineteenth-century experiments
is that the velocity of light is not a vector for Galilean transformations. This mystery will
be solved shortly.
Tensors are generalized vectors. As an example, take the moment of inertia of an object.
Page 89 It specifies the dependence of the angular momentum on the angular velocity. For any
object, doubling the magnitude of angular velocity doubles the magnitude of angular
momentum; however, the two vectors are not parallel to each other if the object is not a
Page 111 sphere. In general, if any two vector quantities are proportional, in the sense that doubling
the magnitude of one vector doubles the magnitude of the other, but without the two Dvipsbugw
vectors being parallel to each other, then the proportionality ‘factor’ is a (second order)
tensor. Like all proportionality factors, tensors have a magnitude. In addition, tensors
have a direction and a shape: they describe the connection between the vectors they relate.
Just as vectors are the simplest quantities with a magnitude and a direction, so tensors
are the simplest quantities with a magnitude and with a direction depending on a second,
chosen direction. Vectors can be visualized as oriented arrows; tensors can be visualized

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 373 n as oriented ellipsoids.* Can you name another example of tensor?
Let us get back to the description of motion. Table 24 shows that in physical systems we
always have to distinguish between the symmetry of the whole Lagrangian – correspond-
ing to the symmetry of the complete pattern – and the representation of the observables
– corresponding to the ribbon multiplets. Since the action must be a scalar, and since all
observables must be tensors, Lagrangians contain sums and products of tensors only in
combinations forming scalars. Lagrangians thus contain only scalar products or general-
izations thereof. In short, Lagrangians always look like

L = α a i b i + β c jk d jk + γ e l mn f l mn + ... (70)

where the indices attached to the variables a, b, c etc. always come in matching pairs to be
summed over. (Therefore summation signs are usually simply left out.) The Greek letters
represent constants. For example, the action of a free point particle in Galilean physics
was given as
S = L dt =
m
∫ 2
v 2 dt ∫ (71)

which is indeed of the form just mentioned. We will encounter many other cases during
our study of motion.**
Copyright © Christoph Schiller November 1997–May 2006

* A rank-n tensor is the proportionality factor between a rank-1 tensor, i.e. between a vector, and an rank-
(n − 1) tensor. Vectors and scalars are rank 1 and rank 0 tensors. Scalars can be pictured as spheres, vectors
as arrows, and rank-2 tensors as ellipsoids. Tensors of higher rank correspond to more and more complex
shapes.
A vector has the same length and direction for every observer; a tensor (of rank 2) has the same determ-
inant, the same trace, and the same sum of diagonal subdeterminants for all observers.
A vector is described mathematically by a list of components; a tensor (of rank 2) is described by a matrix
of components. The rank or order of a tensor thus gives the number of indices the observable has. Can you
Challenge 372 e show this?
** By the way, is the usual list of possible observation viewpoints – namely different positions, different
observation instants, different orientations, and different velocities – also complete for the action (71)? Sur-
Ref. 155 prisingly, the answer is no. One of the first who noted this fact was Niederer, in 1972. Studying the quantum

Dvipsbugw
 198 i galilean motion • 3. the global simplicity of complexity

Galileo already understood that motion is also invariant under change of viewpoints
Page 86 with different velocity. However, the action just given does not reflect this. It took some
years to find out the correct generalization: it is given by the theory of special relativity.
But before we study it, we need to finish the present topic.

Reproducibility, conservation and Noether’s theorem

I will leave my mass, charge and momentum to

“ science.

The reproducibility of observations, i.e. the symmetry under change of instant of time
Graffito

”
Dvipsbugw

or ‘time translation invariance’, is a case of viewpoint-independence. (That is not obvi-

Challenge 375 ny ous; can you find its irreducible representations?) The connection has several important
consequences. We have seen that symmetry implies invariance. It turns out that for con-
tinuous symmetries, such as time translation symmetry, this statement can be made more

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

precise: for any continuous symmetry of the Lagrangian there is an associated conserved
constant of motion and vice versa. The exact formulation of this connection is the the-
orem of Emmy Noether.* She found the result in 1915 when helping Albert Einstein and
David Hilbert, who were both struggling and competing at constructing general relativity.
Ref. 156 However, the result applies to any type of Lagrangian.
Noether investigated continuous symmetries depending on a continuous parameter b.
A viewpoint transformation is a symmetry if the action S does not depend on the value
of b. For example, changing position as

x ( x+b (74)

theory of point particles, he found that even the action of a Galilean free point particle is invariant under
some additional transformations. If the two observers use the coordinates (t, x) and (τ, ξ), the action (71)
Challenge 374 ny is invariant under the transformations

Rx + x0 + vt αt + β
ξ= and τ= with RT R = 1 and αδ − βγ = 1 . (72)
γt + δ γt + δ

where R describes the rotation from the orientation of one observer to the other, v the velocity between the
two observers, and x0 the vector between the two origins at time zero. This group contains two important
special cases of transformations:
Copyright © Christoph Schiller November 1997–May 2006

The connected, static Galilei group ξ = Rx + x0 + vt and τ=t

x αt + β
The transformation group SL(2,R) ξ = and τ= (73)
γt + δ γt + δ

The latter, three-parameter group includes spatial inversion, dilations, time translation and a set of time-
dependent transformations such as ξ = x~t, τ = 1~t called expansions. Dilations and expansions are rarely
mentioned, as they are symmetries of point particles only, and do not apply to everyday objects and systems.
They will return to be of importance later on, however.
* Emmy Noether (b. 1882 Erlangen, d. 1935 Bryn Mayr), German mathematician. The theorem is only a
sideline in her career which she dedicated mostly to number theory. The theorem also applies to gauge
symmetries, where it states that to every gauge symmetry corresponds an identity of the equation of motion,
and vice versa.

Dvipsbugw
 motion and symmetry 199

leaves the action

S0 = ∫ T(v) − U(x) dt (75)

invariant, since S(b) = S 0 . This situation implies that

= p = const ;
∂T
(76)
∂v
Dvipsbugw
in short, symmetry under change of position implies conservation of momentum. The
converse is also true.
Challenge 376 ny In the case of symmetry under shift of observation instant, we find

T + U = const ; (77)

in other words, time translation invariance implies constant energy. Again, the converse

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is also correct. One also says that energy and momentum are the generators of time and
space translations.
The conserved quantity for a continuous symmetry is sometimes called the Noether
charge, because the term charge is used in theoretical physics to designate conserved ex-
tensive observables. So, energy and momentum are Noether charges. ‘Electric charge’,
‘gravitational charge’ (i.e. mass) and ‘topological charge’ are other common examples.
Challenge 377 n What is the conserved charge for rotation invariance?
We note that the expression ‘energy is conserved’ has several meanings. First of all, it
means that the energy of a single free particle is constant in time. Secondly, it means that
the total energy of any number of independent particles is constant. Finally, it means that
the energy of a system of particles, i.e. including their interactions, is constant in time.
Collisions are examples of the latter case. Noether’s theorem makes all of these points at
Challenge 378 e the same time, as you can verify using the corresponding Lagrangians.
But Noether’s theorem also makes, or rather repeats, an even stronger statement: if
energy were not conserved, time could not be defined. The whole description of nature
requires the existence of conserved quantities, as we noticed when we introduced the
Page 39 concepts of object, state and environment. For example, we defined objects as permanent
entities, that is, as entities characterized by conserved quantities. We also saw that the
Page 154 introduction of time is possible only because in nature there are ‘no surprises’. Noether’s
theorem describes exactly what such a ‘surprise’ would have to be: the non-conservation
Copyright © Christoph Schiller November 1997–May 2006

of energy. However, energy jumps have never been observed – not even at the quantum
level.
Since symmetries are so important for the description of nature, Table 25 gives an
overview of all the symmetries of nature we will encounter. Their main properties are also
listed. Except for those marked as ‘approximate’ or ‘speculative’, an experimental proof of
incorrectness of any of them would be a big surprise indeed.

Dvipsbugw
200 i galilean motion • 3. the global simplicity of complexity

TA B L E 25 The symmetries of relativity and quantum theory with their properties; also the complete list
of logical inductions used in the two fields
Symmetry Ty pe S pa c e G r o u p Pos- Con- Va - Main
[num- of ac-topo- si b l e se r v e d cuum/ effect
ber of tion logy rep- q ua n t - m at -
pa r a - r e s e nt- ity ter is
met- ations sy m -
e r s] metric
Dvipsbugw
Geometric or space-time, external, symmetries
Time and space R  R3 space, not scalars, momentum yes/yes allow
translation [4 par.] time compact vectors, and energy everyday
Rotation SO(3) space S2 tensors angular yes/yes communi-
[3 par.] momentum cation

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Galilei boost R3 [3 par.] space, not scalars, velocity of yes/for relativity
time compact vectors, centre of low of motion
tensors mass speeds
Lorentz homogen- space- not tensors, energy- yes/yes constant
eous Lie time compact spinors momentum light speed
SO(3,1) T µν
[6 par.]
Poincaré inhomo- space- not tensors, energy- yes/yes
ISL(2,C) geneous time compact spinors momentum
Lie T µν
[10 par.]
Dilation R+ [1 par.] space- ray n-dimen. none yes/no massless
invariance time continuum particles
Special R4 [4 par.] space- R4 n-dimen. none yes/no massless
conformal time continuum particles
invariance
Conformal [15 par.] space- involved massless none yes/no light cone
invariance time tensors, invariance
spinors
Copyright © Christoph Schiller November 1997–May 2006

Dynamic, interaction-dependent symmetries: gravity

1~r 2 gravity SO(4) config. as SO(4) vector pair perihelion yes/yes closed
[6 par.] space direction orbits
Diffeomorphism [ª par.] space- involved space- local yes/no perihelion
invariance time times energy– shift
momentum
Dynamic, classical and quantum-mechanical motion symmetries

Dvipsbugw
motion and symmetry 201

Symmetry Ty pe S pa c e G r o u p Pos- Con- Va - Main

[num- of ac-topo- si b l e se r v e d cuum/ effect
ber of tion logy rep- q ua n t - m at -
pa r a - r e s e nt- ity ter is
met- ations sy m -
e r s] metric

Motion(‘time’) discrete Hilbert discrete even, odd T-parity yes/no reversibil-

inversion T or phase ity Dvipsbugw
space
Parity(‘spatial’) discrete Hilbert discrete even, odd P-parity yes/no mirror
inversion P or phase world
space exists
Charge global, Hilbert discrete even, odd C-parity yes/no anti-
conjugation C antilinear, or phase particles

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

anti- space exist
Hermitean
CPT discrete Hilbert discrete even CPT-parity yes/yes makes field
or phase theory
space possible
Dynamic, interaction-dependent, gauge symmetries
Electromagnetic [ª par.] space of un- im- un- electric yes/yes massless
classical gauge fields portant important charge light
invariance
Electromagnetic abelian Lie Hilbert circle S1 fields electric yes/yes massless
q.m. gauge inv. U(1) space charge photon
[1 par.]
Electromagnetic abelian Lie space of circle S1 abstract abstract yes/no none
duality U(1) fields
[1 par.]
Weak gauge non- Hilbert as SU(3) particles weak no/
abelian Lie space charge approx.
SU(2)
Copyright © Christoph Schiller November 1997–May 2006

[3 par.]
Colour gauge non- Hilbert as SU(3) coloured colour yes/yes massless
abelian Lie space quarks gluons
SU(3)
[8 par.]
Chiral discrete fermions discrete left, right helicity approxi- ‘massless’
symmetry mately fermions a
Permutation symmetries

Dvipsbugw
202 i galilean motion • 3. the global simplicity of complexity

Symmetry Ty pe S pa c e G r o u p Pos- Con- Va - Main

[num- of ac-topo- si b l e se r v e d cuum/ effect
ber of tion logy rep- q ua n t - m at -
pa r a - r e s e nt- ity ter is
met- ations sy m -
e r s] metric
Particle discrete Fock discrete fermions none n.a./yes Gibbs’
exchange space and bosons paradox
Dvipsbugw
etc.
Selected speculative symmetries of nature
GUT E 8 , SO(10) Hilbert from Lie particles from Lie yes/no coupling
group group constant
conver-
gence

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

N-super- global Hilbert particles, Tmn and N no/no ‘massless’ a
symmetryb sparticles spinors c particles
Q imn
R-parity discrete Hilbert discrete +1, -1 R-parity yes/yes
sfermions,
gauginos
Braid symmetry discrete own discrete unclear unclear yes/maybe unclear
space
Space-time discrete all discrete vacuum unclear yes/maybe fixes
duality particle
masses
Event symmetry discrete space- discrete nature none yes/no unclear
time

For details about the connection between symmetry and induction, see page 673. The explanation
of the terms in the table will be completed in the rest of the walk. The real numbers are denoted
as R.
a. Only approximate; ‘massless’ means that m  m Pl , i.e. that m  22 µg.
b. N = 1 supersymmetry, but not N = 1 supergravity, is probably a good approximation for nature
Copyright © Christoph Schiller November 1997–May 2006

at everyday energies.
c. i = 1 .. N.

In summary, since we can talk about nature we can deduce several of its symmetries, in
particular its symmetry under time and space translations. From nature’s symmetries, us-
ing Noether’s theorem, we can deduce the conserved charges, such as energy or linear and
angular momentum. In other words, the definition of mass, space and time, together with
their symmetry properties, is equivalent to the conservation of energy and momentum.
Conservation and symmetry are two ways to express the same property of nature. To put it
simply, our ability to talk about nature means that energy and momentum are conserved.
In general, the most elegant way to uncover the ‘laws’ of nature is to search for nature’s

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 203

symmetries. In many historical cases, once this connection had been understood, phys-
ics made rapid progress. For example, Albert Einstein discovered the theory of relativity
in this way, and Paul Dirac started off quantum electrodynamics. We will use the same
method throughout our walk; in its third part we will uncover some symmetries which
are even more mind-boggling than those of relativity. Now, though, we will move on to
the next approach to a global description of motion.

Curiosities and fun challenges about motion symmetry

Dvipsbugw
As diet for your brain, a few questions to ponder:
**
What is the path followed by four turtles starting on the four angles of a square, if each
Challenge 379 ny of them continuously walks at the same speed towards the next one?
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 380 n What is the symmetry of a simple oscillation? And of a wave?
**
Challenge 381 n For what systems is motion reversal a symmetry transformation?
**
Challenge 382 ny What is the symmetry of a continuous rotation?

**
A sphere has a tensor for the moment of inertia that is diagonal with three equal numbers.
The same is true for a cube. Can you distinguish spheres and cubes by their rotation
Challenge 383 ny behaviour?
**
Challenge 384 ny Is there a motion in nature whose symmetry is perfect?

Simple motions of extended bodies – oscill ations and

waves
We defined action, and thus change, as the integral of the Lagrangian, and the Lagrangian
Copyright © Christoph Schiller November 1997–May 2006

as the difference between kinetic and potential energy. One of the simplest systems in
nature is a mass m attached to a spring. Its Lagrangian is given by

L = mv 2 − kx 2 ,
1
(78)
2
where k is a quantity characterizing the spring, the so-called spring constant. The Lag-
Challenge 385 e rangian is due to Robert Hooke, in the seventeenth century. Can you confirm it?
The motion that results from this Lagrangian is periodic, as shown in Figure 100. The
Lagrangian describes the oscillation of the spring length. The motion is exactly the same
as that of a long pendulum. It is called harmonic motion, because an object vibrating

Dvipsbugw
 204 i galilean motion • 3. the global simplicity of complexity

position
oscillation
amplitude

time

F I G U R E 100 The simplest oscillation Dvipsbugw

TA B L E 26 Some mechanical frequency values found in nature

O b s e r va t i o n Frequency

Sound frequencies in gas emitted by black holes c. 1 fHz

Precision in measured vibration frequencies of the Sun down to 2 nHz

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Vibration frequencies of the Sun down to c. 300 nHz
Vibration frequencies that disturb gravitational radiation down to 3 µHz
detection
Lowest vibration frequency of the Earth Ref. 157 309 µHz
Resonance frequency of stomach and internal organs (giv- 1 to 10 Hz
ing the ‘sound in the belly’ experience)
Wing beat of tiny fly c. 1000 Hz
Sound audible to young humans 20 Hz to 20 kHz
Sonar used by bats up to over 100 kHz
Sonar used by dolphins up to 150 kHz
Sound frequency used in ultrasound imaging up to 15 MHz
Phonon (sound) frequencies measured in single crystals up to 20 THz and more

rapidly in this way produces a completely pure – or harmonic – musical sound. (The
musical instrument producing the purest harmonic waves is the transverse flute. This
instrument thus gives the best idea of how harmonic motion ‘sounds’.) The graph of a
harmonic or linear oscillation, shown in Figure 100, is called a sine curve; it can be seen as
the basic building block of all oscillations. All other, non-harmonic oscillations in nature
Copyright © Christoph Schiller November 1997–May 2006

Page 205 can be composed from sine curves, as we shall see shortly.
Every oscillating motion continuously transforms kinetic energy into potential energy
and vice versa. This is the case for the tides, the pendulum, or any radio receiver. But many
oscillations also diminish in time: they are damped. Systems with large damping, such as
the shock absorbers in cars, are used to avoid oscillations. Systems with small damping
are useful for making precise and long-running clocks. The simplest measure of damping
is the number of oscillations a system takes to reduce its amplitude to 1~e  1~2.718 times
the original value. This characteristic number is the so-called Q-factor, named after the
abbreviation of ‘quality factor’. A poor Q-factor is 1 or less, an extremely good one is
100 000 or more. (Can you write down a simple Lagrangian for a damped oscillation with
Challenge 386 ny a given Q-factor?) In nature, damped oscillations do not usually keep constant frequency;

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 205

Figure yet to be included

Dvipsbugw
F I G U R E 101 Decomposing a general wave or signal into
harmonic waves

however, for the simple pendulum this remains the case to a high degree of accuracy.
The reason is that for a pendulum, the frequency does not depend significantly on the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

amplitude (as long as the amplitude is smaller than about 20°). This is one reason why
pendulums are used as oscillators in mechanical clocks.
Obviously, for a good clock, the driving oscillation must not only show small damp-
ing, but must also be independent of temperature and be insensitive to other external
influences. An important development of the twentieth century was the introduction of
quartz crystals as oscillators. Technical quartzes are crystals of the size of a few grains of
sand; they can be made to oscillate by applying an electric signal. They have little temper-
ature dependence and a large Q-factor, and therefore low energy consumption, so that
precise clocks can now run on small batteries.
All systems that oscillate also emit waves. In fact, oscillations only appear in exten-
ded systems, and oscillations are only the simplest of motions of extended systems. The
general motion of an extended system is the wave.

Waves and their motion

Waves are travelling imbalances, or, equivalently, travelling oscillations. Waves move,
even though the substrate does not move. Every wave can be seen as a superposition of
harmonic waves. Can you describe the difference in wave shape between a pure harmonic
Challenge 387 e tone, a musical sound, a noise and an explosion? Every sound effect can be thought of
as being composed of harmonic waves. Harmonic waves, also called sine waves or lin-
ear waves, are the building blocks of which all internal motions of an extended body are
Copyright © Christoph Schiller November 1997–May 2006

constructed.
Every harmonic wave is characterized by an oscillation frequency and a propagation
velocity. Low-amplitude water waves show this most clearly.
Waves appear inside all extended bodies, be they solids, liquids, gases or plasmas. In-
side fluid bodies, waves are longitudinal, meaning that the wave motion is in the same
direction as the wave oscillation. Sound in air is an example of a longitudinal wave. Inside
solid bodies, waves can also be transverse; in that case the wave oscillation is perpendicu-
lar to the travelling direction.
Waves appear also on interfaces between bodies: water–air interfaces are a well-known
case. Even a saltwater–freshwater interface, so-called dead water, shows waves: they
can appear even if the upper surface of the water is immobile. Any flight in an aero-

Dvipsbugw
206 i galilean motion • 3. the global simplicity of complexity

TA B L E 27 Some wave velocities

Wa v e Ve l o c i t y

Tsunami around 200 m~s

Sound in most gases 0.3 km~s
Sound in air at 273 K 331 m~s
Sound in air at 293 K 343 m~s
Sound in helium at 293 K 1.1 km~s
Sound in most liquids 1.1 km~s Dvipsbugw
Sound in water at 273 K 1.402 km~s
Sound in water at 293 K 1.482 km~s
Sound in gold 4.5 km~s
Sound in steel 5.790 km~s
Sound in granite 5.8 km~s
Sound in glass 5.9 km~s

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Sound in beryllium 12.8 km~s
Sound in boron up to 15 km~s
Sound in diamond up to 18 km~s
Sound in fullerene (C60 ) up to 26 km~s
Plasma wave velocity in InGaAs 600 km~s
Light in vacuum 2.998 ë 108 m~s

F I G U R E 102 The formation of gravity waves on water

plane provides an opportunity to study the regular cloud arrangements on the interface
between warm and cold air layers in the atmosphere. Seismic waves travelling along the
boundary between the sea floor and the sea water are also well-known. General surface
Copyright © Christoph Schiller November 1997–May 2006

waves are usually neither longitudinal nor transverse, but of a mixed type.
On water surfaces, one classifies waves according to the force that restores the plane
surface. The first type, surface tension waves, plays a role on scales up to a few centimetres.
At longer scales, gravity takes over as the main restoring force and one speaks of gravity
waves. This is the type we focus on here. Gravity waves in water, in contrast to surface
tension waves, are not sinusoidal. This is because of the special way the water moves in
such a wave. As shown in Figure 102, the surface water moves in circles; this leads to the
typical, asymmetrical wave shape with short sharp crests and long shallow troughs. (As
long as there is no wind and the floor below the water is horizontal, the waves are also
symmetric under front-to-back reflection.)
For water gravity waves, as for many other waves, the speed depends on the wavelength.

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 207

Indeed, the speed c of water waves depends on the wavelength λ and on the depth of the
water d in the following way:
¾
c=
дλ 2πd
tanh , (79)
2π λ

where д is the acceleration due to gravity (and an amplitude much smaller than the
wavelength is assumed). The formula shows two limiting regimes. First, short or deep Dvipsbugw
waves appear when the water » depth is larger than half the wavelength; for deep waves,
the phase velocity is c  дλ~2π , thus wavelength dependent, and the group velocity
» or
is about half the phase velocity. Shorter deep waves are thus slower. Secondly, shallow
long waves appear when the depth is less than 5% of the wavelength; in this case, c  дd ,
there is no dispersion, and the group velocity is about the same as the phase velocity.
The most impressive shallow waves are tsunamis, the large waves triggered by submarine
earthquakes. (The Japanese name is composed of tsu, meaning harbour, and nami, mean-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ing wave.) Since tsunamis are shallow waves, they show little dispersion and thus travel
over long distances; they can go round the Earth several times. Typical oscillation times
are between 6 and 60 minutes, giving wavelengths between 70 and 700 km and speeds in
Challenge 388 e the open sea of 200 to 250 m~s, similar to that of a jet plane. Their amplitude on the open
sea is often of the order of 10 cm; however, the amplitude scales with depth d as 1~d 4 and
heights up to 40 m have been measured at the shore. This was the order of magnitude of
the large and disastrous tsunami observed in the Indian Ocean on 26 December 2004.
Waves can also exist in empty space. Both light and gravity waves are examples. The
exploration of electromagnetism and relativity will tell us more about their properties.
Any study of motion must include the study of wave motion. We know from experi-
ence that waves can hit or even damage targets; thus every wave carries energy and mo-
mentum, even though (on average) no matter moves along the wave propagation direc-
tion. The energy E of a wave is the sum of its kinetic and potential energy. The kinetic
energy (density) depends on the temporal change of the displacement u at a given spot:
rapidly changing waves carry a larger kinetic energy. The potential energy (density) de-
pends on the gradient of the displacement, i.e. on its spatial change: steep waves carry
a larger potential energy than shallow ones. (Can you explain why the potential energy
Challenge 389 n does not depend on the displacement itself?) For harmonic waves propagating along the
direction z, each type of energy is proportional to the square of its respective displace-
ment change:
Copyright © Christoph Schiller November 1997–May 2006

Ref. 158

E  ( )2 + v 2 ( )2 .
∂u ∂u
(80)
∂t ∂z
Challenge 390 ny How is the energy density related to the frequency?
The momentum of a wave is directed along the direction of wave propagation. The
momentum value depends on both the temporal and the spatial change of displacement
u. For harmonic waves, the momentum (density) P is proportional to the product of these
two quantities:
Pz 
∂u ∂u
. (81)
∂t ∂z

Dvipsbugw
 208 i galilean motion • 3. the global simplicity of complexity

Interference

Polarisation

Dvipsbugw

Diffraction Refraction

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Damping Dispersion

F I G U R E 103 The six main properties of the motion of waves

When two linear wave trains collide or interfere, the total momentum is conserved
throughout the collision. An important consequence of momentum conservation is that
waves that are reflected by an obstacle do so with an outcoming angle equal to minus the
Challenge 391 n infalling angle. What happens to the phase?
Waves, like moving bodies, carry energy and momentum. In simple terms, if you shout
against a wall, the wall is hit. This hit, for example, can start avalanches on snowy moun-
tain slopes. In the same way, waves, like bodies, can carry also angular momentum. (What
Challenge 392 ny type of wave is necessary for this to be possible?) However, we can distinguish six main
properties that set the motion of waves apart from the motion of bodies.
Copyright © Christoph Schiller November 1997–May 2006

— Waves can add up or cancel each other out; thus they can interpenetrate each other.
These effects, called superposition and interference, are strongly tied to the linearity of
most waves.
— Transverse waves in three dimensions can oscillate in different directions: they show
polarization.
— Waves, such as sound, can go around corners. This is called diffraction.
— Waves change direction when they change medium. This is called refraction.
— Waves can have a frequency-dependent propagation speed. This is called dispersion.
— Often, the wave amplitude decreases over time: waves show damping.
Material bodies in everyday life do not behave in these ways when they move. These six

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 209

wave effects appear because wave motion is the motion of extended entities. The famous
debate whether electrons or light are waves or particles thus requires us to check whether
these effects specific to waves can be observed or not. This is one topic of quantum theory.
Before we study it, can you give an example of an observation that implies that a motion
Challenge 393 n surely cannot be a wave?
As a result of having a frequency f and a propagation velocity v, all sine waves are
characterized by the distance λ between two neighbouring wave crests: this distance is
called the wavelength. All waves obey the basic relation
Dvipsbugw
λf = v . (82)

In many cases the wave velocity v depends on the wavelength of the wave. For example,
this is the case for water waves. This change of speed with wavelength is called dispersion.
In contrast, the speed of sound in air does not depend on the wavelength (to a high degree
of accuracy). Sound in air shows almost no dispersion. Indeed, if there were dispersion

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

for sound, we could not understand each other’s speech at larger distances.
In everyday life we do not experience light as a wave, because the wavelength is only
around one two-thousandth of a millimetre. But light shows all six effects typical of wave
Page 558 motion. A rainbow, for example, can only be understood fully when the last five wave
effects are taken into account. Diffraction and interference can even be observed with
Challenge 394 n your fingers only. Can you tell how?
Like every anharmonic oscillation, every anharmonic wave can be decomposed into
sine waves. Figure 101 gives examples. If the various sine waves contained in a disturbance
propagate differently, the original wave will change in shape while it travels. That is the
reason why an echo does not sound exactly like the original sound; for the same reason,
a nearby thunder and a far-away one sound different.
All systems which oscillate also emit waves. Any radio or TV receiver contains oscillat-
ors. As a result, any such receiver is also a (weak) transmitter; indeed, in some countries
the authorities search for people who listen to radio without permission listening to the ra-
dio waves emitted by these devices. Also, inside the human ear, numerous tiny structures,
the hair cells, oscillate. As a result, the ear must also emit sound. This prediction, made
in 1948 by Tommy Gold, was confirmed only in 1979 by David Kemp. These so-called
otoacoustic emissions can be detected with sensitive microphones; they are presently be-
ing studied in order to unravel the still unknown workings of the ear and in order to
Ref. 159 diagnose various ear illnesses without the need for surgery.
Copyright © Christoph Schiller November 1997–May 2006

Since any travelling disturbance can be decomposed into sine waves, the term ‘wave’ is
used by physicists for all travelling disturbances, whether they look like sine waves or not.
In fact, the disturbances do not even have to be travelling. Take a standing wave: is it a
wave or an oscillation? Standing waves do not travel; they are oscillations. But a standing
wave can be seen as the superposition of two waves travelling in opposite directions. Since
Challenge 395 ny all oscillations are standing waves (can you confirm this?), we can say that all oscillations
are special forms of waves.
The most important travelling disturbances are those that are localized. Figure 101
shows an example of a localized wave group or pulse, together with its decomposition
into harmonic waves. Wave groups are extensively used to talk and as signals for commu-
nication.

Dvipsbugw
 210 i galilean motion • 3. the global simplicity of complexity

Why can we talk to each other? – Huygens’ principle

The properties of our environment often disclose their full importance only when we ask
simple questions. Why can we use the radio? Why can we talk on mobile phones? Why
can we listen to each other? It turns out that a central part of the answer to these questions
is that the space we live has an odd numbers of dimensions.
In spaces of even dimension, it is impossible to talk, because messages do not stop.
This is an important result which is easily checked by throwing a stone into a lake: even
after the stone has disappeared, waves are still emitted from the point at which it entered Dvipsbugw
the water. Yet, when we stop talking, no waves are emitted any more.

– CS – text to be added – CS –

We can also say that Huygens’ principle holds if the wave equation is solved by a circu-
lar wave leaving no amplitude behind it. Mathematicians translate this by requiring that
the evolving delta function δ(c 2 t 2 − r 2 ) satisfies the wave equation, i.e. that ∂ 2t δ = c 2 ∆δ.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The delta function is that strange ‘function’ which is zero everywhere except at the origin,
where it is infinite. A few more properties describe the precise way in which this hap-
pens.* It turns out that the delta function is a solution of the wave equation only if the
space dimension is odd at least three.
In summary, the reason a room gets dark when we switch off the light, is that we live
in a space with a number of dimensions which is odd and larger than one.

Signals
A signal is the transport of information. Every signal is motion of energy. Signals can
be either objects or waves. A thrown stone can be a signal, as can a whistle. Waves are a
more practical form of communication because they do not require transport of matter:
it is easier to use electricity in a telephone wire to transport a statement than to send a
messenger. Indeed, most modern technological advances can be traced to the separation
between signal and matter transport. Instead of transporting an orchestra to transmit
music, we can send radio signals. Instead of sending paper letters we write email messages.
Instead of going to the library we browse the internet.
The greatest advances in communication have resulted from the use of signals to
transport large amounts of energy. That is what electric cables do: they transport energy
without transporting any (noticeable) matter. We do not need to attach our kitchen ma-
Copyright © Christoph Schiller November 1997–May 2006

chines to the power station: we can get the energy via a copper wire.
For all these reasons, the term ‘signal’ is often meant to imply waves only. Voice, sound,
electric signals, radio and light signals are the most common examples of wave signals.
Signals are characterized by their speed and their information content. Both quantities
turn out to be limited. The limit on speed is the central topic of the theory of special
Page 275 relativity.
A simple limit on information content can be expressed when noting that the inform-
ation flow is given by the detailed shape of the signal. The shape is characterized by a
frequency (or wavelength) and a position in time (or space). For every signal – and every

* The main property is ∫ δxdx = 1. In mathematically precise terms, the delta ‘function’ is a distribution.

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 211

Figure to be included

F I G U R E 104 The electrical signals measured in a nerve Dvipsbugw

wave – there is a relation between the time-of-arrival error ∆t and the angular frequency
error ∆ω:
∆t ∆ω E .
1
(83)
2

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

This time–frequency indeterminacy relation expresses that, in a signal, it is impossible
to specify both the time of arrival and the frequency with full precision. The two errors
are (within a numerical factor) the inverse of each other. (One also says that the time-
bandwidth product is always larger than 1~4π.) The limitation appears because on one
hand one needs a wave as similar as possible to a sine wave in order to precisely determine
the frequency, but on the other hand one needs a signal as narrow as possible to precisely
determine its time of arrival. The contrast in the two requirements leads to the limit. The
indeterminacy relation is thus a feature of every wave phenomenon. You might want to
Challenge 396 e test this relation with any wave in your environment.
Similarly, there is a relation between the position error ∆x and the wave vector error
∆k = 2π~∆λ of a signal:
∆x ∆k E .
1
(84)
2
Like the previous case, also this indeterminacy relation expresses that it is impossible
to specify both the position of a signal and its wavelength with full precision. Also this
position–wave-vector indeterminacy relation is a feature of any wave phenomenon.
Every indeterminacy relation is the consequence of a smallest entity. In the case of
waves, the smallest entity of the phenomenon is the period (or cycle, as it used to be
called). Whenever there is a smallest unit in a natural phenomenon, an indeterminacy
Copyright © Christoph Schiller November 1997–May 2006

relation results. We will encounter other indeterminacy relations both in relativity and in
quantum theory. As we will find out, they are due to smallest entities as well.
Whenever signals are sent, their content can be lost. Each of the six characteristics of
waves listed on page 208 can lead to content degradation. Can you provide an example
Challenge 397 ny for each case? The energy, the momentum and all other conserved properties of signals
are never lost, of course. The disappearance of signals is akin to the disappearance of
motion. When motion disappears by friction, it only seems to disappear, and is in fact
transformed into heat. Similarly, when a signal disappears, it only seems to disappear, and
is in fact transformed into noise. (Physical) noise is a collection of numerous disordered
signals, in the same way that heat is a collection of numerous disordered movements.
All signal propagation is described by a wave equation. A famous example is the equa-

Dvipsbugw
 212 i galilean motion • 3. the global simplicity of complexity

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 105 A solitary water wave followed by a motor boat, reconstructing
the discovery by Scott Russel (© Dugald Duncan)

Ref. 160 tion found by Hodgkin and Huxley. It is a realistic approximation for the behaviour of
electrical potential in nerves. Using facts about the behaviour of potassium and sodium
ions, they found an elaborate equation that describes the voltage V in nerves, and thus
the way the signals are propagated. The equation accurately describes the characteristic
voltage spikes measured in nerves, shown in Figure 104. The figure clearly shows that
these waves differ from sine waves: they are not harmonic. Anharmonicity is one result
of nonlinearity. But nonlinearity can lead to even stronger effects.

Solitary waves and solitons

In August 1834, the Scottish engineer John Scott Russell (1808–1882) recorded a strange
observation in a water canal in the countryside near Edinburgh. When a boat pulled
through the channel was suddenly stopped, a strange water wave departed from it. It
consisted of a single crest, about 10 m long and 0.5 m high, moving at about 4 m~s. He fol-
lowed that crest with his horse for several kilometres: the wave died out only very slowly.
He did not observe any dispersion, as is usual in water waves: the width of the crest re-
Copyright © Christoph Schiller November 1997–May 2006

mained constant. Russell then started producing such waves in his laboratory, and extens-
Ref. 161 ively studied their properties. He showed that the speed depended on the amplitude, in
contrast to linear waves. He also found that the depth d of the water canal was an import-
ant parameter. In fact, the speed v, the amplitude A and the width L of these single-crested
waves are related by
¾
¼ 4d 3
v= дd ‹1 +  L=
A
and . (85)
2d 3A

As shown by these expressions, and noted by Russell, high waves are narrow and fast,
whereas shallow waves are slow and wide. The shape of the waves is fixed during their

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 213

Figure to be inserted
Dvipsbugw

F I G U R E 106 Solitons are stable against encounters

motion. Today, these and all other stable waves with a single crest are called solitary waves.
They appear only where the dispersion and the nonlinearity of the system exactly com-
pensate for each other. Russell also noted that the solitary waves in water channels can

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

cross each other unchanged, even when travelling in opposite directions; solitary waves
with this property are called solitons. Solitons are stable against encounters, whereas sol-
itary waves in general are not.
Only sixty years later, in 1895, Korteweg and de Vries found out that solitary waves in
water channels have a shape described by

x − vt
u(x, t) = A sech2 where sech x =
2
, (86)
L ex + e−x
and that the relation found by Russell was due to the wave equation

∂u d 2 ∂ 3 u
» + (1 + u) =0.
1 ∂u 3
+ (87)
дd ∂t 2d ∂x 6 ∂x 3

This equation for the elongation u is called the Korteweg–de Vries equation in their hon-
our.* The surprising stability of the solitary solutions is due to the opposite effect of the
two terms that distinguish the equation from linear wave equations: for the solitary solu-
tions, the nonlinear term precisely compensates for the dispersion induced by the third-
derivative term.
For many decades such solitary waves were seen as mathematical and physical curiosit-
Copyright © Christoph Schiller November 1997–May 2006

ies. But almost a hundred years later it became clear that the Korteweg–de Vries equation
is a universal model for weakly nonlinear waves in the weak dispersion regime, and thus
of basic importance. This conclusion was triggered by Kruskal and Zabusky, who in 1965
proved mathematically that the solutions (86) are unchanged in collisions. This discovery
Ref. 162 prompted them to introduce the term soliton. These solutions do indeed interpenetrate
one another without changing velocity or shape: a collision only produces a small pos-
itional shift for each pulse.

* The equation can be simplified by transforming the variable u; most concisely, it can be rewritten as u t +
u x x x = 6uu x . As long as the solutions are sech functions, this and other transformed versions of the equation
are known by the same name.

Dvipsbugw
 214 i galilean motion • 3. the global simplicity of complexity

Solitary waves play a role in many examples of fluid flows. They are found in ocean cur-
rents; and even the red spot on Jupiter, which was a steady feature of Jupiter photographs
for many centuries, is an example.
Solitary waves also appear when extremely high-intensity sound is generated in solids.
Ref. 163 In these cases, they can lead to sound pulses of only a few nanometres in length. Solitary
light pulses are also used inside certain optical communication fibres, where they provide
Ref. 161 (almost) lossless signal transmission.
Towards the end of the twentieth century a second wave of interest in the mathemat-
ics of solitons arose, when quantum theorists became interested in them. The reason is Dvipsbugw
simple but deep: a soliton is a ‘middle thing’ between a particle and a wave; it has features
of both concepts. For this reason, solitons are now an essential part of any description of
elementary particles, as we will find out later on.

Curiosities and fun challenges about waves and extended bodies

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Society is a wave. The wave moves onward, but

“ the water of which it is composed does not.

Ralph Waldo Emerson, Self-Reliance.

Oscillations, waves and signals are a limitless source of fascination. ”

**
When the frequency of a tone is doubled, one says that the tone is higher by an octave.
Two tones that differ by an octave, when played together, sound pleasant to the ear. Two
other agreeable frequency ratios – or ‘intervals’, as musicians say – are quarts and quints.
Challenge 398 e What are the corresponding frequency ratios? (Note: the answer was one of the oldest
discoveries in physics; it is attributed to Pythagoras, around 500 bce.)
**
An orchestra is playing music in a large hall. At a distance of 30 m, somebody is listening
to the music. At a distance of 3000 km, another person is listening to the music via the
Challenge 399 n radio. Who hears the music first?

**
What is the period of a simple pendulum, i.e. a mass m attached to a massless string of
Challenge 400 ny length l? What is the period if the string is much longer than the radius of the Earth?
Copyright © Christoph Schiller November 1997–May 2006

**
What path is followed by a body moving in a plane, but attached by a spring to a fixed
Challenge 401 n point on the plane?

**
Light is a wave, as we will discover later on. As a result, light reaching the Earth from
space is refracted when it enters the atmosphere. Can you confirm that as a result, stars
Challenge 402 e appear somewhat higher in the night sky than they really are?
**

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 215

air

air water

coin

F I G U R E 107 Shadows and refraction

Dvipsbugw
What are the highest sea waves? This question has been researched systematically only
Ref. 164 recently, using satellites. The surprising result is that sea waves with a height of 25 m and
more are common: there are a few such waves on the oceans at any given time. This result
confirms the rare stories of experienced ship captains and explains many otherwise ship
sinkings.
Surfers may thus get many chances to ride 30 m waves. (The record is just below this

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

size.) But maybe the most impressive waves to surf are those of the Pororoca, a series of
4 m waves that move from the sea into the Amazonas River every spring, against the flow
of the river. These waves can be surfed for tens of kilometres.
**
All waves are damped, eventually. This effect is often frequency-dependent. Can you
Challenge 403 n provide a confirmation of this dependence in the case of sound in air?

**
When you make a hole with a needle in black paper, the hole can be used as a magnifying
Challenge 404 e lens. (Try it.) Diffraction is responsible for the lens effect. By the way, the diffraction of
light by holes was noted by Francesco Grimaldi in the seventeenth century; he deduced
that light is a wave. His observations were later discussed by Newton, who wrongly dis-
missed them.

**
Put a empty cup near a lamp, in such a way that the bottom of the cup remains in the
shadow. When you fill the cup with water, some of the bottom will be lit, because of the
refraction of the light from the lamp. The same effect allows us to build lenses. The same
Page 568 effect is at the basis of instruments such as the telescope.
Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 405 n Are water waves transverse or longitudinal?
**
¼limits the speeds of ships. A surface ship cannot travel (much)
The speed of water waves
faster than about v crit = 0.16дl , where д = 9.8 m~s2 , l is its length, and 0.16 is a number
determined experimentally, called the critical Froude number. This relation is valid for
all vessels, from large tankers (l = 100 m gives v crit = 13 m~s) down to ducks (l = 0.3 m
gives v crit = 0.7 m~s). The critical speed is that of a wave with the same wavelength as
the ship. In fact, moving at higher speeds than the critical value is possible, but requires

Dvipsbugw
 216 i galilean motion • 3. the global simplicity of complexity

much more energy. (A higher speed is also possible if the ship surfs on a wave.) Therefore
all water animals and ships are faster when they swim below the surface – where the limit
due to surface waves does not exist – than when they swim on the surface. For example,
ducks can swim three times as fast under water than on the surface.
Challenge 406 n How far away is the olympic swimming record from the critical value?
**
The group velocity of water waves (in deep water) is less than the velocity of the individual
waves. As a result, when a group of wave crests travels, within the group the crests move Dvipsbugw
from the back to the front, appearing at the back, travelling forward and then dying out
at the front.
**
One can hear the distant sea or a distant highway more clearly in the evening than in
the morning. This is an effect of refraction. Sound speed decreases with temperature. In

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the evening, the ground cools more quickly than the air above. As a result, sound leaving
the ground and travelling upwards is refracted downwards, leading to the long hearing
distance. In the morning, usually the air is cold above and warm below. Sound is refrac-
ted upwards, and distant sound does not reach a listener on the ground. Refraction thus
implies that mornings are quiet, and that one can hear more distant sounds in the even-
ings. Elephants use the sound situation during evenings to communicate over distances
of more than 10 km. (They also use sound waves in the ground to communicate, but that
is another story.)

**
Refraction also implies that there is a sound channel in the ocean, and in the atmosphere.
Sound speed decreases with temperature, and increases with pressure. At an ocean depth
of 1 km, or at an atmospheric height of 13 to 17 km (that is at the top of the tallest cu-
mulonimbus clouds or equivalently, at the middle of the ozone layer) sound has minimal
speed. As a result, sound that starts from that level and tries to leave is channelled back to
it. Whales use the sound channel to communicate with each other with beautiful songs;
Challenge 407 e one can find recordings of these songs on the internet. The military successfully uses
microphones placed at the sound channel in the ocean to locate submarines, and mi-
Ref. 165 crophones on balloons in the atmospheric channel to listen for nuclear explosions. (In
fact, sound experiments conducted by the military are the main reason why whales are
Copyright © Christoph Schiller November 1997–May 2006

deafened and lose their orientation, stranding on the shores. Similar experiments in the
air with high-altitude balloons are often mistaken for flying saucers, as in the famous
Roswell incident.)
**
Ref. 166 Much smaller also animals communicate by sound waves. In 2003, it was found that her-
ring communicate using noises they produce when farting. When they pass wind, the gas
creates a ticking sound whose frequency spectrum reaches up to 20 kHz. One can even
listen to recordings of this sound on the internet. The details of the communication, such
as the differences between males and females, are still being investigated. It is possible
that the sounds may also be used by predators to detect herring, and they might even by

Dvipsbugw
 simple motions of extended bodies – oscill ations and waves 217

used by future fishing vessels.

**
On windy seas, the white wave crests have several important effects. The noise stems
from tiny exploding and imploding water bubbles. The noise of waves on the open sea
is thus the superposition of many small explosions. At the same time, white crests are
the events where the seas absorb carbon dioxide from the atmosphere, and thus reduce
global warming.
Dvipsbugw
**
Challenge 408 n Why are there many small holes in the ceilings of many office buildings?
**
Which quantity determines the wavelength of water waves emitted when a stone is
Challenge 409 ny thrown into a pond?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Ref. 3 Yakov Perelman lists the following four problems in his delightful physics problem book.
(1) A stone falling into a lake produces circular waves. What is the shape of waves
Challenge 410 n produced by a stone falling into a river, where the water flows in one direction?
(2) It is possible to build a lens for sound, in the same way as it is possible to build
Challenge 411 n lenses for light. What would such a lens look like?
Challenge 412 ny (3) What is the sound heard inside a shell?
(4) Light takes about eight minutes to travel from the Sun to the Earth. What con-
Challenge 413 n sequence does this have for a sunrise?
**
Can you describe how a Rubik’s Cube is built? And its generalizations to higher numbers
Challenge 414 n of segments? Is there a limit to the number of segments? These puzzles are even tougher
than the search for a rearrangement of the cube. Similar puzzles can be found in the study
of many mechanisms, from robots to textile machines.
**
Typically, sound produces a pressure variation of 10−8 bar on the ear. How is this determ-
Challenge 415 ny ined?
Copyright © Christoph Schiller November 1997–May 2006

The ear is indeed a sensitive device. It is now known that most cases of sea mammals,
like whales, swimming onto the shore are due to ear problems: usually some military
device (either sonar signals or explosions) has destroyed their ear so that they became
deaf and lose orientation.
**
Infrasound, inaudible sound below 20 Hz, is a modern topic of research. In nature, infra-
sound is emitted by earthquakes, volcanic eruptions, wind, thunder, waterfalls, falling
meteorites and the surf. Glacier motion, seaquakes, avalanches and geomagnetic storms
Ref. 167 also emit infrasound. Human sources include missile launches, traffic, fuel engines and
air compressors.

Dvipsbugw
 218 i galilean motion • 3. the global simplicity of complexity

It is known that high intensities of infrasound lead to vomiting or disturbances of the

sense of equilibrium (140 dB or more for 2 minutes), and even to death (170 dB for 10
minutes). The effects of lower intensities on human health are not yet known.
Infrasound can travel several times around the world before dying down, as the ex-
plosion of the Krakatoa volcano showed in 1883. With modern infrasound detectors, sea
surf can be detected hundreds of kilometres away. Infrasound detectors are even used to
count meteorites at night. Very rarely, meteorites can be heard with the human ear.
** Dvipsbugw
The method used to deduce the sine waves contained in a signal, as shown in Figure 101, is
called the Fourier transformation. It is of importance throughout science and technology.
In the 1980s, an interesting generalization became popular, called the wavelet transforma-
tion. In contrast to Fourier transformations, wavelet transformations allow us to localize
signals in time. Wavelet transformations are used to compress digitally stored images in
Ref. 168 an efficient way, to diagnose aeroplane turbine problems, and in many other applications.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
If you like engineering challenges, here is one that is still open. How can one make a
Challenge 416 r robust and efficient system that transforms the energy of sea waves into electricity?
**
In our description of extended bodies, we assumed that each spot of a body can be fol-
lowed separately throughout its motion. Is this assumption justified? What would happen
Challenge 417 r if it were not?
**
A special type of waves appears in explosions and supersonic flight: shock waves. In a
shock wave, the density or pressure of a gas changes abruptly, on distances of a few mi-
crometers. Studying shock waves is a research field in itself; shock waves determine the
flight of bullets, the snapping of whips and the effects of detonations.

**
Bats fly at night using echolocation. Dolphins also use it. Sonar, used by fishing vessels
to look for fish, copies the system of dolphins. Less well known is that humans have the
Ref. 169 same ability. Have you ever tried to echolocate a wall in a completely dark room? You will
Copyright © Christoph Schiller November 1997–May 2006

be surprised at how easily this is possible. Just make a loud hissing or whistling noise that
Challenge 418 e stops abruptly, and listen to the echo. You will be able to locate walls reliably.

D o extended bodies exist?

We have just discussed the motion of extended bodies in some detail. We have seen that
extended bodies show wave motion. But are extended bodies found in nature? Strangely
enough, this question has been one of the most intensely discussed questions in physics.
Over the centuries, it has reappeared again and again, at each improvement of the descrip-
tion of motion; the answer has alternated between the affirmative and the negative. Many
thinkers have been imprisoned, and many still are being persecuted, for giving answers

Dvipsbugw
 d o extended bodies exist? 219

n=1

n=2

n=5
Dvipsbugw

n=∞

F I G U R E 108 Floors and mountains as fractals (© Paul Martz)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that are not politically correct! In fact, the issue already arises in everyday life.

Mountains and fractals

Whenever we climb a mountain, we follow the outline of its shape. We usually describe
this outline as a curved two-dimensional surface. In everyday life we find that this is
a good approximation. But there are alternative possibilities. The most popular is the
idea that mountains are fractal surfaces. A fractal was defined by Benoit Mandelbrot as
a set that is self-similar under a countable but infinite number of magnification values.*
Page 53 We have already encountered fractal lines. An example of an algorithm for building a
(random) fractal surface is shown on the right side of Figure 108. It produces shapes which
Ref. 170 look remarkably similar to real mountains. The results are so realistic that they are used
in Hollywood movies. If this description were correct, mountains would be extended, but
not continuous.
But mountains could also be fractals of a different sort, as shown in the left side of
Figure 108. Mountain surfaces could have an infinity of small and smaller holes. In fact,
one could also imagine that mountains are described as three-dimensional versions of the
left side of the figure. Mountains would then be some sort of mathematical Swiss cheese.
Can you devise an experiment to decide whether fractals provide the correct description
Copyright © Christoph Schiller November 1997–May 2006

Challenge 419 n for mountains? To settle the issue, a chocolate bar can help.

Can a chocolate bar last forever?

From a drop of water a logician could predict an

“ Atlantic or a Niagara.
Arthur Conan Doyle, A Study in Scarlet

Any child knows how to make a chocolate bar last forever: eat half the remainder every ”
day. However, this method only works if matter is scale-invariant. In other words, the

* For a definition of uncountability, see page 649.

Dvipsbugw
 220 i galilean motion • 3. the global simplicity of complexity

method only works if matter is either fractal, as it then would be scale-invariant for a
discrete set of zoom factors, or continuous, in which case it would be scale-invariant for
any zoom factor. Which case, if either, applies to nature?
Page 53 We have already encountered a fact making continuity a questionable assumption: con-
tinuity would allow us, as Banach and Tarski showed, to multiply food and any other
matter by clever cutting and reassembling. Continuity would allow children to eat the
same amount of chocolate every day, without ever buying a new bar. Matter is thus not
continuous. Now, fractal chocolate is not ruled out in this way; but other experiments
settle the question. Indeed, we note that melted materials do not take up much smaller Dvipsbugw
volumes than solid ones. We also find that even under the highest pressures, materials do
not shrink. Thus matter is not a fractal. What then is its structure?
To get an idea of the structure of matter we can take fluid chocolate, or even just some
oil – which is the main ingredient of chocolate anyway – and spread it out over a large
surface. For example, we can spread a drop of oil onto a pond on a day without rain or
wind; it is not difficult to observe which parts of the water are covered by the oil and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

which are not. A small droplet of oil cannot cover a surface larger than – can you guess
Challenge 420 n the value? Trying to spread the film further inevitably rips it apart. The child’s method
of prolonging chocolate thus does not work for ever: it comes to a sudden end. The oil
experiment shows that there is a minimum thickness of oil films, with a value of about
2 nm. This simple experiment can even be conducted at home; it shows that there is a
smallest size in matter. Matter is made of tiny components. This confirms the observa-
tions made by Joseph Loschmidt* in 1865, who was the first person to measure the size of
the components of matter.** In 1865, it was not a surprise that matter was made of small
components, as the existence of a smallest size – but not its value – had already been
deduced by Galileo, when studying some other simple questions.***

* Joseph Loschmidt (b. 1821 Putschirn, d. 1895 Vienna) Austrian chemist and physicist. The oil experiment
was popularized a few decades later, by Kelvin. It is often claimed that Benjamin Franklin was the first to
conduct the oil experiment; that is wrong. Franklin did not measure the thickness, and did not even consider
the question of the thickness. He did pour oil on water, but missed the most important conclusion that could
be drawn from it. Even geniuses do not discover everything.
** Loschmidt knew that the (dynamic) viscosity of a gas was given by η = ρlv~3, where ρ is the density
of the gas, v the average speed of the components and l their mean free path. With Avogadro’s prediction
(made in 1811 without specifying anyº value) that a volume V of any gas always contains the same number
N of components, one also has l = V ~ 2πNσ 2 , where σ is the cross-section of the components. (The cross-
section is the area of the shadow of an object.) Loschmidt then assumed that when the gas is liquefied, the
Copyright © Christoph Schiller November 1997–May 2006

volume of the liquid is the sum of the volumes of the particles. He then measured all the involved quantities
and determined N. The modern value of N, called Avogadro’s number or Loschmidt’s number, is 6.02 ë 1023
particles in 22.4 l of any gas at standard conditions (today called 1 mol).
*** Galileo was brought to trial because of his ideas about atoms, not about the motion of the Earth, as
is often claimed. To get a clear view of the matters of dispute in the case of Galileo, especially those of
interest to physicists, the best text is the excellent book by Pietro Redondi, Galileo eretico, Einaudi,
1983, translated into English as Galileo Heretic, Princeton University Press, 1987. It is also available in many
other languages. Redondi, a renowned historical scholar and colleague of Pierre Costabel, tells the story of
the dispute between Galileo and the reactionary parts of the Catholic Church. He discovered a document
of that time – the anonymous denunciation which started the trial – that allowed him to show that the
condemnation of Galileo to life imprisonment for his views on the Earth’s motion was organized by his
friend the Pope to protect him from a sure condemnation to death over a different issue.
The reasons for his arrest, as shown by the denunciation, were not his ideas on astronomy and on the

Dvipsbugw
 d o extended bodies exist? 221

How high can animals jump?

Fleas can jump to heights a hundred times their size, humans only to heights about their
Ref. 171 own size. In fact, biological studies yield a simple observation: most animals, regardless
of their size, achieve about the same jumping height of between 0.8 and 2.2 m, whether
they are humans, cats, grasshoppers, apes, horses or leopards. The explanation of this fact
Challenge 421 n takes only two lines. Can you find it?
The above observation seems to be an example of scale invariance. But there are some
interesting exceptions at both ends of the mass range. At the small end, mites and other Dvipsbugw
small insects do not achieve such heights because, like all small objects, they encounter
the problem of air resistance. At the large end, elephants do not jump that high, because
doing so would break their bones. But why do bones break at all?
Why are all humans of about the same size? Why are there no giant adults with a height
of ten metres? Why aren’t there any land animals larger than elephants? The answer yields
the key to understanding the structure of matter. In fact, the materials of which we are

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

made would not allow such changes of scale, as the bones of giants would collapse under
the weight they have to sustain. Bones have a finite strength because their constituents
stick to each other with a finite attraction. Continuous matter – which exists only in car-
toons – could not break at all, and fractal matter would be infinitely fragile. Matter breaks
under finite loads because it is composed of small basic constituents.

Felling trees
The gentle lower slopes of Motion Mountain are covered by trees. Trees are fascinating
structures. Take their size. Why do trees have limited size? Already in the sixteenth cen-
tury, Galileo knew that it is not possible to increase tree height without limits: at some
point a tree would not have the strength to support its own weight. He estimated the max-
imum height to be around 90 m; the actual record, unknown to him at the time, seems to
be 150 m, for the Australian tree Eucalyptus regnans. But why does a limit exist at all? The
answer is the same as for bones: wood has a finite strength because it is not scale invariant;
and it is not scale invariant because it is made of small constituents, namely atoms.*
In fact, the derivation of the precise value of the height limit is more involved. Trees
must not break under strong winds. Wind resistance limits the height-to-thickness ratio
Challenge 422 ny h~d to about 50 for normal-sized trees (for 0.2 m < d < 2 m). Can you say why? Thinner
Copyright © Christoph Schiller November 1997–May 2006

motion of the Earth, but his statements on matter. Galileo defended the view that since matter is not scale
invariant, it must be made of ‘atoms’ or, as he called them, piccolissimi quanti – smallest quanta. This was
and still is a heresy. A true Catholic is still not allowed to believe in atoms. Indeed, the theory of atoms is
not compatible with the change of bread and wine into human flesh and blood, called transsubstantiation,
which is a central tenet of the Catholic faith. In Galileo’s days, church tribunals punished heresy, i.e. deviating
personal opinions, by the death sentence. Despite being condemned to prison in his trial, Galileo published
his last book, written as an old man under house arrest, on the scaling issue. Today, the Catholic Church still
refuses to publish the proceedings and other documents of the trial. Its officials carefully avoid the subject
of atoms, as any statement on this subject would make the Catholic Church into a laughing stock. In fact,
quantum theory, named after the term used by Galileo, has become the most precise description of nature
yet.
Ref. 172 * There is another important limiting factor: the water columns inside trees must not break. Both factors
seem to yield similar limiting heights.

Dvipsbugw
 222 i galilean motion • 3. the global simplicity of complexity

three
mono-
atomic
steps

lamp eye

photograph to be included
Dvipsbugw
F I G U R E 109 Atoms exist: rotating an F I G U R E 110 Atomic steps in broken gallium
aluminium rod leads to brightness arsenide crystals can be seen under a light
oscillations microscope

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

trees are limited in height to less than 10 m by the requirement that they return to the
Ref. 173 vertical after being bent by the wind.
Such studies of natural constraints also answer the question of why trees are made
from wood and not, for example, from steel. You could check for yourself that the max-
imum height of a column of a given mass is determined by the ratio E~ρ 2 between the
Challenge 423 n elastic module and the square of the mass density. Wood is actually the material for which
Ref. 174 this ratio is highest. Only recently have material scientists managed to engineer slightly
better ratios with fibre composites.
Why do materials break at all? All observations yield the same answer and confirm
Galileo’s reasoning: because there is a smallest size in materials. For example, bodies un-
der stress are torn apart at the position at which their strength is minimal. If a body were
completely homogeneous, it could not be torn apart; a crack could not start anywhere. If
a body had a fractal Swiss-cheese structure, cracks would have places to start, but they
would need only an infinitesimal shock to do so.
A simple experiment that shows that solids have a smallest size is shown in Figure 109.
A cylindrical rod of pure, single crystal aluminium shows a surprising behaviour when
it is illuminated from the side: its brightness depends on how the rod is oriented, even
though it is completely round. This angular dependence is due to the atomic arrangement
of the aluminium atoms in the rod.
It is not difficult to confirm experimentally the existence of smallest size in solids. It is
Copyright © Christoph Schiller November 1997–May 2006

sufficient to break a single crystal, such as a gallium arsenide wafer, in two. The breaking
surface is either completely flat or shows extremely small steps, as shown in Figure 110.
Challenge 424 ny These steps are visible under a normal light microscope. (Why?) It turns out that all the
step heights are multiples of a smallest height: its value is about 0.2 nm. The existence of
a smallest height, corresponding to the height of an atom, contradicts all possibilities of
scale invariance in matter.

The sound of silence

Climbing the slopes of Motion Mountain, we arrive in a region of the forest covered with
deep snow. We stop for a minute and look around. It is dark; all the animals are asleep;
there is no wind and there are no sources of sound. We stand still, without breathing, and

Dvipsbugw
 d o extended bodies exist? 223

listen to the silence. (You can have this experience also in a sound studio such as those
used for musical recordings, or in a quiet bedroom at night.) In situations of complete
silence, the ear automatically becomes more sensitive*; we then have a strange experience.
We hear two noises, a lower- and a higher-pitched one, which are obviously generated
inside the ear. Experiments show that the higher note is due to the activity of the nerve
cells in the inner ear. The lower note is due to pulsating blood streaming through the
head. But why do we hear a noise at all?
Many similar experiments confirm that whatever we do, we can never eliminate noise
from measurements. This unavoidable type of noise is called shot noise in physics. The Dvipsbugw
statistical properties of this type of noise actually correspond precisely to what would be
expected if flows, instead of being motions of continuous matter, were transportation of
a large number of equal, small and discrete entities. Thus, simply listening to noise proves
that electric current is made of electrons, that air and liquids are made of molecules, and
that light is made of photons. In a sense, the sound of silence is the sound of atoms. Shot
noise would not exist in continuous systems.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Little hard balls
I prefer knowing the cause of a single thing to

“ being king of Persia.

Precise observations show that matter is neither continuous nor a fractal: matter is made
Democritus

”
of smallest basic particles. Galileo, who deduced their existence by thinking about giants
and trees, called them ‘smallest quanta.’ Today they are called ‘atoms’, in honour of a fam-
ous argument of the ancient Greeks. Indeed, 2500 years ago, the Greeks asked the follow-
ing question. If motion and matter are conserved, how can change and transformation
exist? The philosophical school of Leucippus and Democritus of Abdera** studied two
particular observations in special detail. They noted that salt dissolves in water. They also
noted that fish can swim in water. In the first case, the volume of water does not increase
when the salt is dissolved. In the second case, when fish advance, they must push water
aside. They deduced that there is only one possible explanation that satisfies observations
and also reconciles conservation and transformation: nature is made of void and of small,
hard, indivisible and conserved particles.*** In this way any example of motion, change
or transformation is due to rearrangements of these particles; change and conservation
are reconciled.
Copyright © Christoph Schiller November 1997–May 2006

* The human ear can detect pressure variations at least as small as 20 µPa.
** Leucippus of Elea (Λευκιππος) (c. 490 to c. 430 bce ), Greek philosopher; Elea was a small town south
of Naples. It lies in Italy, but used to belong to the Magna Graecia. Democritus (∆εµοκριτος) of Abdera
(c. 460 to c. 356 or 370 bce ), also a Greek philosopher, was arguably the greatest philosopher who ever lived.
Together with his teacher Leucippus, he was the founder of the atomic theory; Democritus was a much
admired thinker, and a contemporary of Socrates. The vain Plato never even mentions him, as Democritus
was a danger to his own fame. Democritus wrote many books which have been lost; they were not copied
during the Middle Ages because of his scientific and rational world view, which was felt to be a danger by
religious zealots who had the monopoly on the copying industry.
*** The story is told by Lucrece, or Titus Lucretius Carus, in his famous text De natura rerum, around 50
bce. Especially if we imagine particles as little balls, we cannot avoid calling this a typically male idea. (What
Challenge 425 d would be the female approach?)

Dvipsbugw
 224 i galilean motion • 3. the global simplicity of complexity

light laser
detector diode

vertical
piezo
controller tip

Dvipsbugw
horizontal sample
piezo
controller

F I G U R E 111 The principle, and a simple

realization, of an atomic force microscope

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In short, since matter is hard, has a shape and is divisible, Leucippus and Democritus
imagined it as being made of atoms. Atoms are particles which are hard, have a shape, but
are indivisible. In other words, the Greeks imagined nature as a big Lego set. Lego pieces
are first of all hard or impenetrable, i.e. repulsive at very small distances. They are attractive
at small distances: they remain stuck together. Finally, they have no interaction at large
distances. Atoms behave in the same way. (Actually, what the Greeks called ‘atoms’ partly
corresponds to what today we call ‘molecules’. The latter term was invented by Amadeo
Avogadro in 1811 in order to clarify the distinction. But we can forget this detail for the
moment.)
Since atoms are so small, it took many years before all scientists were convinced by the
experiments showing their existence. In the nineteenth century, the idea of atoms was
Page 240 beautifully verified by the discovery of the ‘laws’ of chemistry and those of gas behaviour.
Later on, the noise effects were discovered.
Nowadays, with advances in technology, single atoms can be seen, photographed, holo-
Ref. 175, Ref. 176 grammed, counted, touched, moved, lifted, levitated, and thrown around. And indeed,
like everyday matter, atoms have mass, size, shape and colour. Single atoms have even
Ref. 177 been used as lamps and lasers.
Modern researchers in several fields have fun playing with atoms in the same way that
children play with Lego. Maybe the most beautiful demonstration of these possibilities is
Copyright © Christoph Schiller November 1997–May 2006

provided by the many applications of the atomic force microscope. If you ever have the
opportunity to see one, do not miss it!* It is a simple device which follows the surface of
an object with an atomically sharp needle; such needles, usually of tungsten, are easily
Ref. 178 manufactured with a simple etching method. The changes in the height of the needle
along its path over the surface are recorded with the help of a deflected light ray. With a
little care, the atoms of the object can be felt and made visible on a computer screen. With
special types of such microscopes, the needle can be used to move atoms one by one to
specified places on the surface. It is also possible to scan a surface, pick up a given atom
Ref. 179 and throw it towards a mass spectrometer to determine what sort of atom it is.

* A cheap version costs only a few thousand euro, and will allow you to study the difference between a silicon

Dvipsbugw
 d o extended bodies exist? 225

no image yet
Dvipsbugw
F I G U R E 112 The atoms on the surface of F I G U R E 113 The result of moving helium
a silicon crystal mapped with an atomic atoms on a metallic surface (© IBM)
force microscope

Incidentally, the construction of atomic force microscopes is only a small improve-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ment on what nature is building already by the millions; when we use our ears to listen,
we are actually detecting changes in eardrum position of about 1 nm. In other words, we
all have two ‘atomic force microscopes’ built into our heads.
In summary, matter is not scale invariant: in particular, it is neither smooth nor fractal.
Matter is made of atoms. Different types of atoms, as well as their various combinations,
produce different types of substances. Pictures from atomic force microscopes show that
the size and arrangement of atoms produce the shape and the extension of objects, con-
firming the Lego model of matter.* As a result, the description of the motion of extended
objects can be reduced to the description of the motion of their atoms. Atomic motion
will be a major theme in the following pages. One of its consequences is especially im-
portant: heat.

Curiosities and fun challenges about fluids and solids

Before we continue, a few puzzles are due. They indicate the range of phenomena that the
motion of extended bodies encompasses.

**
How much water is necessary to moisten the air in a room in winter? At 0°C, the vapour
pressure of water is 6 mbar, 20°C it is 23 mbar. As a result, heating air in the winter gives
Copyright © Christoph Schiller November 1997–May 2006

Challenge 426 e at most a humidity of 25%. To increase the humidity by 50%, one thus needs about 1 litre
of water per 100 m3 .
**

wafer – crystalline – a flour wafer – granular-amorphous – and consecrated wafer.

* Studying matter in even more detail yields the now well-known idea that matter, at higher and higher
magnifications, is made of molecules, atoms, nuclei, protons and neutrons, and finally, quarks. Atoms also
contain electrons. A final type of matter, neutrinos, is observed coming from the Sun and from certain types
of radioactive materials. Even though the fundamental bricks have become smaller with time, the basic idea
remains: matter is made of smallest entities, nowadays called elementary particles. In the second part of our
Page 921 mountain ascent we will explore this idea in detail. Appendix C lists the measured properties of all known
elementary particles.

Dvipsbugw
 226 i galilean motion • 3. the global simplicity of complexity

You are in a boat on a pond with a stone, a bucket of water and a piece of wood. What
happens to the water level of the pond after you throw the stone in it? After you throw
Challenge 427 n the water into the pond? After you throw the piece of wood?
**
What is the maximum length of a vertically hanging wire? Could a wire be lowered from
Challenge 428 n a suspended geostationary satellite down to the Earth? This would mean we could realize
a space ‘lift’. How long would the cable have to be? How heavy would it be? How would
you build such a system? What dangers would it face? Dvipsbugw

**
Matter is made of atoms. Over the centuries the stubborn resistance of many people to
this idea has lead to the loss of many treasures. For over a thousand years, people thought
that genuine pearls could be distinguished from false ones by hitting them with a hammer:
only false pearls would break. However, all pearls break. (Also diamonds break in this

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

situation.) As a result, all the most beautiful pearls in the world have been smashed to
pieces.
**
Put a rubber air balloon over the end of a bottle and let it hang inside the bottle. How
Challenge 429 e much can you blow up the balloon inside the bottle?
**
Put a small paper ball into the neck of a horizontal bottle and try to blow it into the bottle.
Challenge 430 e The paper will fly towards you. Why?
**
It is possible to blow an egg from one egg-cup to a second one just behind it. Can you
Challenge 431 e perform this trick?

**
In the seventeenth century, engineers who needed to pump water faced a challenge. To
pump water from mine shafts to the surface, no water pump managed more than 10 m
of height difference. For twice that height, one always needed two pumps in series, con-
Challenge 432 n nected by an intermediate reservoir. Why? How then do trees manage to pump water
Copyright © Christoph Schiller November 1997–May 2006

upwards for larger heights?

**
Comic books have difficulties with the concept of atoms. Could Asterix really throw Ro-
mans into the air using his fist? Are Lucky Luke’s precise revolver shots possible? Can
Spiderman’s silk support him in his swings from building to building? Can the Roadrun-
ner stop running in three steps? Can the Sun be made to stop in the sky by command?
Can space-ships hover using fuel? Take any comic-book hero and ask yourself whether
Challenge 433 e matter made of atoms would allow him the feats he seems capable of. You will find that
most cartoons are comic precisely because they assume that matter is not made of atoms,
but continuous! In a sense, atoms make life a serious adventure.

Dvipsbugw
 d o extended bodies exist? 227

water

F I G U R E 114 What is your personal stone-skipping record?

** Dvipsbugw
When hydrogen and oxygen are combined to form water, the amount of hydrogen needed
is exactly twice the amount of oxygen, if no gas is to be left over after the reaction. How
Challenge 434 n does this observation confirm the existence of atoms?
**
Challenge 435 n How are alcohol-filled chocolate pralines made? Note that the alcohol is not injected into

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

them afterwards, because there would be no way to keep the result tight enough.

**
How often can a stone jump when it is thrown over the surface of water? The present world
record was achieved in 2002: 40 jumps. More information is known about the previous
Ref. 180 world record, achieved in 1992: a palm-sized, triangular and flat stone was thrown with
a speed of 12 m~s (others say 20 m~s) and a rotation speed of about 14 revolutions per
second along a river, covering about 100 m with 38 jumps. (The sequence was filmed
with a video recorder from a bridge.)
What would be necessary to increase the number of jumps? Can you build a machine
Challenge 436 r that is a better thrower than yourself?
**
The biggest component of air is nitrogen (about 78 %). The second biggest component is
Challenge 437 n oxygen (about 21 %). What is the third biggest one?

**
Water can flow uphill: Heron’s fountain shows this most clearly. Heron of Alexandria
(c. 10 to c. 70) described it 2000 years ago; it is easily built at home, using some plastic
Challenge 438 n bottles and a little tubing. How does it work?
Copyright © Christoph Schiller November 1997–May 2006

**
A light bulb is placed, underwater, in a stable steel cylinder with a diameter of 16 cm. A
Fiat Cinquecento (500 kg) is placed on a piston pushing onto the water surface. Will the
Challenge 439 n bulb resist?
**
Challenge 440 ny What is the most dense gas? The most dense vapour?

**
Every year, the Institute of Maritime Systems of the University of Rostock organizes a

Dvipsbugw
 228 i galilean motion • 3. the global simplicity of complexity

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 115 Heron’s fountain

contest. The challenge is to build a paper boat with the highest carrying capacity. The
paper boat must weigh at most 10 g; the carrying capacity is measured by pouring lead
small shot onto it, until the boat sinks. The 2002 record stands at 2.6 kg. Can you achieve
Challenge 441 e this value? (For more information, see the http://www.paperboat.de website.)

**
A modern version of an old question – already posed by Daniel Colladon (1802–1893) –
is the following. A ship of mass m in a river is pulled by horses walking along the riverbank
attached by ropes. If the river is of superfluid helium, meaning that there is no friction
between ship and river, what energy is necessary to pull the ship upstream along the river
Challenge 442 n until a height h has been gained?

**
The Swiss professor Auguste Piccard (1884–1962) was a famous explorer of the strato-
sphere. He reached a height of 16 km in his aerostat. Inside the airtight cabin hanging
under his balloon, he had normal air pressure. However, he needed to introduce several
ropes attached at the balloon into the cabin, in order to be able to pull them, as they con-
Copyright © Christoph Schiller November 1997–May 2006

trolled his balloon. How did he get the ropes into the cabin while preventing air from
Challenge 443 n leaving the cabin?

**
A human cannot breathe at any depth under water, even if he has a tube going to the
surface. At a few metres of depth, trying to do so is inevitably fatal! Even at a depth of
Challenge 444 n 60 cm only, the human body can only breathe in this way for a few minutes. Why?
**
A human in air falls with a limiting speed of about 180 km~h, depending on clothing. How
Challenge 445 ny long does it take to fall from 3000 m to 200 m?

Dvipsbugw
 d o extended bodies exist? 229

**
Several humans have survived free falls from aeroplanes for a thousand metres or more,
Challenge 446 n even though they had no parachute. How was this possible?
**
Liquid pressure depends on height. If the average human blood pressure at the height of
Challenge 447 n the heart is 13.3 kPa, can you guess what it is inside the feet when standing?
** Dvipsbugw
The human heart pumps blood at a rate of about 0.1 l~s. A capillary has the diameter of a
red blood cell, around 7 µm, and in it the blood moves at a speed of half a millimetre per
Challenge 448 n second. How many capillaries are there in a human?
**
A few drops of tea usually flow along the underside of the spout of a teapot (or fall onto the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

table). This phenomenon has even been simulated using supercomputer simulations of
Ref. 181 the motion of liquids, by Kistler and Scriven, using the Navier–Stokes equations. Teapots
are still shedding drops, though.
**
The best giant soap bubbles can be made by mixing 1.5 l of water, 200 ml of corn syrup
and 450 ml of washing-up liquid. Mix everything together and then let it rest for four
hours. You can then make the largest bubbles by dipping a metal ring of up to 100 mm
Challenge 449 n diameter into the mixture. But why do soap bubbles burst?
**
Can humans start earthquakes? What would happen if all the 1000 million Indians were
Challenge 450 n to jump at the same time from the kitchen table to the floor?
In fact, several strong earthquakes have been triggered by humans. This has happened
when water dams have been filled, or when water has been injected into drilling holes.
It has been suggested that the extraction of deep underground water also causes earth-
quakes. If this is confirmed, a sizeable proportion of all earthquakes could be human-
triggered.
**
Copyright © Christoph Schiller November 1997–May 2006

Challenge 451 n How can a tip of a stalactite be distinguished from a tip of a stalagmite? Does the differ-
ence exist also for icicles?

**
A drop of water that falls into a pan containing hot oil dances on the surface for a consid-
erable time, if the oil is above 220°C. Cooks test the temperature of oil in this way. Why
Challenge 452 ny does this so-called Leidenfrost effect* take place?
**

* It is named after Johann Gottlieb Leidenfrost (1715–1794), German physician.

Dvipsbugw
 230 i galilean motion • 3. the global simplicity of complexity

Dvipsbugw
F I G U R E 116 Which funnel is faster?

How much more weight would your bathroom scales show if you stood on them in a
Challenge 453 n vacuum?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Challenge 454 n Why don’t air molecules fall towards the bottom of the container and stay there?
**
Challenge 455 n Which of the two water funnels in Figure 116 is emptied more rapidly? Apply energy
Ref. 182 conservation to the fluid’s motion (also called Bernoulli’s ‘law’) to find the answer.
**
As we have seen, fast flow generates an underpressure. How do fish prevent their eyes
Challenge 456 n from popping when they swim rapidly?
**
Golf balls have dimples for the same reasons that tennis balls are hairy and that shark and
dolphin skin is not flat: deviations from flatness reduce the flow resistance because many
Challenge 457 ny small eddies produce less friction than a few large ones. Why?

**
One of the most complex extended bodies is the human body. In modern simulations
of the behaviour of humans in car accidents, the most advanced models include ribs,
vertebrae, all other bones and the various organs. For each part, its specific deformation
Copyright © Christoph Schiller November 1997–May 2006

properties are taken into account. With such models and simulations, the protection of
passengers and drivers in cars can be optimized.

**
Glass is a solid. Nevertheless, many textbooks say that glass is a liquid. This error has been
propagated for about a hundred years, probably originating from a mistranslation of a
sentence in a German textbook published in 1933 by Gustav Tamman, Der Glaszustand.
Challenge 458 n Can you give at least three reasons why glass is a solid and not a liquid?

**
The recognized record height reached by a helicopter is 12 442 m above sea level, though

Dvipsbugw
 d o extended bodies exist? 231

12 954 m has also been claimed. (The first height was reached in 1972, the second in 2002,
both by French pilots in French helicopters.) Why, then, do people still continue to use
their legs in order to reach the top of Mount Sagarmatha, the highest mountain in the
Challenge 459 n world?
**
A loosely knotted sewing thread lies on the surface of a bowl filled with water. Putting
a bit of washing-up liquid into the area surrounded by the thread makes it immediately
Challenge 460 e become circular. Why? Dvipsbugw

**
The deepest hole ever drilled into the Earth is 12 km deep. In 2003, somebody proposed
to enlarge such a hole and then to pour millions of tons of liquid iron into it. He claims
that the iron would sink towards the centre of the Earth. If a measurement device com-
munication were dropped into the iron, it could send its observations to the surface using

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 461 n sound waves. Can you give some reasons why this would not work?

**
Challenge 462 n How can you put a handkerchief under water using a glass, while keeping it dry?
**
Are you able to blow a ping pong ball out of a funnel? What happens if you blow through
a funnel towards a burning candle?

**
The economic power of a nation has long been associated with its capacity to produce
high-quality steel. Indeed, the Industrial Revolution started with the mass production of
steel. Every scientist should know the basics facts about steel. Steel is a combination of
iron and carbon to which other elements, mostly metals, may be added as well. One can
distinguish three main types of steel, depending on the crystalline structure. Ferritic steels
have a body-centred cubic structure, austenitic steels have a face-centred cubic structure,
and martensitic steels have a body-centred tetragonal structure. Table 28 gives further
details.
**
Copyright © Christoph Schiller November 1997–May 2006

A simple phenomenon which requires a complex explanation is the cracking of a whip.

Ref. 183 Since the experimental work of Peter Krehl it has been known that the whip cracks when
Challenge 463 ny the tip reaches a velocity of twice the speed of sound. Can you imagine why?
**
The fall of a leaf, with its complex path, is still a topic of investigation. We are far from
being able to predict the time a leaf will take to reach the ground; the motion of the air
around a leaf is not easy to describe. One of the simplest phenomena of hydrodynamics
remains one of its most difficult problems.

**

Dvipsbugw
 232 i galilean motion • 3. the global simplicity of complexity

TA B L E 28 Steel types, properties and uses

Ferritic steel Austenitic steel Martensitic steel

‘usual’ steel ‘soft’ steel hardened steel, brittle

body centred cubic (bcc) face centred cubic (fcc) body centred tetragonal (bct)
iron and carbon iron, chromium, nickel, carbon steel and alloys
manganese, carbon
Examples Dvipsbugw
construction steel most stainless (18/8 Cr/Ni) knife edges
steels
car sheet steel kitchenware drill surfaces
ship steel food industry spring steel, crankshafts
12 % Cr stainless ferrite Cr/V steels for nuclear
reactors

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Properties
phases described by the phases described by the phases described by the
iron-carbon phase diagram Schaeffler diagram iron-carbon diagram and the
TTT (time–temperature
transformation) diagram
in equilibrium at RT some alloys in equilibrium at not in equilibrium at RT, but
RT stable
mechanical properties and mechanical properties and mechanical properties and
grain size depend on heat grain size depend on grain size strongly depend on
treatment thermo-mechanical heat treatment
pre-treatment
hardened by reducing grain hardened by cold working hard anyway – made by laser
size, by forging, by increasing only irradiation, induction heating,
carbon content or by nitration etc.
grains of ferrite and paerlite, grains of austenite grains of martensite
with cementite (Fe3 C)
ferromagnetic not magnetic or weakly ferromagnetic
magnetic
Copyright © Christoph Schiller November 1997–May 2006

Fluids exhibit many interesting effects. Soap bubbles in air are made of a thin spherical
film of liquid with air on both sides. In 1932, anti-bubbles, thin spherical films of air with
Ref. 184 liquid on both sides, were first observed. In 2004, the Belgian physicist Stéphane Dorbolo
and his team showed that it is possible to produce them in simple experiments, and in
particular, in Belgian beer.
**
A bicycle chain is an extended object with no stiffness. However, if it is made to rotate
rapidly, it gets dynamical stiffness, and can roll down an inclined plane. This surprising
effect can be seen on the http://www.iwf.de/Navigation/Projekte/LNW/Pohl/index.asp
website.

Dvipsbugw
 what can move in nature? 233

TA B L E 29 Extensive quantities in nature, i.e. quantities that flow and accumulate

Domain Extensive Current Intens- Energy R e s i s ta n c e

q ua n t i t y ive flow to
q ua n t i t y transport
(energy (flow (driving (power) (intensity
carrier) intensity) strength) of entropy
genera-
tion) Dvipsbugw
Rivers mass m mass flow m~t height P = дh m~t R m = дht~m
difference дh [m2 ~s kg]
Gases volume V volume flow V ~t pressure p P = pV ~t R V = pt~V
[kg~s m5 ]
Mechanics momentum p force F = dp~dt velocity v P = vF R p = t~m

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

[s~kg]
angular torque angular P = ωM R L = t~mr 2
momentum L M = dL~dt velocity ω [s~kg m2 ]
Chemistry amount of substance flow chemical P = µ In R n = µt~n
substance n I n = dn~dt potential µ [Js~mol2 ]
Thermo- entropy S entropy flow temperature P = T I S R S = Tt~S
dynamics I S = dS~dt T [K2 ~W]
Light like all massless radiation, it can flow but cannot accumulate
Electricity charge q electrical current electrical P=UI R = U~I
I = dq~dt potential U [Ω]
Magnetism no accumulable magnetic sources are found in nature
Nuclear extensive quantities exist, but do not appear in everyday life
physics
Gravitation empty space can move and flow, but the motion is not observed in everyday life

**
Copyright © Christoph Schiller November 1997–May 2006

Mechanical devices are not covered in this text. There is a lot of progress in the area even
Ref. 185 at present. For example, people have built robots that are able to ride a unicycle. But even
Ref. 186 the physics of human unicycling is not simple.

What can move in nature?

Before we continue to the next way to describe motion globally, we will have a look at the
possibilities of motion in everyday life. One overview is given in Table 29. The domains
that belong to everyday life – motion of fluids, of matter, of matter types, of heat, of light
and of charge – are the domains of continuum physics.
Within continuum physics, there are three domains we have not yet studied: the mo-

Dvipsbugw
 234 i galilean motion • 3. the global simplicity of complexity

tion of charge and light, called electrodynamics, the motion of heat, called thermodynam-
ics, and the motion of the vacuum. Once we have explored these domains, we will have
completed the first step of our description of motion: continuum physics. In continuum
physics, motion and moving entities are described with continuous quantities that can
take any value, including arbitrarily small or arbitrarily large values.
But nature is not continuous. We have already seen that matter cannot be indefinitely
divided into ever-smaller entities. In fact, we will discover that there are precise experi-
ments that provide limits to the observed values for every domain of continuum physics.
There is a limit to mass, to speed, to angular momentum, to force, to entropy and to Dvipsbugw
change of charge. The consequences of these discoveries form the second step in our de-
scription of motion: quantum theory and relativity. Quantum theory is based on lower
limits; relativity is based on upper limits. The third and last step of our description of
motion will be formed by the unification of quantum theory and general relativity.
Every domain of physics, regardless of which one of the above steps it belongs to, de-
scribes change in terms two quantities: energy, and an extensive quantity characteristic

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 187 of the domain. An observable quantity is called extensive if it increases with system size.
Table 29 provides an overview. The intensive and extensive quantities corresponding to
what in everyday language is called ‘heat’ are temperature and entropy.

How d o objects get warm?

We continue our short stroll through the field of global descriptions of motion with an
overview of heat and the main concepts associated with it. For our purposes we only need
Ref. 188 to know the basic facts about heat. The main points that are taught in school are almost
sufficient.
Macroscopic bodies, i.e. bodies made of many atoms, have temperature. The temper-
ature of a macroscopic body is an aspect of its state. It is observed that any two bodies in
contact tend towards the same temperature: temperature is contagious. In other words,
temperature describes an equilibrium situation. The existence and contagiousness of tem-
perature is often called the zeroth principle of thermodynamics. Heating is the increase of
temperature.
How is temperature measured? The eighteenth century produced the clearest answer:
temperature is best defined and measured by the expansion of gases. For the simplest, so-
called ideal gases, the product of pressure p and volume V is proportional to temperature:

pV  T .
Copyright © Christoph Schiller November 1997–May 2006

(88)

The proportionality constant is fixed by the amount of gas used. (More about it shortly.)
The ideal gas relation allows us to determine temperature by measuring pressure and
volume. This is the way (absolute) temperature has been defined and measured for about
Ref. 189 a century. To define the unit of temperature, one only has to fix the amount of gas used. It
Page 1154 is customary to fix the amount of gas at 1 mol; for oxygen this is 32 g. The proportionality
constant, called the ideal gas constant R, is defined to be R = 8.3145 J~mol K. This number
has been chosen in order to yield the best approximation to the independently defined
Celsius temperature scale. Fixing the ideal gas constant in this way defines 1 K, or one
Kelvin, as the unit of temperature. In simple terms, a temperature increase of one Kelvin

Dvipsbugw
 temperature 235

is defined as the temperature increase that makes the volume of an ideal gas increase –
Challenge 464 ny keeping the pressure fixed – by a fraction of 1/273.15 or 0.3661 %.
In general, if one needs to determine the temperature of an object, one takes a mole
of gas, puts it in contact with the object, waits a while, and then measures the pressure
and the volume of the gas. The ideal gas relation (88) then gives the temperature. Most im-
portantly, the ideal gas relation shows that there is a lowest temperature in nature, namely
that temperature at which an ideal gas would have a vanishing volume. That would hap-
pen at T = 0 K, i.e. at −273.15°C. Obviously, other effects, like the volume of the atoms
themselves, prevent the volume of the gas from ever reaching zero. The third principle of Dvipsbugw
thermodynamics provides another reason why this is impossible.

Temperature
The temperature achieved by a civilization can be used as a measure of its technological
achievements. One can define the Bronze Age (1.1 kK, 3500 bce) , the Iron Age (1.8 kK,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1000 bce), the Electric Age (3 kK from c. 1880) and the Atomic Age (several MK, from
1944) in this way. Taking into account also the quest for lower temperatures, one can
Ref. 190 define the Quantum Age (4 K, starting 1908).
Heating implies flow of energy. For example, friction heats up and slows down mov-
ing bodies. In the old days, the ‘creation’ of heat by friction was even tested experiment-
ally. It was shown that heat could be generated from friction, just by continuous rubbing,
without any limit; this ‘creation’ implies that heat is not a material fluid extracted from
the body – which in this case would be consumed after a certain time – but something
else. Indeed, today we know that heat, even though it behaves in some ways like a fluid, is
due to disordered motion of particles. The conclusion of these studies is simple. Friction
is the transformation of mechanical energy into thermal energy.
To heat 1 kg of water by 1 K by friction, 4.2 kJ of mechanical energy must be trans-
formed through friction. The first to measure this quantity with precision was, in 1842,
the German physician Julius Robert Mayer (1814–1878). He regarded his experiment as
proof of the conservation of energy; indeed, he was the first person to state energy con-
servation! It is something of an embarrassment to modern physics that a medical doctor
was the first to show the conservation of energy, and furthermore, that he was ridiculed
by most physicists of his time. Worse, conservation of energy was accepted only when
it was repeated many years later by two authorities: Hermann von Helmholtz – himself
also a physician turned physicist – and William Thomson, who also cited similar, but
Copyright © Christoph Schiller November 1997–May 2006

later experiments by James Joule.* All of them acknowledged Mayer’s priority. Publicity
by William Thomson eventually led to the naming of the unit of energy after Joule.
In short, the sum of mechanical energy and thermal energy is constant. This is usu-
ally called the first principle of thermodynamics. Equivalently, it is impossible to produce
mechanical energy without paying for it with some other form of energy. This is an im-
portant statement, because among others it means that humanity will stop living one day.

* Hermann von Helmholtz (b. 1821 Potsdam, d. 1894 Berlin), important Prussian scientist. William Thom-
son (later William Kelvin) (1824–1907), important Irish physicist. James Prescott Joule (1818–1889), Eng-
lish physicist. Joule is pronounced so that it rhymes with ‘cool’, as his descendants like to stress. (The pro-
nunciation of the name ‘Joule’ varies from family to family.)

Dvipsbugw
236 i galilean motion • 3. the global simplicity of complexity

TA B L E 30 Some temperature values

O b s e r va t i o n Te m p e r at u r e

Lowest, but unattainable, temperature 0 K = −273.15°C

In the context of lasers, it sometimes makes sense to talk about
negative temperature.
Temperature a perfect vacuum would have at Earth’s surface 40 zK
Page 871 Dvipsbugw
Sodium gas in certain laboratory experiments – coldest matter 0.45 nK
system achieved by man and possibly in the universe
Temperature of neutrino background in the universe c. 2 K
Temperature of photon gas background (or background radi- 2.7 K
ation) in the universe
Liquid helium 4.2 K

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Oxygen triple point 54.3584 K
Liquid nitrogen 77 K
Coldest weather ever measured (Antarctic) 185 K = −88°C
Freezing point of water at standard pressure 273.15 K = 0.00°C
Triple point of water 273.16 K = 0.01°C
Average temperature of the Earth’s surface 287.2 K
Smallest uncomfortable skin temperature 316 K (10 K above normal)
Interior of human body 310.0  0.5 K = 36.8  0.5°C
Hottest weather measured 331 K = 58°C
Boiling point of water at standard pressure 373.13 K or 99.975°C
Liquid bronze c. 1100 K
Liquid, pure iron 1810 K
Freezing point of gold 1337.33 K
Light bulb filament 2.9 kK
Earth’s centre 4 kK
Sun’s surface 5.8 kK
Air in lightning bolt 30 kK
Hottest star’s surface (centre of NGC 2240) 250 kK
Space between Earth and Moon (no typo) up to 1 MK
Copyright © Christoph Schiller November 1997–May 2006

Sun’s centre 20 MK
Inside the JET fusion tokamak 100 MK
Centre of hottest stars 1 GK
Maximum temperature of systems without electron–positron ca. 6 GK
pair generation
Universe when it was 1 s old 100 GK
Hagedorn temperature 1.9 TK
Heavy ion collisions – highest man-made value up to 3.6 TK
Planck temperature – nature’s upper temperature limit 1032 K

Dvipsbugw
 temperature 237

Indeed, we live mostly on energy from the Sun; since the Sun is of finite size, its energy
Challenge 465 n content will eventually be consumed. Can you estimate when this will happen?
There is also a second (and the mentioned third) principle of thermodynamics, which
will be presented later on. The study of these topics is called thermostatics if the systems
concerned are at equilibrium, and thermodynamics if they are not. In the latter case, we
distinguish situations near equilibrium, when equilibrium concepts such as temperature
can still be used, from situations far from equilibrium, such as self-organization, where
Page 255 such concepts often cannot be applied.
Does it make sense to distinguish between thermal energy and heat? It does. Many Dvipsbugw
older texts use the term ‘heat’ to mean the same as thermal energy. However, this is con-
fusing; in this text, ‘heat’ is used, in accordance with modern approaches, as the everyday
term for entropy. Both thermal energy and heat flow from one body to another, and both
accumulate. Both have no measurable mass.* Both the amount of thermal energy and the
amount of heat inside a body increase with increasing temperature. The precise relation
will be given shortly. But heat has many other interesting properties and stories to tell. Of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

these, two are particularly important: first, heat is due to particles; and secondly, heat is
at the heart of the difference between past and future. These two stories are intertwined.

Entropy
– It’s irreversible.

“ – Like my raincoat!
Mel Brooks, Spaceballs, 1987

Every domain of physics describes change in terms of two quantities: energy, and an ex- ”
Ref. 187 tensive quantity characteristic of the domain. Even though heat is related to energy, the
quantity physicists usually call heat is not an extensive quantity. Worse, what physicists
call heat is not the same as what we call heat in our everyday speech. The extensive quant-
ity corresponding to what we call ‘heat’ in everyday speech is called entropy.** Entropy
describes heat in the same way as momentum describes motion. When two objects differ-
ing in temperature are brought into contact, an entropy flow takes place between them,
like the flow of momentum that take place when two objects of different speeds collide.
Let us define the concept of entropy more precisely and explore its properties in some
more detail.
Entropy measures the degree to which energy is mixed up inside a system, that is, the
degree to which energy is spread or shared among the components of a system. Therefore,
Copyright © Christoph Schiller November 1997–May 2006

entropy adds up when identical systems are composed into one. When two litre bottles
of water at the same temperature are poured together, the entropy of the water adds up.
Like any other extensive quantity, entropy can be accumulated in a body; it can flow
into or out of bodies. When water is transformed into steam, the entropy added into
the water is indeed contained in the steam. In short, entropy is what is called ‘heat’ in

* This might change in future, when mass measurements improve in precision, thus allowing the detection
Page 314 of relativistic effects. In this case, temperature increase may be detected through its related mass increase.
However, such changes are noticeable only with twelve or more digits of precision in mass measurements.
** The term ‘entropy’ was invented by the German physicist Rudolph Clausius (1822–1888) in 1865. He
formed it from the Greek ἐν ‘in’ and τρόπος ‘direction’, to make it sound similar to ‘energy’. It has always had
the meaning given here.

Dvipsbugw
 238 i galilean motion • 3. the global simplicity of complexity

TA B L E 31 Some measured entropy values

Pro cess/System E n t r o p y va l u e

Melting of 1 kg of ice 1.21 kJ~K kg = 21.99 J~K mol

Water under standard conditions 70.1 J~K mol
Boiling of 1 kg of liquid water at 101.3 kPa 6.03 kJ~K= 110 J~K mol
Iron under standard conditions 27.2 J~K mol
Oxygen under standard conditions 161.1 J~K mol
Dvipsbugw

everyday speech.
In contrast to several other important extensive quantities, entropy is not conserved.
The sharing of energy in a system can be increased, for example by heating it. However,
entropy is ‘half conserved’: in closed systems, entropy does not decrease; mixing cannot

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

be undone. What is called equilibrium is simply the result of the highest possible mixing.
In short, the entropy in a closed system increases until it reaches the maximum possible
value.
When a piece of rock is detached from a mountain, it falls, tumbles into the valley,
heating up a bit, and eventually stops. The opposite process, whereby a rock cools and
tumbles upwards, is never observed. Why? The opposite motion does not contradict any
Challenge 466 ny rule or pattern about motion that we have deduced so far.
Rocks never fall upwards because mountains, valleys and rocks are made of many
particles. Motions of many-particle systems, especially in the domain of thermostatics,
are called processes. Central to thermostatics is the distinction between reversible pro-
cesses, such as the flight of a thrown stone, and irreversible processes, such as the afore-
mentioned tumbling rock. Irreversible processes are all those processes in which friction
and its generalizations play a role. They are those which increase the sharing or mixing of
energy. They are important: if there were no friction, shirt buttons and shoelaces would
Ref. 191 not stay fastened, we could not walk or run, coffee machines would not make coffee, and
Page 806 maybe most importantly of all, we would have no memory.
Irreversible processes, in the sense in which the term is used in thermostatics, trans-
form macroscopic motion into the disorganized motion of all the small microscopic com-
ponents involved: they increase the sharing and mixing of energy. Irreversible processes
are therefore not strictly irreversible – but their reversal is extremely improbable. We can
Copyright © Christoph Schiller November 1997–May 2006

say that entropy measures the ‘amount of irreversibility’: it measures the degree of mixing
or decay that a collective motion has undergone.
Entropy is not conserved. Entropy – ‘heat’ – can appear out of nowhere, since energy
sharing or mixing can happen by itself. For example, when two different liquids of the
same temperature are mixed – such as water and sulphuric acid – the final temperature
of the mix can differ. Similarly, when electrical current flows through material at room
temperature, the system can heat up or cool down, depending on the material.
The second principle of thermodynamics states that ‘entropy ain’t what it used to be.’
More precisely, the entropy in a closed system tends towards its maximum. Here, a closed
system is a system that does not exchange energy or matter with its environment. Can
Challenge 467 ny you think of an example?

Dvipsbugw
 temperature 239

Entropy never decreases. Everyday life shows that in a closed system, the disorder
increases with time, until it reaches some maximum. To reduce disorder, we need effort,
i.e. work and energy. In other words, in order to reduce the disorder in a system, we
need to connect the system to an energy source in some clever way. Refrigerators need
electrical current precisely for this reason.
Because entropy never decreases, white colour does not last. Whenever disorder in-
creases, the colour white becomes ‘dirty’, usually grey or brown. Perhaps for this reason
white objects, such as white clothes, white houses and white underwear, are valued in our
society. White objects defy decay. Dvipsbugw
Entropy allows to define the concept of equilibrium more precisely as the state of max-
imum entropy, or maximum energy sharing.

Flow of entropy
We know from daily experience that transport of an extensive quantity always involves

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

friction. Friction implies generation of entropy. In particular, the flow of entropy itself
produces additional entropy. For example, when a house is heated, entropy is produced
in the wall. Heating means to keep a temperature difference ∆T between the interior and
the exterior of the house. The heat flow J traversing a square metre of wall is given by

J = κ∆T = κ(Ti − Te ) (89)

where κ is a constant characterizing the ability of the wall to conduct heat. While con-
ducting heat, the wall also produces entropy. The entropy production σ is proportional to
the difference between the interior and the exterior entropy flows. In other words, one
has
(Ti − Te )2
σ= =κ
J J
− . (90)
Te Ti Ti Te

Note that we have assumed in this calculation that everything is near equilibrium in each
slice parallel to the wall, a reasonable assumption in everyday life. A typical case of a good
wall has κ = 1 W~m2 K in the temperature range between 273 K and 293 K. With this value,
one gets an entropy production of

σ = 5 ë 10−3 W~m2 K . (91)

Copyright © Christoph Schiller November 1997–May 2006

Can you compare the amount of entropy that is produced in the flow with the amount that
Challenge 468 ny is transported? In comparison, a good goose-feather duvet has κ = 1.5 W~m2 K, which in
shops is also called 15 tog.*
There are two other ways, apart from heat conduction, to transport entropy: convection,
used for heating houses, and radiation, which is possible also through empty space. For
* That unit is not as bad as the official (not a joke) BthU ë h~sqft~cm~°F used in some remote provinces of
our galaxy.
The insulation power of materials is usually measured by the constant λ = κd which is independent of
the thickness d of the insulating layer. Values in nature range from about 2000 W~K m for diamond, which
is the best conductor of all, down to between 0.1 W~K m and 0.2 W~K m for wood, between 0.015 W~K m
and 0.05 W~K m for wools, cork and foams, and the small value of 5 ë 10−3 W~K m for krypton gas.

Dvipsbugw
 240 i galilean motion • 3. the global simplicity of complexity

F I G U R E 117 The basic idea of statistical

Dvipsbugw
mechanics about gases

example, the Earth radiates about 1.2 W~m2 K into space, in total thus about 0.51 PW~K.
The entropy is (almost) the same that the Earth receives from the Sun. If more entropy
had to be radiated away than received, the temperature of the surface of the Earth would
have to increase. This is called the greenhouse effect. (It is also called global warming.) Let’s

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

hope that it remains small in the near future.

Do isolated systems exist?

In all our discussions so far, we have assumed that we can distinguish the system under
investigation from its environment. But do such isolated or closed systems, i.e. systems not
interacting with their environment, actually exist? Probably our own human condition
was the original model for the concept: we do experience having the possibility to act
independently of our environment. An isolated system may be simply defined as a system
not exchanging any energy or matter with its environment. For many centuries, scientists
saw no reason to question this definition.
The concept of an isolated system had to be refined somewhat with the advent of
quantum mechanics. Nevertheless, the concept provides useful and precise descriptions
of nature also in that domain. Only in the third part of our walk will the situation change
drastically. There, the investigation of whether the universe is an isolated system will lead
Challenge 469 n to surprising results. (What do you think?)* We’ll take the first steps towards the answer
shortly.

Why do balloons take up space? – The end of continuity

Heat properties are material-dependent. Studying them should therefore enable us to
Copyright © Christoph Schiller November 1997–May 2006

understand something about the constituents of matter. Now, the simplest materials of
all are gases.** Gases need space: an amount of gas has pressure and volume. Indeed, it
did not take long to show that gases could not be continuous. One of the first scientists
to think about gases as made up of atoms was Daniel Bernoulli.*** Bernoulli reasoned

* A strange hint: your answer is almost surely wrong.

** By the way, the word gas is a modern construct. It was coined by the Brussels alchemist and physician
Johan Baptista van Helmont (1579–1644), to sound similar to ‘chaos’. It is one of the few words which have
been invented by one person and then adopted all over the world.
*** Daniel Bernoulli (b. 1700 Bâle, d. 1782 Bâle), important Swiss mathematician and physicist. His father
Johann and his uncle Jakob were famous mathematicians, as were his brothers and some of his nephews.
Daniel Bernoulli published many mathematical and physical results. In physics, he studied the separation

Dvipsbugw
 temperature 241

F I G U R E 118 Which balloon wins?

Dvipsbugw

that if atoms are small particles, with mass and momentum, he should be able to make
quantitative predictions about the behaviour of gases, and check them with experiment.
If the particles fly around in a gas, then the pressure of a gas in a container is produced
by the steady flow of particles hitting the wall. It was then easy to conclude that if the
particles are assumed to behave as tiny, hard and perfectly elastic balls, the pressure p,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 470 ny volume V and temperature T must be related by

pV =
3k
NT (92)
2
where N is the number of particles contained in the gas. (The Boltzmann constant k,
one of the fundamental constants of nature, is defined below.) A gas made of particles
with such textbook behaviour is called an ideal gas. Relation (92) has been confirmed by
experiments at room and higher temperatures, for all known gases.
Bernoulli thus derived the gas relation, with a specific prediction
for the proportionality constant, from the single assumption that
gases are made of small massive constituents. This derivation provides
a clear argument for the existence of atoms and for their behaviour as
normal, though small objects. (Can you imagine how N might be de-
Challenge 471 ny termined experimentally?)
The ideal gas model helps us to answer questions such as the one
illustrated in Figure 118. Two identical rubber balloons, one filled up
to a larger size than the other, are connected via a pipe and a valve. Daniel Bernoulli
Challenge 472 n The valve is opened. Which one deflates?
The ideal gas relation states that hotter gases, at given pressure, need more volume.
Copyright © Christoph Schiller November 1997–May 2006

Challenge 473 e The relation thus explains why winds and storms exist, why hot air balloons rise, why
car engines work, why the ozone layer is destroyed by certain gases, or why during the
extremely hot summer of 2001 in the south of Turkey, oxygen maks were necessary to
walk outside around noon.
Now you can take up the following challenge: how can you measure the weight of a
Challenge 474 n car or a bicycle with a ruler only?

of compound motion into translation and rotation. In 1738 he published the Hydrodynamique, in which
he deduced all results from a single principle, namely the conservation of energy. The so-called Bernoulli’s
principle states that (and how) the pressure of a fluid decreases when its speed increases. He studied the tides
and many complex mechanical problems, and explained the Boyle–Mariotte gas law. For his publications he
won the prestigious prize of the French Academy of Sciences – a forerunner of the Nobel Prize – ten times.

Dvipsbugw
 242 i galilean motion • 3. the global simplicity of complexity

The picture of gases as being made of hard constituents without any long-distance
interactions breaks down at very low temperatures. However, the ideal gas relation (92)
can be improved to overcome these limitations by taking into account the deviations due
Ref. 192 to interactions between atoms or molecules. This approach is now standard practice and
allows us to measure temperatures even at extremely low values. The effects observed
below 80 K, such as the solidification of air, frictionless transport of electrical current, or
Ref. 193 frictionless flow of liquids, form a fascinating world of their own, the beautiful domain
Page 844, page 847 of low-temperature physics; it will be explored later on.
Dvipsbugw
Brownian motion
It is easy to observe, under a microscope, that small particles (such as pollen) in a liquid
never come to rest. They seem to follow a random zigzag movement. In 1827, the English
botanist Robert Brown (1773–1858) showed with a series of experiments that this obser-
vation is independent of the type of particle and of the type of liquid. In other words,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Brown had discovered a fundamental noise in nature. Around 1860, this motion was
attributed to the molecules of the liquid colliding with the particles. In 1905 and 1906,
Ref. 194 Marian von Smoluchowski and, independently, Albert Einstein argued that this theory
could be tested experimentally, even though at that time nobody was able to observe mo-
lecules directly. The test makes use of the specific properties of thermal noise.
It had already been clear for a long time that if molecules, i.e. indivisible matter
particles, really existed, then heat had to be disordered motion of these constituents
and temperature had to be the average energy per degree of freedom of the constituents.
Bernoulli’s model of Figure 117 implies that for monoatomic gases the kinetic energy Tkin
Challenge 475 ny per particle is given by
Tkin = kT
3
(93)
2

where T is temperature. The so-called Boltzmann constant k = 1.4 ë 10−23 J~K is the stand-
ard conversion factor between temperature and energy.*At a room temperature of 293 K,
the kinetic energy is thus 6 zJ.
Using relation (93) to calculate the speed of air molecules at room temperature yields
Challenge 476 ny values of several hundred metres per second. Why then does smoke from a candle take
so long to diffuse through a room? Rudolph Clausius (1822–1888) answered this question
in the mid-nineteenth century: diffusion is slowed by collisions with air molecules, in the
same way as pollen particles collide with molecules in liquids.
Copyright © Christoph Schiller November 1997–May 2006

At first sight, one could guess that the average distance the pollen particle has moved
after n collisions should be zero, because the molecule velocities are random. However,
this is wrong, as experiment shows.

* The important Austrian physicist Ludwig Boltzmann (b. 1844 Vienna, d. 1906 Duino) is most famous for
his work on thermodynamics, in which he explained all thermodynamic phenomena and observables, in-
cluding entropy, as results of the behaviour of molecules. Planck named the Boltzmann constant after his
investigations. He was one of the most important physicists of the late nineteenth century and stimulated
many developments that led to quantum theory. It is said that Boltzmann committed suicide partly because
of the resistance of the scientific establishment to his ideas. Nowadays, his work is standard textbook mater-
ial.

Dvipsbugw
 temperature 243

probability density evolution

b

b
Dvipsbugw

F I G U R E 119 Example paths for particles in Brownian motion and their displacement distribution

An average square displacement, written `d 2 e, is observed for the pollen particle. It

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

cannot be predicted in which direction the particle will move, but it does move. If the
distance the particle moves after one collision is l, the average square displacement after
Challenge 477 ny n collisions is given, as you should be able to show yourself, by

`d 2 e = nl 2 . (94)

For molecules with an average velocity v over time t this gives

`d 2 e = nl 2 = vl t . (95)

In other words, the average square displacement increases proportionally with time. Of
course, this is only valid if the liquid is made of separate molecules. Repeatedly measuring
the position of a particle should give the distribution shown in Figure 119 for the probab-
ility that the particle is found at a given distance from the starting point. This is called the
Ref. 195 (Gaussian) normal distribution. In 1908, Jean Perrin* performed extensive experiments
in order to test this prediction. He found that equation (95) corresponded completely
with observations, thus convincing everybody that Brownian motion is indeed due to
collisions with the molecules of the surrounding liquid, as Smoluchowski and Einstein
had predicted.** Perrin received the 1926 Nobel Prize for these experiments.
Copyright © Christoph Schiller November 1997–May 2006

Einstein also showed that the same experiment could be used to determine the number
of molecules in a litre of water (or equivalently, the Boltzmann constant k). Can you work
Challenge 478 d out how he did this?

* Jean Perrin (1870–1942), important French physicist, devoted most of his career to the experimental proof
of the atomic hypothesis and the determination of Avogadro’s number; in pursuit of this aim he perfected
the use of emulsions, Brownian motion and oil films. His Nobel Prize speech (http://nobelprize.org/physics/
laureates/1926/perrin-lecture.html) tells the interesting story of his research. He wrote the influential book
Les atomes and founded the Centre National de la Recherche Scientifique. He was also the first to speculate,
in 1901, that an atom is similar to a small solar system.
Ref. 196 ** In a delightful piece of research, Pierre Gaspard and his team showed in 1998 that Brownian motion is
Page 259 also chaotic, in the strict physical sense given later on.

Dvipsbugw
 244 i galilean motion • 3. the global simplicity of complexity

TA B L E 32 Some typical entropy values per particle at

standard temperature and pressure as multiples of the
Boltzmann constant

M at e r i a l Entropy per
pa r t i c l e

Monoatomic solids 0.3 k to 10 k

Diamond 0.29 k
Graphite 0.68 k
Dvipsbugw
Lead 7.79 k
Monoatomic gases 15-25 k
Helium 15.2 k
Radon 21.2 k
Diatomic gases 15 k to 30 k
Polyatomic solids 10 k to 60 k

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Polyatomic liquids 10 k to 80 k
Polyatomic gases 20 k to 60 k
Icosane 112 k

Entropy and particles

Once it had become clear that heat and temperature are due to the motion of microscopic
particles, people asked what entropy was microscopically. The answer can be formulated
in various ways. The two most extreme answers are:
— Entropy is the expected number of yes-or-no questions, multiplied by k ln 2, the an-
swers of which would tell us everything about the system, i.e. about its microscopic
state.
— Entropy measures the (logarithm of the) number W of possible microscopic states.
A given macroscopic state can have many microscopic realizations. The logarithm of
this number, multiplied by the Boltzmann constant k, gives the entropy.*
In short, the higher the entropy, the more microstates are possible. Through either
of these definitions, entropy measures the quantity of randomness in a system. In other
words, it measures the transformability of energy: higher entropy means lower transform-
Copyright © Christoph Schiller November 1997–May 2006

ability. Alternatively, entropy measures the freedom in the choice of microstate that a sys-
tem has. High entropy means high freedom of choice for the microstate. For example,
when a molecule of glucose (a type of sugar) is produced by photosynthesis, about 40
bits of entropy are released. This means that after the glucose is formed, 40 additional
yes-or-no questions must be answered in order to determine the full microscopic state
of the system. Physicists often use a macroscopic unit; most systems of interest are large,
and thus an entropy of 1023 bits is written as 1 J~K.**

* When Max Planck went to Austria to search for the anonymous tomb of Boltzmann in order to get him
buried in a proper grave, he inscribed the formula S = k ln W on the tombstone. (Which physicist would
finance the tomb of another, nowadays?)
Challenge 479 ny ** This is only approximate. Can you find the precise value?

Dvipsbugw
 temperature 245

To sum up, entropy is thus a specific measure for the characterization of disorder of
Ref. 197 thermal systems. Three points are worth making here. First of all, entropy is not the meas-
ure of disorder, but one measure of disorder. It is therefore not correct to use entropy as
a synonym for the concept of disorder, as is often done in the popular literature. Entropy
is only defined for systems that have a temperature, in other words, only for systems that
are in or near equilibrium. (For systems far from equilibrium, no measure of disorder
has been found yet; probably none is possible.) In fact, the use of the term entropy has
degenerated so much that sometimes one has to call it thermodynamic entropy for clarity.
Secondly, entropy is related to information only if information is defined also as Dvipsbugw
−k ln W. To make this point clear, take a book with a mass of one kilogram. At room
temperature, its entropy content is about 4 kJ~K. The printed information inside a book,
say 500 pages of 40 lines with each containing 80 characters out of 64 possibilities, corres-
ponds to an entropy of 4 ë 10−17 J~K. In short, what is usually called ‘information’ in every-
day life is a negligible fraction of what a physicist calls information. Entropy is defined
using the physical concept of information.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Finally, entropy is also not a measure for what in normal life is called the complexity
Ref. 198 of a situation. In fact, nobody has yet found a quantity describing this everyday notion.
Challenge 480 ny The task is surprisingly difficult. Have a try!
In summary, if you hear the term entropy used with a different meaning than S =
k ln W, beware. Somebody is trying to get you, probably with some ideology.

The minimum entropy of nature – the quantum of information

Before we complete our discussion of thermostatics we must point out in another way
the importance of the Boltzmann constant k. We have seen that this constant appears
whenever the granularity of matter plays a role; it expresses the fact that matter is made
of small basic entities. The most striking way to put this statement is the following: There
is a smallest entropy in nature. Indeed, for all systems, the entropy obeys

SE
k
. (96)
2
This result is almost 100 years old; it was stated most clearly (with a different numerical
Ref. 199 factor) by the Hungarian–German physicist Leo Szilard. The same point was made by the
Ref. 200 French physicist Léon Brillouin (again with a different numerical factor). The statement
can also be taken as the definition of the Boltzmann constant.
Copyright © Christoph Schiller November 1997–May 2006

The existence of a smallest entropy in nature is a strong idea. It eliminates the possibil-
ity of the continuity of matter and also that of its fractality. A smallest entropy implies that
matter is made of a finite number of small components. The limit to entropy expresses
the fact that matter is made of particles.* The limit to entropy also shows that Galilean
physics cannot be correct: Galilean physics assumes that arbitrarily small quantities do
exist. The entropy limit is the first of several limits to motion that we will encounter until
we finish the second part of our ascent. After we have found all limits, we can start the
third and final part, leading to unification.

* The minimum entropy implies that matter is made of tiny spheres; the minimum action, which we will
encounter in quantum theory, implies that these spheres are actually small clouds.

Dvipsbugw
 246 i galilean motion • 3. the global simplicity of complexity

The existence of a smallest quantity implies a limit on the precision of measurement.

Measurements cannot have infinite precision. This limitation is usually stated in the form
of an indeterminacy relation. Indeed, the existence of a smallest entropy can be rephrased
as an indeterminacy relation between the temperature T and the inner energy U of a
system:

∆U E .
1 k
∆ (97)
T 2
Dvipsbugw
Ref. 201 This relation* was given by Niels Bohr; it was discussed by Werner Heisenberg, who
Page 1079 called it one of the basic indeterminacy relations of nature. The Boltzmann constant (di-
Ref. 203 vided by 2) thus fixes the smallest possible entropy value in nature. For this reason, Gilles
Ref. 200 Cohen-Tannoudji calls it the quantum of information and Herbert Zimmermann calls it
the quantum of entropy.
The relation (97) points towards a more general pattern. For every minimum value for

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

an observable, there is a corresponding indeterminacy relation. We will come across this
several times in the rest of our adventure, most importantly in the case of the quantum
Page 704 of action and Heisenberg’s indeterminacy relation.
The existence of a smallest entropy has numerous consequences. First of all, it sheds
light on the third principle of thermodynamics. A smallest entropy implies that absolute
zero temperature is not achievable. Secondly, a smallest entropy explains why entropy
values are finite instead of infinite. Thirdly, it fixes the absolute value of entropy for every
system; in continuum physics, entropy, like energy, is only defined up to an additive con-
stant. The entropy limit settles all these issues.
The existence of a minimum value for an observable implies that an indeterminacy re-
lation appears for any two quantities whose product yields that observable. For example,
entropy production rate and time are such a pair. Indeed, an indeterminacy relation con-
nects the entropy production rate P = dS~dt and the time t:

∆P ∆t E
k
. (98)
2
From this and the previous relation (97) it is possible to deduce all of statistical physics,
Ref. 203, Ref. 202 i.e., the precise theory of thermostatics and thermodynamics. We will not explore this
further here. (Can you show that the zeroth principle follows from the existence of a
Copyright © Christoph Schiller November 1997–May 2006

Challenge 481 ny smallest entropy?) We will limit ourselves to one of the cornerstones of thermodynamics:
the second principle.

Why can’t we remember the future?

It’s a poor sort of memory which only works

Page 45
“ backwards.
Lewis Carroll, Alice in Wonderland

When we first discussed time, we ignored the difference between past and future. But ob- ”
viously, a difference exists, as we do not have the ability to remember the future. This is

Ref. 202 * It seems that the historical value for the right hand side, given by k, has to be corrected to k~2.

Dvipsbugw
 temperature 247

TA B L E 33 Some minimum flow values found in nature

O b s e r va t i o n M i n i m u m va l u e

Matter flow one molecule, one atom or one particle

Volume flow one molecule, one atom or one particle
Momentum flow Planck’s constant divided by wavelength
Angular momentum flow Planck’s constant
Chemical amount of substance one molecule, one atom or one particle
Entropy flow minimum entropy Dvipsbugw
Charge flow elementary charge
Light flow Planck’s constant divided by wavelength

not a limitation of our brain alone. All the devices we have invented, such as tape record-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ers, photographic cameras, newspapers and books, only tell us about the past. Is there a
way to build a video recorder with a ‘future’ button? Such a device would have to solve a
Challenge 482 ny deep problem: how would it distinguish between the near and the far future? It does not
take much thought to see that any way to do this would conflict with the second principle
of thermodynamics. That is unfortunate, as we would need precisely the same device to
Challenge 483 ny show that there is faster-than-light motion. Can you find the connection?
In summary, the future cannot be remembered because entropy in closed systems
tends towards a maximum. Put even more simply, memory exists because the brain is
made of many particles, and so the brain is limited to the past. However, for the most
simple types of motion, when only a few particles are involved, the difference between
past and future disappears. For few-particle systems, there is no difference between times
gone by and times approaching. We could say that the future differs from the past only
in our brain, or equivalently, only because of friction. Therefore the difference between
the past and the future is not mentioned frequently in this walk, even though it is an es-
sential part of our human experience. But the fun of the present adventure is precisely to
overcome our limitations.

Is everything made of particles?

A physicist is the atom’s way of knowing about

“ atoms.
Copyright © Christoph Schiller November 1997–May 2006

George Wald

”
Ref. 204

Historically, the study of statistical mechanics has been of fundamental importance for
physics. It provided the first demonstration that physical objects are made of interacting
particles. The story of this topic is in fact a long chain of arguments showing that all the
properties we ascribe to objects, such as size, stiffness, colour, mass density, magnetism,
thermal or electrical conductivity, result from the interaction of the many particles they
consist of. The discovery that all objects are made of interacting particles has often been
called the main result of modern science.
Page 233 How was this discovery made? Table 29 listed the main extensive quantities used in
physics. Extensive quantities are able to flow. It turns out that all flows in nature are com-

Dvipsbugw
 248 i galilean motion • 3. the global simplicity of complexity

posed of elementary processes, as shown in Table 33. We have seen that the flow of mass,
volume, charge, entropy and substance are composed. Later, quantum theory will show
the same for the flow of linear and angular momentum. All flows are made of particles.
This success of this idea has led many people to generalize it to the statement:
‘Everything we observe is made of parts.’ This approach has been applied with success
Ref. 205 to chemistry with molecules, materials science and geology with crystals, electricity with
electrons, atoms with elementary particles, space with points, time with instants, light
with photons, biology with cells, genetics with genes, neurology with neurons, mathem-
atics with sets and relations, logic with elementary propositions, and even to linguist- Dvipsbugw
ics with morphemes and phonemes. All these sciences have flourished on the idea that
everything is made of related parts. The basic idea seems so self-evident that we find it
Challenge 484 ny difficult even to formulate an alternative. Just try!
However, in the case of the whole of nature, the idea that nature is a sum of related
Page 1046 parts is incorrect. It turns out to be a prejudice, and a prejudice so entrenched that it
retarded further developments in physics in the latter decades of the twentieth century. In

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

particular, it does not apply to elementary particles or to space-time. Finding the correct
description for the whole of nature is the biggest challenge of our adventure, as it requires
a complete change in thinking habits. There is a lot of fun ahead.
Jede Aussage über Komplexe läßt sich in eine

“ Aussage über deren Bestandteile und in

diejenigen Sätze zerlegen, welche die Komplexe
vollständig beschreiben.*
Ludwig Wittgenstein, Tractatus, 2.0201

Why stones can be neither smooth nor fractal, nor made of

little hard balls
”
The exploration of temperature yields another interesting result. Researchers first stud-
ied gases, and measured how much energy was needed to heat them by 1 K. The result
is simple: all gases share only a few values, when the number of molecules N is taken
into account. Monoatomic gases (in a container with constant volume) require 3N k~2,
diatomic gases (and those with a linear molecule) 5N k~2, and almost all other gases 3N k,
Page 242 where k = 1.4 ë 10−23 J~K is the Boltzmann constant.
The explanation of this result was soon forthcoming: each thermodynamic degree of
freedom** contributes the energy kT~2 to the total energy, where T is the temperature. So
the number of degrees of freedom in physical bodies is finite. Bodies are not continuous,
nor are they fractals: if they were, their specific thermal energy would be infinite. Matter
Copyright © Christoph Schiller November 1997–May 2006

is indeed made of small basic entities.

All degrees of freedom contribute to the specific thermal energy. At least, this is what
classical physics predicts. Solids, like stones, have 6 thermodynamic degrees of freedom
and should show a specific thermal energy of 3N k. At high temperatures, this is indeed
observed. But measurements of solids at room temperature yield lower values, and the
* Every statement about complexes can be resolved into a statement about their constituents and into the
propositions that describe the complexes completely.
** A thermodynamic degree of freedom is, for each particle in a system, the number of dimensions in which
it can move plus the number of dimensions in which it is kept in a potential. Atoms in a solid have six,
particles in monoatomic gases have only three; particles in diatomic gases or rigid linear molecules have
Ref. 206 five. The number of degrees of freedom of larger molecules depends on their shape.

Dvipsbugw
 temperature 249

match
head Dvipsbugw

F I G U R E 120
The fire pump

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

lower the temperature, the lower the values become. Even gases show values lower than
those just mentioned, when the temperature is sufficiently low. In other words, molecules
and atoms behave differently at low energies: atoms are not immutable little hard balls.
The deviation of these values is one of the first hints of quantum theory.

Curiosities and fun challenges about heat

Even though heat is disordered motion, it follows simple rules. Some of them are surpris-
ing.

**
Compression of air increases its temperature. This is shown directly by the fire pump, a
variation of a bicycle pump, shown in Figure 120. (For a working example, see the website
http://www.tn.tudelft.nl/cdd). A match head at the bottom of an air pump made of trans-
parent material is easily ignited by the compression of the air above it. The temperature
of the air after compression is so high that the match head ignites spontaneously.

**
If heat really is disordered motion of atoms, a big problem appears. When two atoms
collide head-on, in the instant of smallest distance, neither atom has velocity. Where does
the kinetic energy go? Obviously, it is transformed into potential energy. But that implies
Copyright © Christoph Schiller November 1997–May 2006

that atoms can be deformed, that they have internal structure, that they have parts, and
thus that they can in principle be split. In short, if heat is disordered atomic motion, atoms
are not indivisible! In the nineteenth century this argument was put forward in order to
show that heat cannot be atomic motion, but must be some sort of fluid. But since we
know that heat really is kinetic energy, atoms must indeed be divisible, even though their
name means ‘indivisible’. We do not need an expensive experiment to show this.
**
Not only gases, but also most other materials expand when the temperature rises. As
a result, the electrical wires supported by pylons hang much lower in summer than in
Challenge 485 n winter. True?

Dvipsbugw
 250 i galilean motion • 3. the global simplicity of complexity

1 2 3 4

Dvipsbugw
F I G U R E 121 Can you boil water in this paper cup?

**
Ref. 207 The following is a famous Fermi problem. Given that a human corpse cools down in four

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 486 ny hours after death, what is the minimum number of calories needed per day in our food?

**
The energy contained in thermal motion is not negligible. A 1 g bullet travelling at the
speed of sound has a kinetic energy of only 0.01 kcal.
**
3
Challenge 487 n How does a typical, 1500 m hot-air balloon work?

**
If you do not like this text, here is a proposal. You can use the paper to make a cup, as
shown in Figure 121, and boil water in it over an open flame. However, to succeed, you
Challenge 488 n have to be a little careful. Can you find out in what way?
**
Mixing 1 kg of water at 0°C and 1 kg of water at 100°C gives 2 kg of water at 50°C. What
Challenge 489 ny is the result of mixing 1 kg of ice at 0°C and 1 kg of water at 100°C?
**
Ref. 208 The highest recorded air temperature in which a man has survived is 127°C. This was
Copyright © Christoph Schiller November 1997–May 2006

tested in 1775 in London, by the secretary of the Royal Society, Charles Blagden, together
with a few friends, who remained in a room at that temperature for 45 minutes. Interest-
ingly, the raw steak which he had taken in with him was cooked (‘well done’) when he
and his friends left the room. What condition had to be strictly met in order to avoid
Challenge 490 n cooking the people in the same way as the steak?
**
Challenge 491 n Why does water boil at 99.975°C instead of 100°C?

**
Challenge 492 n Can you fill a bottle precisely with 1  10 −30
kg of water?

Dvipsbugw
 temperature 251

invisible pulsed
laser beam
emitting sound

Dvipsbugw
laser
cable
to amplifier

F I G U R E 122 The invisible

loudspeaker

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
One gram of fat, either butter or human fat, contains 38 kJ of chemical energy (or, in
ancient units more familiar to nutritionists, 9 kcal). That is the same value as that of petrol.
Challenge 493 n Why are people and butter less dangerous than petrol?

**
In 1992, the Dutch physicist Martin van der Mark invented a loudspeaker which worked
the heating of air by heating air with a laser beam. He demonstrated that with the right
wavelength and with a suitable modulation of the intensity, a laser beam in air can gen-
erate sound, . The effect at the basis of this device, called the photoacoustic effect, appears
in many materials. The best wavelength for air is in the infrared domain, on one of the
few absorption lines of water vapour. In other words, a properly modulated infrared laser
beam that shines through the air generates sound. The light can be emitted from a small
matchbox-sized semiconductor laser hidden in the ceiling and shining downwards. The
sound is emitted in all directions perpendicular to the beam. Since infrared laser light is
not visible, Martin van der Mark thus invented an invisible loudspeaker! Unfortunately,
the efficiency of present versions is still low, so that the power of the speaker is not yet
Copyright © Christoph Schiller November 1997–May 2006

sufficient for practical applications. Progress in laser technology should change this, so
that in the future we should be able to hear sound that is emitted from the centre of an
otherwise empty room.
**
A famous exam question: How can you measure the height of a building with a barometer,
Challenge 494 n a rope and a ruler? Find at least six different ways.
**
What is the approximate probability that out of one million throws of a coin you get
Challenge 495 ny exactly 500 000 heads and as many tails?

Dvipsbugw
 252 i galilean motion • 3. the global simplicity of complexity

º
You may want to use Stirling’s formula n!  2πn (n~e)n to calculate the result.*

**
Challenge 496 n Does it make sense to talk about the entropy of the universe?
**
Challenge 497 ny Can a helium balloon lift the tank which filled it?
** Dvipsbugw
All friction processes, such as osmosis, diffusion, evaporation, or decay, are slow. They
take a characteristic time. It turns out that any (macroscopic) process with a time-scale is
irreversible. This is no real surprise: we know intuitively that undoing things always takes
more time than doing them. That is again the second principle of thermodynamics.
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

It turns out that storing information is possible with negligible entropy generation. How-
Ref. 209 ever, erasing information requires entropy. This is the main reason why computers, as well
as brains, require energy sources and cooling systems, even if their mechanisms would
otherwise need no energy at all.
**
When mixing hot rum and cold water, how does the increase in entropy due to the mixing
Challenge 498 ny compare with the entropy increase due to the temperature difference?

**
Why aren’t there any small humans, e.g. 10 mm in size, as in many fairy tales? In fact,
Challenge 499 n there are no warm-blooded animals of that size. Why not?
**
Shining a light onto a body and repeatedly switching it on and off produces sound. This
is called the photoacoustic effect, and is due to the thermal expansion of the material.
By changing the frequency of the light, and measuring the intensity of the noise, one
reveals a characteristic photoacoustic spectrum for the material. This method allows us
to detect gas concentrations in air of one part in 109 . It is used, among other methods, to
study the gases emitted by plants. Plants emit methane, alcohol and acetaldehyde in small
Copyright © Christoph Schiller November 1997–May 2006

quantities; the photoacoustic effect can detect these gases and help us to understand the
processes behind their emission.

**
What is the rough probability that all oxygen molecules in the air would move away from
Challenge 500 ny a given city for a few minutes, killing all inhabitants?
**
»
(2n + 1~3)π (n~e)n . Another
º º
* There are many improvements to Stirling’s formula. A simple one is n! 
is 2πn (n~e)n e ~(12n+1) < n! < 2πn (n~e)n e ~(12n) .
1 1

Dvipsbugw
 temperature 253

If you pour a litre of water into the sea, stir thoroughly through all the oceans and then
Challenge 501 ny take out a litre of the mixture, how many of the original atoms will you find?

**
Challenge 502 ny How long would you go on breathing in the room you are in if it were airtight?
**
Challenge 503 ny What happens if you put some ash onto a piece of sugar and set fire to the whole? (Warn-
ing: this is dangerous and not for kids.) Dvipsbugw

**
Entropy calculations are often surprising. For a system of N particles with two states
each, there are Wall = 2 N states. For its most probable configuration, with exactly half the
particles in one state, and the other half in the other state, we have Wmax = N!~((N~2)!)2 .
Now, for a macroscopic system of particles, we might typically have N = 1024 . That gives
Wall Q Wmax ; indeed, the former is 1012 times larger than the latter. On the other hand, we

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 504 ny find that ln Wall and ln Wmax agree for the first 20 digits! Even though the configuration
with exactly half the particles in each state is much more rare than the general case, where
Challenge 505 ny the ratio is allowed to vary, the entropy turns out to be the same. Why?
**
If heat is due to motion of atoms, our built-in senses of heat and cold are simply detectors
Challenge 506 ny of motion. How could they work?
By the way, the senses of smell and taste can also be seen as motion detectors, as they
Challenge 507 ny signal the presence of molecules flying around in air or in liquids. Do you agree?
**
The Moon has an atmosphere, although an extremely thin one, consisting of sodium (Na)
and potassium (K). This atmosphere has been detected up to nine Moon radii from its sur-
face. The atmosphere of the Moon is generated at the surface by the ultraviolet radiation
Challenge 508 n from the Sun. Can you estimate the Moon’s atmospheric density?
**
Does it make sense to add a line in Table 29 for the quantity of physical action? A column?
Challenge 509 ny Why?
Copyright © Christoph Schiller November 1997–May 2006

**
Diffusion provides a length scale. For example, insects take in oxygen through their skin.
As a result, the interiors of their bodies cannot be much more distant from the surface
than about a centimetre. Can you list some other length scales in nature implied by diffu-
Challenge 510 n sion processes?
**
Rising warm air is the reason why many insects are found in tall clouds in the evening.
Many insects, especially that seek out blood in animals, are attracted to warm and humid
air.

Dvipsbugw
 254 i galilean motion • 3. the global simplicity of complexity

cold air

hot air

room temperature air

F I G U R E 123 The Wirbelrohr or Ranque–Hilsch vortex Dvipsbugw

tube

**
Thermometers based on mercury can reach 750°C. How is this possible, given that mer-
Challenge 511 n cury boils at 357°C?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Challenge 512 n What does a burning candle look like in weightless conditions?
**
It is possible to build a power station by building a large chimney, so that air heated by
the Sun flows upwards in it, driving a turbine as it does so. It is also possible to make a
power station by building a long vertical tube, and letting a gas such as ammonia rise into
it which is then liquefied at the top by the low temperatures in the upper atmosphere; as
it falls back down a second tube as a liquid – just like rain – it drives a turbine. Why are
Challenge 513 n such schemes, which are almost completely non-polluting, not used yet?
**
One of the most surprising devices ever invented is the Wirbelrohr or Ranque–Hilsch
vortex tube. By blowing compressed air at room temperature into it at its midpoint, two
flows of air are formed at its ends. One is extremely cold, easily as low as −50°C, and one
extremely hot, up to 200°C. No moving parts and no heating devices are found inside.
Challenge 514 n How does it work?
**
Copyright © Christoph Schiller November 1997–May 2006

It is easy to cook an egg in such a way that the white is hard but the yolk remains liquid.
Challenge 515 n Can you achieve the opposite?

**
Thermoacoustic engines, pumps and refrigerators provide many strange and fascinating
applications of heat. For example, it is possible to use loud sound in closed metal cham-
bers to move heat from a cold place to a hot one. Such devices have few moving parts and
Ref. 210 are being studied in the hope of finding practical applications in the future.

**
Challenge 516 ny Does a closed few-particle system contradict the second principle of thermodynamics?

Dvipsbugw
 self-organization and chaos 255

Figures to be added Dvipsbugw

F I G U R E 124 Examples of self-organization for sand

**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What happens to entropy when gravitation is taken into account? We carefully left grav-
itation out of our discussion. In fact, many problems appear – just try to think about
the issue. For example, Jakob Bekenstein has discovered that matter reaches its highest
Challenge 517 ny possible entropy when it forms a black hole. Can you confirm this?
**
The numerical values (but not the units!) of the Boltzmann constant k = 1.38 ë 10−23 J~K
and the combination h~ce agree in their exponent and in their first three digits, where h is
Challenge 518 ny Planck’s constant and e is the electron charge. Can you dismiss this as mere coincidence?

Self-organization and chaos

To speak of non-linear physics is like calling

Ref. 211
“ zoology the study of non-elephant animals.
Stanislaw Ulam

In our list of global descriptions of motion, the high point is the study of self-organization. ”
Self-organization is the appearance of order. Order is a term that includes shapes, such as
Ref. 212 the complex symmetry of snowflakes; patterns, such as the stripes of zebras; and cycles,
such as the creation of sound when singing. Every example of what we call beauty is a
Challenge 519 n combination of shapes, patterns and cycles. (Do you agree?) Self-organization can thus
Copyright © Christoph Schiller November 1997–May 2006

be called the study of the origin of beauty.

The appearance of order is a general observation across nature. Fluids in particular ex-
hibit many phenomena where order appears and disappears. Examples include the more
or less regular flickering of a burning candle, the flapping of a flag in the wind, the regular
stream of bubbles emerging from small irregularities in the surface of a champagne glass,
and the regular or irregular dripping of a water tap.
The appearance of order is found from the cell differentiation in an embryo inside a
woman’s body; the formation of colour patterns on tigers, tropical fish and butterflies; the
symmetrical arrangements of flower petals; the formation of biological rhythms; and so
on.
All growth processes are self-organization phenomena. Have you ever pondered the

Dvipsbugw
 256 i galilean motion • 3. the global simplicity of complexity

TA B L E 34 Sand patterns in the sea and on land

Pa t t e r n Period Amplitude Origin

sand banks 2 to 10 km 2 to 20 m tides

sand waves 100 to 800 m 5m tides
megaribbles 1m 0.1 m tides
ribbles 5 cm 5 mm waves
singing sand 95 to 105 Hz up to 105 dB wind on sand dunes, ava-
lanches making the dune vi- Dvipsbugw
brate

incredible way in which teeth grow? A practically inorganic material forms shapes in the
upper and the lower rows fitting exactly into each other. How this process is controlled
is still a topic of research. Also the formation, before and after birth, of neural networks

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

in the brain is another process of self-organization. Even the physical processes at the
basis of thinking, involving changing electrical signals, is to be described in terms of self-
organization.
Biological evolution is a special case of growth. Take the evolution of animal shapes.
It turns out that snake tongues are forked because that is the most efficient shape for
Ref. 213 following chemical trails left by prey and other snakes of the same species. (Snakes smell
with help of the tongue.) The fixed numbers of fingers in human hands or of petals of
Page 698 flowers are also consequences of self-organization.
Many problems of self-organization are mechanical problems: for example, the form-
ation of mountain ranges when continents move, the creation of earthquakes, or the cre-
ation of regular cloud arrangements in the sky. It can be fascinating to ponder, during
an otherwise boring flight, the mechanisms behind the formation of the clouds you see
Challenge 520 e from the aeroplane.
Studies into the conditions required for the appearance or disappearance of order have
shown that their description requires only a few common concepts, independently of the
details of the physical system. This is best seen looking at a few examples.
All the richness of self-organization reveals itself in the study of plain sand. Why do
sand dunes have ripples, as does the sand floor at the bottom of the sea? We can also
study how avalanches occur on steep heaps of sand and how sand behaves in hourglasses,
in mixers, or in vibrating containers. The results are often surprising. For example, as
Copyright © Christoph Schiller November 1997–May 2006

Ref. 214 recently as 1996 Paul Umbanhowar and his colleagues found that when a flat container
holding tiny bronze balls (around 0.165 mm in diameter) is shaken up and down in va-
cuum at certain frequencies, the surface of this bronze ‘sand’ forms stable heaps. They are
shown in Figure 125. These heaps, so-called oscillons, also bob up and down. Oscillons
can move and interact with one another.
Oscillons in sand are simple example for a general effect in nature: discrete systems
with nonlinear interactions can exhibit localized excitations. This fascinating topic is just
Ref. 215 beginning to be researched. It might well be that it will yield results relevant to our under-
standing of elementary particles.
Sand shows many other pattern-forming processes. A mixture of sand and sugar, when
poured onto a heap, forms regular layered structures that in cross section look like zebra

Dvipsbugw
 self-organization and chaos 257

n = 21

Dvipsbugw

n = 23

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

time
F I G U R E 125 Oscillons formed by shaken F I G U R E 126 Magic numbers: 21 spheres, when
bronze balls; horizontal size is about 2 cm swirled in a dish, behave differently from
(© Paul Umbanhowar) non-magic numbers, like 23, of spheres (redrawn
from photographs, © Karsten Kötter)

stripes. Horizontally rotating cylinders with binary mixtures inside them separate the
mixture out over time. Or take a container with two compartments separated by a 1 cm
wall. Fill both halves with sand and rapidly shake the whole container with a machine.
Over time, all the sand will spontaneously accumulate in one half of the container. As
another example of self-organization in sand, people have studied the various types of
Ref. 216 sand dunes that ‘sing’ when the wind blows over them. In fact, the behaviour of sand
and dust is proving to be such a beautiful and fascinating topic that the prospect of each
human returning dust does not look so grim after all.
Another simple and beautiful example of self-organization is the effect discovered in
Ref. 217 1999 by Karsten Kötter and his group. They found that the behaviour of a set of spheres
swirled in a dish depends on the number of spheres used. Usually, all the spheres get
continuously mixed up. But for certain ‘magic’ numbers, such as 21, stable ring patterns
emerge, for which the outside spheres remain outside and the inside ones remain inside.
Copyright © Christoph Schiller November 1997–May 2006

The rings, best seen by colouring the spheres, are shown in Figure 126.
These and many other studies of self-organizing systems have changed our understand-
ing of nature in a number of ways. First of all, they have shown that patterns and shapes
are similar to cycles: all are due to motion. Without motion, and thus without history,
Ref. 218 there is no order, neither patterns nor shapes. Every pattern has a history; every pattern
is a result of motion.
Secondly, patterns, shapes and cycles are due to the organized motion of large num-
bers of small constituents. Systems which self-organize are always composite: they are
cooperative structures.
Thirdly, all these systems obey evolution equations which are nonlinear in the config-
uration variables. Linear systems do not self-organize. Many self-organizing systems also

Dvipsbugw
 258 i galilean motion • 3. the global simplicity of complexity

show chaotic motion.

Fourthly, the appearance and disappearance of order depends on the strength of a
driving force, the so-called order parameter. Often, chaotic motion appears when the driv-
ing is increased beyond the value necessary for the appearance of order. An example of
chaotic motion is turbulence, which appears when the order parameter, which is propor-
tional to the speed of the fluid, is increased to high values.
Moreover, all order and all structure appears when two general types of motion com-
pete with each other, namely a ‘driving’, energy-adding process, and a ‘dissipating’, brak-
ing mechanism. Thermodynamics plays a role in all self-organization. Self-organizing Dvipsbugw
systems are always dissipative systems, and are always far from equilibrium. When the
driving and the dissipation are of the same order of magnitude, and when the key beha-
viour of the system is not a linear function of the driving action, order may appear.*
All self-organizing systems at the onset of order appearance can be described by equa-
tions for the pattern amplitude A of the general form

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

= λA − µSAS2 A + κ ∆A + higher orders .
∂A(t, x)
(99)
∂t
Here, the – possibly complex – observable A is the one that appears when order appears,
such as the oscillation amplitude or the pattern amplitude. The first term λA is the driving
term, in which λ is a parameter describing the strength of the driving. The next term is
a typical nonlinearity in A, with µ a parameter that describes its strength, and the third
term κ ∆A = κ(∂ 2 A~∂x 2 + ∂ 2 A~∂y 2 + ∂ 2 A~∂z 2 ) is a typical dissipative (and diffusive)
term.
One can distinguish two main situations. In cases where the dissipative term plays no
role (κ = 0), one finds that when the driving parameter λ increases above zero, a temporal
Challenge 521 ny oscillation appears, i.e. a stable cycle with non-vanishing amplitude. In cases where the
diffusive term does play a role, equation (99) describes how an amplitude for a spatial
oscillation appears when the driving parameter λ becomes positive, as the solution A = 0
Challenge 522 ny then becomes spatially unstable.
In both cases, the onset of order is called a bifurcation, because at this critical value
of the driving parameter λ the situation with amplitude zero, i.e. the homogeneous (or
unordered) state, becomes unstable, and the ordered state becomes stable. In nonlinear
systems, order is stable. This is the main conceptual result of the field. Equation (99) and its
numerous variations allow us to describe many phenomena, ranging from spirals, waves,
Copyright © Christoph Schiller November 1997–May 2006

Ref. 219 hexagonal patterns, and topological defects, to some forms of turbulence. For every phys-
ical system under study, the main task is to distil the observable A and the parameters λ,
µ and κ from the underlying physical processes.
Self-organization is a vast field which is yielding new results almost by the week. To

* To describe the ‘mystery’ of human life, terms like ‘fire’, ‘river’ or ‘tree’ are often used as analogies. These
are all examples of self-organized systems: they have many degrees of freedom, have competing driving
and braking forces, depend critically on their initial conditions, show chaos and irregular behaviour, and
sometimes show cycles and regular behaviour. Humans and human life resemble them in all these respects;
thus there is a solid basis to their use as metaphors. We could even go further and speculate that pure beauty
is pure self-organization. The lack of beauty indeed often results from a disturbed equilibrium between
external braking and external driving.

Dvipsbugw
 self-organization and chaos 259

oscillation, quasiperiodic
fixed point limit cycle motion chaotic motion

configuration variables configuration variables Dvipsbugw

F I G U R E 127 Examples of different types of motion in configuration space

state value
system 1

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

system 2

time

F I G U R E 128 Sensitivity to initial conditions

discover new topics of study, it is often sufficient to keep one’s eye open; most effects are
Challenge 523 ny comprehensible without advanced mathematics. Good hunting!
Most systems that show self-organization also show another type of motion. When the
driving parameter of a self-organizing system is increased to higher and higher values,
order becomes more and more irregular, and in the end one usually finds chaos. For
physicists, c ha o T c motion is the most irregular type of motion.* Chaos can be defined
i

independently of self-organization, namely as that motion of systems for which small

changes in initial conditions evolve into large changes of the motion (exponentially with
time), as shown in Figure 128. More precisely, chaos is irregular motion characterized by a
positive Lyapounov exponent. The weather is such a system, as are dripping water-taps, the
fall of dice, and many other common systems. For example, research on the mechanisms
by which the heart beat is generated has shown that the heart is not an oscillator, but a
chaotic system with irregular cycles. This allows the heart to be continuously ready for
demands for changes in beat rate which arise once the body needs to increase or decrease
Copyright © Christoph Schiller November 1997–May 2006

Ref. 220 its efforts.

Incidentally, can you give a simple argument to show that the so-called butterfly effect
Challenge 524 n does not exist? This ‘effect’ is often cited in newspapers: the claim is that nonlinearities
imply that a small change in initial conditions can lead to large effects; thus a butterfly
wing beat is alleged to be able to induce a tornado. Even though nonlinearities do indeed
lead to growth of disturbances, the butterfly effect has never been observed; it does not

* On the topic of chaos, see the beautiful book by H.-O. Peitgen, H. Jürgens & D. Saupe, Chaos and
Fractals, Springer Verlag, 1992. It includes stunning pictures, the necessary mathematical background, and
some computer programs allowing personal exploration of the topic. ‘Chaos’ is an old word: according to
Greek mythology, the first goddess, Gaia, i.e. the Earth, emerged from the chaos existing at the beginning.
She then gave birth to the other gods, the animals and the first humans.

Dvipsbugw
 260 i galilean motion • 3. the global simplicity of complexity

exist.
There is chaotic motion also in machines: chaos appears in the motion of trains on the
rails, in gear mechanisms, and in fire-fighter’s hoses. The precise study of the motion in
Challenge 525 ny a zippo cigarette lighter will probably also yield an example of chaos. The mathematical
description of chaos – simple for some textbook examples, but extremely involved for
others – remains an important topic of research.
All the steps from disorder to order, quasiperiodicity and finally to chaos, are examples
of self-organization. These types of motion, illustrated in Figure 127, are observed in many
fluid systems. Their study should lead, one day, to a deeper understanding of the mysteries Dvipsbugw
Ref. 221 of turbulence. Despite the fascination of this topic, we will not explore it further, because
it does not lead towards the top of Motion Mountain.
But self-organization is of interest also for a more general reason. It is sometimes said
that our ability to formulate the patterns or rules of nature from observation does not
imply the ability to predict all observations from these rules. According to this view, so-
called ‘emergent’ properties exist, i.e. properties appearing in complex systems as some-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

thing new that cannot be deduced from the properties of their parts and their interactions.
(The ideological backdrop to this view is obvious; it is the last attempt to fight the determ-
inism.) The study of self-organization has definitely settled this debate. The properties of
water molecules do allow us to predict Niagara Falls.* Similarly, the diffusion of signal
molecules do determine the development of a single cell into a full human being: in par-
ticular, cooperative phenomena determine the places where arms and legs are formed;
they ensure the (approximate) right–left symmetry of human bodies, prevent mix-ups of
connections when the cells in the retina are wired to the brain, and explain the fur pat-
terns on zebras and leopards, to cite only a few examples. Similarly, the mechanisms at
the origin of the heart beat and many other cycles have been deciphered.
Self-organization provides general principles which allow us in principle to predict the
behaviour of complex systems of any kind. They are presently being applied to the most
complex system in the known universe: the human brain. The details of how it learns to
coordinate the motion of the body, and how it extracts information from the images in
the eye, are being studied intensely. The ongoing work in this domain is fascinating. If
Challenge 527 ny you plan to become a scientist, consider taking this path.
Such studies provide the final arguments that confirm what J. Offrey de la Mettrie
in 1748 stated and explored in his famous book L’homme machine: humans are complex
machines. Indeed, the lack of understanding of complex systems in the past was due
mainly to the restrictive teaching of the subject of motion, which usually concentrated –
Copyright © Christoph Schiller November 1997–May 2006

as we do in this walk – on examples of motion in simple systems. Even though the subject
of self-organization provides fascinating insights, and will do so for many years to come,

* Already small versions of Niagara Falls, namely dripping water taps, show a large range of cooperative
Ref. 222 phenomena, including the chaotic, i.e. non-periodic, fall of water drops. This happens when the water flow
Challenge 526 ny has the correct value, as you can verify in your own kitchen. Several cooperative fluid phenomena have been
simulated even on the molecular level.

Dvipsbugw
 self-organization and chaos 261

water
pipe Dvipsbugw

pearls

λ finger

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 129 The F I G U R E 130 Water F I G U R E 131 A braiding
wavy surface of pearls water stream (©
icicles Vakhtang Putkaradze)

we now leave it. We continue with our own adventure exploring the basics of motion.*
Ich sage euch: man muss noch Chaos in sich

“ haben, um einen tanzenden Stern gebären zu

können. Ich sage euch: ihr habt noch Chaos in
euch.
Friedrich Nietzsche, Also sprach Zarathustra.

Curiosities and fun challenges about self-organization

Every example of a pattern or of beauty contains a physical challenge:
”
**
All icicles have a wavy surface, with a crest-to-crest distance of about 1 cm, as shown in
Copyright © Christoph Schiller November 1997–May 2006

Figure 129. The distance is determined by the interplay between water flow and surface
Challenge 528 ny cooling. How?

**
When wine is made to swirl in a wine glass, after the motion has calmed down, the wine
flowing down the glass walls forms little arcs. Can you explain in a few words what forms
Challenge 529 ny them?
**

Ref. 223 * An important case of self-organization is humour.

Dvipsbugw
 262 i galilean motion • 3. the global simplicity of complexity

How does the average distance between cars parked along a street change over time, as-
Challenge 530 d suming a constant rate of cars leaving and arriving?

**
When a fine stream of water leaves a water tap, putting a finger in the stream leads to a
Challenge 531 d wavy shape, as shown in Figure 130. Why?
**
When water emerges from a oblong opening, the stream forms a braid pattern, as shown Dvipsbugw
in Figure 131. This effect results from the interplay and competition between inertia and
Ref. 224 surface tension: inertia tends to widen the stream, while surface tension tends to narrow
it. Predicting the distance from one narrow region to the next is still a topic of research.
If the experiment is done in free air, without a plate, one usually observes an additional
effect: there is a chiral braiding at the narrow regions, induced by the asymmetries of the
water flow. You can observe this effect in the toilet! Scientific curiosity knows no limits:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 532 ny are you a right-turner or a left-turner, or both? On every day?

**
Gerhard Müller has discovered a simple but beautiful way to observe self-organization in
solids. His system also provides a model for a famous geological process, the formation of
hexagonal columns in basalt, such as the Devil’s Staircase in Ireland. Similar formations
Ref. 225 are found in many other places of the Earth. Just take some rice flour or corn starch,
mix it with about half the same amount of water, put the mixture into a pan and dry it
Challenge 533 e with a lamp. Hexagonal columns form. The analogy works because the drying of starch
and the cooling of lava are diffusive processes governed by the same equations, because
the boundary conditions are the same, and because both materials respond with a small
reduction in volume.
**
Water flow in pipes can be laminar (smooth) or turbulent (irregular and disordered).
The transition depends on the diameter d of the pipe and the speed v of the water. The
transition usually happens when the so-called Reynolds number – defined as R = vd~η
(η being the kinematic viscosity of the water, around 1 mm2 ~s) – becomes greater than
about 2000. However, careful experiments show that with proper handling, laminar flows
can be produced up to R = 100 000. A linear analysis of the equations of motion of the
Copyright © Christoph Schiller November 1997–May 2006

fluid, the Navier–Stokes equations, even predicts stability of laminar flow for all Reynolds
numbers. This riddle was solved only in the years 2003 and 2004. First, a complex math-
ematical analysis showed that the laminar flow is not always stable, and that the transition
to turbulence in a long pipe occurs with travelling waves. Then, in 2004, careful experi-
ments showed that these travelling waves indeed appear when water is flowing through
Ref. 226 a pipe at large Reynolds numbers.
**
For some beautiful pictures on self-organization in fluids, see the http://serve.me.nus.edu.
sg/limtt website. Among others, it shows that a circular vortex can ‘suck in’ a second one
behind it, and that the process can then repeat.

Dvipsbugw
 from the limitations of physics to the limits of motion 263

**
Also dance is an example of self-organization. Self-organization takes part in the brain.
Like for all complex movements, learning then is often a challenge. Nowadays there are
beautiful books that tell how physics can help you improve your dancing skills and the
Ref. 227 grace of your movements.

4. from the limitations of physics to the limits of

Dvipsbugw
motion
I only know that I know nothing.

“ Socrates, as cited by Plato

Socrates’ saying applies also to Galilean physics, despite its general success in engineering ”
and in the description of everyday life. We will now give a short overview of the limita-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tions of the field.

Research topics in classical dynamics

Even though the science of mechanics is now several hundred years old, research into its
details is still continuing.
— We have already mentioned above the issue of the stability of the solar system. The
long-term future of the planets is unknown. In general, the behaviour of few-body sys-
Ref. 228 tems interacting through gravitation is still a research topic of mathematical physics.
Answering the simple question of how long a given set of bodies gravitating around
each other will stay together is a formidable challenge. The history of this so-called
many-body problem is long and involved. Interesting progress has been achieved, but
the final answer still eludes us.
— Many challenges remain in the fields of self-organization, of nonlinear evolution equa-
tions, and of chaotic motion; and they motivate numerous researchers in mathematics,
physics, chemistry, biology, medicine and the other sciences.
— Perhaps the toughest of all problems in physics is how to describe turbulence. When
the young Werner Heisenberg was asked to continue research on turbulence, he re-
fused – rightly so – saying it was too difficult; he turned to something easier and dis-
covered quantum mechanics instead. Turbulence is such a vast topic, with many of its
Copyright © Christoph Schiller November 1997–May 2006

concepts still not settled, that despite the number and importance of its applications,
only now, at the beginning of the twenty-first century, are its secrets beginning to be
Ref. 229 unravelled. It is thought that the equations of motion describing fluids, the so-called
Navier–Stokes equations, are sufficient to understand turbulence.* But the mathemat-
ics behind them is mind-boggling. There is even a prize of one million dollars offered
by the Clay Mathematics Institute for the completion of certain steps on the way to
solving the equations.

* They are named after Claude Navier (b. 1785 Dijon, d. 1836 Paris), important French engineer and bridge
builder, and Georges Gabriel Stokes (b. 1819 Skreen, d. 1903 Cambridge), important Irish physicist and
mathematician.

Dvipsbugw
 264 i galilean motion • 4. from the limitations of physics

What is contact?
Democritus declared that there is a unique sort

“ of motion: that ensuing from collision.

Simplicius, Commentary on the Physics of
Aristotle, 42, 10

”
Ref. 230

Of the questions unanswered by classical physics, the details of contact and collisions are
among the most pressing. Indeed, we defined mass in terms of velocity changes during
Page 77 collisions. But why do objects change their motion in such instances? Why are collisions Dvipsbugw
between two balls made of chewing gum different from those between two stainless-steel
balls? What happens during those moments of contact?
Contact is related to material properties, which in turn influence motion in a complex
way. The complexity is such that the sciences of material properties developed independ-
ently from the rest of physics for a long time; for example, the techniques of metallurgy
(often called the oldest science of all) of chemistry and of cooking were related to the prop-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

erties of motion only in the twentieth century, after having been independently pursued
for thousands of years. Since material properties determine the essence of contact, we
need knowledge about matter and about materials to understand the notion of mass, and
thus of motion. The second part of our mountain ascent will reveal these connections.

Precision and accuracy

When we started climbing Motion Mountain, we stated that to gain height means to in-
crease the precision of our description of nature. To make even this statement itself more
precise, we distinguish between two terms: precision is the degree of reproducibility; ac-
curacy is the degree of correspondence to the actual situation. Both concepts apply to
measurements,* to statements and to physical concepts.
Appendix B At present, the record number of digits ever measured for a physical quantity is 14.
Why so few? Classical physics doesn’t provide an answer. What is the maximum number
of digits we can expect in measurements; what determines it; and how can we achieve
it? These questions are still open at this point in our ascent; they will be covered in the
second part of it.
On the other hand, statements with false accuracy abound. What should we think
of a car company – Ford – who claim that the drag coefficient c w of a certain model
Challenge 534 n is 0.375? Or of the official claim that the world record in fuel consumption for cars is
2315.473 km~l? Or of the statement that 70.3 % of all citizens share a certain opinion?
Copyright © Christoph Schiller November 1997–May 2006

One lesson we learn from investigations into measurement errors is that we should never
provide more digits for a result than we can put our hand into fire for.
Is it possible to draw or produce a rectangle for which the ratio of lengths is a real num-
Challenge 535 n ber, e.g. of the form 0.131520091514001315211420010914..., whose digits encode a book? (A
simple method would code a space as 00, the letter ‘a’ as 01, ‘b’ as 02, ‘c’ as 03, etc. Even
more interestingly, could the number be printed inside its own book?)
In our walk we aim for precision and accuracy, while avoiding false accuracy. There-
fore, concepts have mainly to be precise, and descriptions have to be accurate. Any in-

* For measurements, both precision and accuracy are best described by their standard deviation, as explained
in Appendix B, on page 1164.

Dvipsbugw
 to the limits of motion 265

accuracy is a proof of lack of understanding. To put it bluntly, ‘inaccurate’ means wrong.

Increasing the accuracy and precision of our description of nature implies leaving behind
us all the mistakes we have made so far. That is our aim in the following.

Can all of nature be described in a book?

Let us have some fun with a paradox related to our adventure. If a perfect physics pub-
lication describing all of nature existed, it must also describe itself, its own production –
including its author – and most important of all, its own contents. Is this possible? Using Dvipsbugw
the concept of information, we can state that such a book should contain all information
contained in the universe, including the information in the book itself. Is this possible?
If nature requires an infinitely long book to be fully described, such a publication ob-
viously cannot exist. In this case, only approximate descriptions of nature are possible.
If nature requires a finite amount of information for its description, then the universe
cannot contain more information than is already contained in the book. This would imply

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that the rest of the universe would not add to the information already contained in the
book. It seems that the entropy of the book and the entropy of the universe must be
similar. This is possible, but seems somewhat unlikely.
We note that the answer to this puzzle also implies the answer to another puzzle:
whether a brain can contain a full description of nature. In other words, the question
is: can humans understand nature? We do believe so. In other words, we seem to believe
something rather unlikely: that the universe does not contain more information than
what our brain could contain or even contains already. However, this conclusion is not
correct. The terms ‘universe’ and ‘information’ are not used correctly in this reasoning,
Page 1049 as you might want to verify. We will solve this puzzle later in our adventure. Until then,
Challenge 536 e do make up your own mind.

Why is measurement possible?

In the description of gravity given so far, the one that everybody learns – or should learn
– at school, acceleration is connected to mass and distance via a = GM~r 2 . That’s all.
But this simplicity is deceiving. In order to check whether this description is correct, we
have to measure lengths and times. However, it is impossible to measure lengths and time
intervals with any clock or any ruler based on the gravitational interaction alone! Try to
Challenge 537 n conceive such an apparatus and you will be inevitably be disappointed. You always need
a non-gravitational method to start and stop the stopwatch. Similarly, when you measure
Copyright © Christoph Schiller November 1997–May 2006

length, e.g. of a table, you have to hold a ruler or some other device near it. The interaction
necessary to line up the ruler and the table cannot be gravitational.
A similar limitation applies even to mass measurements. Try to measure mass using
Challenge 538 n gravitation alone. Any scale or balance needs other – usually mechanical, electromagnetic
or optical – interactions to achieve its function. Can you confirm that the same applies
Challenge 539 n to speed and to angle measurements? In summary, whatever method we use, in order to
measure velocity, length, time, and mass, interactions other than gravity are needed. Our
ability to measure shows that gravity is not all there is.

Dvipsbugw
266 i galilean motion • 4. from the limitations of physics

Is motion unlimited?
Galilean physics does not explain the ability to measure. In fact, it does not even explain
the existence of standards. Why do objects have fixed lengths? Why do clocks work with
regularity? Galilean physics cannot explain these observations.
Galilean physics also makes no clear statements on the universe as a whole. It seems
to suggest that it is infinite. Finitude does not fit with the Galilean description of mo-
tion. Galilean physics is thus limited in its explanations because it disregards the limits
of motion. Dvipsbugw
We also note that the existence of infinite speeds in nature would not allow us to
define time sequences. Clocks would then be impossible. In other words, a description of
nature that allows unlimited speeds is not precise. Precision requires limits. To achieve
the highest possible precision, we need to discover all limits to motion. So far, we have
discovered only one: there is a smallest entropy. We now turn to another, more striking
one: the limit for speed. To understand this limit, we will explore the most rapid motion

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

we know: the motion of light.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
bibliography 267

Biblio graphy

138 Giuseppe Fumagalli, Chi l’ha detto?, Hoepli, 1983. Cited on page 173.
139 The beautiful story of the south-pointing carriage is told in Appendix B of James Foster
& J.D. Nightingale, A Short Course in General Relativity, Springer Verlag, 2nd edition,
1998. Such carriages have existed in China, as told by the great sinologist Joseph Needham,
but their construction is unknown. The carriage described by Foster and Nightingale is the
one reconstructed in 1947 by George Lancaster, a British engineer. Cited on page 174. Dvipsbugw
140 See for example Z. Ghahramani, Building blocks of movement, Nature 407, pp. 682–683,
2000. Researchers in robot control are also interested in such topics. Cited on page 174.
141 G. Gu tierrez, C. Fehr, A. Calzadill a & D. Figueroa, Fluid flow up the wall of
a spinning egg, American Journal of Physics 66, pp. 442–445, 1998. Cited on page 175.
142 A historical account is given in Wolf gang Yourgray & Stanley Mandelstam,
Variational Principles in Dynamics and Quantum Theory, Dover, 1968. Cited on pages 176
and 182.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

143 Max Päsler, Prinzipe der Mechanik, Walter de Gruyter & Co., 1968. Cited on page 180.
144 The relations between possible Lagrangians are explained by Herbert Goldstein, Clas-
sical Mechanics, 2nd edition, Addison-Wesley, 1980. Cited on page 181.
145 C.G. Gray, G. Karl & V.A. Novikov, From Maupertius to Schrödinger. Quantization
of classical variational principles, American Journal of Physics 67, pp. 959–961, 1999. Cited
on page 183.
146 The Hemingway statement is quoted by Marlene Dietrich in Aaron E. Hotchner, Papa
Hemingway, Random House, 1966, in part 1, chapter 1. Cited on page 183.
147 J.A. Moore, An innovation in physics instruction for nonscience majors, American
Journal of Physics 46, pp. 607–612, 1978. Cited on page 183.
148 See e.g. Al an P. B oss, Extrasolar planets, Physics Today 49, pp. 32–38. September 1996.
The most recent information can be found at the ‘Extrasolar Planet Encyclopaedia’ main-
tained at http://www.obspm.fr/planets by Jean Schneider at the Observatoire de Paris. Cited
on page 186.
149 A good review article is by David W. Hughes, Comets and Asteroids, Contemporary
Physics 35, pp. 75–93, 1994. Cited on page 186.
150 G.B. West, J.H. Brown & B.J. Enquist, A general model for the origin of allometric
scaling laws in biology, Science 276, pp. 122–126, 4 April 1997, with a comment on page 34 of
the same issue. The rules governing branching properties of blood vessels, of lymph systems
and of vessel systems in plants are explained. For more about plants, see also the paper G.B.
Copyright © Christoph Schiller November 1997–May 2006

West, J.H. Brown & B.J. Enquist, A general model for the structure and allometry of
plant vascular systems, Nature 400, pp. 664–667, 1999. Cited on page 187.
151 J.R. Banavar, A. Martin & A. R inald o, Size and form in efficient transportation
networks, Nature 399, pp. 130–132, 1999. Cited on page 188.
152 John Mansley Robinson, An Introduction to Early Greek Philosophy, Houghton Muffin
1968, chapter 5. Cited on page 189.
153 See e.g. B. B ower, A child’s theory of mind, Science News 144, pp. 40–41. Cited on page
191.
154 The most beautiful book on this topic is the text by Branko Grünbaum & G.C. Shep-
hard, Tilings and Patterns, W.H. Freeman and Company, New York, 1987. It has been trans-
lated into several languages and republished several times. Cited on page 192.

Dvipsbugw
268 i galilean motion

155 U. Niederer, The maximal kinematical invariance group of the free Schrödinger equation,
Helvetica Physica Acta 45, pp. 802–810, 1972. See also the introduction by O. Jahn & V.V.
Sreedhar, The maximal invariance group of Newton’s equations for a free point particle,
http://www.arxiv.org/abs/math-ph/0102011. Cited on page 197.
156 The story is told in the interesting biography of Einstein by A. Pais, ‘Subtle is the Lord...’ –
The Science and the Life of Albert Einstein, Oxford University Press, 1982. Cited on page 198.
157 W. Zürn & R. Widmer-Schnidrig, Globale Eigenschwingungen der Erde, Physik
Journal 1, pp. 49–55, 2002. Cited on page 204.
Dvipsbugw
158 N. Gau thier, What happens to energy and momentum when two oppositely-moving
wave pulses overlap?, American Journal of Physics 71, pp. 787–790, 2003. Cited on page 207.
159 An informative and modern summary of present research about the ear and the details of
its function is http://www.physicsweb.org/article/world/15/5/8. Cited on page 209.
160 A.L. Hod gkin & A.F. Huxley, A quantitative description of membrane current and
its application to conduction and excitation in nerve, Journal of Physiology 117, pp. 500–
544, 1952. This famous paper of theoretical biology earned the authors the Nobel Prize in

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Medicine in 1963. Cited on page 212.
161 T. Filippov, The Versatile Soliton, Springer Verlag, 2000. See also J.S. Russel, Report of
the Fourteenth Meeting of the British Association for the Advancement of Science, Murray,
London, 1844, pp. 311–390. Cited on pages 212 and 214.
162 N.J. Z abusky & M.D. Kruskal, Interaction of solitons in a collisionless plasma and the
recurrence of initial states, Physical Review Letters 15, pp. 240–243, 1965. Cited on page 213.
163 O. Muskens, De kortste knal ter wereld, Nederlands tijdschrift voor natuurkunde pp. 70–
73, 2005. Cited on page 214.
164 E. Heller, Freak waves: just bad luck, or avoidable?, Europhysics News pp. 159–161, Septem-
ber/October 2005, downloadable at www.europhysicsnews.org. Cited on page 215.
165 For more about the ocean sound channel, see the novel by Tom Cl ancy, The Hunt
for Red October. See also the physics script by R.A. Muller, Government secrets
of the oceans, atmosphere, and UFOs, http://web.archive.org/web/*/http://muller.lbl.gov/
teaching/Physics10/chapters/9-SecretsofUFOs.html 2001. Cited on page 216.
166 B. Wilson, R.S. Batty & L.M. Dill, Pacific and Atlantic herring produce burst pulse
sounds, Biology Letters 271, number S3, 7 February 2004. Cited on page 216.
167 See for example the article by G. Fritsch, Infraschall, Physik in unserer Zeit 13, pp. 104–
110, 1982. Cited on page 217.
168 Wavelet transformations were developed by the French mathematicians Alex Grossmann,
Copyright © Christoph Schiller November 1997–May 2006

Jean Morlet and Thierry Paul. The basic paper is A. Grossmann, J. Morlet & T. Paul,
Integral transforms associated to square integrable representations, Journal of Mathematical
Physics 26, pp. 2473–2479, 1985. For a modern introduction, see Stéphane Mall at, A
Wavelet Tour of Signal Processing, Academic Press, 1999. Cited on page 218.
169 Jay Ingram, The Velocity of Honey, Viking, 2o03. Cited on page 218.
170 M. Ausloos, Proceedings of the Royal Society in London A 400, pp. 331–350, 1985. Cited
on page 219.
171 T.A. McMahob & J. Tyler B onner, Form und Leben – Konstruktion vom Reißbrett
der Natur, Spektrum Verlag, 1985. Cited on page 221.
172 G.W. Koch, S.C. Sillett, G.M. Jennings & S.D. Davis, The limits to tree height,
Nature 428, pp. 851–854, 2004. Cited on page 221.

Dvipsbugw
bibliography 269

173 A simple article explaining the tallness of trees is A. Mineyev, Trees worthy of Paul
Bunyan, Quantum pp. 4–10, January–February 1994. (Paul Bunyan is a mythical giant lum-
berjack who is the hero of the early frontier pioneers in the United States.) Note that the
transport of liquids in trees sets no limits on their height, since water is pumped up along
tree stems (except in spring, when it is pumped up from the roots) by evaporation from the
leaves. This works almost without limits because water columns, when nucleation is care-
fully avoided, can be put under tensile stresses of over 100 bar, corresponding to 1000 m.
See also P. Nobel, Plant Physiology, Academic Press, 2nd Edition, 1999. Cited on page
222. Dvipsbugw
174 Such information can be taken from the excellent overview article by M.F. Ashby, On the
engineering properties of materials, Acta Metallurgica 37, pp. 1273–1293, 1989. The article
explains the various general criteria which determine the selection of materials, and gives
numerous tables to guide the selection. Cited on page 222.
175 For a photograph of a single Barium atom – named Astrid – see Hans Dehmelt, Experi-
ments with an isolated subatomic particle at rest, Reviews of Modern Physics 62, pp. 525–530,
1990. For an earlier photograph of a Barium ion, see W. Neuhauser, M. Hohenstatt,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

P.E. Toschek & H. Dehmelt, Localized visible Ba + mono-ion oscillator, Physical Re-
view A 22, pp. 1137–1140, 1980. Cited on page 224.
176 Holograms of atoms were first produced by Hans-Werner Fink & al., Atomic resolu-
tion in lens-less low-energy electron holography, Physical Review Letters 67, pp. 1543–1546,
1991. Cited on page 224.
177 A single–atom laser was built in 1994 by K. An, J.J. Childs, R.R. Dasari & M.S. Feld,
Microlaser: a laser with one atom in an optical resonator, Physical Review Letters 73, p. 3375,
1994. Cited on page 224.
178 See for example C. Schiller, A.A. Koomans, T.L. van Rooy, C. Schönenberger
& H.B. Elswijk, Decapitation of tungsten field emitter tips during sputter sharpening,
Surface Science Letters 339, pp. L925–L930, 1996. Cited on page 224.
179 U. Weierstall & J.C.H. Spence, An STM with time-of-flight analyzer for atomic spe-
cies identification, MSA 2000, Philadelphia, Microscopy and Microanalysis 6, Supplement 2,
p. 718, 2000. Cited on page 224.
180 Lydéric B ocquet, The physics of stone skipping, American Journal of Physics 17, pp. 150–
155, 2003. The present recod holder is Kurt Steiner, with 40 skips. See http://pastoneskipping.
com/steiner.htm and http://www.stoneskipping.com. The site http://www.yeeha.net/nassa/
guin/g2.html is by the a previous world record holder, Jerdome Coleman–McGhee. Cited
on page 227.
181 S.F. Kistler & L.E. Scriven, The teapot effect: sheetforming flows with deflection, wet-
Copyright © Christoph Schiller November 1997–May 2006

ting, and hysteresis, Journal of Fluid Mechanics 263, pp. 19–62, 1994. Cited on page 229.
182 E. Holl ander, Over trechters en zo ..., Nederlands tijdschrift voor natuurkunde 68, p. 303,
2002. Cited on page 230.
183 P. Krehl, S. Engemann & D. Schwenkel, The puzzle of whip cracking – uncovered
by a correlation of whip-tip kinematics with shock wave emission, Shock Waves 8, pp. 1–9,
1998. The authors used high-speed cameras to study the motion of the whip. A new aspect
has been added by A. Goriely & T. McMillen, Shape of a cracking whip, Physical
Review Letters 88, p. 244301, 2002. This article focuses on the tapered shape of the whip.
However, the neglection of the tuft – a piece at the end of the whip which is required to
make it crack – in the latter paper shows that there is more to be discovered still. Cited on
page 231.

Dvipsbugw
270 i galilean motion

184 S. Dorbolo, H. Caps & N. Vandewalle, Fluid instabilities in the birth and death of
antibubbles, New Journal of Physics 5, p. 161, 2003. Cited on page 232.
185 Z. Sheng & K. Yamafuji, Realization of a Human Riding a Unicycle by a Robot, Proceed-
ings of the 1995 IEEE International Conference on Robotics and Automation, Vol. 2, pp. 1319
– 1326, 1995. Cited on page 233.
186 On human uncicycling, see Jack Wiley, The Complete Book of Unicycling, Lodi, 1984, and
Sebastian Hoeher, Einradfahren und die Physik, Reinbeck, 1991. Cited on page 233.
187 F. Herrmann, Mengenartige Größen im Physikunterricht, Physikalische Blätter 54,
Dvipsbugw
pp. 830–832, September 1998. See also his lecture notes on general introductory physics
on the website http://www-tfp.physik.uni-karlsruhe.de/~didaktik. Cited on pages 234
and 237.
188 Thermostatics is difficult to learn also because it was not discovered in a systematic way.
See C. Truesdell, The Tragicomical History of Thermodynamics 1822–1854, Springer Ver-
lag, 1980. An excellent advanced textbook on thermostatics and thermodynamics is Linda
R eichl, A Modern Course in Statistical Physics, Wiley, 2nd edition, 1998. Cited on page

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

234.
189 Gas expansion was the main method used for the definition of the official temperature scale.
Only in 1990 were other methods introduced officially, such as total radiation thermometry
(in the range 140 K to 373 K), noise thermometry (2 K to 4 K and 900 K to 1235 K), acoustical
thermometry (around 303 K), magnetic thermometry (0.5 K to 2.6 K) and optical radiation
thermometry (above 730 K). Radiation thermometry is still the central method in the range
from about 3 K to about 1000 K. This is explained in detail in R.L. Rusby, R.P. Hudson,
M. Durieux, J.F. Schooley, P.P.M. Steur & C.A. Swenson, The basis of the ITS-
90, Metrologia 28, pp. 9–18, 1991. On the water boiling point see also page 1172. Cited on
page 234.
190 See for example the captivating text by Gino Segrè, A Matter of Degrees: What Temperat-
ure Reveals About the Past and Future of Our Species, Planet and Universe, Viking, New York,
2002. Cited on page 235.
191 B. Polster, What is the best way to lace your shoes?, Nature 420, p. 476, 5 December 2002.
Cited on page 238.
192 See for example the article by H. Preston-Thomas, The international temperature scale
of 1990 (ITS-90), Metrologia 27, pp. 3–10, 1990, and the errata H. Preston-Thomas, The
international temperature scale of 1990 (ITS-90), Metrologia 27, p. 107, 1990, Cited on page
242.
193 For an overview, see Christian Enss & Siegfried Hunklinger, Low-Temperature
Physics, Springer, 2005. Cited on page 242.
Copyright © Christoph Schiller November 1997–May 2006

194 The famous paper on Brownian motion which contributed so much to Einstein’s fame is A.
Einstein, Über die von der molekularkinetischen Theorie der Wärme geforderte Bewe-
gung von in ruhenden Flüssigkeiten suspendierten Teilchen, Annalen der Physik 17, pp. 549–
560, 1905. In the following years, Einstein wrote a series of further papers elaborating on this
topic. For example, he published his 1905 Ph.D. thesis as A. Einstein, Eine neue Bestim-
mung der Moleküldimensionen, Annalen der Physik 19, pp. 289–306, 1906, and he corrected
a small mistake in A. Einstein, Berichtigung zu meiner Arbeit: ‘Eine neue Bestimmung
der Moleküldimensionen’, Annalen der Physik 34, pp. 591–592, 1911, where, using new data,
he found the value 6.6 ë 1023 for Avogadro’s number. Cited on page 242.
195 The first tests of the prediction were performed by J. Perrin, Comptes Rendus de l’Académie
des Sciences 147, pp. 475–476, and pp. 530–532, 1908. He masterfully sums up the whole

Dvipsbugw
bibliography 271

discussion in Jean Perrin, Les atomes, Librarie Félix Alcan, Paris, 1913. Cited on page
243.
196 Pierre Gaspard & al., Experimental evidence for microscopic chaos, Nature 394, p. 865,
27 August 1998. Cited on page 243.
197 These points are made clearly and forcibly, as is his style, by N.G. van Kampen, Entropie,
Nederlands tijdschrift voor natuurkunde 62, pp. 395–396, 3 December 1996. Cited on page
245.
198 This is a disappointing result of all efforts so far, as Grégoire Nicolis always stresses in his
university courses. Seth Lloyd has compiled a list of 31 proposed definitions of complexity, Dvipsbugw
containing among others, fractal dimension, grammatical complexity, computational com-
plexity, thermodynamic depth. See, for example, a short summary in Scientific American
p. 77, June 1995. Cited on page 245.
199 Minimal entropy is discussed by L. Szil ard, Über die Entropieverminderung in einem
thermodynamischen System bei Eingriffen intelligenter Wesen, Zeitschrift für Physik 53,
pp. 840–856, 1929. This classic paper can also be found in English translation in his collected

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

works. Cited on page 245.
200 G. Cohen-Tannoudji, Les constantes universelles, Pluriel, Hachette, 1998. See also L.
Brillouin, Science and Information Theory, Academic Press, 1962. Cited on pages 245
and 246.
201 See for example A.E. Shalyt-Margolin & A.Ya. Tregubovich, Generalized uncer-
tainty relation in thermodynamics, http://www.arxiv.org/abs/gr-qc/0307018, or J. Uffink
& J. van Lith-van Dis, Thermodynamic uncertainty relations, Foundations of Physics
29, p. 655, 1999. Cited on page 246.
202 H.W. Zimmermann, Particle entropies and entropy quanta IV: the ideal gas, the second
law of thermodynamics, and the P-t uncertainty relation, Zeitschrift für physikalische Chemie
217, pp. 55–78, 2003, and H.W. Zimmermann, Particle entropies and entropy quanta V:
the P-t uncertainty relation, Zeitschrift für physikalische Chemie 217, pp. 1097–1108, 2003.
Cited on page 246.
203 B. L avenda, Statistical Physics: A Probabilistic Approach, Wiley-Interscience, New York,
1991. Cited on page 246.
204 The quote is found in the introduction by George Wald to the text by L awrence J.
Henderson, The Fitness of the Environment, Macmillan, New York, 1913, reprinted 1958.
Cited on page 247.
205 A fascinating introduction to chemistry is the text by John Emsley, Molecules at an Ex-
hibition, Oxford University Press, 1998. Cited on page 248.
206 An excellent introduction into the physics of heat is the book by Linda R eichl, A Modern
Copyright © Christoph Schiller November 1997–May 2006

Course in Statistical Physics, Wiley, 2nd edition, 1998. Cited on page 248.
207 See V.L. Telegdi, Enrico Fermi in America, Physics Today 55, pp. 38–43, June 2002. Cited
on page 250.
208 K. Schmidt-Nielsen, Desert Animals: Physiological Problems of Heat and Water, Oxford
University Press, 1964. Cited on page 250.
209 Why entropy is created when information is erased, but not when it is acquired, is explained
in Charles H. Bennett & Rolf L andauer, Fundamental Limits of Computation,
Scientific American 253:1, pp. 48–56, 1985. The conclusion: we should pay to throw the news-
paper away, not to buy it. Cited on page 252.
210 See, for example, G. Swift, Thermoacoustic engines and refrigerators, Physics Today 48,
pp. 22–28, July 1995. Cited on page 254.

Dvipsbugw
272 i galilean motion

211 Quoted in D. Campbell, J. Cru tchfield, J. Farmer & E. Jen, Experimental math-
ematics: the role of computation in nonlinear science, Communications of the Association of
Computing Machinery 28, pp. 374–384, 1985. Cited on page 255.
212 For more about the shapes of snowflakes, see the famous book by W.A. Bentlye & W.J.
Humphreys, Snow Crystals, Dover Publications, New York, 1962. This second printing of
the original from 1931 shows a large part of the result of Bentley’s lifelong passion, namely
several thousand photographs of snowflakes. Cited on page 255.
213 K. Schwenk, Why snakes have forked tongues, Science 263, pp. 1573–1577, 1994. Cited on
page 256. Dvipsbugw
214 P.B. Umbanhowar, F. Melo & H.L. Swinney, Localized excitations in a vertically
vibrated granular layer, Nature 382, pp. 793–796, 29 August 1996. Cited on page 256.
215 D.K. Campbell, S. Fl ach & Y.S. Kivshar, Localizing energy through nonlinearity
and discreteness, Physics Today 57, pp. 43–49, January 2004. Cited on page 256.
216 B. Andreotti, The song of dunes as a wave-particle mode locking, Physical Review Letters
92, p. 238001, 2004. Cited on page 257.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

217 K. Kötter, E. Goles & M. Markus, Shell structures with ‘magic numbers’ of spheres
in a swirled disk, Physical Review E 60, pp. 7182–7185, 1999. Cited on page 257.
218 A good introduction is the text by Daniel Walgraef, Spatiotemporal Pattern Formation,
With Examples in Physics, Chemistry and Materials Science, Springer 1996. Cited on page
257.
219 For an overview, see the Ph.D. thesis by Joceline Lega, Défauts topologiques associés à la
brisure de l’invariance de translation dans le temps, Université de Nice, 1989. Cited on page
258.
220 An idea of the fascinating mechanisms at the basis of the heart beat is given by A. Babloy-
antz & A. Destexhe, Is the normal heart a periodic oscillator?, Biological Cybernetics
58, pp. 203–211, 1989. Cited on page 259.
221 For a short, modern overview of turbulence, see L.P. Kadanoff, A model of turbulence,
Physics Today 48, pp. 11–13, September 1995. Cited on page 260.
222 For a clear introduction, see T. Schmidt & M. Mahrl, A simple mathematical model of
a dripping tap, European Journal of Physics 18, pp. 377–383, 1997. Cited on page 260.
223 An overview of science humour can be found in the famous anthology compiled by R.L.
Weber, edited by E. Mend oza, A Random Walk in Science, Institute of Physics, 1973. It
is also available in several expanded translations. Cited on page 261.
224 K. Mertens, V. Pu tkaradze & P. Vorobieff, Braiding patterns on an inclined plane,
Copyright © Christoph Schiller November 1997–May 2006

Nature 430, p. 165, 2004. Cited on page 262.

225 G. Müller, Starch columns: analog model for basalt columns, Journal of Geophysical Re-
search 103, pp. 15239–15253, 1998. Cited on page 262.
226 B. Hof, C.W.H. van Doorne, J. Westerweel, F.T.M. Nieuwstadt, H. Wedin,
R. Kerswell, F. Waleffe, H. Faisst & B. Eckhardt, Experimental observation of
nonlinear traveling waves in turbulent pipe flow, Science 305, pp. 1594–1598, 2004. Cited
on page 262.
227 A fascinating book on the topic is Kenneth L aws & Martha Swope, Physics and the
Art of Dance: Understanding Movement, Oxford University Press 2002. Cited on page 263.
228 J.J. Lissauer, Chaotic motion in the solar system, Reviews of Modern Physics 71, pp. 835–
845, 1999. Cited on page 263.

Dvipsbugw
bibliography 273

229 The present state of our understanding of turbulence is described in ... Cited on page 263.
230 See Jean-Paul Dumont, Les écoles présocratiques, Folio Essais, Gallimard, 1991, p. 426.
Cited on page 264.

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 C h a pter II

SPEC IAL R EL ATI V I T Y

Dvipsbugw

There are limitations on motion that are missed by the Galilean description. The first
limitation we discover is the existence of a maximal speed in nature. The maximum speed
implies many fascinating results: it leads to observer-varying time and length intervals,
to an intimate relation between mass and energy, and to the existence of event horizons.
We explore them now.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

5. maximum speed, observers at rest, and motion of
light
Fama nihil est celerius.*

“ ”
L ight is indispensable for a precise description of motion. To check whether a
ine or a path of motion is straight, we must look along it. In other words, we use
ight to define straightness. How do we decide whether a plane is flat? We look across
it,** again using light. How do we measure length to high precision? With light. How do
we measure time to high precision? With light: once it was light from the Sun that was
Page 1154 used; nowadays it is light from caesium atoms.
In other words, light is important because it is the standard for undisturbed motion.
Physics would have evolved much more rapidly if, at some earlier time, light propagation
had been recognized as the ideal example of motion.
But is light really a phenomenon of motion? This was already known in ancient Greece,
from a simple daily phenomenon, the shadow. Shadows prove that light is a moving en-
tity, emanating from the light source, and moving in straight lines.*** The obvious con-
clusion that light takes a certain amount of time to travel from the source to the surface
Copyright © Christoph Schiller November 1997–May 2006

* ‘Nothing is faster than rumour.’ This common sentence is a simplified version of Virgil’s phrase: fama,
malum qua non aliud velocius ullum. ‘Rumour, the evil faster than all.’ From the Aeneid, book IV, verses 173
and 174.
** Note that looking along the plane from all sides is not sufficient for this: a surface that a light beam touches
right along its length in all directions does not need to be flat. Can you give an example? One needs other
Challenge 540 n methods to check flatness with light. Can you specify one?
*** Whenever a source produces shadows, the emitted entities are called rays or radiation. Apart from light,
other examples of radiation discovered through shadows were infrared rays and ultraviolet rays, which em-
anate from most light sources together with visible light, and cathode rays, which were found to be to the
motion of a new particle, the electron. Shadows also led to the discovery of X-rays, which again turned out
to be a version of light, with high frequency. Channel rays were also discovered via their shadows; they turn
out to be travelling ionized atoms. The three types of radioactivity, namely α-rays (helium nuclei), β-rays

Dvipsbugw
 276 ii special rel ativity • 5. speed, rest and light

Jupiter and Io
(second measurement)

Earth (second
measurement)

Dvipsbugw
Sun Earth (first Jupiter and Io
measurement) (first measurement)

F I G U R E 132 Rømer’s method of measuring the speed of light

showing the shadow had already been reached by the Greek thinker Empedocles (c. 490

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 231
to c. 430 bce ).
We can confirm this result with a different, equally simple, but subtle argument. Speed
can be measured. Therefore the perfect speed, which is used as the implicit measurement
standard, must have a finite value. An infinite velocity standard would not allow meas-
Challenge 541 n urements at all. In nature, the lightest entities move with the highest speed. Light, which
is indeed light, is an obvious candidate for motion with perfect but finite speed. We will
confirm this in a minute.
A finite speed of light means that whatever we see is a message from the past. When
we see the stars, the Sun or a loved one, we always see an image of the past. In a sense,
nature prevents us from enjoying the present – we must therefore learn to enjoy the past.
The speed of light is high; therefore it was not measured until 1676, even though many,
including Galileo, had tried to do so earlier. The first measurement method was worked
out by the Danish astronomer Ole Rømer* when he was studying the orbits of Io and
the other moons of Jupiter. He obtained an incorrect value for the speed of light because
he used the wrong value for their distance from Earth. However, this was quickly correc-
ted by his peers, including Newton himself. You might try to deduce his method from
Challenge 542 n Figure 132. Since that time it has been known that light takes a bit more than 8 minutes
to travel from the Sun to the Earth. This was confirmed in a beautiful way fifty years later,
Page 101 in 1726, by the astronomer James Bradley. Being English, Bradley thought of the ‘rain
method’ to measure the speed of light.
Copyright © Christoph Schiller November 1997–May 2006

Ref. 232
How can we measure the speed of falling rain? We walk rapidly with an umbrella,
measure the angle α at which the rain appears to fall, and then measure our own velocity

(again electrons), and γ-rays (high-energy X-rays) also produce shadows. All these discoveries were made
between 1890 and 1910: those were the ‘ray days’ of physics.
* Ole (Olaf) Rømer (1644 Aarhus – 1710 Copenhagen), Danish astronomer. He was the teacher of the
Dauphin in Paris, at the time of Louis XIV. The idea of measuring the speed of light in this way was due
to the Italian astronomer Givanni Cassini, whose assistant Rømer had been. Rømer continued his measure-
ments until 1681, when Rømer had to leave France, like all protestants (such as Christiaan Huygens), so that
his work was interrupted. Back in Denmark, a fire destroyed all his measurement notes. As a result, he was
not able to continue improving the precision of his method. Later he became an important administrator
and reformer of the Danish state.

Dvipsbugw
 maximum speed, observers at rest, and motion of light 277

rain's perspective light's perspective

rain light

c
c
earth v
v
Sun

Dvipsbugw

human perspective human perspective

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

α v
Sun
v

F I G U R E 133 The rain method of measuring the speed of light

v. As shown in Figure 133, the speed c of the rain is then given by

c = v~ tan α . (100)

The same measurement can be made for light; we just need to measure the angle at which
the light from a star above Earth’s orbit arrives at the Earth. Because the Earth is moving
relative to the Sun and thus to the star, the angle is not a right one. This effect is called
the aberration of light; the angle is found most easily by comparing measurements made
six months apart. The value of the angle is 20.5 ′′ ; nowadays it can be measured with
a precision of five decimal digits. Given that the speed of the Earth around the Sun is
v = 2πR~T = 29.7 km~s, the speed of light must therefore be c = 3.00 ë 108 m~s.* This is

* Umbrellas were not common in Britain in 1726; they became fashionable later, after being introduced from
Copyright © Christoph Schiller November 1997–May 2006

China. The umbrella part of the story is made up. In reality, Bradley had his idea while sailing on the Thames,
when he noted that on a moving ship the apparent wind has a different direction from that on land. He had
observed 50 stars for many years, notably Gamma Draconis, and during that time he had been puzzled by
the sign of the aberration, which was opposite to the effect he was looking for, namely the star parallax. Both
the parallax and the aberration for a star above the ecliptic make them describe a small ellipse in the course
Challenge 543 n of an Earth year, though with different rotation senses. Can you see why?
By the way, it follows from special relativity that the formula (100) is wrong, and that the correct formula
Challenge 544 n is c = v~ sin α; can you see why?
To determine the speed of the Earth, we first have to determine its distance from the Sun. The simplest
method is the one by the Greek thinker Aristarchos of Samos (c. 310 to c. 230 bce ). We measure the angle
between the Moon and the Sun at the moment when the Moon is precisely half full. The cosine of that angle
gives the ratio between the distance to the Moon (determined, for example, by the methods of page 117)
Challenge 545 n and the distance to the Sun. The explanation is left as a puzzle for the reader.

Dvipsbugw
 278 ii special rel ativity • 5. speed, rest and light

half-silvered
mirror

mirror light
source

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 134 Fizeau’s set-up to measure the speed of light (© AG Didaktik
und Geschichte der Physik, Universität Oldenburg)

an astonishing value, especially when compared with the highest speed ever achieved by
a man-made object, namely the Voyager satellites, which travel at 52 Mm~h = 14 km~s,
with the growth of children, about 3 nm~s, or with the growth of stalagmites in caves,
about 0.3 pm~s. We begin to realize why measurement of the speed of light is a science
in its own right.
The first precise measurement of the speed of light was made in 1849 by the French
physicist Hippolyte Fizeau (1819–1896). His value was only 5 % greater than the modern
one. He sent a beam of light towards a distant mirror and measured the time the light
took to come back. How did Fizeau measure the time without any electric device? In fact,
Page 57 he used the same ideas that are used to measure bullet speeds; part of the answer is given
Challenge 547 n in Figure 134. (How far away does the mirror have to be?) A modern reconstruction of
Ref. 235 his experiment by Jan Frercks has achieved a precision of 2 %. Today, the experiment is
much simpler; in the chapter on electrodynamics we will discover how to measure the
Page 560 speed of light using two standard UNIX or Linux computers connected by a cable.
The speed of light is so high that it is even difficult to prove that it is finite. Perhaps the
most beautiful way to prove this is to photograph a light pulse flying across one’s field of
view, in the same way as one can photograph a car driving by or a bullet flying through
Copyright © Christoph Schiller November 1997–May 2006

The angle in question is almost a right angle (which would yield an infinite distance), and good instru-
Ref. 233 ments are needed to measure it with precision, as Hipparchos noted in an extensive discussion of the prob-
lem around 130 bce. Precise measurement of the angle became possible only in the late seventeenth century,
when it was found to be 89.86°, giving a distance ratio of about 400. Today, thanks to radar measurements
Page 1167 of planets, the distance to the Sun is known with the incredible precision of 30 metres. Moon distance vari-
Challenge 546 n ations can even be measured to the nearest centimetre; can you guess how this is achieved?
Ref. 234 Aristarchos also determined the radius of the Sun and of the Moon as multiples of those of the Earth.
Aristarchos was a remarkable thinker: he was the first to propose the heliocentric system, and perhaps the
first to propose that stars were other, faraway suns. For these ideas, several of his contemporaries proposed
that he should be condemned to death for impiety. When the Polish monk and astronomer Nicolaus Co-
pernicus (1473–1543) again proposed the heliocentric system two thousand years later, he did not mention
Aristarchus, even though he got the idea from him.

Dvipsbugw
 maximum speed, observers at rest, and motion of light 279

red
shutter
switch
beam

path of light pulse

Dvipsbugw
10 mm

F I G U R E 135 A photograph of a light pulse moving from right to left through a bottle
with milky water, marked in millimetres (© Tom Mattick)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 136 A consequence of the finiteness of the
speed of light

Ref. 236 the air. Figure 135 shows the first such photograph, produced in 1971 with a standard
off-the-shelf reflex camera, a very fast shutter invented by the photographers, and, most
noteworthy, not a single piece of electronic equipment. (How fast does such a shutter have
Challenge 548 n to be? How would you build such a shutter? And how would you make sure it opened at
the right instant?)
A finite speed of light also implies that a rapidly rotating light beam behaves as shown
Copyright © Christoph Schiller November 1997–May 2006

as in Figure 136. In everyday life, the high speed of light and the slow rotation of light-
houses make the effect barely noticeable.
In short, light moves extremely rapidly. It is much faster than lightning, as you might
Challenge 549 n like to check yourself. A century of increasingly precise measurements of the speed have
culminated in the modern value

c = 299 792 458 m~s. (101)

In fact, this value has now been fixed exactly, by definition, and the metre has been defined
in terms of c. Table 35 gives a summary of what is known today about the motion of light.
Two surprising properties were discovered in the late nineteenth century. They form the

Dvipsbugw
 280 ii special rel ativity • 5. speed, rest and light

TA B L E 35 Properties of the motion of light

O b s e r va t i o n s a b o u t l i g h t

Light can move through vacuum.

Light transports energy.
Light has momentum: it can hit bodies.
Light has angular momentum: it can rotate bodies.
Light moves across other light undisturbed. Dvipsbugw
Light in vacuum always moves faster than any material body does.
The speed of light, its true signal speed, is the forerunner speed. Page 579
In vacuum its value is 299 792 458 m~s.
The proper speed of light is infinite. Page 297
Shadows can move without any speed limit.
Light moves in a straight line when far from matter.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

High-intensity light is a wave.
Light beams are approximations when the wavelength is neglected.
In matter, both the forerunner speed and the energy speed of light are lower than in vacuum.
In matter, the group velocity of light pulses can be zero, positive, negative or infinite.

Ref. 237 basis of special relativity.

Can one play tennis using a laser pulse as the ball and mirrors
as rackets?
Et nihil est celerius annis.*

“ Ovid, Metamorphoses.

We all know that in order to throw a stone as far as possible, we run as we throw it; we ”
know instinctively that in that case the stone’s speed with respect to the ground is higher.
However, to the initial astonishment of everybody, experiments show that light emitted
from a moving lamp has the same speed as light emitted from a resting one. Light (in
vacuum) is never faster than light; all light beams have the same speed. Many specially
Ref. 238 designed experiments have confirmed this result to high precision. The speed of light can
be measured with a precision of better than 1 m~s; but even for lamp speeds of more than
Copyright © Christoph Schiller November 1997–May 2006

Challenge 550 n 290 000 000 m~s no differences have been found. (Can you guess what lamps were used?)
In everyday life, we know that a stone arrives more rapidly if we run towards it. Again,
for light no difference has been measured. All experiments show that the velocity of light
has the same value for all observers, even if they are moving with respect to each other
or with respect to the light source. The speed of light is indeed the ideal, perfect measure-
ment standard.**

* ‘Nothing is faster than the years.’ Book X, verse 520.

** An equivalent alternative term for the speed of light is ‘radar speed’ or ‘radio speed’; we will see below
Page 563 why this is the case.
The speed of light is also not far from the speed of neutrinos. This was shown most spectacularly by the

Dvipsbugw
 maximum speed, observers at rest, and motion of light 281

Ref. 241 There is also a second set of experimental evidence for the
constancy of the speed of light. Every electromagnetic device,
such as an electric toothbrush, shows that the speed of light is
Page 536 constant. We will discover that magnetic fields would not res-
ult from electric currents, as they do every day in every motor
and in every loudspeaker, if the speed of light were not constant.
This was actually how the constancy was first deduced, by sev-
eral researchers. Only after understanding this, did the German–
Swiss physicist Albert Einstein* show that the constancy is also Dvipsbugw
in agreement with the motion of bodies, as we will do in this
section. The connection between electric toothbrushes and re-
Page 536 lativity will be described in the chapter on electrodynamics.** In Albert Einstein
simple terms, if the speed of light were not constant, observers
would be able to move at the speed of light. Since light is a wave, such observers would
see a wave standing still. However, electromagnetism forbids the such a phenomenon.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Therefore, observers cannot reach the speed of light.
In summary, the velocity v of any physical system in nature (i.e., any localized mass or
energy) is bound by
vDc. (102)

This relation is the basis of special relativity; in fact, the full theory of special relativity
is contained in it. Einstein often regretted that the theory was called ‘Relativitätstheorie’
or ‘theory of relativity’; he preferred the name ‘Invarianztheorie’ or ‘theory of invariance’,
Ref. 244 but was not able to change the name.
observation of a supernova in 1987, when the flash and the neutrino pulse arrived a 12 seconds apart. (It
is not known whether the difference is due to speed differences or to a different starting point of the two
flashes.) What is the first digit for which the two speed values could differ, knowing that the supernova was
Challenge 551 n 1.7 ë 105 light years away?
Experiments also show that the speed of light is the same in all directions of space, to at least 21 digits of
Ref. 239 precision. Other data, taken from gamma ray bursts, show that the speed of light is independent of frequency,
Ref. 240 to at least 20 digits of precision.
* Albert Einstein (b. 1879 Ulm, d. 1955 Princeton); one of the greatest physicists ever. He published three
important papers in 1905, one about Brownian motion, one about special relativity, and one about the idea
of light quanta. Each paper was worth a Nobel Prize, but he was awarded the prize only for the last one. Also
in 1905, he proved the famous formula E 0 = mc 2 (published in early 1906), possibly triggered by an idea of
Page 318 Olinto de Pretto. Although Einstein was one of the founders of quantum theory, he later turned against it.
Copyright © Christoph Schiller November 1997–May 2006

His famous discussions with his friend Niels Bohr nevertheless helped to clarify the field in its most counter-
intuitive aspects. He explained the Einstein–de Haas effect which proves that magnetism is due to motion
inside materials. In 1915 and 1916, he published his highest achievement: the general theory of relativity,
one of the most beautiful and remarkable works of science.
Being Jewish and famous, Einstein was a favourite target of attacks and discrimination by the National
Socialist movement; in 1933 he emigrated to the USA. He was not only a great physicist, but also a great
Ref. 242 thinker; his collection of thoughts about topics outside physics are worth reading.
Anyone interested in emulating Einstein should know that he published many papers, and that many
of them were wrong; he would then correct the results in subsequent papers, and then do so again. This
happened so frequently that he made fun of himself about it. Einstein realizes the famous definition of a
genius as a person who makes the largest possible number of mistakes in the shortest possible time.
** For information about the influences of relativity on machine design, see the interesting textbook by Van
Ref. 243 Bladel.

Dvipsbugw
 282 ii special rel ativity • 5. speed, rest and light

The constancy of the speed of light is in complete contrast with Galilean mechanics,
and proves that the latter is wrong at high velocities. At low velocities the description
remains good, because the error is small. But if we want a description valid at all velocities,
we have to discard Galilean mechanics. For example, when we play tennis we use the fact
that by hitting the ball in the right way, we can increase or decrease its speed. But with
light this is impossible. Even if we take an aeroplane and fly after a light beam, it still
moves away with the same speed. Light does not behave like cars. If we accelerate a bus
we are driving, the cars on the other side of the road pass by with higher and higher
speeds. For light, this is not so: light always passes by with the same speed.* Dvipsbugw
Why is this result almost unbelievable, even though the measurements show it un-
ambiguously? Take two observers O and Ω (pronounced ‘omega’) moving with relative
velocity v, such as two cars on opposite sides of the street. Imagine that at the moment
they pass each other, a light flash is emitted by a lamp in O. The light flash moves through
positions x(t) for O and through positions ξ(τ) (pronounced ‘xi of tau’) for Ω. Since the
speed of light is the same for both, we have

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

=c= .
x ξ
(103)
t τ

However, in the situation described, we obviously have x x ξ. In other words, the con-
stancy of the speed of light implies that t x τ, i.e. that time is different for observers moving
Challenge 552 e relative to each other. Time is thus not unique. This surprising result, which has been con-
Ref. 245 firmed by many experiments, was first stated clearly in 1905 by Albert Einstein. Though
many others knew about the invariance of c, only the young Einstein had the courage to
say that time is observer-dependent, and to face the consequences. Let us do so as well.
Already in 1895, the discussion of viewpoint invariance had been called the theory of
relativity by Henri Poincaré.** Einstein called the description of motion without gravity
Ref. 241 the theory of special relativity, and the description of motion with gravity the theory of
general relativity. Both fields are full of fascinating and counter-intuitive results. In par-
ticular, they show that everyday Galilean physics is wrong at high speeds.
The speed of light is a limit speed. We stress that we are not talking of the situation
where a particle moves faster than the speed of light in matter, but still slower than the
speed of light in vacuum. Moving faster than the speed of light in matter is possible. If the
particle is charged, this situation gives rise to the so-called Čerenkov radiation. It corres-
ponds to the V-shaped wave created by a motor boat on the sea or the cone-shaped shock
Copyright © Christoph Schiller November 1997–May 2006

wave around an aeroplane moving faster than the speed of sound. Čerenkov radiation is
regularly observed; for example it is the cause of the blue glow of the water in nuclear re-
actors. Incidentally, the speed of light in matter can be quite low: in the centre of the Sun,

* Indeed, even with the current measurement precision of 2 ë 10−13 , we cannot discern any changes of the
Ref. 239 speed of light with the speed of the observer.
** Henri Poincaré (1854–1912), important French mathematician and physicist. Poincaré was one of the
most productive men of his time, advancing relativity, quantum theory, and many parts of mathematics.
The most beautiful and simple introduction to relativity is still that given by Albert Einstein himself, for
example in Über die spezielle und allgemeine Relativitätstheorie, Vieweg, 1997, or in The Meaning of Relativity,
Methuen, London, 1951. It has taken a century for books almost as beautiful to appear, such as the text by
Ref. 246, Ref. 247 Taylor and Wheeler.

Dvipsbugw
 maximum speed, observers at rest, and motion of light 283

the speed of light is estimated to be only around 10 km~year, and even in the laboratory,
Ref. 248, Ref. 249 for some materials, it has been found to be as low as 0.3 m~s. In the following, when we
use the term ‘speed of light’, we mean the speed of light in vacuum. The speed of light in
air in smaller than that in vacuum only by a fraction of a percent, so that in most cases,
the difference can be neglected.

Special relativity in a few lines

The speed of light is constant for all observers. We Dvipsbugw
can thus deduce all relations between what two dif- first
t
Ref. 250 ferent observers measure with the help of Figure 137. observer second
or clock
It shows two observers moving with constant speed observer
or clock
against each other in space-time, with the first send-
ing a light flash to the second, from where it is reflec- k2 T
ted back to the first. Since light speed is constant, light light

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is the only way to compare time and space coordin-
ates for two distant observers. Two distant clocks (like t1 = (k2+1)T/2 t2 = kT
two distant metre bars) can only be compared, or
synchronized, using light or radio flashes. Since light
speed is constant, light paths are parallel in such dia- T
grams.
A constant relative speed between two observers O
implies that a constant factor k relates the time co- x
Challenge 553 n ordinates of events. (Why is the relation linear?) If a
flash starts at a time T as measured for the first ob- F I G U R E 137 A drawing containing
server, it arrives at the second at time kT, and then most of special relativity
back again at the first at time k 2 T. The drawing shows
Challenge 554 n that ¾
c+v v k2 − 1
k= or = . (104)
c−v c k2 + 1

Page 284 This factor will appear again in the Doppler effect.*
The figure also shows that the time coordinate t 1 assigned by the first observer to the
moment in which the light is reflected is different from the coordinate t 2 assigned by the
second observer. Time is indeed different for two observers in relative motion. Figure 138
Copyright © Christoph Schiller November 1997–May 2006

illustrates the result.

The time dilation factor between the two time coordinates is found from Figure 137 by
comparing the values t 1 and t 2 ; it is given by

=¼ = γ(v) .
t1 1
(105)
t2 1− v2
c2

Time intervals for a moving observer are shorter by this factor γ; the time dilation factor
is always larger than 1. In other words, moving clocks go slower. For everyday speeds the

* The explanation of relativity using the factor k is often called k-calculus.

Dvipsbugw
 284 ii special rel ativity • 5. speed, rest and light

one moving watch

first second
time time

two fixed watches

F I G U R E 138 Moving clocks go slow

Dvipsbugw

Challenge 555 e effect is tiny. That is why we do not detect time differences in everyday life. Nevertheless,
Galilean physics is not correct for speeds near that of light. The same factor γ also appears
in the formula E = γmc 2 , which we will deduce below. Expression (104) or (105) is the
only piece of mathematics needed in special relativity: all other results derive from it.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

If a light flash is sent forward starting from the second observer and reflected back,
he will make the same statement: for him, the first clock is moving, and also for him, the
moving clock goes slower. Each of the observers observes that the other clock goes slower.
The situation is similar to that of two men comparing the number of steps between two
identical ladders that are not parallel. A man on either ladder will always observe that the
steps of the other ladder are shorter. For another analogy, take two people moving away
from each other: each of them notes that the other gets smaller as their distance increases.
Naturally, many people have tried to find arguments to avoid the strange conclusion
that time differs from observer to observer. But none have succeeded, and experimental
results confirm this conclusion. Let us have a look at some of them.

Acceleration of light and the Doppler effect

Light can be accelerated. Every mirror does this! We will see in the chapter on electromag-
netism that matter also has the power to bend light, and thus to accelerate it. However, it
Page 567 will turn out that all these methods only change the direction of propagation; none has the
power to change the speed of light in a vacuum. In short, light is an example of a motion
Challenge 556 n that cannot be stopped. There are only a few other such examples. Can you name one?
What would happen if we could accelerate light to higher speeds? For this to be pos-
sible, light would have to be made of particles with non-vanishing mass. Physicists call
such particles massive particles. If light had mass, it would be necessary to distinguish
Copyright © Christoph Schiller November 1997–May 2006

the ‘massless energy speed’ c from the speed of light c L , which would be lower and would
depend on the kinetic energy of those massive particles. The speed of light would not be
constant, but the massless energy speed would still be so. Massive light particles could
be captured, stopped and stored in a box. Such boxes would make electric illumination
unnecessary; it would be sufficient to store some daylight in them and release the light,
slowly, during the following night, maybe after giving it a push to speed it up.*
Physicists have tested the possibility of massive light in quite some detail. Observations
Ref. 251, Ref. 252 now put any possible mass of light (particles) at less than 1.3 ë 10−52 kg from terrestrial

* Incidentally, massive light would also have longitudinal polarization modes. This is in contrast to observa-
tions, which show that light is polarized exclusively transversally to the propagation direction.

Dvipsbugw
 maximum speed, observers at rest, and motion of light 285

sender receiver

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

receiver θr
light x
signal

z
sender θs

v x

F I G U R E 139 The set-up for the observation of the Doppler effect

experiments, and at less than 4 ë 10−62 kg from astrophysical arguments (which are a bit
less strict). In other words, light is not heavy, light is light.
But what happens when light hits a moving mirror? If the speed of light does not
change, something else must. The situation is akin to that of a light source moving with re-
spect to the receiver: the receiver will observe a different colour from that observed by the
sender. This is called the Doppler effect. Christian Doppler* was the first to study the fre-
quency shift in the case of sound waves – the well-known change in whistle tone between
Copyright © Christoph Schiller November 1997–May 2006

approaching and departing trains – and to extend the concept to the case of light waves.
As we will see later on, light is (also) a wave, and its colour is determined by its frequency,
or equivalently, by its wavelength λ. Like the tone change for moving trains, Doppler real-
ized that a moving light source produces a colour at the receiver that is different from the
colour at the source. Simple geometry, and the conservation of the number of maxima
Challenge 557 e and minima, leads to the result

* Christian Andreas Doppler (b. 1803 Salzburg, d. 1853 Venezia), Austrian physicist. Doppler studied the
effect named after him for sound and light. In 1842 he predicted (correctly) that one day we would be able
to use the effect to measure the motion of distant stars by looking at their colours.

Dvipsbugw
 286 ii special rel ativity • 5. speed, rest and light

=» (1 − cos θ r ) = γ (1 − cos θ r ) .
λr 1 v v
(106)
λs 1 − v 2 ~c 2 c c

The variables v and θ r in this expression are defined in Figure 139. Light from an approach-
ing source is thus blue-shifted, whereas light from a departing source is red-shifted. The
first observation of the Doppler effect for light was made by Johannes Stark* in 1905,
who studied the light emitted by moving atoms. All subsequent experiments confirmed
the calculated colour shift within measurement errors; the latest checks have found agree-
Dvipsbugw
Ref. 253 ment to within two parts per million. In contrast to sound waves, a colour change is also
found when the motion is transverse to the light signal. Thus, a yellow rod in rapid mo-
tion across the field of view will have a blue leading edge and a red trailing edge prior to
the closest approach to the observer. The colours result from a combination of the longit-
udinal (first-order) Doppler shift and the transverse (second-order) Doppler shift. At a
particular angle θ unshifted the colours will be the same. (How does the wavelength change
Challenge 558 n in the purely transverse case? What is the expression for θ unshifted in terms of v?)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The colour shift is used in many applications. Almost all solid bodies are mirrors for
radio waves. Many buildings have doors that open automatically when one approaches.
A little sensor above the door detects the approaching person. It usually does this by
measuring the Doppler effect of radio waves emitted by the sensor and reflected by the
Page 563 approaching person. (We will see later that radio waves and light are manifestations of
the same phenomenon.) So the doors open whenever something moves towards them.
Police radar also uses the Doppler effect, this time to measure the speed of cars.**
The Doppler effect also makes it possible to measure the velocity of light sources.
Indeed, it is commonly used to measure the speed of distant stars. In these cases, the
Doppler shift is often characterized by the red-shift number z, defined with the help of
wavelength λ or frequency F by
¾
c+v
z= = −1=
∆λ f S
−1 . (107)
λ fR c−v

Challenge 560 n Can you imagine how the number z is determined? Typical values for z for light sources
in the sky range from −0.1 to 3.5, but higher values, up to more than 10, have also been
Challenge 561 n found. Can you determine the corresponding speeds? How can they be so high?
In summary, whenever one tries to change the speed of light, one only manages to
change its colour. That is the Doppler effect.
Copyright © Christoph Schiller November 1997–May 2006

Page 133 We know from classical physics that when light passes a large mass, such as a star, it is
Challenge 562 n deflected. Does this deflection lead to a Doppler shift?

* Johannes Stark (1874–1957), discovered in 1905 the optical Doppler effect in channel rays, and in 1913
the splitting of spectral lines in electrical fields, nowadays called the Stark effect. For these two discoveries
he received the 1919 Nobel Prize for physics. He left his professorship in 1922 and later turned into a full-
blown National Socialist. A member of the NSDAP from 1930 onwards, he became known for aggressively
criticizing other people’s statements about nature purely for ideological reasons; he became rightly despised
by the academic community all over the world.
Challenge 559 n ** At what speed does a red traffic light appear green?

Dvipsbugw
 maximum speed, observers at rest, and motion of light 287

The difference between light and sound

The Doppler effect for light is much more important than the Doppler effect for sound.
Even if the speed of light were not yet known to be constant, this effect alone would prove
that time is different for observers moving relative to each other. Why? Time is what we
read from our watch. In order to determine whether another watch is synchronized with
our own one, we look at both watches. In short, we need to use light signals to synchron-
Ref. 254 ize clocks. Now, any change in the colour of light moving from one observer to another
necessarily implies that their watches run differently, and thus that time is different for Dvipsbugw
the two of them. One way to see this is to note that also a light source is a clock – ‘ticking’
very rapidly. So if two observers see different colours from the same source, they meas-
ure different numbers of oscillations for the same clock. In other words, time is different
for observers moving against each other. Indeed, equation (104) implies that the whole
of relativity follows from the full Doppler effect for light. (Can you confirm that the con-
nection between observer-dependent frequencies and observer-dependent time breaks

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 563 n down in the case of the Doppler effect for sound?)
Why does the behaviour of light imply special relativity, while that of sound in air does
not? The answer is that light is a limit for the motion of energy. Experience shows that
there are supersonic aeroplanes, but there are no superluminal rockets. In other words,
the limit v D c is valid only if c is the speed of light, not if c is the speed of sound in air.
However, there is at least one system in nature where the speed of sound is indeed a
limit speed for energy: the speed of sound is the limit speed for the motion of dislocations
Page 987 in crystalline solids. (We discuss this in detail later on.) As a result, the theory of special
relativity is also valid for such dislocations, provided that the speed of light is replaced
everywhere by the speed of sound! Dislocations obey the Lorentz transformations, show
Ref. 255 length contraction, and obey the famous energy formula E = γmc 2 . In all these effects
the speed of sound c plays the same role for dislocations as the speed of light plays for
general physical systems.
If special relativity is based on the statement that nothing can move faster than light,
this statement needs to be carefully checked.

Can one shoot faster than one’s shadow?

Quid celerius umbra?*

Challenge 564 e
“ ”
for Lucky Luke to achieve the feat shown in Figure 140, his bullet has to move faster than
the speed of light. (What about his hand?) In order to emulate Lucky Luke, we could take
Copyright © Christoph Schiller November 1997–May 2006

the largest practical amount of energy available, taking it directly from an electrical power
station, and accelerate the lightest ‘bullets’ that can be handled, namely electrons. This ex-
periment is carried out daily in particle accelerators such as the Large Electron Positron
ring, the LEP, of 27 km circumference, located partly in France and partly in Switzerland,
near Geneva. There, 40 MW of electrical power (the same amount used by a small city)
accelerates electrons and positrons to energies of over 16 nJ (104.5 GeV) each, and their
Ref. 256 speed is measured. The result is shown in Figure 141: even with these impressive means

* ‘What is faster than the shadow?’ A motto often found on sundials.

Dvipsbugw
 288 ii special rel ativity • 5. speed, rest and light

Dvipsbugw

F I G U R E 140 Lucky Luke

it is impossible to make electrons move more rapidly than light. (Can you imagine a way

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 565 e to measure energy and speed separately?) The speed–energy relation of Figure 141 is a
Page 313 consequence of the maximum speed, and is deduced below. These and many similar ob-
servations thus show that there is a limit to the velocity of objects. Bodies (and radiation)
cannot move at velocities higher than the speed of light.* The accuracy of Galilean mech-
anics was taken for granted for more than three centuries, so that nobody ever thought
of checking it; but when this was finally done, as in Figure 141, it was found to be wrong.
The people most unhappy with this
limit are computer engineers: if the speed 1
2 TGal = m v2
limit were higher, it would be possible v 2
to make faster microprocessors and thus 2
faster computers; this would allow, for c 1
example, more rapid progress towards T = m c2 ( – 1)
1 - v2/c 2
the construction of computers that un-
derstand and use language.
The existence of a limit speed runs T
counter to Galilean mechanics. In fact, F I G U R E 141 Experimental values (dots) for the
it means that for velocities near that of electron velocity v as function of their kinetic
light, say about 15 000 km~s or more, the energy T, compared with the prediction of Galilean
expression mv 2 ~2 is not equal to the kin- physics (blue) and that of special relativity (red)
etic energy T of the particle. In fact, such
Copyright © Christoph Schiller November 1997–May 2006

high speeds are rather common: many families have an example in their home. Just calcu-
late the speed of electrons inside a television, given that the transformer inside produces
Challenge 566 n 30 kV.

* There are still people who refuse to accept these results, as well as the ensuing theory of relativity. Every
physicist should enjoy the experience, at least once in his life, of conversing with one of these men. (Strangely,
no woman has yet been reported as belonging to this group of people.) This can be done, for example, via
Ref. 257 the internet, in the sci.physics.relativity newsgroup. See also the http://www.crank.net website. Crackpots
are a fascinating lot, especially since they teach the importance of precision in language and in reasoning,
which they all, without exception, neglect. Encounters with several of them provided the inspiration for this
chapter.

Dvipsbugw
 maximum speed, observers at rest, and motion of light 289

t
second
first
observer third
observer
(e.g. train) observer
(e.g. Earth)
(e.g. stone)

kseT

kteT
Dvipsbugw
T

O
x

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 142 How to deduce the
composition of velocities

The observation of speed of light as a limit speed for objects is easily seen to be a
consequence of its constancy. Bodies that can be at rest in one frame of reference obviously
move more slowly than the maximum velocity (light) in that frame. Now, if something
moves more slowly than something else for one observer, it does so for all other observers
Challenge 567 d as well. (Trying to imagine a world in which this would not be so is interesting: funny
things would happen, such as things interpenetrating each other.) Since the speed of light
is the same for all observers, no object can move faster than light, for every observer.
We follow that the maximum speed is the speed of massless entities. Electromagnetic
waves, including light, are the only known entities that can travel at the maximum speed.
Gravitational waves are also predicted to achieve maximum speed. Though the speed
of neutrinos cannot be distinguished experimentally from the maximum speed, recent
Ref. 258 experiments suggest that they do have a tiny mass.
Conversely, if a phenomenon exists whose speed is the limit speed for one observer,
Challenge 568 e then this limit speed must necessarily be the same for all observers. Is the connection
Challenge 569 r between limit property and observer invariance generally valid in nature? Copyright © Christoph Schiller November 1997–May 2006

The composition of velocities

If the speed of light is a limit, no attempt to exceed it can succeed. This implies that when
velocities are composed, as when one throws a stone while running, the values cannot
simply be added. If a train is travelling at velocity v te relative to the Earth, and somebody
throws a stone inside it with velocity v st relative to the train in the same direction, it is
usually assumed as evident that the velocity of the stone relative to the Earth is given by
v se = v st + v te . In fact, both reasoning and measurement show a different result.
The existence of a maximum speed, together with Figure 142, implies that the k-factors
must satisfy k se = k st k te .* Then we only have to insert the relation (104) between each

* By taking the (natural) logarithm of this equation, one can define a quantity, the rapidity, that measures

Dvipsbugw
 290 ii special rel ativity • 5. speed, rest and light

Challenge 570 e k-factor and the respective speed to get

v st + v te
v se = .
1 + v st v te ~c 2
(108)

Challenge 571 e This is called the velocity composition formula. The result is never larger than c and is
always smaller than the naive sum of the velocities.* Expression (108) has been confirmed
Page 313, page 536 by all of the millions of cases for which it has been checked. You may check that it reduces
Ref. 252 to the naive sum for everyday life values. Dvipsbugw

Observers and the principle of special relativity

Special relativity is built on a simple principle:
U The maximum speed of energy transport is the same for all observers.
Ref. 260 Or, as Hendrik Lorentz** liked to say, the equivalent:
U The speed v of a physical system is bound by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

vDc (109)

for all observers, where c is the speed of light.

This independence of the speed of light from the observer was checked with high preci-
Ref. 261 sion by Michelson and Morley*** in the years from 1887 onwards. It has been confirmed
in all subsequent experiments; the most precise to date, which achieved a precision of
Ref. 262 10−14 is shown in Figure 143.
In fact, special relativity is also confirmed by all the precision experiments that were
performed before it was formulated. You can even confirm it yourself at home. The way
Page 536 to do this is shown in the section on electrodynamics.
The existence of a limit speed has several interesting consequences. To explore them,
let us keep the rest of Galilean physics intact.**** The limit speed is the speed of light. It
is constant for all observers. This constancy implies:
the speed and is additive.
Ref. 259 * One can also deduce the Lorentz transformation directly from this expression.
** Hendrik Antoon Lorentz (b. 1853 Arnhem, d. 1928 Haarlem) was, together with Boltzmann and Kelvin,
one of the most important physicists of his time. He deduced the so-called Lorentz transformation and the
Lorentz contraction from Maxwell’s equation for the electrodynamic field. He was the first to understand,
Copyright © Christoph Schiller November 1997–May 2006

long before quantum theory confirmed the idea, that Maxwell’s equations for the vacuum also describe
matter and all its properties, as long as moving charged point particles – the electrons – are included. He
showed this in particular for the dispersion of light, for the Zeeman effect, for the Hall effect and for the
Faraday effect. He gave the correct description of the Lorentz force. In 1902, he received the physics Nobel
Prize, together with Pieter Zeeman. Outside physics, he was active in the internationalization of scientific
collaborations. He was also instrumental in the creation of the largest human-made structures on Earth: the
polders of the Zuyder Zee.
*** Albert Abraham Michelson (b. 1852 Strelno, d. 1931 Pasadena), Prussian–Polish–US-American physi-
cist, awarded the Nobel Prize in physics in 1907. Michelson called the set-up he devised an interferometer, a
term still in use today. Edward William Morley (1838–1923), US-American chemist, was Michelson’s friend
and long-time collaborator.
Page 86 **** This point is essential. For example, Galilean physics states that only relative motion is physical. Galilean
physics also excludes various mathematically possible ways to realize a constant light speed that would con-

Dvipsbugw
 maximum speed, observers at rest, and motion of light 291

AOM Power
30 driver servo

beat frequency change [Hz ]

Fiber
20 Frequency
FC counter
PZT AOM

angle/3 [deg]
Laser 1 PD
10 Nd: YAG FC
T °C
FC
0 T °C Laser 2 Res B PD
Nd: YAG Fiber BS
PZT AOM DBM
10 FC
∑ PD
PD
Frequency Local
20 servo oscillator PD Res A
0 100 200 300 400 500 600 Frequency Local
DBM
servo oscillator
time since begin of rotation [s]
∑
AOM Power
driver
servo Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 143 The result, the schematics and the cryostat set-up for the most precise
Michelson–Morley experiment performed to date (© Stephan Schiller)

— In a closed free-floating room, there is no way to tell the speed of the room.
— There is no notion of absolute rest (or space): rest (like space) is an observer-dependent
concept.*
— Time depends on the observer; time is not absolute.
More interesting and specific conclusions can be drawn when two additional conditions
are assumed. First, we study situations where gravitation can be neglected. (If this not the
case, we need general relativity to describe the system.) Secondly, we also assume that the
data about the bodies under study – their speed, their position, etc. – can be gathered
Copyright © Christoph Schiller November 1997–May 2006

without disturbing them. (If this not the case, we need quantum theory to describe the
system.)
To deduce the precise way in which the different time intervals and lengths measured
by two observers are related to each other, we take an additional simplifying step. We start

tradict everyday life.

Einstein’s original 1905 paper starts from two principles: the constancy of the speed of light and the
equivalence of all inertial observers. The latter principle had already been stated in 1632 by Galileo; only the
constancy of the speed of light was new. Despite this fact, the new theory was named – by Poincaré – after
Ref. 244 the old principle, instead of calling in ‘invariance theory’, as Einstein would have preferred.
Challenge 572 n * Can you give the argument leading to this deduction?

Dvipsbugw
 292 ii special rel ativity • 5. speed, rest and light

observer (greek)
v

light
c

observer (roman)

F I G U R E 144 Two inertial observers and a

beam of light

Galilean physics special relativity Dvipsbugw

t τ t τ

L L

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

O, Ω x, ξ O, Ω x

F I G U R E 145 Space-time diagrams for light seen from two

different observers using coordinates (t, x) and (τ, ξ)

with a situation where no interaction plays a role. In other words, we start with relativistic
kinematics of bodies moving without disturbance.
If an undisturbed body is observed to travel along a straight line with a constant ve-
locity (or to stay at rest), one calls the observer inertial, and the coordinates used by the
observer an inertial frame of reference. Every inertial observer is itself in undisturbed
motion. Examples of inertial observers (or frames) thus include – in two dimensions –
those moving on a frictionless ice surface or on the floor inside a smoothly running train
or ship; for a full example – in all three spatial dimensions – we can take a cosmonaut
travelling in a space-ship as long as the engine is switched off. Inertial observers in three
dimensions might also be called free-floating observers. They are thus not so common.
Challenge 573 e Non-inertial observers are much more common. Can you confirm this? Inertial observ-
ers are the most simple ones, and they form a special set:
— Any two inertial observers move with constant velocity relative to each other (as longs
as gravity plays no role, as assumed above).
Copyright © Christoph Schiller November 1997–May 2006

— All inertial observers are equivalent: they describe the world with the same equations.
Because it implies the lack of absolute space and time, this statement was called the
principle of relativity by Henri Poincaré. However, the essence of relativity is the exist-
ence of a limit speed.
To see how measured length and space intervals change from one observer to the other,
we assume two inertial observers, a Roman one using coordinates x, y, z and t, and a
Greek one using coordinates ξ, υ, ζ and τ,* that move with velocity v relative to each
other. The axes are chosen in such a way that the velocity points in the x-direction. The

* They are read as ‘xi’, ‘upsilon’, ‘zeta’ and ‘tau’. The names, correspondences and pronunciations of all Greek
letters are explained in Appendix A.

Dvipsbugw
 maximum speed, observers at rest, and motion of light 293

constancy of the speed of light in any direction for any two observers means that for the
motion of light the coordinate differentials are related by

0 = (cdt)2 − (dx)2 − (dy)2 − (dz)2 = (cdτ)2 − (dξ)2 − (dυ)2 − (dζ)2 . (110)

Assume also that a flash lamp at rest for the Greek observer, thus with dξ = 0, produces
two flashes separated by a time interval dτ. For the Roman observer, the flash lamp moves
with speed v, so that dx = vdt. Inserting this into the previous expression, and assuming
linearity and speed direction independence for the general case, we find that intervals are Dvipsbugw
Challenge 574 e related by

dτ + vdξ~c 2
dt = γ(dτ + vdξ~c 2 ) = » with v = dx~dt
1 − v 2 ~c 2
dξ + vdτ
dx = γ(dξ + vdτ) = »
1 − v 2 ~c 2

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

dy = dυ
dz = dζ . (111)

These expressions describe how length and time intervals measured by different observ-
ers are related. At relative speeds v that are small compared to the velocity of light, such
as occur in everyday life, the time intervals are essentially equal; the stretch factor or re-
lativistic correction or relativistic contraction γ is then equal to 1 for all practical purposes.
However, for velocities near that of light the measurements of the two observers give
different values. In these cases, space and time mix, as shown in Figure 145.
The expressions (111) are also strange in another respect. When two observers look
Challenge 575 n at each other, each of them claims to measure shorter intervals than the other. In other
words, special relativity shows that the grass on the other side of the fence is always shorter
– if one rides along beside the fence on a bicycle and if the grass is inclined. We explore
this bizarre result in more detail shortly.
The stretch factor γ is equal to 1 for most practical purposes in everyday life. The largest
value humans have ever achieved is about 2 ë 105 ; the largest observed value in nature is
Challenge 576 n about 1012 . Can you imagine where they occur?
Once we know how space and time intervals change, we can easily deduce how coordin-
ates change. Figures 144 and 145 show that the x coordinate of an event L is the sum of
Copyright © Christoph Schiller November 1997–May 2006

two intervals: the ξ coordinate and the length of the distance between the two origins. In
other words, we have
ξ = γ(x − vt) and v =
dx
. (112)
dt
Using the invariance of the space-time interval, we get

τ = γ(t − xv~c 2 ) . (113)

Henri Poincaré called these two relations the Lorentz transformations of space and time

Dvipsbugw
 294 ii special rel ativity • 5. speed, rest and light

after their discoverer, the Dutch physicist Hendrik Antoon Lorentz.* In one of the most
Ref. 263 beautiful discoveries of physics, in 1892 and 1904, Lorentz deduced these relations from
Page 546 the equations of electrodynamics, where they had been lying, waiting to be discovered,
since 1865.** In that year James Clerk Maxwell had published the equations in order to
describe everything electric and magnetic. However, it was Einstein who first understood
that t and τ, as well as x and ξ, are equally correct and thus equally valid descriptions of
space and time.
The Lorentz transformation describes the change of viewpoint from one inertial frame
to a second, moving one. This change of viewpoint is called a (Lorentz) boost. The formu- Dvipsbugw
lae (112) and (113) for the boost are central to the theories of relativity, both special and
general. In fact, the mathematics of special relativity will not get more difficult than that:
if you know what a square root is, you can study special relativity in all its beauty.
Many alternative formulae for boosts have been explored, such as expressions in which
Ref. 264 the relative acceleration of the two observers is included, as well as the relative velocity.
However, they had all to be discarded after comparing their predictions with experi-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

mental results. Before we have a look at such experiments, we continue with a few logical
deductions from the boost relations.

What is space-time?
Von Stund’ an sollen Raum für sich und Zeit für

“ sich völlig zu Schatten herabsinken und nur

noch eine Art Union der beiden soll
Selbstständigkeit bewaren.***
Hermann Minkowski.

The Lorentz transformations tell us something important: that space and time are two
aspects of the same basic entity. They ‘mix’ in different ways for different observers. This
fact is commonly expressed by stating that time is the fourth dimension. This makes sense
”
because the common basic entity – called space-time – can be defined as the set of all
events, events being described by four coordinates in time and space, and because the set
Challenge 577 n of all events has the properties of a manifold.**** (Can you confirm this?)
In other words, the existence of a maximum speed in nature forces us to introduce
a space-time manifold for the description of nature. In the theory of special relativity,
the space-time manifold is characterized by a simple property: the space-time interval di
Ref. 265 between two nearby events, defined as
Copyright © Christoph Schiller November 1997–May 2006

v2
di 2 = c 2 dt 2 − dx 2 − dy 2 − dz 2 = c 2 dt 2 (1 − ), (114)
c2

* For information about Hendrik Antoon Lorentz, see page 290.

** The same discovery had been published first in 1887 by the German physicist Woldemar Voigt (1850–
1919); Voigt – pronounced ‘Fohgt’ – was also the discoverer of the Voigt effect and the Voigt tensor. Inde-
pendently, in 1889, the Irishman George F. Fitzgerald also found the result.
*** ‘Henceforth space by itself and time by itself shall completely fade into shadows and only a kind of union
of the two shall preserve autonomy.’ This famous statement was the starting sentence of Minkowski’s 1908
talk at the meeting of the Gesellschaft für Naturforscher und Ärzte.
Page 1214 **** The term ‘manifold’ is defined in Appendix D.

Dvipsbugw
 maximum speed, observers at rest, and motion of light 295

is independent of the (inertial) observer. Such a space-time is also called Minkowski

space-time, after Hermann Minkowski* the teacher of Albert Einstein; he was the first,
in 1904, to define the concept of space-time and to understand its usefulness and import-
ance.
The space-time interval di of equation (114) has a simple interpretation. It is the time
measured by an observer moving from event (t, x) to event (t + dt, x + dx), the so-called
proper time, multiplied by c. If we neglect the factor c, we could simply call it wristwatch
time.
We live in a Minkowski space-time, so to speak. Minkowski space-time exists inde- Dvipsbugw
pendently of things. And even though coordinate systems can be different from observer
to observer, the underlying entity, space-time, is still unique, even though space and time
by themselves are not.
How does Minkowski space-time differ from Galilean space-time, the combination of
everyday space and time? Both space-times are manifolds, i.e. continuum sets of points,
both have one temporal and three spatial dimensions, and both manifolds have the topo-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 578 n logy of the punctured sphere. (Can you confirm this?) Both manifolds are flat, i.e. free of
curvature. In both cases, space is what is measured with a metre rule or with a light ray,
and time is what is read from a clock. In both cases, space-time is fundamental; it is and
remains the background and the container of things and events.
The central difference, in fact the only one, is that Minkowski space-time, in contrast to
the Galilean case, mixes space and time, and in particular, does so differently for observers
with different speeds, as shown in Figure 145. That is why time is an observer-dependent
concept.
The maximum speed in nature thus forces us to describe motion with space-time. That
is interesting, because in space-time, speaking in simple terms, motion does not exist. Mo-
tion exists only in space. In space-time, nothing moves. For each point particle, space-
time contains a world-line. In other words, instead of asking why motion exists, we can
equivalently ask why space-time is criss-crossed by world-lines. At this point, we are still
far from answering either question. What we can do is to explore how motion takes place.

Can we travel to the past? – Time and causality

We know that time is different for different observers. Does time nevertheless order
events in sequences? The answer given by relativity is a clear ‘yes and no’. Certain sets
of events are not naturally ordered by time; others sets are. This is best seen in a space-
Copyright © Christoph Schiller November 1997–May 2006

time diagram.
Clearly, two events can be placed in sequence only if one event is the cause of the other.
But this connection can only apply if the events exchange energy (e.g. through a signal).
In other words, a relation of cause and effect between two events implies that energy or
signals can travel from one event to the other; therefore, the speed connecting the two
events must not be larger than the speed of light. Figure 146 shows that event E at the
origin of the coordinate system can only be influenced by events in quadrant IV (the past
light cone, when all space dimensions are included), and can itself influence only events

* Hermann Minkowski (1864–1909), German mathematician. He had developed similar ideas to Einstein,
but the latter was faster. Minkowski then developed the concept of space-time. Minkowski died suddenly at
the age of 44.

Dvipsbugw
 296 ii special rel ativity • 5. speed, rest and light

time t

ne

lig
II
lig

ne
T future T

co

ht
ht

co
future

ht

co
co

ht
lig

ne
ne

lig
III I
E elsewhere E elsewhere
elsewhere space y

Dvipsbugw
IV x

past
past

F I G U R E 146 A space-time diagram for a moving object T seen from an inertial observer O in the case
of one and two spatial dimensions

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

in quadrant II (the future light cone). Events in quadrants I and III neither influence nor
are influenced by event E. The light cone defines the boundary between events that can
be ordered with respect to their origin – namely those inside the cone – and those that
cannot – those outside the cones, happening elsewhere for all observers. (Some people
call all the events happening elsewhere the present.) So, time orders events only partially.
For example, for two events that are not causally connected, their temporal order (or their
simultaneity) depends on the observer!
In particular, the past light cone gives the complete set of events that can influence
what happens at the origin. One says that the origin is causally connected only to the past
light cone. This statement reflects the fact that any influence involves transport of energy,
and thus cannot travel faster than the speed of light. Note that causal connection is an
invariant concept: all observers agree on whether or not it applies to two given events.
Challenge 579 n Can you confirm this?
A vector inside the light cone is called timelike; one on the light cone is called lightlike
or null; and one outside the cone is called spacelike. For example, the world-line of an
observer, i.e. the set of all events that make up its past and future history, consists of time-
like events only. Time is the fourth dimension; it expands space to space-time and thus
Copyright © Christoph Schiller November 1997–May 2006

‘completes’ space-time. This is the relevance of the fourth dimension to special relativity,
no more and no less.
Special relativity thus teaches us that causality and time can be defined only because
light cones exist. If transport of energy at speeds faster than that of light did exist, time
could not be defined. Causality, i.e. the possibility of (partially) ordering events for all
observers, is due to the existence of a maximal speed.
If the speed of light could be surpassed in some way, the future could influence the past.
Challenge 580 n Can you confirm this? In such situations, one would observe acausal effects. However,
there is an everyday phenomenon which tells that the speed of light is indeed maximal:
our memory. If the future could influence the past, we would also be able to remember the
future. To put it in another way, if the future could influence the past, the second principle

Dvipsbugw
 curiosities of special rel ativity 297

of thermodynamics would not be valid and our memory would not work.* No other data
from everyday life or from experiments provide any evidence that the future can influence
the past. In other words, time travel to the past is impossible. How the situation changes in
quantum theory will be revealed later on. Interestingly, time travel to the future is possible,
as we will see shortly.

Curiosities of special rel ativity

Faster than light: how far can we travel? Dvipsbugw
How far away from Earth can we travel, given that the trip should not last more than
a lifetime, say 80 years, and given that we are allowed to use a rocket whose speed can
approach the speed of light as closely as desired? Given the time t we are prepared to
spend in a rocket, given the speed v of the rocket and assuming optimistically that it can
accelerate and decelerate in a negligible amount of time, the distance d we can move away
Challenge 581 e is given by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

d=»
vt
. (115)
1 − v 2 ~c 2

The distance d is larger than ct already for v A 0.71c, and, if v is chosen large enough, it
increases beyond all bounds! In other words, relativity does not limit the distance we can
travel in a lifetime, and not even the distance we can travel in a single second. We could,
in principle, roam the entire universe in less than a second. In situations such as these it
makes sense to introduce the concept of proper velocity w, defined as

w = d~t = » =γv.
v
(116)
1 − v 2 ~c 2

As we have just seen, proper velocity is not limited by the speed of light; in fact the proper
velocity of light itself is infinite.**

Synchronization and time travel – can a mother stay younger

than her own daughter?
A maximum speed implies that time is is different for different observers moving relative
to each other. So we have to be careful about how we synchronize clocks that are far
apart, even if they are at rest with respect to each other in an inertial reference frame. For
Copyright © Christoph Schiller November 1997–May 2006

example, if we have two similar watches showing the same time, and if we carry one of
them for a walk and back, they will show different times afterwards. This experiment has

* Another related result is slowly becoming common knowledge. Even if space-time had a nontrivial shape,
such as a cylindrical topology with closed time-like curves, one still would not be able to travel into the
Ref. 266 past, in contrast to what many science fiction novels suggest. This is made clear by Stephen Blau in a recent
pedagogical paper.
** Using proper velocity, the relation given in equation (108) for the superposition of two velocities wa = γ a va
Challenge 582 e and wb = γ b vb simplifies to
w sY = γ a γ b (v a + v bY ) and w sÙ = w bÙ , (117)
where the signs Y and Ù designate the component in the direction of and the component perpendicular to
Ref. 267 va , respectively. One can in fact express all of special relativity in terms of ‘proper’ quantities.

Dvipsbugw
 298 ii special rel ativity • 5. speed, rest and light

first
twin

trip of
Earth second twin
time Dvipsbugw
time
comparison
and
first change of
twin rocket

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 147 The twin paradox

Ref. 268, Ref. 269 actually been performed several times and has fully confirmed the prediction of special
relativity. The time difference for a person or a watch in an aeroplane travelling around the
Earth once, at about 900 km~h, is of the order of 100 ns – not very noticeable in everyday
life. In fact, the delay is easily calculated from the expression

=γ.
t
(118)
t′
Human bodies are clocks; they show the elapsed time, usually called age, by various
changes in their shape, weight, hair colour, etc. If a person goes on a long and fast trip, on
her return she will have aged less than a second person who stayed at her (inertial) home.
The most famous illustration of this is the famous twin paradox (or clock paradox).
An adventurous twin jumps on a relativistic rocket that leaves Earth and travels for many
years. Far from Earth, he jumps on another relativistic rocket going the other way and
returns to Earth. The trip is illustrated in Figure 147. At his arrival, he notes that his
twin brother on Earth is much older than himself. Can you explain this result, especially
the asymmetry between the two brothers? This result has also been confirmed in many
Copyright © Christoph Schiller November 1997–May 2006

Ref. 270 experiments.

Special relativity thus confirms, in a surprising fashion, the well-known observation
that those who travel a lot remain younger. The price of the retained youth is, however,
that everything around one changes very much more quickly than if one is at rest with
the environment.
The twin paradox can also be seen as a confirmation of the possibility of time travel
to the future. With the help of a fast rocket that comes back to its starting point, we can
arrive at local times that we would never have reached within our lifetime by staying
home. Alas, we can never return to the past.*
Ref. 271 * There are even special books on time travel, such as the well researched text by Nahin. Note that the concept

Dvipsbugw
 curiosities of special rel ativity 299

One of the simplest experiments confirming

the prolonged youth of fast travellers involves the higher atmosphere

counting of muons. Muons are particles that are

continuously formed in the upper atmosphere
Page 897 by cosmic radiation. Muons at rest (with respect
to the measuring clock) have a finite half-life of
2.2 µs (or, at the speed of light, 660 m). After this
amount of time, half of the muons have decayed. high
This half-life can be measured using simple muon counter Dvipsbugw
counters. In addition, there exist special counters
that only count muons travelling within a certain decays
speed range, say from 0.9950c to 0.9954c. One
can put one of these special counters on top of
a mountain and put another in the valley below, low
counter
as shown in Figure 148. The first time this exper-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

iment was performed, the height difference was F I G U R E 148 More muons than expected
Ref. 272 1.9 km. Flying 1.9 km through the atmosphere at arrive at the ground because fast travel
the mentioned speed takes about 6.4 µs. With the keeps them young
half-life just given, a naive calculation finds that
Challenge 583 n only about 13% of the muons observed at the top should arrive at the lower site. However,
it is observed that about 82% of the muons arrive below. The reason for this result is the
relativistic time dilation. Indeed, at the mentioned speed, muons experience a time dif-
ference of only 0.62 µs during the travel from the mountain top to the valley. This shorter
time yields a much lower number of lost muons than would be the case without time
dilation; moreover, the measured percentage confirms the value of the predicted time
Challenge 584 n dilation factor γ within experimental errors, as you may want to check. A similar effect
is seen when relativistic muons are produced in accelerators.
Half-life dilation has also been found for many other decaying systems, such as pions,
hydrogen atoms, neon atoms and various nuclei, always confirming the predictions of
special relativity. Since all bodies in nature are made of particles, the ‘youth effect’ of high
speeds (usually called ‘time dilation’) applies to bodies of all sizes; indeed, it has not only
Ref. 252 been observed for particles, but also for lasers, radio transmitters and clocks.
If motion leads to time dilation, a clock on the Equator, constantly running around
the Earth, should go slower than one at the poles. However, this prediction, which was
Ref. 273 made by Einstein himself, is incorrect. The centrifugal acceleration leads to a reduction in
Copyright © Christoph Schiller November 1997–May 2006

gravitational acceleration that exactly cancels the increase due to the velocity. This story
serves as a reminder to be careful when applying special relativity in situations involving
gravity. Special relativity is only applicable when space-time is flat, not when gravity is
present.
In short, a mother can stay younger than her daughter. We can also conclude that we
cannot synchronize clocks at rest with respect to each other simply by walking, clock in
hand, from one place to another. The correct way to do so is to exchange light signals.
Challenge 585 n Can you describe how?

of time travel has to be clearly defined; otherwise one has no answer to the clerk who calls his office chair a
time machine, as sitting on it allows him to get to the future.

Dvipsbugw
 300 ii special rel ativity • 5. speed, rest and light

observations
observations by the pilot
by the farmer

pilot
time Dvipsbugw
farmer
time

plane ends
barn ends

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 149 The observations of the pilot and the barn owner

A precise definition of synchronization allows us to call two distant events simultan-

eous. In addition, special relativity shows that simultaneity depends on the observer. This
is confirmed by all experiments performed so far.
However, the mother’s wish is not easy to fulfil. Let us imagine that a mother is acceler-
ated in a spaceship away from Earth at 10 m~s2 for ten years, then decelerates at 10 m~s2
for another ten years, then accelerates for ten additional years towards the Earth, and fi-
nally decelerates for ten final years in order to land safely back on our planet. The mother
has taken 40 years for the trip. She got as far as 22 000 light years from Earth. At her
return on Earth, 44 000 years have passed. All this seems fine, until we realize that the
necessary amount of fuel, even for the most efficient engine imaginable, is so large that
Challenge 586 e the mass returning from the trip is only one part in 2 ë 1019 . The necessary amount of fuel
does not exist on Earth.

Length contraction
The length of an object measured by an observer attached to the object is called its proper
Copyright © Christoph Schiller November 1997–May 2006

length. According to special relativity, the length measured by an inertial observer passing
by is always smaller than the proper length. This result follows directly from the Lorentz
Challenge 587 e transformations.
For a Ferrari driving at 300 km~h or 83 m~s, the length is contracted by 0.15 pm: less
than the diameter of a proton. Seen from the Sun, the Earth moves at 30 km~s; this gives
a length contraction of 6 cm. Neither of these effects has ever been measured. But larger
effects could be. Let us explore some examples.
Imagine a pilot flying through a barn with two doors, one at each end. The plane
is slightly longer than the barn, but moves so rapidly that its relativistically contracted
length is shorter than the length of the barn. Can the farmer close the barn (at least for
a short time) with the plane completely inside? The answer is positive. But why can the

Dvipsbugw
 curiosities of special rel ativity 301

ski ski
h
trap trap

F I G U R E 150 The observations of the trap digger and of the snowboarder, as

(misleadingly) published in the literature

h
Dvipsbugw

v
B rope F

g < h v(t) v(t)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 151 Does the conducting glider keep the F I G U R E 152 What happens to
lamp lit at large speeds? the rope?

pilot not say the following: relative to him, the barn is contracted; therefore the plane
does not fit inside the barn? The answer is shown in Figure 149. For the farmer, the doors
close (and reopen) at the same time. For the pilot, they do not. For the farmer, the pilot is
in the dark for a short time; for the pilot, the barn is never dark. (That is not completely
Challenge 588 n true: can you work out the details?)
We now explore some variations of the general case. Can a rapid snowboarder fall into
a hole that is a bit shorter than his board? Imagine him boarding so fast that the length
contraction factor γ = d~d ′ is 4.* For an observer on the ground, the snowboard is four
times shorter, and when it passes over the hole, it will fall into it. However, for the boarder,
it is the hole which is four times shorter; it seems that the snowboard cannot fall into it.
Ref. 274 More careful analysis shows that, in contrast to the observation of the hole digger, the
snowboarder does not experience the board’s shape as fixed: while passing over the hole,
the boarder observes that the board takes on a parabolic shape and falls into the hole,
Challenge 590 e as shown in Figure 150. Can you confirm this? In other words, shape is not an observer-
invariant concept. (However, rigidity is observer-invariant, if defined properly; can you
Challenge 591 n confirm this?)
This explanation, though published, is not correct, as Harald van Lintel and Christian
Copyright © Christoph Schiller November 1997–May 2006

Ref. 275 Gruber have pointed out. One should not forget to estimate the size of the effect. At re-
lativistic speeds the time required for the hole to affect the full thickness of the board
cannot be neglected. The snowboarder only sees his board take on a parabolic shape if
it is extremely thin and flexible. For usual boards moving at relativistic speeds, the snow-
Challenge 592 ny boarder has no time to fall any appreciable height h or to bend into the hole before passing
it. Figure 150 is so exaggerated that it is incorrect. The snowboarder would simply speed
over the hole.
The paradoxes around length contraction become even more interesting in the case of
Ref. 276 a conductive glider that makes electrical contact between two rails, as shown in Figure 151.
Challenge 589 n * Even the Earth contracts in its direction of motion around the Sun. Is the value measurable?

Dvipsbugw
 302 ii special rel ativity • 5. speed, rest and light

The two rails are parallel, but one rail has a gap that is longer than the glider. Can you
work out whether a lamp connected in series stays lit when the glider moves along the
Challenge 593 n rails with relativistic speed? (Make the simplifying and not fully realistic assumption that
electrical current flows as long and as soon as the glider touches the rails.) Do you get
the same result for all observers? And what happens when the glider is longer than the
detour? (Warning: this problem gives rise to heated debates!) What is unrealistic in this
experiment?
Ref. 277 Another example of length contraction appears when two objects, say two cars, are
connected over a distance d by a straight rope, as shown in Figure 152 Imagine that both Dvipsbugw
are at rest at time t = 0 and are accelerated together in exactly the same way. The observer
» the same distance apart. On the other hand,
at rest will maintain that the two cars remain
the rope needs to span a distance d = d~ 1 − v 2 ~c 2 , and thus has to expand when the
′

two cars are accelerating. In other words, the rope will break. Is this prediction confirmed
Challenge 594 n by observers on each of the two cars?
A funny – but quite unrealistic – example of length contraction is that of a submarine

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 278 moving horizontally. Imagine that the resting submarine has tuned its weight to float in
water without any tendency to sink or to rise. Now the submarine moves (possibly with
relativistic speed). The captain observes the water outside to be Lorentz contracted; thus
the water is denser and he concludes that the submarine will rise. A nearby fish sees the
submarine to be contracted, thus denser than water, and concludes that the submarine
Challenge 595 n will sink. Who is wrong, and what is the buoyancy force? Alternatively, answer the fol-
Challenge 596 n lowing question: why is it impossible for a submarine to move at relativistic speed?
In summary, length contraction is can almost never be realistically observed for mac-
roscopic bodies. However, it does play an important role for images.

Relativistic films – aberration and Doppler effect

We have encountered several ways in which observations change when an observer
moves at high speed. First of all, Lorentz contraction and aberration lead to distorted
images. Secondly, aberration increases the viewing angle beyond the roughly 180 degrees
that humans are used to in everyday life. A relativistic observer who looks in the direc-
tion of motion sees light that is invisible for a resting observer, because for the latter, it
comes from behind. Thirdly, the Doppler effect produces colour-shifted images. Fourthly,
the rapid motion changes the brightness and contrast of the image: the so-called search-
light effect. Each of these changes depends on the direction of sight; they are shown in
Copyright © Christoph Schiller November 1997–May 2006

Figure 154.
Modern computers enable us to simulate the observations made by rapid observers
with photographic quality, and even to produce simulated films.* The images of Figure 153
are particularly helpful in allowing us to understand image distortion. They show the
viewing angle, the circle which distinguish objects in front of the observer from those
behind the observer, the coordinates of the observer’s feet and the point on the horizon

* See for example images and films at http://www.anu.edu.au/Physics/Searle/ by Anthony Searle, at

http://www.tat.physik.uni-tuebingen.de/~weiskopf/gallery/index.html by Daniel Weiskopf, at http://www.
itp.uni-hannover.de/~dragon/stonehenge/stone1.htm by Norbert Dragon and Nicolai Mokros, or at http://
www.tempolimit-lichtgeschwindigkeit.de by Hanns Ruder’s group.

Dvipsbugw
curiosities of special rel ativity 303

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

F I G U R E 153 Flying through twelve vertical columns (shown in the two uppermost images) with 0.9
times the speed of light as visualized by Nicolai Mokros and Norbert Dragon, showing the effect of
speed and position on distortions (© Nicolai Mokros)

Dvipsbugw
 304 ii special rel ativity • 5. speed, rest and light

F I G U R E 154 Flying through three straight and vertical columns with 0.9 times the speed of light as Dvipsbugw
visualized by Daniel Weiskopf: on the left with the original colours; in the middle including the
Doppler effect; and on the right including brightness effects, thus showing what an observer would
actually see (© Daniel Weiskopf )

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 155 What a researcher standing and one running rapidly through a corridor observe
Copyright © Christoph Schiller November 1997–May 2006

(ignoring colour effects) (© Daniel Weiskopf )

toward which the observer is moving. Adding these markers in your head when watching
other pictures or films may help you to understand more clearly what they show.
We note that the shape of the image seen by a moving observer is a distorted version
of that seen by one at rest at the same point. A moving observer, however, does not see
different things than a resting one at the same point. Indeed, light cones are independent
of observer motion.
The Lorentz contraction is measurable; however, it cannot be photographed. This sur-
Ref. 279 prising distinction was discovered only in 1959. Measuring implies simultaneity at the ob-

Dvipsbugw
 curiosities of special rel ativity 305

ject’s position; photographing implies simultaneity at the observer’s position. On a photo-

graph, the Lorentz contraction is modified by the effects due to different light travel times
from the different parts of an object; the result is a change in shape that is reminiscent of,
but not exactly the same as, a rotation. The total deformation is an angle-dependent aber-
Page 277 ration. We discussed aberration at the beginning of this section. Aberration transforms
circles into circles: such transformations are called conformal.
The images of Figure 155, produced by Daniel Weiskopf, also include the Doppler effect
and the brightness changes. They show that these effects are at least as striking as the
distortion due to aberration. Dvipsbugw
This leads to the ‘pearl necklace paradox’. If the relativistic motion transforms spheres
into spheres, and rods into shorter rods, what happens to a pearl necklace moving along
Challenge 597 n its own long axis? Does it get shorter?
There is much more to be explored using relativistic films. For example, the author
Challenge 598 r predicts that films of rapidly rotating spheres in motion will reveal interesting effects.
Also in this case, optical observation and measurement results will differ. For certain

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

combinations of relativistic rotations and relativistic boosts, it is predicted* that the sense
of rotation (clockwise or anticlockwise) will differ for different observers. This effect will
play an interesting role in the discussion of unification.

Which is the best seat in a bus?

Ref. 277 Let us explore another surprise of special relativity. Imagine two twins inside two identic-
ally accelerated cars, one in front of the other, starting from standstill at time t = 0, as
described by an observer at rest with respect to both of them. (There is no connecting
rope now.) Both cars contain the same amount of fuel. We easily deduce that the accele-
ration of the two twins stops, when the fuel runs out, at the same time in the frame of
Challenge 599 e the outside observer. In addition, the distance between the cars has remained the same
all along for the outside observer, and the two cars continue rolling with an identical con-
stant velocity v, as long as friction is negligible. If we call the events at which the front car
and back car engines switch off f and b, their time coordinates in the outside frame are
related simply by t f = t b . By using the Lorentz transformations you can deduce for the
Challenge 600 ny frame of the freely rolling twins the relation

t b = γ∆x v~c 2 + t f , (119) Copyright © Christoph Schiller November 1997–May 2006

which means that the front twin has aged more than the back twin! Thus, in accelerated
systems, ageing is position-dependent.
For choosing a seat in a bus, though, this result does not help. It is true that the best
seat in an accelerating bus is the back one, but in a decelerating bus it is the front one. At
the end of a trip, the choice of seat does not matter.
Is it correct to deduce that people on high mountains age faster than people in valleys,
Challenge 601 n so that living in a valley helps postponing grey hair?

* In July 2005.

Dvipsbugw
 306 ii special rel ativity • 5. speed, rest and light

average speed: c/2

time time
average speed: c/3
t'

moving t'
judge

J light signal
moving Dvipsbugw
x'
judge
J
light signal x'

space space

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 156 For the athlete on the left, the judge moving in the opposite
direction sees both feet off the ground at certain times, but not for the
athlete on the right

How fast can one walk?

To walk means to move the feet in such a way that at least one of them is on the ground at
any time. This is one of the rules athletes have to follow in Olympic walking competitions;
they are disqualified if they break it. A student athlete was thinking about the theoretical
maximum speed he could achieve in the Olympics. The ideal would be that each foot
accelerates instantly to (almost) the speed of light. The highest walking speed is achieved
by taking the second foot off the ground at exactly the same instant at which the first is put
down. By ‘same instant’, the student originally meant ‘as seen by a competition judge at
rest with respect to Earth’. The motion of the feet is shown in the left diagram of Figure 156;
it gives a limit speed for walking of half the speed of light. But then the student noticed
that a moving judge will see both feet off the ground and thus disqualify the athlete for
Ref. 280 running. To avoid disqualification by any judge, the second foot has to wait for a light
signal from the first. The limit speed for Olympic walking is thus only one third of the
Copyright © Christoph Schiller November 1997–May 2006

speed of light.

Is the speed of shadow greater than the speed of light?

Actually, motion faster than light does exist and is even rather common. Special relativity
only constrains the motion of mass and energy. However, non-material points or non-
energy-transporting features and images can move faster than light. There are several
Page 297 simple examples. To be clear, we are not talking about proper velocity, which in these
Challenge 602 n cases cannot be defined anyway. (Why?)
The following examples show speeds that are genuinely higher than the speed of light
in vacuum.
Consider the point marked X in Figure 157, the point at which scissors cut paper. If

Dvipsbugw
 curiosities of special rel ativity 307

The Beatles

The Beatles

v Dvipsbugw
X

The Beatles

F I G U R E 157 A simple example of motion that is F I G U R E 158 Another

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

faster than light example of faster-than-light
motion

the scissors are closed rapidly enough, the point moves faster than light. Similar examples
can also be found in every window frame, and in fact in any device that has twisting parts.
Another example of superluminal motion is a music record – an old-fashioned LP
– disappearing into its sleeve, as shown in Figure 158. The point where the edge of the
record meets the edge of the sleeve can travel faster than light.
Another example suggests itself when we remember that we live on a spherical planet.
Imagine you lie on the floor and stand up. Can you show that the initial speed with which
Challenge 603 n the horizon moves away from you can be larger than that of light?
Finally, a standard example is the motion of a spot of light produced by shining a laser
beam onto the Moon. If the laser is moved, the spot can easily move faster than light. The
same applies to the light spot on the screen of an oscilloscope when a signal of sufficiently
high frequency is fed to the input.
All these are typical examples of the speed of shadows, sometimes also called the speed
of darkness. Both shadows and darkness can indeed move faster than light. In fact, there
Copyright © Christoph Schiller November 1997–May 2006

Challenge 604 n is no limit to their speed. Can you find another example?
In addition, there is an ever-increasing number of experimental set-ups in which the
phase velocity or even the group velocity of light is higher than c. They regularly make
headlines in the newspapers, usually along the lines of ‘light moves faster than light’. We
Page 577 will discuss this surprising phenomenon in more detail later on. In fact, these cases can
also be seen – with some imagination – as special cases of the ‘speed of shadow’ phe-
nomenon.
For a different example, imagine we are standing at the exit of a tunnel of length l. We
see a car, whose speed we know to be v, entering the other end of the tunnel and driving
towards us. We know that it entered the tunnel because the car is no longer in the Sun or
because its headlights were switched on at that moment. At what time t, after we see it

Dvipsbugw
 308 ii special rel ativity • 5. speed, rest and light

time
observer
emitted or reflected light

tachyon

Dvipsbugw

light cone

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

space

F I G U R E 159 Hypothetical space-time diagram for

tachyon observation

entering the tunnel, does it drive past us? Simple reasoning shows that t is given by

t = l~v − l~c . (120)

In other words, the approaching car seems to have a velocity v appr of

v appr = =
l vc
, (121)
t c−v

which is higher than c for any car velocity v higher than c~2. For cars this does not happen
too often, but astronomers know a type of bright object in the sky called a quasar (a
contraction of ‘quasi-stellar object’), which sometimes emits high-speed gas jets. If the
emission is in or near the direction of the Earth, its apparent speed – even the purely
transverse component – is higher than c. Such situations are now regularly observed with
telescopes.
Copyright © Christoph Schiller November 1997–May 2006

Ref. 281
Note that to a second observer at the entrance of the tunnel, the apparent speed of the
car moving away is given by
v leav =
vc
, (122)
c+v

which is never higher than c~2. In other words, objects are never seen departing with
more than half the speed of light.
The story has a final twist. We have just seen that motion faster than light can be ob-
served in several ways. But could an object moving faster than light be observed at all?
Surprisingly, it could be observed only in rather unusual ways. First of all, since such
an imaginary object, usually called a tachyon, moves faster than light, we can never see

Dvipsbugw
 curiosities of special rel ativity 309

R v G
u
O w

F I G U R E 160 If O’s stick is parallel

to R’s and R’s is parallel to G’s, then Dvipsbugw
O’s stick and G’s stick are not

it approaching. If it can be seen at all, a tachyon can only be seen departing. Seeing a
tachyon would be similar to hearing a supersonic jet. Only after a tachyon has passed
nearby, assuming that it is visible in daylight, could we notice it. We would first see a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

flash of light, corresponding to the bang of a plane passing with supersonic speed. Then
we would see two images of the tachyon, appearing somewhere in space and departing
in opposite directions, as can be deduced from Figure 159. Even if one of the two images
were approaching us, it would be getting fainter and smaller. This is, to say the least, rather
unusual behaviour. Moreover, if you wanted to look at a tachyon at night, illuminating
it with a torch, you would have to turn your head in the direction opposite to the arm
with the torch! This requirement also follows from the space-time diagram: can you see
Challenge 605 e why? Nobody has ever seen such phenomena. Tachyons, if they existed, would be strange
Ref. 282 objects: they would accelerate when they lose energy, a zero-energy tachyon would be the
Page 316 fastest of all, with infinite speed, and the direction of motion of a tachyon depends on the
motion of the observer. No object with these properties has ever been observed. Worse, as
we just saw, tachyons would seem to appear from nothing, defying laws of conservation;
and note that, just as tachyons cannot be seen in the usual sense, they cannot be touched
either, since both processes are due to electromagnetic interactions, as we will see later in
our ascent of Motion Mountain. Tachyons therefore cannot be objects in the usual sense.
In the second part of our adventure we will show that quantum theory actually rules out
the existence of (real) tachyons. However, quantum theory also requires the existence of
‘virtual’ tachyons, as we will discover.

Parallel to parallel is not parallel – Thomas rotation

Copyright © Christoph Schiller November 1997–May 2006

Relativity has strange consequences indeed. Any two observers can keep a stick parallel
to the other’s, even if they are in motion with respect to each other. But strangely, given a
a chain of sticks for which any two adjacent ones are parallel, the first and the last sticks
will not generally be parallel. In particular, they never will be if the motions of the various
observers are in different directions, as is the case when the velocity vectors form a loop.
The simplest set-up is shown in Figure 160. In special relativity, a general concatenation
Ref. 283 of pure boosts does not give a pure boost, but a boost plus a rotation. As a result, the
endpoints of chains of parallel sticks are usually not parallel.
An example of this effect appears in rotating motion. If we walk in a fast circle holding
a stick, always keeping the stick parallel to the direction it had just before, at the end of the
circle the stick will have an angle with respect to the original direction. Similarly, the axis

Dvipsbugw
 310 ii special rel ativity • 5. speed, rest and light

of a rotating body circling a second body will not be pointing in the same direction after
one turn. This effect is called Thomas precession, after Llewellyn Thomas, who discovered
it in 1925, a full 20 years after the birth of special relativity. It had escaped the attention of
dozens of other famous physicists. Thomas precession is important in the inner working
of atoms; we will return to it in a later section of our adventure. These surprising phenom-
ena are purely relativistic, and are thus measurable only in the case of speeds comparable
to that of light.

A never-ending story – temperature and relativity Dvipsbugw

The literature on temperature is confusing. Albert Einstein and Wolfgang Pauli agreed
on the following result: the temperature T seen by an observer moving with speed v is
related to the temperature T0 measured by the observer at rest with respect to the heat
bath via »
T = T0 1 − v 2 ~c 2 . (123)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

A moving observer thus always measures lower values than a resting one.
In 1908, Max Planck used this expression, together with the corresponding transform-
ation for heat, to deduce that the entropy is invariant under Lorentz transformations.
Being the discoverer of the Boltzmann constant k, Planck proved in this way that the
constant is a relativistic invariant.
Not all researchers agree on the expression. Others maintain that T and T0 should
be interchanged in the temperature transformation. Also, powers other than the simple
Ref. 284 square root have been proposed. The origin of these discrepancies is simple: temperature
is only defined for equilibrium situations, i.e. for baths. But a bath for one observer is not
a bath for the other. For low speeds, a moving observer sees a situation that is almost a
heat bath; but at higher speeds the issue becomes tricky. Temperature is deduced from
the speed of matter particles, such as atoms or molecules. For moving observers, there
is no good way to measure temperature. The naively measured temperature value even
depends on the energy range of matter particles that is measured! In short, thermal equi-
librium is not an observer-invariant concept. Therefore, no temperature transformation
formula is correct. (With certain additional assumptions, Planck’s expression does seem
to hold, however.) In fact, there are not even any experimental observations that would al-
low such a formula to be checked. Realizing such a measurement is a challenge for future
experimenters – but not for relativity itself.
Copyright © Christoph Schiller November 1997–May 2006

Rel ativistic mechanics

Because the speed of light is constant and velocities do not add up, we need to rethink
the definitions of mass, momentum and energy. We thus need to recreate a theory of
mechanics from scratch.

Mass in relativity
Page 77 In Galilean physics, the mass ratio between two bodies was defined using collisions; it

Dvipsbugw
 rel ativistic mechanics 311

was given by the negative inverse of the velocity change ratio

=−
m2 ∆v 1
. (124)
m1 ∆v 2

However, experiments show that the expression must be different for speeds near that of
Challenge 606 ny light. In fact, experiments are not needed: thinking alone can show this. Can you do so?
There is only one solution to this problem. The two Galilean conservation theorems
Ref. 285 P i m i v i = const for momentum and P i m i = const for mass have to be changed into Dvipsbugw

Q γ i m i v i = const (125)
i

and
Q γ i m i = const . (126)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

i

These expressions, which will remain valid throughout the rest of our ascent of Motion
Mountain, imply, among other things, that teleportation is not possible in nature. (Can
Challenge 607 n you confirm this?) Obviously, in order to recover Galilean physics, the relativistic correc-
tion (factors) γ i have to be almost equal to 1 for everyday velocities, that is, for velocities
nowhere near the speed of light.
Even if we do not know the value of the relativistic correction factor, we can deduce it
from the collision shown in Figure 161.
In the first frame of reference (A) we have
γv mv = γ V MV and γv m + m = γ V M. From the ob- Observer A
servations of the second frame of reference (B) we m m
Challenge 608 e deduce that V composed with V gives v, in other before: v
words, that after:
v=
2V V
.
1 + V ~c
2 2
(127) M

When these equations are combined, the relativ- Observer B

istic correction γ is found to depend on the mag- before:
nitude of the velocity v through m
V V m
after:
Copyright © Christoph Schiller November 1997–May 2006

γv = »
1 M
. (128)
1 − v 2 ~c 2 F I G U R E 161 An inelastic collision of two
identical particles seen from two
different inertial frames of reference
With this expression, and a generalization of
the situation of Galilean physics, the mass ratio
between two colliding particles is defined as the ratio

∆(γ 2 v 2 )
=−
m1
.
∆(γ 1 v 1 )
(129)
m2

(We do not give here the generalized mass definition, mentioned in the chapter on Ga-

Dvipsbugw
 312 ii special rel ativity • 5. speed, rest and light

before

pA
A B

after
θ
pA ϕ

rule: ϕ+θ = 90° Dvipsbugw

B

F I G U R E 162 A useful rule for playing

non-relativistic snooker

Page 79 lilean mechanics, that is based on acceleration ratios, because it contains some subtleties,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

which we will discover shortly.) The correction factors γ i ensure that the mass defined
by this equation is the same as the one defined in Galilean mechanics, and that it is the
same for all types of collision a body may have.* In this way, mass remains a quantity
characterizing the difficulty of accelerating a body, and it can still be used for systems of
bodies as well.
Following the example of Galilean physics, we call the quantity

p = γmv (130)

the (linear) relativistic (three-) momentum of a particle. Again, the total momentum is a
conserved quantity for any system not subjected to external influences, and this conserva-
tion is a direct consequence of the way mass is defined.
For low speeds, or γ  1, relativistic momentum is the same as that of Galilean physics,
and is proportional to velocity. But for high speeds, momentum increases faster than
velocity, tending to infinity when approaching light speed.

Why relativistic snooker is more difficult

There is a well-known property of collisions between a moving sphere or particle and a
resting one of the same mass that is important when playing snooker, pool or billiards.
After such a collision, the two spheres will depart at a right angle from each other, as
Copyright © Christoph Schiller November 1997–May 2006

shown in Figure 162.

However, experiments show that the right angle rule does not apply to relativistic col-
lisions. Indeed, using the conservation of momentum and a bit of dexterity you can cal-
Challenge 610 ny culate that
tan θ tan φ =
2
, (131)
γ+1

where the angles are defined in Figure 163. It follows that the sum φ + θ is smaller than
a right angle in the relativistic case. Relativistic speeds thus completely change the game

Challenge 609 e * The results below also show that γ = 1 + T~mc 2 , where T is the kinetic energy of a particle.

Dvipsbugw
 rel ativistic mechanics 313

θ
ϕ

accelerator beam target detector

Dvipsbugw
F I G U R E 163 The dimensions of detectors in particle accelerators are based on the relativistic
snooker angle rule

of snooker. Indeed, every accelerator physicist knows this: for electrons or protons, these
angles can easily be deduced from photographs taken in cloud chambers, which show

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the tracks left by particles when they move through them. All such photographs confirm
Ref. 252 the above expression. In fact, the shapes of detectors are chosen according to expression
(131), as sketched in Figure 163. If the formula – and relativity – were wrong, most of these
detectors would not work, as they would miss most of the particles after the collision. In
fact, these experiments also prove the formula for the composition of velocities. Can you
Challenge 611 ny show this?

Mass is concentrated energy

Let us go back to the collinear and inelastic collision of Figure 161. What is the mass M
Challenge 612 n of the final system? Calculation shows that
»
M~m = 2(1 + γv ) A 2 . (132)

In other words, the mass of the final system is larger than the sum of the two original
masses m. In contrast to Galilean mechanics, the sum of all masses in a system is not a
conserved quantity. Only the sum P i γ i m i of the corrected masses is conserved.
Relativity provides the solution to this puzzle. Everything falls into place if, for the
energy E of an object of mass m and velocity v, we use the expression Copyright © Christoph Schiller November 1997–May 2006

mc 2
E = γmc 2 = » , (133)
1 − v 2 ~c 2

applying it both to the total system and to each component. The conservation of the cor-
rected mass can then be read as the conservation of energy, simply without the factor c 2 .
In the example of the two identical masses sticking to each other, the two particles are
thus each described by mass and energy, and the resulting system has an energy E given
by the sum of the energies of the two particles. In particular, it follows that the energy E 0
of a body at rest and its mass m are related by

E 0 = mc 2 , (134)

Dvipsbugw
 314 ii special rel ativity • 5. speed, rest and light

which is perhaps the most beautiful and famous discovery of modern physics. Since c 2 is
so large, we can say that mass is concentrated energy. In other words, special relativity says
that every mass has energy, and that every form of energy in a system has mass. Increasing
the energy of a system increases its mass, and decreasing the energy content decreases the
mass. In short, if a bomb explodes inside a closed box, the mass, weight and momentum
of the box are the same before and after the explosion, but the combined mass of the
debris inside the box will be smaller than before. All bombs – not only nuclear ones –
thus take their energy from a reduction in mass. In addition, every action of a system –
such a caress, a smile or a look – takes its energy from a reduction in mass. Dvipsbugw
The kinetic energy T is thus given by

1 ë 3 v4 1 ë 3 ë 5 v6
T = γmc 2 − mc 2 = mv 2 +
1
m + + ... (135)
2 2 ë 4 c2 2 ë 4 ë 6 c4
Challenge 613 e (using the binomial theorem) which reduces to the Galilean value only for low speeds.
The mass–energy equivalence E = γmc 2 implies that taking any energy from matter

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

results in a mass decrease. When a person plays the piano, thinks or runs, its mass de-
creases. When a cup of tea cools down or when a star shines, its mass decreases. The
mass–energy equivalence pervades all of nature.
By the way, we should be careful to distinguish the transformation of mass into energy
from the transformation of matter into energy. The latter is much more rare. Can you
Challenge 614 n give some examples?
The mass–energy relation (133) means the death of many science fiction fantasies. It
implies that there are no undiscovered sources of energy on or near Earth. If such sources
existed, they would be measurable through their mass. Many experiments have looked
for, and are still looking for, such effects with a negative result. There is no free energy in
nature.*
The mass–energy relation m = E 0 ~c 2 also implies that one needs about 90 thousand
million kJ (or 21 thousand million kcal) to increase one’s weight by one single gram. Of
course, dieticians have slightly different opinions on this matter! In fact, humans do get
their everyday energy from the material they eat, drink and breathe by reducing its com-
bined mass before expelling it again. However, this chemical mass defect appearing when
fuel is burned cannot yet be measured by weighing the materials before and after the
reaction: the difference is too small, because of the large conversion factor involved. In-
deed, for any chemical reaction, bond energies are about 1 aJ (6 eV) per bond; this gives
Copyright © Christoph Schiller November 1997–May 2006

a weight change of the order of one part in 1010 , too small to be measured by weigh-
ing people or determining mass differences between food and excrement. Therefore, for
everyday chemical processes mass can be taken to be constant, in accordance with Ga-
lilean physics.
Modern methods of mass measurement of single molecules have made it possible to
measure the chemical mass defect by comparing the mass of a single molecule with
that of its constituent atoms. David Pritchard’s group has developed so-called Penning

* There may be two extremely diluted, yet undiscovered, form of energy, called dark matter and (confusingly)
Page 442 dark energy, scattered throughout the universe. They are deduced from (quite difficult) mass measurements.
The issue has not yet been finally resolved.

Dvipsbugw
 rel ativistic mechanics 315

traps, which allow masses to be determined from the measurement of frequencies; the
attainable precision of these cyclotron resonance experiments is sufficient to confirm
Ref. 286 ∆E 0 = ∆mc 2 for chemical bonds. In the future, increased precision will even allow bond
energies to be determined in this way with precision. Since binding energy is often radi-
ated as light, we can say that these modern techniques make it possible to weigh light.
Thinking about light and its mass was the basis for Einstein’s first derivation of the
mass–energy relation. When an object emits two equal light beams in opposite directions,
its energy decreases by the emitted amount. Since the two light beams are equal in energy
and momentum, the body does not move. If we describe the same situation from the Dvipsbugw
Challenge 615 ny viewpoint of a moving observer, we see again that the rest energy of the object is

E 0 = mc 2 . (136)

In summary, all physical processes, including collisions, need relativistic treatment

whenever the energy involved is a sizeable fraction of the rest energy.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Every energy increase produces a mass increase. Therefore also heating a body makes
it heavier. However, this effect is so small that nobody has measured it up to this day. It
is a challenge for experiments of the future to do this one day.
How are energy and momentum related? The definitions of momentum (130) and en-
Challenge 616 e ergy (133) lead to two basic relations. First of all, their magnitudes are related by

m 2 c 4 = E 2 − p2 c 2 (137)

for all relativistic systems, be they objects or, as we will see below, radiation. For the mo-
mentum vector we get the other important relation

p=
E
v, (138)
c2
Challenge 617 e which is equally valid for any type of moving energy, be it an object or a beam or a pulse of
radiation.* We will use both relations often in the rest of our ascent of Motion Mountain,
including the following discussion.

Collisions, virtual objects and tachyons

We have just seen that in relativistic collisions the conservation of total energy and mo-
Copyright © Christoph Schiller November 1997–May 2006

mentum are intrinsic consequences of the definition of mass. Let us now have a look at
collisions in more detail, using these new concepts. A collision is a process, i.e. a series of
events, for which
— the total momentum before the interaction and after the interaction is the same;
— the momentum is exchanged in a small region of space-time;
— for small velocities, the Galilean description is valid.
In everyday life an impact, i.e. a short-distance interaction, is the event at which both
objects change momentum. But the two colliding objects are located at different points

* In 4-vector notation, we can write v~c = P~P0 , where P0 = E~c.

Dvipsbugw
 316 ii special rel ativity • 5. speed, rest and light

time t τ

E'2 p'2
E'1 p'1
E
p
E2 p2
E1 p1

Dvipsbugw
object 1
object 2 object 1 object 2

space x ξ

F I G U R E 164 Space-time diagram of a collision for two observers

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 287 when this happens. A collision is therefore described by a space-time diagram such as the
left-hand one in Figure 164, reminiscent of the Orion constellation. It is easy to check that
Challenge 618 e the process described by such a diagram is a collision according to the above definition.
The right-hand side of Figure 164 shows the same process seen from another, Greek,
frame of reference. The Greek observer says that the first object has changed its mo-
mentum before the second one. That would mean that there is a short interval when
momentum and energy are not conserved!
The only way to make sense of the situation is to assume that there is an exchange of a
third object, drawn with a dotted line. Let us find out what the properties of this object are.
If we give numerical subscripts to the masses, energies and momenta of the two bodies,
Challenge 619 e and give them a prime after the collision, the unknown mass m obeys

1 − v 1 v 1′
m 2 c 4 = (E 1 − E 1′ )2 − (p 1 − p′1 )2 c 2 = 2m 12 c 4 − 2E 1 E 1′ ( )<0. (139)
c2
This is a strange result, because it means that the unknown mass is an imaginary num-
ber!!* On top of that, we also see directly from the second graph that the exchanged object
moves faster than light. It is a tachyon, from the Greek ταχύς ‘rapid’. In other words, colli-
sions involve motion that is faster than light! We will see later that collisions are indeed the
Copyright © Christoph Schiller November 1997–May 2006

only processes where tachyons play a role in nature. Since the exchanged objects appear
only during collisions, never on their own, they are called virtual objects, to distinguish
them from the usual, real objects, which can move freely without restriction.** We will
study their properties later on, when we come to discuss quantum theory.
»
»
* It is usual to change the mass–energy and mass–momentum relation of tachyons to E = mc 2 ~ v 2 ~c 2 − 1
and p = mv~ v 2 ~c 2 − 1 ; this amounts to a redefinition of m. After the redefinition, tachyons have real
mass. The energy and momentum relations show that tachyons lose energy and momentum when they get
faster. (Provocatively, a single tachyon in a box could provide us with all the energy we need.) Both signs
for the energy and momentum relations must be retained, because otherwise the equivalence of all inertial
observers would not be generated. Tachyons thus do not have a minimum energy or a minimum momentum.
** More precisely, a virtual particle does not obey the relation m 2 c 4 = E 2 − p2 c 2 , valid for real particles.

Dvipsbugw
 rel ativistic mechanics 317

In nature, a tachyon is always a virtual object. Real objects are always bradyons – from
the Greek βραδύς ‘slow’ – or objects moving slower than light. Note that tachyons, despite
their high velocity, do not allow the transport of energy faster than light; and that they
do not violate causality if and only if they are emitted or absorbed with equal probability.
Challenge 620 ny Can you confirm all this?
When we study quantum theory, we will also discover that a general contact interac-
tion between objects is described not by the exchange of a single virtual object, but by
a continuous stream of virtual particles. For standard collisions of everyday objects, the
interaction turns out to be electromagnetic. In this case, the exchanged particles are vir- Dvipsbugw
tual photons. In other words, when one hand touches another, when it pushes a stone, or
when a mountain supports the trees on it, streams of virtual photons are continuously
Page 733 exchanged.
There is an additional secret hidden in collisions. In the right-hand side of Figure 164,
the tachyon is emitted by the first object and absorbed by the second one. However, it
Challenge 621 n is easy to imagine an observer for which the opposite happens. In short, the direction

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of travel of a tachyon depends on the observer! In fact, this is a hint about antimatter.
In space-time diagrams, matter and antimatter travel in opposite directions. Also the
connection between relativity and antimatter will become more apparent in quantum
Page 759 theory.

Systems of particles – no centre of mass

Relativity also forces us to eliminate the cherished concept of centre of mass. We can see
this already in the simplest example possible: that of two equal objects colliding.
Figure 165 shows that from the viewpoint in which one of two colliding particles is at
rest, there are at least three different ways to define the centre of mass. In other words, the
Ref. 288 centre of mass is not an observer-invariant concept. We can deduce from the figure that
the concept only makes sense for systems whose components move with small velocities
relative to each other. For more general systems, centre of mass is not uniquely definable.
Will this hinder us in our ascent? No. We are more interested in the motion of single
particles than that of composite objects or systems.

Why is most motion so slow?

For most everyday systems, the time intervals measured by two different observers are
practically equal; only at large relative speeds, typically at more than a few per cent of the
Copyright © Christoph Schiller November 1997–May 2006

speed of light, is there a noticeable difference. Most such situations are microscopic. We
have already mentioned the electrons inside a television tube or inside a particle accel-
erator. The particles making up cosmic radiation are another example: their high energy
has produced many of the mutations that are the basis of evolution of animals and plants
on this planet. Later we will discover that the particles involved in radioactivity are also
relativistic.
But why don’t we observe any rapid macroscopic bodies? Moving bodies, including
observers, with relativistic velocities have a property not found in everyday life: when
they are involved in a collision, part of their energy is converted into new matter via
E = γmc 2 . In the history of the universe this has happened so many times that practically
all the bodies still in relativistic motion are microscopic particles.

Dvipsbugw
 318 ii special rel ativity • 5. speed, rest and light

A CM-0 B
v v

transformed CM
A CM-1 B

v=0 v 2v/(1+v2/c2 )

Dvipsbugw
geometrical CM
A CM-2 B

v=0 2 2 2v/(1+v2/c2 )
v/(1+v /c )

momentum CM

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

A CM-3 B

v=0 2 2 1/2 2v/(1+v2/c2 )

v/(1- v /c )

F I G U R E 165 There is no way to define a relativistic centre

of mass

A second reason for the disappearance of rapid relative motion is radiation damping.
Challenge 622 n Can you imagine what happens to charges during collisions, or in a bath of light?
In short, almost all matter in the universe moves with small velocity relative to other
matter. The few known counter-examples are either very old, such as the quasar jets men-
tioned above, or stop after a short time. The huge energies necessary for macroscopic
relativistic motion are still found in supernova explosions, but they cease to exist after
only a few weeks. In summary, the universe is mainly filled with slow motion because it
Page 450 is old. We will determine its age shortly.

The history of the mass–energy equivalence formula of de

Pretto and Einstein
Albert Einstein took several months after his first paper on special relativity to deduce
Copyright © Christoph Schiller November 1997–May 2006

the expression
E = γmc 2 (140)

which is often called the most famous formula of physics. He published it in a second,
Ref. 241 separate paper towards the end of 1905. Arguably, the formula could have been discovered
thirty years earlier, from the theory of electromagnetism.
In fact, at least one person did deduce the result before Einstein. In 1903 and 1904,
before Einstein’s first relativity paper, a little-known Italian engineer, Olinto De Pretto,
was the first to calculate, discuss and publish the formula E = mc 2 .* It might well be that

* Umberto Bartocci, mathematics professor of the University of Perugia in Italy, published the details of

Dvipsbugw
 rel ativistic mechanics 319

lig

ne
future T

ht

co
co

ht
ne

lig
E elsewhere y

Dvipsbugw
x

past

F I G U R E 166 The space-time diagram of a

moving object T

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Einstein got the idea for the formula from De Pretto, possibly through his friend Michele
Besso or other Italian-speaking friends he met when he visited his parents, who were
living in Italy at the time. Of course, the value of Einstein’s efforts is not diminished by
this.
Ref. 289 In fact, a similar formula had also been deduced in 1904 by Friedrich Hasenöhrl and
published again in Annalen der Physik in 1905, before Einstein, though with an incorrect
numerical prefactor, due to a calculation mistake. The formula E = mc 2 is also part of
several expressions in two publications in 1900 by Henri Poincaré. The real hero in the
story might well be Tolver Preston, who discussed the equivalence of mass and energy
already in 1875, in his book Physics of the Ether. The mass-energy equivalence was thus
indeed floating in the air, only waiting to be discovered.
In the 1970s there was a similar story: a simple relation between the gravitational acce-
leration and the temperature of the vacuum was discovered. The result had been waiting
to be discovered for over 50 years. Indeed, a number of similar, anterior results were
found in the libraries. Could other simple relations be hidden in modern physics waiting
Challenge 623 n to be found?

4-vectors
Copyright © Christoph Schiller November 1997–May 2006

To describe motion consistently for all observers, we have to introduce some new quant-
ities. First of all, motion of particles is seen as a sequence of events. To describe events
with precision, we use event coordinates, also called 4-coordinates. These are written as

X = (ct, x) = (ct, x, y, z) = X i . (141)

In this way, an event is a point in four-dimensional space-time, and is described by four

coordinates. The coordinates are called the zeroth, namely time X 0 = ct, the first, usually

this surprising story in several papers. The full account is found in his book Umberto Bartocci, Albert
Einstein e Olinto De Pretto: la vera storia della formula più famosa del mondo, Ultreja, Padova, 1998.

Dvipsbugw
 320 ii special rel ativity • 5. speed, rest and light

called X 1 = x, the second, X 2 = y, and the third, X 3 = z. One can then define a distance d
between events as the length of the difference vector. In fact, one usually uses the square
of the length, to avoid those unwieldy square roots. In special relativity, the magnitude
(‘squared length’) of a vector is always defined through

XX = X 0 2 − X 1 2 − X 2 2 − X 3 2 = ct 2 − x 2 − y 2 − z 2 = X a X a = η ab X a X b = η ab X a X b .(142)

In this equation we have introduced for the first time two notations that are useful in
Dvipsbugw
relativity. First of all, we automatically sum over repeated indices. Thus, X a X a means the
sum of all products X a X a as a ranges over all indices. Secondly, for every 4-vector X
we distinguish two ways to write the coordinates, namely coordinates with superscripts
and coordinates with subscripts. (In three dimensions, we only use subscripts.) They are
related by the following general relation

X a = η ab X b = (ct, −x, −y, −z) , (143)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where we have introduced the so-called metric η ab , an abbreviation of the matrix*

’ 1 0 0 0 “
– —
= η ab =– — .
0 −1 0 0
– —
ab
η (144)
0 0 −1 0
” 0 0 0 −1 •

Don’t panic: this is all, and it won’t get more difficult! We now go back to physics.
The magnitude of a position or distance vector, also called the space-time interval, is
essentially the proper time times c. The proper time is the time shown by a clock moving
in a straight line and with constant velocity from the starting point to the end point in
space-time. The difference from the usual 3-vectors is that the magnitude of the interval
can be positive, negative or even zero. For example, if the start and end points in space-
time require motion with the speed of light, the proper time is zero (this is required for
Page 296 null vectors). If the motion is slower than the speed of light, the squared proper time is
positive and the distance is timelike. For negative intervals and thus imaginary proper
times, the distance is spacelike.** A simplified overview is given by Figure 166.
Now we are ready to calculate and measure motion in four dimensions. The meas-
urements are based on one central idea. We cannot define the velocity of a particle as
Copyright © Christoph Schiller November 1997–May 2006

the derivative of its coordinates with respect to time, since time and temporal sequences
depend on the observer. The solution is to define all observables with respect to the just-
mentioned proper time τ, which is defined as the time shown by a clock attached to the ob-
ject. In relativity, motion and change are always measured with respect to clocks attached
to the moving system. In particular, the relativistic velocity or 4-velocity U of a body is

* Note that 30 % of all physics textbooks use the negative of η as the metric, the so-called spacelike convention,
and thus have opposite signs in this definition. In this text, as in 70 % of all physics texts, we use the timelike
convention.
** In the latter case, the negative of the magnitude, which is a positive number, is called the squared proper
distance. The proper distance is the length measured by an odometer as the object moves along.

Dvipsbugw
 rel ativistic mechanics 321

thus defined as the rate of change of the event coordinates or 4-coordinates X = (ct, x)
with respect to proper time, i.e. as

U = dX~dτ . (145)

The coordinates X are measured in the coordinate system defined by the inertial observer
chosen. The value of the velocity U depends on the observer or coordinate system used;
so the velocity depends on the observer, as it does in everyday life. Using dt = γ dτ and
thus Dvipsbugw
= =γ , where as usual γ = »
dx dx dt dx 1
, (146)
dτ dt dτ dt 1 − v 2 ~c 2

we get the relation with the 3-velocity v = dx~dt:

u 0 = γc , u i = γv i or U = (γc, γv) . (147)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

For small velocities we have γ  1, and then the last three components of the 4-velocity
are those of the usual, Galilean 3-velocity. For the magnitude of the 4-velocity U we find
UU = U a U a = η ab U a U b = c 2 , which is therefore independent of the magnitude of the
3-velocity v and makes it a timelike vector, i.e. a vector inside the light cone.*
Note that the magnitude of a 4-vector can be zero even though all its components
are different from zero. Such a vector is called null. Which motions have a null velocity
Challenge 625 n vector?
Similarly, the relativistic acceleration or 4-acceleration B of a body is defined as

B = dU~dτ = d2 X~dτ 2 . (149)

Using dγ~dτ = γdγ~dt = γ 4 va~c 2 , we get the following relations between the four com-
Ref. 290 ponents of B and the 3-acceleration a = dv~dt:

(va)v i
B0 = γ 4 B i = γ2 a i + γ4
va
, . (150)
c c2

The magnitude b of the 4-acceleration is easily found via BB = η cd B c B d = −γ 4 (a 2 +

γ 2 (va)2 ~c 2 ) = −γ 6 (a 2 − (v  a)2 ~c 2 ). Note that it does depend on the value of the
Copyright © Christoph Schiller November 1997–May 2006

* In general, a 4-vector is defined as a quantity (h 0 , h 1 , h 2 , h 3 ), which transforms as

h′0 = γ V (h 0 − h 1 V ~c)
h 1 = γ V (h 1 − h 0 V ~c)
′

h′2 = h 2
′
h3 = h3 (148)

when changing from one inertial observer to another moving with a relative velocity V in the x direction; the
corresponding generalizations for the other coordinates are understood. This relation allows one to deduce
the transformation laws for any 3-vector. Can you deduce the velocity composition formula (108) from this
Challenge 624 n definition, applying it to 4-velocity?

Dvipsbugw
 322 ii special rel ativity • 5. speed, rest and light

time
(E/c , p)

space

Dvipsbugw

F I G U R E 167 Energy–momentum is tangent to

the world line

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

3-acceleration a. The magnitude of the 4-acceleration is also called the proper accelera-
tion because B2 = −a 2 if v = 0. (What is the connection between 4-acceleration and
Challenge 626 n 3-acceleration for an observer moving with the same speed as the object?) We note
that 4-acceleration lies outside the light cone, i.e. that it is a spacelike vector, and that
BU = η cd B c U d = 0, which means that the 4-acceleration is always perpendicular to the
4-velocity.* We also note that accelerations, in contrast to velocities, cannot be called re-
lativistic: the difference between b i and a i , or between their two magnitudes, does not
depend on the value of a i , but only on the value of the speed v. In other words, acceler-
ations require relativistic treatment only when the involved velocities are relativistic. If
the velocities involved are low, even the highest accelerations can be treated with Galilean
methods.
When the acceleration a is parallel to the velocity v, we get B = γ 3 a; when a is perpen-
Page 330 dicular to v, as in circular motion, we get B = γ 2 a. We will use this result below.

4-momentum
To describe motion, we also need the concept of momentum. The 4-momentum is defined
as
P = mU (153)
Copyright © Christoph Schiller November 1997–May 2006

* Similarly, the relativistic jerk or 4-jerk J of a body is defined as

J = dB~dτ = d2 U~dτ 2 . (151)

Challenge 627 ny For the relation with the 3-jerk j = da~dt we then get

γ5 (va)2 γ5 (va)2 v i
J = (J 0 , J i ) = Œ (jv + a 2 + 4γ 2 2 ) , γ 3 j i + 2 ((jv)v i + a 2 v i + 4γ 2 + 3(va)a i ) ‘ (152)
c c c c2

Challenge 628 ny which we will use later on. Surprisingly, J does not vanish when j vanishes. Why not?

Dvipsbugw
 rel ativistic mechanics 323

and is therefore related to the 3-momentum p by

P = (γmc, γmv) = (E~c, p) . (154)

For this reason 4-momentum is also called the energy–momentum 4-vector. In short, the
4-momentum of a body is given by mass times 4-displacement per proper time. This is the
simplest possible definition of momentum and energy. The concept was introduced by
Max Planck in 1906. The energy–momentum 4-vector, also called momenergy, like the
4-velocity, is tangent to the world line of a particle. This connection, shown in Figure 167, Dvipsbugw
follows directly from the definition, since

(E~c, p) = (γmc, γmv) = m(γc, γv) = m(dt~dτ, dx~dτ) . (155)

The (square of the) length of momenergy, namely PP = η ab P a P b , is by definition the

same for all inertial observers; it is found to be

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

E 2 ~c 2 − p2 = m 2 c 2 , (156)

thus confirming a result given above. We have already mentioned that energies or situ-
ations are called relativistic if the kinetic energy T = E − E 0 is not negligible when com-
pared to the rest energy E 0 = mc 2 . A particle whose kinetic energy is much higher than
its rest mass is called ultrarelativistic. Particles in accelerators or in cosmic rays fall into
Challenge 629 n this category. (What is their energy–momentum relation?)
In contrast to Galilean mechanics, relativity implies an absolute zero for the energy.
One cannot extract more energy than mc 2 from a system of mass m. In particular, a zero
value for potential energy is fixed in this way. In short, relativity shows that energy is
bounded from below.
Note that by the term ‘mass’ m we always mean what is sometimes called the rest mass.
This name derives from the bad habit of many science fiction and secondary-school books
of calling the product γm the relativistic mass. Workers in the field usually (but not unan-
Ref. 291 imously) reject this concept, as did Einstein himself, and they also reject the often-heard
expression that ‘(relativistic) mass increases with velocity’. Relativistic mass and energy
would then be two words for the same concept: this way to talk is at the level of the tabloid
press.
Not all Galilean energy contributes to mass. Potential energy in an outside field does
Copyright © Christoph Schiller November 1997–May 2006

not. Relativity forces us into precise energy bookkeeping. ‘Potential energy’ in relativity
is an abbreviation for ‘energy reduction of the outside field’.
Can you show that for two particles with momenta P1 and P2 , one has P1 P2 = m 1 E 2 =
Challenge 630 n M 2 E 1 = c 2 γv 12 m 1 m 2 , where v 12 is their relative velocity?

4-force
The 4-force K is defined as
K = dP~dτ = mB . (157)

Dvipsbugw
 324 ii special rel ativity • 5. speed, rest and light

Therefore force remains equal to mass times acceleration in relativity. From the definition
Ref. 290, Ref. 292 of K we deduce the relation with 3-force f = dp~dt = md(γv)~dt, namely*

K = (K 0 , K i ) = (γ 4 mva~c, γ 2 ma i + γ 4 v i )=( , γ ) = (γ , γf) .

mva γ dE dp fv
(158)
c 2 c dt dt c
Challenge 632 e The 4-force, like the 4-acceleration, is orthogonal to the 4-velocity. The meaning of the
zeroth component of the 4-force can easily be discerned: it is the power required to accel-
erate the object. One has KU = c 2 dm~dτ = γ 2 (dE~dt−fv): this is the proper rate at which Dvipsbugw
the internal energy of a system increases. The product KU vanishes only for rest-mass-
conserving forces. Particle collisions that lead to reactions do not belong to this class.
In everyday life, the rest mass is preserved, and then one gets the Galilean expression
fv = dE~dt.

Rotation in relativity

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

If at night we turn around our own axis while looking at the sky, the stars move with a
velocity much higher than that of light. Most stars are masses, not images. Their speed
should be limited by that of light. How does this fit with special relativity?
This example helps to clarify in an-
other way what the limit velocity ac- C
tually is. Physically speaking, a rotat-
ing sky does not allow superluminal A v
energy transport, and thus does not
v
contradict the concept of a limit v'
speed. Mathematically speaking, the B v'
speed of light limits relative velocities D
only between objects that come near F I G U R E 168 On the definition of relative velocity
to each other, as shown on the left of
Figure 168. To compare velocities of distant objects is only possible if all velocities in-
volved are constant in time; this is not the case in the present example. The differential
version of the Lorentz transformations make this point particularly clear. In many gen-
eral cases, relative velocities of distant objects can be higher than the speed of light. We
Page 307 encountered one example earlier, when discussing the car in the tunnel, and we will en-
Page 337 counter a few more examples shortly.
With this clarification, we can now briefly consider rotation in relativity. The first ques-
Copyright © Christoph Schiller November 1997–May 2006

tion is how lengths and times change in a rotating frame of reference. You may want
to check that an observer in a rotating frame agrees with a non-rotating colleague on
the radius of a rotating body; however, both find that the rotating body, even if it is ri-
Challenge 633 e gid, has a circumference different from the one it had before it started rotating. Sloppily
speaking, the value of π changes for rotating observers. The ratio between the circumfer-
Challenge 634 e ence c and the radius r turns out to be c~r = 2πγ: it increases with rotation speed. This

* Some authors define 3-force as dp~dτ; then K looks slightly different. In any case, it is important to note that
in relativity, 3-force f = dp~dt is indeed proportional to 3-acceleration a; however, force and acceleration are
Challenge 631 n not parallel to each other. In fact, for rest-mass-preserving forces one finds f = γma + (fv)v~c 2 . In contrast,
in relativity 3-momentum is not proportional to 3-velocity, although it is parallel to it.

Dvipsbugw
 rel ativistic mechanics 325

Ref. 293 counter-intuitive result is often called Ehrenfest’s paradox. Among other things, it shows
that space-time for an observer on a rotating disc is not the Minkowski space-time of
special relativity.
Rotating bodies behave strangely in many ways. For ex-
ample, one gets into trouble when one tries to synchronize O3 O2 O
1
clocks mounted on a rotating circle,as shown in Figure 169 On
If one starts synchronizing the clock at O2 with that at O1 , On–1
and so on, continuing up to clock On , one finds that the last
clock is not synchronized with the first. This result reflects Dvipsbugw
the change in circumference just mentioned. In fact, a care-
ful study shows that the measurements of length and time
intervals lead all observers Ok to conclude that they live in
a rotating space-time. Rotating discs can thus be used as an
introduction to general relativity, where this curvature and F I G U R E 169 Observers on a
its effects form the central topic. More about this in the next rotating object

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

chapter.
Is angular velocity limited? Yes: the tangential speed in an inertial frame of reference
cannot exceed that of light. The limit thus depends on the size of the body in question.
Challenge 635 ny That leads to a neat puzzle: can one see objects rotating very rapidly?
We mention that 4-angular momentum is defined naturally as

l ab = x a pb − x b p a . (159)

In other words, 4-angular momentum is a tensor, not a vector, as shown by its two indices.
Challenge 636 ny Angular momentum is conserved in special relativity. The moment of inertia is naturally
defined as the proportionality factor between angular velocity and angular momentum.
Obviously, for a rotating particle, the rotational energy is part of the rest mass. You may
Challenge 637 ny want to calculate the fraction for the Earth and the Sun. It is not large. By the way, how
Challenge 638 ny would you determine whether a microscopic particle, too small to be seen, is rotating?
In relativity, rotation and translation combine in strange ways. Imagine a cylinder in
uniform rotation along its axis, as seen by an observer at rest. As Max von Laue has dis-
cussed, the cylinder will appear twisted to an observer moving along the rotation axis.
Challenge 639 e Can you confirm this?
Here is a last puzzle about rotation. Velocity is relative; this means that the measured
Challenge 640 ny value depends on the observer. Is this the case also for angular velocity?
Copyright © Christoph Schiller November 1997–May 2006

Wave motion
In Galilean physics, a wave is described by a wave vector and a frequency. In special
relativity, the two are combined in the wave 4-vector, given by

L= ( , n) ,
1 ω
(160)
λ c

Dvipsbugw
 326 ii special rel ativity • 5. speed, rest and light

where λ is the wavelength, ω the wave velocity, and n the normed direction vector. Sup-
pose an observer with 4-velocity U finds that a wave L has frequency ν. Show that

ν = LU (161)

Challenge 641 ny must be obeyed. Interestingly, the wave velocity ω transforms in a different way than
Ref. 246 particle velocity except in the case ω = c. Also the aberration formula for wave motion
Challenge 642 ny differs from that for particles, except in the case ω = c.
Dvipsbugw
The action of a free particle – how do things move?
If we want to describe relativistic motion of a free particle in terms of an extremal prin-
Page 176 ciple, we need a definition of the action. We already know that physical action is a meas-
ure of the change occurring in a system. For an inertially moving or free particle, the only
change is the ticking of its proper clock. As a result, the action of a free particle will be

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

proportional to the elapsed proper time. In order to get the standard unit of energy times
time, or Js, for the action, the first guess for the action of a free particle is

S = −mc 2
τ2
∫τ1
dτ , (162)

where τ is the proper time along its path. This is indeed the correct expression. It im-
plies conservation of (relativistic) energy and momentum, as the change in proper time
Challenge 643 ny is maximal for straight-line motion with constant velocity. Can you confirm this? Indeed,
in nature, all particles move in such a way that their proper time is maximal. In other
words, we again find that in nature things change as little as possible. Nature is like a
wise old man: its motions are as slow as possible. If you prefer, every change is maximally
effective. As we mentioned before, Bertrand Russell called this the law of cosmic laziness.
The expression (162) for the action is due to Max Planck. In 1906, by exploring it in
detail, he found that the quantum of action ħ, which he had discovered together with the
Boltzmann constant, is a relativistic invariant (like the Boltzmann constant k). Can you
Challenge 644 ny imagine how he did this?
The action can also be written in more complex, seemingly more frightening ways.
These equivalent ways to write it are particularly appropriate to prepare for general re-
lativity:
¾
Copyright © Christoph Schiller November 1997–May 2006

»
S= ∫ L dt = −mc ∫ dt = −mc u a u a dτ = −mc
t2 1 τ2 dx a dx bs2
2
t1 γ τ1
∫ s1
∫
ds ds
ds , η ab
(163)
where s is some arbitrary, but monotonically increasing, function of τ, such as τ itself. As
usual, the metric η α β of special relativity is

’ 1 0 0 0 “
– —
= η ab =– — .
0 −1 0 0
– —
ab
η (164)
0 0 −1 0
” 0 0 0 −1 •

Dvipsbugw
 rel ativistic mechanics 327

You can easily confirm the form of the action (163) by deducing the equation of motion
Challenge 645 ny in the usual way.
In short, nature is in not a hurry: every object moves in a such way that its own clock
shows the longest delay possible, compared with any alternative motion nearby.* This
general principle is also valid for particles under the influence of gravity, as we will see
in the section on general relativity, and for particles under the influence of electric or
magnetic interactions. In fact, it is valid in all cases of (macroscopic) motion found in
nature. For the moment, we just note that the longest proper time is realized when the
Challenge 647 ny difference between kinetic and potential energy is minimal. (Can you confirm this?) For Dvipsbugw
the Galilean case, the longest proper time thus implies the smallest average difference
between the two energy types. We thus recover the principle of least action in its Galilean
formulation.
Page 176 Earlier on, we saw that the action measures the change going on in a system. Special re-
lativity shows that nature minimizes change by maximizing proper time. In nature, proper
time is always maximal. In other words, things move along paths of maximal ageing. Can

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 648 ny you explain why ‘maximal ageing’ and ‘cosmic laziness’ are equivalent?
We thus again find that nature is the opposite of a Hollywood movie: nature changes
in the most economical way possible. The deeper meaning of this result is left to your
personal thinking: enjoy it!

Conformal transformations – why is the speed of light constant?

The distinction between space and time in special relativity depends on the observer. On
the other hand, all inertial observers agree on the position, shape and orientation of the
light cone at a point. Thus, in the theory of relativity, the light cones are the basic physical
‘objects’. Given the importance of light cones, we might ask if inertial observers are the
only ones that observe the same light cones. Interestingly, it turns out that other observers
do as well.
The first such category of observers are those using units of measurement in which all
time and length intervals are multiplied by a scale factor λ. The transformations among
these points of view are given by
x a ( λx a (165)

and are called dilations.

A second category of additional observers are found by applying the so-called special
Copyright © Christoph Schiller November 1997–May 2006

conformal transformations. These are compositions of an inversion

xa
xa ( (166)
x2

with a translation by a vector b a , namely

xa ( xa + ba , (167)

* If neutrinos were massless, the action (163) would not be applicable for them. Why? Can you find an
Challenge 646 ny alternative for this (admittedly academic) case?

Dvipsbugw
 328 ii special rel ativity • 5. speed, rest and light

and a second inversion. Thus the special conformal transformations are

xa + ba x 2 xa xa
xa ( or ( 2 + ba . (168)
1 + 2b a x a + b 2 x 2 x 2 x

These transformations are called conformal because they do not change angles of (infin-
Challenge 649 ny itesimally) small shapes, as you may want to check. They therefore leave the form (of in-
finitesimally small objects) unchanged. For example, they transform infinitesimal circles
into infinitesimal circles. They are called special because the full conformal group includes Dvipsbugw
the dilations and the inhomogeneous Lorentz transformations as well.*
Note that the way in which special conformal transformations leave light cones invari-
Challenge 651 ny ant is rather subtle.
Since dilations do not commute with time translations, there is no conserved quantity
associated with this symmetry. (The same is true of Lorentz boosts.) In contrast, rotations
and spatial translations do commute with time translations and thus do lead to conserved

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

quantities.
In summary, vacuum is conformally invariant – in the special sense just mentioned
– and thus also dilation invariant. This is another way to say that vacuum alone is not
sufficient to define lengths, as it does not fix a scale factor. As we would expect, matter
is necessary to do so. Indeed, (special) conformal transformations are not symmetries of
situations containing matter. Only vacuum is conformally invariant; nature as a whole is
not.
However, conformal invariance, or the invariance of light cones, is sufficient to al-
low velocity measurements. Conformal invariance is also necessary for velocity measure-
Challenge 652 ny ments, as you might want to check.
We have seen that conformal invariance implies inversion symmetry: that is, that the
large and small scales of a vacuum are related. This suggest that the constancy of the speed
of light is related to the existence of inversion symmetry. This mysterious connection
gives us a glimpse of the adventures we will encounter in the third part of our ascent of
Motion Mountain. Conformal invariance turns out to be an important property that will
lead to some incredible insights.**

Challenge 650 ny * The set of all special conformal transformations forms a group with four parameters; adding dilations
and the inhomogeneous Lorentz transformations one gets fifteen parameters for the full conformal group.
The conformal group is locally isomorphic to SU(2,2) and to the simple group SO(4,2): these concepts are
Copyright © Christoph Schiller November 1997–May 2006

Page 1194 explained in Appendix D. Note that all this is true only for four space-time dimensions; in two dimensions
– the other important case, especially in string theory – the conformal group is isomorphic to the group of
arbitrary analytic coordinate transformations, and is thus infinite-dimensional.
** The conformal group does not appear only in the kinematics of special relativity: it is the symmetry
group of all physical interactions, such as electromagnetism, provided that all the particles involved have
zero mass, as is the case for the photon. A field that has mass cannot be conformally invariant; therefore
conformal invariance is not an exact symmetry of all of nature. Can you confirm that a mass term mφ 2 in a
Challenge 653 ny Lagrangian is not conformally invariant?
However, since all particles observed up to now have masses that are many orders of magnitude smaller
than the Planck mass, it can be said that they have almost vanishing mass; conformal symmetry can then
be seen as an approximate symmetry of nature. In this view, all massive particles should be seen as small
corrections, or perturbations, of massless, i.e. conformally invariant, fields. Therefore, for the construction
of a fundamental theory, conformally invariant Lagrangians are often assumed to provide a good starting

Dvipsbugw
 accelerating observers 329

Accelerating observers
So far, we have only studied what inertial, or free-flying, observers say to each other when
they talk about the same observation. For example, we saw that moving clocks always
run slow. The story gets even more interesting when one or both of the observers are
accelerating.
One sometimes hears that special relativity cannot be used to describe accelerating
observers. That is wrong, just as it is wrong to say that Galilean physics cannot be used for
accelerating observers. Special relativity’s only limitation is that it cannot be used in non- Dvipsbugw
flat, i.e. curved, space-time. Accelerating bodies do exist in flat space-times, and therefore
they can be discussed in special relativity.
As an appetizer, let us see what an acceler-
ating, Greek, observer says about the clock
Ref. 294 of an inertial, Roman, one, and vice versa. observer (Greek) v
Assume that the Greek observer, shown in
Figure 170, moves along the path x(t), as light

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

c
observed by the inertial Roman one. In gen-
eral, the Roman/Greek clock rate ratio is observer (Roman)
given by ∆τ~∆t = (τ 2 − τ 1 )~(t 2 − t 1 ). Here F I G U R E 170 The simplest situation for an
the Greek coordinates are constructed with inertial and an accelerated observer
a simple procedure: take the two sets of
events defined by t = t 1 and t = t 2 , and let τ 1 and τ 2 be the points where these sets
intersect the time axis of the Greek observer.* We assume that the Greek observer is iner-
tial and moving with velocity v as observed by the Roman one. The clock ratio of a Greek
observer is then given by

∆τ dτ »
= = 1 − v 2 ~c 2 =
1
, (169)
∆t dt γv

Challenge 654 ny a formula we are now used to. We find again that moving clocks run slow.
Ref. 294 For accelerated motions, the differential version of the above reasoning is necessary.
The Roman/Greek clock rate ratio is again dτ~dt, and τ and τ + dτ are calculated in the
same way from the times t and t + dt. Assume again that the Greek observer moves along
the path x(t), as measured by the Roman one. We find directly that

τ = t − x(t)v(t)~c 2
Copyright © Christoph Schiller November 1997–May 2006

(170)

and thus
τ + dτ = (t + dt) − [x(t) − dtv(t)][v(t) + dta(t)]~c 2 . (171)

Together, these equations yield

‘dτ~dt’ = γv (1 − vv~c 2 − xa~c 2 ) . (172)

approximation.
* These sets form what mathematicians call hypersurfaces.

Dvipsbugw
 330 ii special rel ativity • 5. speed, rest and light

This shows that accelerated clocks can run fast or slow, depending on their position x and
the sign of their acceleration a. There are quotes in the above equation because we can
see directly that the Greek observer notes

‘dt~dτ’ = γv , (173)

which is not the inverse of equation (172). This difference becomes most apparent in the
simple case of two clocks with the same velocity, one of which has a constant acceleration
д towards the origin, whereas the other moves inertially. We then have Dvipsbugw

‘dτ~dt’ = 1 + дx~c 2 (174)

and
‘dt~dτ’ = 1 . (175)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We will discuss this situation shortly. But first we must clarify the concept of acceleration.

Acceleration for inertial observers

Accelerations behave differently from velocities under change of viewpoint. Let us first
take the simple case in which the object and two inertial observers all move along the
x-axis. If the Roman inertial observer measures an acceleration a = dv~dt = d2 x~dt 2 ,
and the Greek observer, also inertial, measures an acceleration α = dω~dτ = d2 ξ~dτ 2 , we
Ref. 247 get
γv3 a = γ 3ω α . (176)

This relation shows that accelerations are not Lorentz invariant, unless the velocities are
small compared to the speed of light. This is in contrast to our everyday experience, where
accelerations are independent of the speed of the observer.
Expression (176) simplifies if the accelerations are measured at a time t at which ω
vanishes – i.e. if they are measured by the so-called comoving inertial observer. In that
case the acceleration relation is given by

a c = aγv3 (177)

and the acceleration a c = α is also called proper acceleration, as its value describes what
Copyright © Christoph Schiller November 1997–May 2006

the Greek, comoving observer feels: proper acceleration describes the experience of being
pushed into the back of the accelerating seat.
In general, the observer’s speed and the acceleration are not parallel. We can calculate
Ref. 295 how the value of 3-acceleration a measured by a general inertial observer is related to the
value ac measured by the comoving observer using expressions (150) and (148). We get
the generalization of (177):
vac = vaγv3 (178)

Dvipsbugw
 accelerating observers 331

and
(1 − γv )(vac )v γv (vac )v
a= Œac − ‘ .
1
− (179)
γv2 v2 c2

Squaring yields the relation

(ac v)2
a2 = Œa ‘
1 2
− (180)
γv4 c c2
Dvipsbugw
Page 321 which we know already in a slightly different form. It shows (again) that the comoving
or proper 3-acceleration is always larger than the 3-acceleration measured by an outside
inertial observer. The faster the outside inertial observer is moving, the smaller the accele-
Challenge 655 e ration he observes. Acceleration is not a relativistic invariant. The expression also shows
that whenever the speed is perpendicular to the acceleration, a boost yields a factor γv2 ,
whereas a speed parallel to the acceleration gives the already mentioned γv3 dependence.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We see that acceleration complicates many issues, and it requires a deeper investiga-
tion. To keep matters simple, from now on we only study constant accelerations. Interest-
Page 476 ingly, this situation serves also as a good introduction to black holes and, as we will see
shortly, to the universe as a whole.

Accelerating frames of reference

How do we check whether we live in an inertial frame of reference? Let us first define the
term. An inertial frame (of reference) has two defining properties. First, lengths and dis-
tances measured with a ruler are described by Euclidean geometry. In other words, rulers
behave as they do in daily life. In particular, distances found by counting how many rulers
(rods) have to be laid down end to end to reach from one point to another – the so-called
rod distances – behave as in everyday life. For example, they obey Pythagoras’ theorem in
the case of right-angled triangles. Secondly, the speed of light is constant. In other words,
any two observers in that frame, independent of their time and of the position, make the
following observation: the ratio c between twice the rod distance between two points and
the time taken by light to travel from one point to the other and back is always the same.
Equivalently, an inertial frame is one for which all clocks always remain synchronized
and whose geometry is Euclidean. In particular, in an inertial frame all observers at fixed
coordinates always remain at rest with respect to each other. This last condition is, how-
Copyright © Christoph Schiller November 1997–May 2006

ever, a more general one. There are other, non-inertial, situations where this is still the
case.
Non-inertial frames, or accelerating frames, are a useful concept in special relativity. In
fact, we all live in such a frame. We can use special relativity to describe it in the same
way that we used Galilean physics to describe it at the beginning of our journey.
A general frame of reference is a continuous set of observers remaining at rest with
respect to each other. Here, ‘at rest with respect to each other’ means that the time for a
light signal to go from one observer to another and back again is constant over time, or
equivalently, that the rod distance between the two observers is constant. Any frame of
reference can therefore also be called a rigid collection of observers. We therefore note
that a general frame of reference is not the same as a set of coordinates; the latter is usu-

Dvipsbugw
 332 ii special rel ativity • 5. speed, rest and light

τ
t

n
II

izo
ξ

r
ho
re
tu
fu
Ω

O
III c2/g x
I

pa
Dvipsbugw

st
ho
IV

rizo
n
F I G U R E 171 The hyperbolic motion of an
rectilinearly, uniformly accelerating observer Ω

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ally not rigid. If all the rigidly connected observers have constant coordinate values, we
speak of a rigid coordinate system. Obviously, these are the most useful when it comes to
describing accelerating frames of reference.*
Ref. 297 Note that if two observers both move with a velocity v, as measured in some inertial
frame, they observe that they are at rest with respect to each other only if this velocity
Challenge 656 ny is constant. Again we find, as above, that two people tied to each other by a rope, and at
a distance such that the rope is under tension, will see the rope break (or hang loose) if
they accelerate together to (or decelerate from) relativistic speeds in precisely the same
way. Relativistic acceleration requires careful thinking.
An observer who always feels the same force on his body is called uniformly accelerat-
ing. More precisely, a uniformly accelerating observer is an observer whose acceleration
at every moment, measured by the inertial frame with respect to which the observer is
at rest at that moment, always has the same value B. It is important to note that uniform
acceleration is not uniformly accelerating when always observed from the same inertial
frame. This is an important difference from the Galilean case.
For uniformly accelerated motion in the sense just defined, we need

B ë B = −д 2 (181)
Copyright © Christoph Schiller November 1997–May 2006

Ref. 298 where д is a constant independent of t. The simplest case is uniformly accelerating motion
that is also rectilinear, i.e. for which the acceleration a is parallel to v at one instant of time
Challenge 657 ny and (therefore) for all other times as well. In this case we can write, using 3-vectors,

Ref. 296 * There are essentially only two other types of rigid coordinate frames, apart from the inertial frames:

— The frame ds 2 = dx 2 + dy 2 + dz 2 − c 2 dt 2 (1 + дk x k ~c 2 )2 with arbitrary, but constant, acceleration of the

origin. The acceleration is a = −g(1 + gx~c 2 ).
— The uniformly rotating frame ds 2 = dx 2 + dy 2 + dz 2 + 2ω(−y dx + x dy)dt − (1 − r 2 ω 2 ~c 2 )dt. Here the
z-axis is the rotation axis, and r 2 = x 2 + y 2 .

Dvipsbugw
 accelerating observers 333

γ3a = g =g.
dγv
or (182)
dt

Taking the direction we are talking about to be the x-axis, and solving for v(t), we get

v=¼
дt
, (183)
д2 t2
1 + c2

where it was assumed that v(0) = 0. We note that for small times we get v = дt and for
Dvipsbugw
large times v = c, both as expected. The momentum of the accelerated observer increases
Challenge 658 ny linearly with time, again as expected. Integrating, we find that the accelerated observer
moves along the path ¾
c2 д2 t 2
x(t) = 1+ 2 , (184)
д c

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where it is assumed that x(0) = c 2 ~д, in order to keep the expression simple. Because of
this result, visualized in Figure 171, a rectilinearly and uniformly accelerating observer is
said to undergo hyperbolic motion. For small times, the world-line reduces to the usual
x = дt 2 ~2 + x 0 , whereas for large times it is x = ct, as expected. The motion is thus
uniformly accelerated only for the moving body itself, not for an outside observer.
The proper time τ of the accelerated observer is related to the time t of the inertial
frame in the usual way by dt = γdτ. Using the expression for the velocity v(t) of equation
Ref. 298, Ref. 299 (183) we get*
c2
t = sinh and x = cosh
c дτ дτ
(185)
д c д c

for the relationship between proper time τ and the time t and position x measured by
the external, inertial Roman observer. We will encounter this relation again during our
study of black holes.
Does all this sound boring? Just imagine accelerating on a motorbike at д = 10 m~s2
for the proper time τ of 25 years. That would bring you beyond the end of the known
universe! Isn’t that worth a try? Unfortunately, neither motorbikes nor missiles that ac-
Challenge 659 n celerate like this exist, as their fuel tanks would have to be enormous. Can you confirm
this?
Copyright © Christoph Schiller November 1997–May 2006

Ref. 300 * Use your favourite mathematical formula collection – every student should have one – to deduce this. The
hyperbolic sine and the»hyperbolic cosine are defined by sinh y = (e»
y
− e−y )~2 and cosh y = (e y + e−y )~2.
They imply that ∫ dy~ y + a = arsinh y~a = Arsh y~a = ln(y + y 2 + a 2 ).
2 2

Dvipsbugw
 334 ii special rel ativity • 5. speed, rest and light

For uniform acceleration, the coordinates transform as

t = ( + ) sinh
c ξ дτ
д c c
c2
x = ( + ξ) cosh
дτ
д c
y=υ
z=ζ, (186) Dvipsbugw

where τ now is the time coordinate in the Greek frame. We note also that the space-time
interval dσ satisfies

dσ 2 = (1 + дξ~c 2 )2 c 2 dτ 2 − dξ 2 − dυ 2 − dζ 2 = c 2 dt 2 − dx 2 − dy 2 − dz 2 , (187)

and since for dτ = 0 distances are given by Pythagoras’ theorem, the Greek reference

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 301 frame is indeed rigid.
After this forest of formulae, let’s tackle a simple question, shown in Figure 171. The
inertial, Roman observer O sees the Greek observer Ω departing with acceleration д,
moving further and further away, following equation (184). What does the Greek observer
say about his Roman colleague? With all the knowledge we have now, that is easy to
answer to answer. At each point of his trajectory Ω sees that O has the coordinate τ = 0
Challenge 660 e (can you confirm this?), which means that the distance to the Roman observer, as seen
by Greek one, is the same as the space-time interval OΩ. Using expression (184), we see
Ref. 302 that this is » º
d OΩ = ξ 2 = x 2 − c 2 t 2 = c 2 ~д , (188)

which, surprisingly enough, is constant in time! In other words, the Greek observer will
observe that he stays at a constant distance from the Roman one, in complete contrast to
what the Roman observer says. Take your time to check this strange result in some other
way. We will need it again later on, to explain why the Earth does not explode. (Can you
Challenge 661 n guess how that is related to this result?)
The composition theorem for accelerations is more complex than for velocities. The
Ref. 303 best explanation of this was published by Mishra. If we call a nm the acceleration of sys-
tem n by observer m, we are seeking to express the object acceleration a 01 as function
Copyright © Christoph Schiller November 1997–May 2006

of the value a 02 measured by the other observer, the relative acceleration a 12 , and the
proper acceleration a 22 of the other observer: see Figure 172. Here we will only study
one-dimensional situations, where all observers and all objects move along one axis. (For
clarity, we also write v 11 = v and v 02 = u.) In Galilean physics we have the general connec-
Challenge 662 e tion
a 01 = a 02 − a 12 + a 22 (189)

Dvipsbugw
 accelerating observers 335

a11 : proper acceleration

v11 = 0

y
Observer 1
x
a22 : proper acceleration
v22 = 0
Dvipsbugw
v0n : object speed seen by observer n
Observer 2
x a0n : object acceleration
Object seen by observer n

F I G U R E 172 The definitions necessary to deduce the composition behaviour of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

accelerations

because accelerations behave simply. In special relativity, one gets

(1 − v 2 ~c 2 )3~2 (1 − u 2 ~c 2 )(1 − v 2 ~c 2 )−1~2 (1 − u 2 ~c 2 )(1 − v 2 ~c 2 )3~2

a 01 = a 02 − +
(1 − uv~c 2 )3 (1 − uv~c 2 )2 (1 − uv~c 2 )3
a 12 a 22

(190)
Challenge 663 ny and you might enjoy checking the expression.
Page 312 Can you state how the acceleration ratio enters into the definition of mass in special
Challenge 664 ny relativity?

Event horizons
There are many surprising properties of accelerated motion. Of special interest is the
trajectory, in the coordinates ξ and τ of the rigidly accelerated frame, of an object located
Challenge 665 ny at the departure point x = x 0 = c 2 ~д at all times t. One gets the two relations*

c2
ξ=− (1 − sech )
дτ
д c
dξ~dτ = −c sech
дτ дτ
tanh .
Copyright © Christoph Schiller November 1997–May 2006

(192)
c c

These equations are strange. For large times τ the coordinate ξ approaches the limit value
−c 2 ~д and dξ~dτ approaches zero. The situation is similar to that of a car accelerating
away from a woman standing on a long road. Seen from the car, the woman moves away;

* The functions appearing above, the hyperbolic secant and the hyperbolic tangent, are defined using the
expressions from the footnote on page 333:

1 sinh y
sech y = and tanh y = . (191)
cosh y cosh y

Dvipsbugw
 336 ii special rel ativity • 5. speed, rest and light

τ
t

n
II

izo
ξ

r
ho
re
tu
fu
Ω

O
III c2/g x
I

pa
Dvipsbugw

st
ho
IV

rizo
n
F I G U R E 173 Hyperbolic motion and event
horizons

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

however, after a while, the only thing one notices is that she is slowly approaching the
horizon. In Galilean physics, both the car driver and the woman on the road see the
other person approaching their horizon; in special relativity, only the accelerated observer
makes this observation.
A graph of the situation helps to clarify the result. In Figure 173 we can see that light
emitted from any event in regions II and III cannot reach the Greek observer. Those
events are hidden from him and cannot be observed. Strangely enough, however, light
from the Greek observer can reach region II. The boundary between the part of space-
time that can be observed and the part that cannot is called the event horizon. In relativity,
event horizons act like one-way gates for light and other signals. For completeness, the
Challenge 666 ny graph also shows the past event horizon. Can you confirm that event horizons are black?
So, not all events observed in an inertial frame of reference can be observed in a uni-
formly accelerating frame of reference. Uniformly accelerating frames of reference pro-
duce event horizons at a distance −c 2 ~д. For example, a person who is standing can never
see further than this distance below his feet.
By the way, is it true that a light beam cannot catch up with an observer in hyperbolic
Challenge 667 n motion, if the observer has a sufficient headstart?
Here is a more advanced challenge, which prepares us for general relativity. What is
Copyright © Christoph Schiller November 1997–May 2006

Challenge 668 n the shape of the horizon seen by a uniformly accelerated observer?

Acceleration changes colours

We saw earlier that a moving receiver sees different colours from the sender. So far, we
discussed this colour shift, or Doppler effect, for inertial motion only. For accelerating
frames the situation is even stranger: sender and receiver do not agree on colours even
Ref. 298, Ref. 304 if they are at rest with respect to each other. Indeed, if light is emitted in the direction of
the acceleration, the formula for the space-time interval gives

д0 x 2 2 2
dσ 2 = ‹1 +  c dt (193)
c2

Dvipsbugw
 accelerating observers 337

in which д0 is the proper acceleration of an observer located at x = 0. We can deduce in

Challenge 669 ny a straightforward way that
=1− 2 =
fr дr h 1
(194)
fs c Š1 + дs2h  c

where h is the rod distance between the source and the receiver, and where дs = д0 ~(1 +
д0 x s ~c 2 ) and дr = д0 ~(1 + дo x r ~c 2 ) are the proper accelerations measured at the source
and at the detector. In short, the frequency of light decreases when light moves in the Dvipsbugw
direction of acceleration. By the way, does this have an effect on the colour of trees along
Challenge 670 n their vertical extension?
The formula usually given, namely

=1− 2 ,
fr дh
(195)
fs c

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is only correct to a first approximation. In accelerated frames of reference, we have to
be careful about the meaning of every quantity. For everyday accelerations, however, the
Challenge 671 ny differences between the two formulae are negligible. Can you confirm this?

Can light move faster than c?

What speed of light does an accelerating observer measure? Using expression (195) above,
an accelerated observer deduces that

v light = c (1 + )
дh
(196)
c2
which is higher than c for light moving in front of or ‘above’ him, and lower than c for
light moving behind or ‘below’ him. This strange result follows from a basic property of
any accelerating frame of reference. In such a frame, even though all observers are at rest
with respect to each other, clocks do not remain synchronized. This change of the speed
of light has also been confirmed by experiment.* Thus, the speed of light is only constant
when it is defined as c = dx~dt, and if dx and dt are measured with a ruler located at
a point inside the interval dx and a clock read off during the interval dt. If the speed
of light is defined as ∆x~∆t, or if the ruler defining distances or the clock measuring
Copyright © Christoph Schiller November 1997–May 2006

times is located away from the propagating light, the speed of light is different from c for
accelerating observers! This is the same effect you can experience when you turn around
your vertcial axis at night: the star velocities you observe are much higher than the speed
of light.
Note that this result does not imply that signals or energy can be moved faster than c.
Challenge 672 n You may want to check this for yourself.
In fact, all these effects are negligible for distances l that are much less than c 2 ~a. For
an acceleration of 9.5 m~s2 (about that of free fall), distances would have to be of the order

Page 410 * The propagation delays to be discussed in the chapter on general relativity can be seen as confirmations of
this effect.

Dvipsbugw
 338 ii special rel ativity • 5. speed, rest and light

time
clock 1 clock 2

t3

t2

Dvipsbugw

t1

space

F I G U R E 174 Clocks and the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

measurement of the speed of
light as two-way velocity

of one light year, or 9.5 ë 1012 km, in order for any sizable effects to be observed. In short,
c is the speed of light relative to nearby matter only.
By the way, everyday gravity is equivalent to a constant acceleration. So, why then
Challenge 673 n distant objects, such as stars, move faster than light, following expression (196)?

What is the speed of light?

We have seen that the speed of light, as usually defined, is given by c only if either the
observer is inertial or the observer measures the speed of light passing nearby (rather
than light passing at a distance). In short, the speed of light has to be measured locally.
But this condition does not eliminate all subtleties.
An additional point is often forgotten. Usually, length is measured by the time it takes
light to travel. In such a case the speed of light will obviously be constant. But how does
one check the constancy? One needs to eliminate length measurements. The simplest way
to do this is to reflect light from a mirror, as shown in Figure 174. The constancy of the
speed of light implies that if light goes up and down a short straight line, then the clocks
at the two ends measure times given by
Copyright © Christoph Schiller November 1997–May 2006

t 3 − t 1 = 2 (t 2 − t 1 ) . (197)

Here it is assumed that the clocks have been synchronised according to the prescription
on page 299. If the factor were not exactly two, the speed of light would not be constant.
In fact, all experiments so far have yielded a factor of two, within measurement errors.*

* The subtleties of the one-way and two-way speed of light will remain a point of discussion for a long time.
Many experiments are explained and discussed in Ref. 252. Zhang says in his summary on page 171, that the
one-way velocity of light is indeed independent of the light source; however, no experiment really shows that
Ref. 305 it is equal to the two-way velocity. Moreover, most so called ‘one-way’ experiments are in fact still ‘two-way’

Dvipsbugw
 accelerating observers 339

This result is sometimes expressed by saying that it is impossible to measure the one-
Challenge 674 n way velocity of light; only the two-way velocity of light is measurable. Do you agree?

Limits on the length of solid bodies

An everyday solid object breaks when some part of it moves with respect to some other
part with more than the speed of sound c of the material.* For example, when an object
hits the floor and its front end is stopped within a distance d, the object breaks at the
latest when Dvipsbugw
v 2 2d
E . (198)
c2 l
In this way, we see that we can avoid the breaking of fragile objects by packing them into
foam rubber – which increases the stopping distance – of roughly the same thickness as
the object’s size. This may explain why boxes containing presents are usually so much
larger than their contents!

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The fracture limit can also be written in a different way. To avoid breaking, the accele-
ration a of a solid body with length l must obey

l a < c2 , (199)

where c is the speed of sound, which is the speed limit for the material parts of solids. Let
Ref. 306 us now repeat the argument in relativity, using the speed of light instead of that of sound.
Imagine accelerating the front of a solid body with some proper acceleration a. The back
end cannot move with an acceleration α equal or larger than infinity, or if one prefers, it
Challenge 675 n cannot move with more than the speed of light. A quick check shows that therefore the
length l of a solid body must obey

l α < c 2 ~2 , (200)

where c is now the speed of light. The speed of light thus limits the size of solid bodies. For
example, for 9.8 m~s2 , the acceleration of good motorbike, this expression gives a length
limit of 9.2 Pm, about a light year. Not a big restriction: most motorbikes are shorter.
However, there are other, more interesting situations. The highest accelerations achiev-
able today are produced in particle accelerators. Atomic nuclei have a size of a few femo-
Challenge 676 ny tometres. Can you deduce at which energies they break when smashed together in an
Copyright © Christoph Schiller November 1997–May 2006

accelerator? In fact, inside a nucleus, the nucleons move with accelerations of the order
of v 2 ~r  ħ 2 ~m 2 r 3  1031 m~s2 ; this is one of the highest values found in nature.
Note that Galilean physics and relativity produce a similar conclusion: a limiting speed,
be it that of sound or that of light, makes it impossible for solid bodies to be rigid. When
we push one end of a body, the other end always moves a little bit later.
What does this mean for the size of elementary particles? Take two electrons a distance
d apart, and call their size l. The acceleration due to electrostatic repulsion then leads to

experiments (see his page 150).

* The (longitudinal) speed of sound is about 5.9 km~s for glass, iron or steel; about 4.5 km~s for gold; and
about 2 km~s for lead. Other sound speeds are given on page 206.

Dvipsbugw
 340 ii special rel ativity • 5. speed, rest and light

Challenge 677 ny an upper limit for their size given by

4πε 0 c 2 d 2 m
l< . (201)
e2
The nearer electrons can get, the smaller they must be. The present experimental limit
gives a size smaller than 10−19 m. Can electrons be exactly point-like? We will come back
to this question during our study of general relativity and quantum theory.
Dvipsbugw
Special rel ativity in four sentences
This section of our ascent of Motion Mountain can be quickly summarized.
— All (free floating) observers find that there is a unique, perfect velocity in nature,
namely a common maximum energy velocity, which is realized by massless radiation
such as light or radio signals, but cannot be achieved by material systems.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

— Therefore, even though space-time is the same for every observer, times and lengths
vary from one observer to another, as described by the Lorentz transformations (112)
and (113), and as confirmed by experiment.
— Collisions show that a maximum speed implies that mass is concentrated energy, and
that the total energy of a body is given by E = γmc 2 , as again confirmed by experiment.
— Applied to accelerated objects, these results lead to numerous counter-intuitive con-
sequences, such as the twin paradox, the appearance of event horizons and the appear-
ance of short-lived tachyons in collisions.
Special relativity shows that motion, though limited in speed, is relative, defined using
the propagation of light, conserved, reversible and deterministic.

Could the speed of light vary?

The speed of massless light is the limit speed. Assuming that all light is indeed massless,
could the speed of light still change from place to place, or as time goes by? This tricky
question still makes a fool out of many physicists. The first answer is usually a loud: ‘Yes,
of course! Just look at what happens when the value of c is changed in formulae.’ (In fact,
there have even been attempts to build ‘variable speed of light theories’.) However, this
often-heard statement is wrong.
Since the speed of light enters into our definition of time and space, it thus enters,
Copyright © Christoph Schiller November 1997–May 2006

even if we do not notice it, into the construction of all rulers, all measurement standards
and all measuring instruments. Therefore there is no way to detect whether the value
actually varies. No imaginable experiment could detect a variation of the limit speed, as
Challenge 678 n the limit speed is the basis for all measurements. ‘That is intellectual cruelty!’, you might
say. ‘All experiments show that the speed of light is invariant; we had to swallow one
counter-intuitive result after another to accept the constancy of the speed of light, and
now we are supposed to admit that there is no other choice?’ Yes, we are. That is the irony
of progress in physics. The observer-invariance of the speed of light is counter-intuitive
and astonishing when compared to the lack of observer-invariance at everyday, Galilean
speeds. But had we taken into account that every speed measurement is – whether we like
it or not – a comparison with the speed of light, we would not have been astonished by

Dvipsbugw
special rel ativity in four sentences 341

the constancy of the speed of light; rather, we would have been astonished by the strange
properties of small speeds.
In short, there is in principle no way to check the invariance of a standard. To put it
another way, the truly surprising aspect of relativity is not the invariance of c; it is the
disappearance of c from the formulae of everyday motion.

What happens near the speed of light?

As one approaches the speed of light, the quantities in the Lorentz transformation diverge. Dvipsbugw
A division by zero is impossible: indeed, neither masses nor observers can move at the
speed of light. However, this is only half the story.
No observable actually diverges in nature. Approaching the speed of light as nearly
as possible, even special relativity breaks down. At extremely large Lorentz contractions,
there is no way to ignore the curvature of space-time; indeed, gravitation has to be taken
into account in those cases. Near horizons, there is no way to ignore the fluctuations of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

speed and position; quantum theory has to be taken into account there. The exploration
of these two limitations define the next two stages of our ascent of Motion Mountain.
At the start of our adventure, during our exploration of Galilean physics, once we had
defined the basic concepts of velocity, space and time, we turned our attention to gravit-
ation. The invariance of the speed of light has forced us to change these basic concepts.
We now return to the study of gravitation in the light of this invariance.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
342 ii special rel ativity

Biblio graphy

231 Aristotle, On sense and the sensible, section 1, part 1, 350 bce. Cited in Jean-Paul
Dumont, Les écoles présocratiques, Folio Essais, Gallimard, p. 157, 1991. Cited on page 276.
232 The history of the measurement of the speed of light can be found in chapter 19 of the text
by Francis A. Jenkins & Harvey E. White, Fundamentals of Optics, McGraw-Hill,
New York, 1957. Cited on page 276.
233 On the way to perform such measurements, see Sydney G. Brewer, Do-it-yourself As- Dvipsbugw
tronomy, Edinburgh University Press, 1988. Kepler himself never measured the distances of
planets to the Sun, but only ratios of planetary distances. The parallax of the Sun from two
points of the Earth is at most 8.79 ′′ ; it was first measured in the eighteenth century. Cited
on page 278.
234 Aristarchos, On the sizes and the distances of the Sun and the Moon, c. 280 bce, in
Michael J. Crowe, Theories of the World From Antiquity to the Copernican Revolution,
Dover, 1990. Cited on page 278.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

235 J. Frercks, Creativity and technology in experimentation: Fizeau’s terrestrial determina-
tion of the speed of light, Centaurus 42, pp. 249–287, 2000. See also the beautiful website on
reconstrutions of historical science experiments at http://www.uni-oldenburg.de/histodid/
forschung/nachbauten. Cited on page 278.
236 The way to make pictures of light pulses with an ordinary photographic camera, without
any electronics, is described by M.A. Duguay & A.T. Mattick, Ultrahigh speed pho-
tography of picosecond light pulses and echoes, Applied Optics 10, pp. 2162–2170, 1971. The
picture on page 279 is taken from it. Cited on page 279.
237 You can learn the basics of special relativity with the help of the web, without referring to
any book, using the http://physics.syr.edu/research/relativity/RELATIVITY.html web page
as a starting point. This page mentions most of the English-language relativity resources
available on the web. Links in other languages can be found with search engines. Cited on
page 280.
238 Observations of gamma ray bursts show that the speed of light does not depend on the
lamp speed to within one part in 1020 , as shown by K. Brecher, Bulletin of the American
Physical Society 45, 2000. He assumed that both sides of the burster emit light. The large
speed difference and the pulse sharpness then yield this result. See also his older paper K.
Brecher, Is the speed of light independent of the source?, Physics Letters 39, pp. 1051–
1054, Errata 1236, 1977. Measuring the light speed from rapidly moving stars is another way.
Some of these experiments are not completely watertight, however. There is a competing
Copyright © Christoph Schiller November 1997–May 2006

theory of electrodynamics, due to Ritz, which maintains that the speed of light is c only
when measured with respect to the source; The light from stars, however, passes through
the atmosphere, and its speed might thus be reduced to c.
The famous experiment with light emitted from rapid pions at CERN is not subject to this
criticism. It is described in T. Alväger, J.M. Bailey, F.J.M. Farley, J. Kjellman &
I. Wallin, Test of the second postulate of relativity in the GeV region, Physics Letters 12,
pp. 260–262, 1964. See also T. Alväger & al., Velocity of high-energy gamma rays, Arkiv
för Fysik 31, pp. 145–157, 1965.
Another precise experiment at extreme speeds is described by G.R. Kalbfleisch, N.
Baggett, E.C. Fowler & J. Alspector, Experimental comparison of neutrino, anti-
neutrino, and muon velocities, Physical Review Letters 43, pp. 1361–1364, 1979. Cited on
page 280.

Dvipsbugw
bibliography 343

239 See e.g. C. Will, Theory and Experiment in Gravitational Physics, Revised edition, Cam-
bridge University Press, 1993. Cited on pages 281 and 282.
240 B.E. Schaefer, Severe limits on variations of the speed of light with frequency, Physical
Review Letters 82, pp. 4964–4966, 21 June 1999. Cited on page 281.
241 The beginning of the modern theory of relativity is the famous paper by Albert Ein-
stein, Zur Elektrodynamik bewegter Körper, Annalen der Physik 17, pp. 891–921, 1905. It
still well worth reading, and every physicist should have done so. The same can be said of the
famous paper, probably written after he heard of Olinto De Pretto’s idea, found in Albert
Einstein, Ist die Trägheit eines Körpers von seinem Energieinhalt abhängig?, Annalen der
Dvipsbugw
Physik 18, pp. 639–641, 1905. See also the review Albert Einstein, Über das Relativität-
sprinzip und die aus demselben gezogenen Folgerungen, Jahrbuch der Radioaktivität und
Elektronik 4, pp. 411–462, 1907. These papers are now available in many languages. A later,
unpublished review is available in facsimile and with an English translation as Albert
Einstein, Hanoch Gutfreund, ed., Einstein’s 1912 Manuscript on the Theory of Relativity,
George Braziller, 2004. Cited on pages 281, 282, and 318.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

242 Albert Einstein, Mein Weltbild, edited by Carl Selig, Ullstein Verlag, 1998. Cited
on page 281.
243 Jean van Bl adel, Relativity and Engineering, Springer, 1984. Cited on page 281.
244 Albrecht Fölsing, Albert Einstein – eine Biographie, Suhrkamp p. 237, 1993. Cited on
pages 281 and 291.
245 R.J. Kennedy & E.M. Thorndike, Experimental establishment of the relativity of time,
Physical Review 42, pp. 400–418, 1932. See also H.E. Ives & G.R. Stilwell, An experi-
mental study of the rate of a moving atomic clock, Journal of the Optical Society of America
28, pp. 215–226, 1938, and 31, pp. 369–374, 1941. For a modern, high-precision versions, see
C. Braxmeier, H. Müller, O. Pradl, J. Mlynek, A. Peters & S. Schiller, New
tests of relativity using a cryogenic optical resonator, Physical Review Letters 88, p. 010401,
2002. The newest result is in P. Antonini, M. Okhapkin, E. Göklü & S. Schiller,
Testing the constancy of the speed of light with rotating cryogenic optical resonators, Phys-
ical Review A 71, p. 050101, 2005, or http://www.arxiv.org/abs/gr-qc/0504109. Cited on page
282.
246 Edwin F. Taylor & John A. Wheeler, Spacetime Physics – Introduction to Special
Relativity, second edition, Freeman, 1992. See also Nick M.J. Woodhouse, Special Re-
lativity, Springer, 2003. Cited on pages 282 and 326.
247 Wolf gang R indler, Relativity – Special, General and Cosmological, Oxford University
Press, 2001. A beautiful book by one of the masters of the field. Cited on pages 282 and 330.
Copyright © Christoph Schiller November 1997–May 2006

248 The slowness of the speed of light inside stars is due to the frequent scattering of photons
by the star matter. The most common estimate for the Sun is an escape time of 40 000 to 1
million years, but estimates between 17 000 years and 50 million years can be found in the
literature. Cited on page 283.
249 L. Vestergaard Hau, S.E. Harris, Z. Du tton & C.H. Behroozi, Light speed
reduction to 17 meters per second in an ultracold atomic gas, Nature 397, pp. 594–598, 1999.
See also Ref. 25. Cited on page 283.
250 The method of explaining special relativity by drawing a few lines on paper is due to Her-
mann B ondi, Relativity and Common Sense: A New Approach to Einstein, Dover, New
York, 1980. See also Dierck-Ekkehard Liebscher, Relativitätstheorie mit Zirkel und
Lineal, Akademie-Verlag Berlin, 1991. Cited on page 283.

Dvipsbugw
344 ii special rel ativity

251 Rod S. L akes, Experimental limits on the photon mass and cosmic vector potential, Phys-
ical Review Letters 80, pp. 1826–1829, 1998. The speed of light is independent of frequency
within a factor of 6ë10−21 , as was shown from gamma ray studies by B.E. Schaefer, Severe
limits on variations of the speed of light with frequency, Physical Review Letters 82, pp. 4964–
4966, 1999. Cited on page 284.
252 An overview of experimental results is given in Yuan Zhong Zhang, Special Relativity
and its Experimental Foundations, World Scientific, 1998. Cited on pages 284, 290, 299, 313,
338, and 345.
253 R.W. McGowan & D.M. Giltner, New measurement of the relativistic Doppler shift Dvipsbugw
in neon, Physical Review Letters 70, pp. 251–254, 1993. Cited on page 286.
254 The present record for clock synchronization seems to be 1 ps for two clocks distant 3 km
from each other. See A. Valencia, G. Scarcelli & Y. Shih, Distant clock synchron-
ization using entangled photon pairs, Applied Physics Letters 85, pp. 2655–2657, 2004, or
http://www.arxiv.org/abs/quant-ph/0407204. Cited on page 287.
255 J. Frenkel & T. Kontorowa, Über die Theorie der plastischen Verformung, Physikalis-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

che Zeitschrift der Sowietunion 13, p. 1, 1938. F.C. Frank, On the equations of motion of
crystal dislocations, Proceedings of the Physical Society A 62, pp. 131–134, 1949. J. Eshelby,
Uniformly moving dislocations, Proceedings of the Physical Society A 62, pp. 307–314, 1949.
See also G. Leibfried & H. Dietze, Zeitschrift für Physik 126, p. 790, 1949. A general in-
troduction can be found in A. Seeger & P. Schiller, Kinks in dislocation lines and their
effects in internal friction in crystals, Physical Acoustics 3A, W.P. Mason, ed., Academic
Press, 1966. See also the textbooks by Frank R.N. Nabarro, Theory of Crystal Dislo-
cations, Oxford University Press, 1967, or J.P. Hirth & J. Lothe, Theory of Dislocations,
McGraw Hill, 1968. Cited on page 287.
256 This beautiful graph is taken from Z.G.T. Guiragossian, G.B. Rothbart, M.R.
Yearian, R. Gearhart & J.J. Murray, Relative velocity measurements of electrons
and gamma rays at 15 GeV, Physical Review Letters 34, pp. 335–338, 1975. Cited on page
287.
257 To find out more about the best-known crackpots, and their ideas, send an email to
[email protected] with the one-line body ‘subscribe psychoceramics’. Cited on page
288.
258 The speed of neutrinos is the same as that of light to 9 decimal digits. This is explained
by Leo Stod olsky, The speed of light and the speed of neutrinos, Physics Letters B 201,
p. 353, 1988. An observation of a small mass for the neutrino has been published by the Ja-
panese Super-Kamiokande collaboration, in Y. Fukuda & al., Evidence for oscillation of
atmospheric neutrinos, Physical Review Letters 81, pp. 1562–1567, 1998. The newer results
Copyright © Christoph Schiller November 1997–May 2006

published by the Canadian Sudbury Neutrino Observatory, as Q.R. Ahmad & al., Direct
evidence for neutrino flavor transformation from neutral-current interactions in the Sud-
bury Neutrino Observatory, Physical Review Letters 89, p. 011301, 2002, also confirm that
neutrinos have a mass in the 1 eV region. Cited on pages 289 and 1258.
259 B. Rothenstein & G. Eckstein, Lorentz transformations directly from the speed of
light, American Journal of Physics 63, p. 1150, 1995. See also the comment by E. Kapuścik,
Comment on “Lorentz transformations directly from the speed of light,” by B. Rothenstein
and G. Eckstein, American Journal of Physics 65, p. 1210, 1997. Cited on page 290.
260 See e.g. the 1922 lectures by Lorentz at Caltech, published as H.A. Lorentz, Problems of
Modern Physics, edited by H. Bateman, Ginn and Company, page 99, 1927. Cited on page
290.

Dvipsbugw
bibliography 345

261 A.A. Michelson & E.W. Morley, On the relative motion of the Earth and the lumini-
ferous ether, American Journal of Science (3rd series) 34, pp. 333–345, 1887. Michelson pub-
lished many other papers on the topic after this one. Cited on page 290.
262 S. Stephan, P. Antonini & M. Okhapkin, A precision test of the isotropy of
the speed of light using rotating cryogenic resonators, http://www.arxiv.org/abs/physics/
0510169. Cited on page 290.
263 H.A. Lorentz, De relative beweging van de aarde en dem aether, Amst. Versl. 1, p. 74,
1892, and also H.A. Lorentz, Electromagnetic phenomena in a system moving with any
velocity smaller than that of light, Amst. Proc. 6, p. 809, 1904, or Amst. Versl. 12, p. 986, 1904. Dvipsbugw
Cited on page 294.
264 A general refutation of such proposals is discussed by S.R. Mainwaring & G.E. Sted-
man, Accelerated clock principles, Physical Review A 47, pp. 3611–3619, 1993. Experiments
on muons at CERN in 1968 showed that accelerations of up to 1020 m~s2 have no effect, as
explained by D.H. Perkins, Introduction to High Energy Physics, Addison-Wesley, 1972,
or by J. Bailey & al., Il Nuovo Cimento 9A, p. 369, 1972. Cited on page 294.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

265 W. R indler, General relativity before special relativity: an unconventional overview of
relativity theory, American Journal of Physics 62, pp. 887–893, 1994. Cited on page 294.
266 Steven K. Bl au, Would a topology change allow Ms. Bright to travel backward in time?,
American Journal of Physics 66, pp. 179–185, 1998. Cited on page 297.
267 On the ‘proper’ formulation of relativity, see for example D. Hestenes, Proper particle
mechanics, Journal of Mathematical Physics 15, pp. 1768–1777, 1974. Cited on page 297.
268 The simple experiment to take a precise clock on a plane, fly it around the world and then
compare it with an identical one left in place was first performed by J.C. Hafele & R.E.
Keating, Around-the-world atomic clocks: predicted relativistic time gains, Science 177,
pp. 166–167, and Around-the-world atomic clocks: observed relativistic time gains, pp. 168–
170, 14 July 1972. See also Ref. 252. Cited on page 298.
269 A readable introduction to the change of time with observers, and to relativity in general,
is Roman U. Sexl & Herbert Kurt Schmidt, Raum-Zeit-Relativität, 2. Auflage,
Vieweg & Sohn, Braunschweig, 1991. Cited on page 298.
270 Most famous is the result that moving muons stay younger, as shown for example by D.H.
Frisch & J.B. Smith, Measurement of the relativistic time dilation using µ-mesons, Amer-
ican Journal of Physics 31, pp. 342–355, 1963. For a full pedagogical treatment of the twin
paradox, see E. Sheld on, Relativistic twins or sextuplets?, European Journal of Physics 24,
pp. 91–99, 2003. Cited on page 298.
271 Paul J. Nahin, Time Machines – Time Travel in Physics, Metaphysics and Science Fiction,
Copyright © Christoph Schiller November 1997–May 2006

Springer Verlag and AIP Press, second edition, 1999. Cited on page 298.
272 The first muon experiment was B. Rossi & D.B. Hall, Variation of the rate of decay of
mesotrons with momentum, Physical Review 59, pp. 223–228, 1941. ‘Mesotron’ was the old
name for muon. Cited on page 299.
273 A. Harvey & E. Schucking, A small puzzle from 1905, Physics Today, pp. 34–36, March
2005. Cited on page 299.
274 W. R indler, Length contraction paradox, American Journal of Physics 29, pp. 365–366,
1961. For a variation without gravity, see R. Shaw, Length contraction paradox, American
Journal of Physics 30, p. 72, 1962. Cited on page 301.
275 H. van Lintel & C. Gruber, The rod and hole paradox re-examined, European Journal
of Physics 26, pp. 19–23, 2005. Cited on page 301.

Dvipsbugw
346 ii special rel ativity

276 This situation is discussed by G.P. Sastry, Is length contraction paradoxical?, American
Journal of Physics 55, 1987, pp. 943–946. This paper also contains an extensive literature list
covering variants of length contraction paradoxes. Cited on page 301.
277 S.P. B oughn, The case of the identically accelerated twins, American Journal of Physics 57,
pp. 791–793, 1989. Cited on pages 302 and 305.
278 J.M. Supplee, Relativistic buoyancy, American Journal of Physics 57 1, pp. 75–77, January
1989. See also G.E.A. Matsas, Relativistic Arquimedes law for fast moving bodies and the
general-relativistic resolution of the ‘submarine paradox’, Physical Review D 68, p. 027701,
2003, or http://www.arxiv.org/abs/gr-qc/0305106. Cited on page 302. Dvipsbugw
279 The distinction was first published by J. Terrell, Invisibility of Lorentz contraction, Phys-
ical Review 116, pp. 1041–1045, 1959, and R. Penrose, The apparent shape of a relativistic-
ally moving sphere, Proceedings of the Cambridge Philosophical Society 55, pp. 137–139, 1959.
Cited on page 304.
280 G.R. Rybicki, Speed limit on walking, American Journal of Physics 59, pp. 368–369, 1991.
Cited on page 306.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

281 The first examples of such astronomical observations were provided by A.R. Whitney &
al., Quasars revisited: rapid time variations observed via very-long-baseline interferometry,
Science 173, pp. 225–230, 1971, and by M.H. Cohen & al., The small-scale structure of
radio galaxies and quasi-stellar sources at 3.8 centimetres, Astrophysical Journal 170, pp. 207–
217, 1971. See also T.J. Pearson, S.C. Unwin, M.H. Cohen, R.P. Linfield, A.C.S.
R eadhead, G.A. Seielstad, R.S. Simon & R.C. Walker, Superluminal expansion
of quasar 3C 273, Nature 290, pp. 365–368, 1981. An overview is given in J.A. Zensus & T.J.
Pearson, editors, Superluminal radio sources, Cambridge University Press, 1987. Another
measurement, using very long baseline interferometry with radio waves, was shown on the
cover of Nature: I.F. Mirabel & L.F. Rodriguez, A superluminal source in the galaxy,
Nature 371, pp. 46–48, 1994. A more recent example was reported in Science News 152, p. 357,
6 December 1997.
Pedagogical explanations are given by D.C. Gabuzda, The use of quasars in teaching
introductory special relativity, American Journal of Physics 55, pp. 214–215, 1987, and by
Edwin F. Taylor & John A. Wheeler, Spacetime Physics – Introduction to Special
Relativity, second edition, Freeman, 1992, pages 89-92. This excellent book was mentioned
already in the text. Cited on page 308.
282 O.M. Bil aniuk & E.C. Sudarshan, Particles beyond the light barrier, Physics Today
22, pp. 43–51, 1969, and O.M.P. Bil anuk, V.K. Deshpande & E.C.G. Sudarshan,
‘Meta’ relativity, American Journal of Physics 30, pp. 718–723, 1962. See also E. R ecami,
editor, Tachyons, Monopoles and Related Topics, North-Holland, Amsterdam, 1978. Cited
Copyright © Christoph Schiller November 1997–May 2006

on page 309.
283 J.P. Costell a, B.H.J. McKell ar, A.A. R awlinson & G.J. Stephenson, The
Thomas rotation, American Journal of Physics 69, pp. 837–847, 2001. Cited on page 309.
284 See for example S.S. Costa & G.E.A. Matsas, Temperature and relativity, preprint avail-
able at http://www.arxiv.org/abs/gr-qc/9505045. Cited on page 310.
285 R.C. Tolman & G.N. Lewis, The principle of relativity and non-Newtonian mechanics,
Philosophical Magazine 18, pp. 510–523, 1909, and R.C. Tolman, Non-Newtonian mech-
anics: the mass of a moving body, Philosophical Magazine 23, pp. 375–380, 1912. Cited on
page 311.
286 This information is due to a private communication by Frank DiFilippo; part of the story
is given in F. DiFilippo, V. Natarajan, K.R. B oyce & D.E. Pritchard, Accurate

Dvipsbugw
bibliography 347

atomic masses for fundamental metrology, Physical Review Letters 73, pp. 1481–1484, 1994.
These measurements were performed with Penning traps; a review of the possibilities they
offer is given by R.C. Thompson, Precision measurement aspects of ion traps, Measure-
ment Science and Technology 1, pp. 93–105, 1990. The most important experimenters in the
field of single particle levitation were awarded the Nobel Prize in 1989. One of the Nobel
Prize lectures can be found in W. Paul, Electromagnetic traps for neutral and charged
particles, Reviews of Modern Physics 62, pp. 531–540, 1990. Cited on page 315.
287 J.L. Synge, Relativity: The Special Theory, North-Holland, 1956, pp. 208–213. More about
antiparticles in special relativity can be found in J.P. Costell a, B.H.J. McKell ar & Dvipsbugw
A.A. R awlinson, Classical antiparticles, American Journal of Physics 65, pp. 835–841,
1997. See also Ref. 302. Cited on page 316.
288 A. Papapetrou, Drehimpuls- und Schwerpunktsatz in der relativistischen Mechanik,
Praktika Acad. Athenes 14, p. 540, 1939, and A. Papapetrou, Drehimpuls- und Schwer-
punktsatz in der Diracschen Theorie, Praktika Acad. Athenes 15, p. 404, 1940. See also
M.H.L. Pryce, The mass-centre in the restricted theory of relativity and its connexion
with the quantum theory of elementary particles, Proceedings of the Royal Society in Lon-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

don, A 195, pp. 62–81, 1948. Cited on page 317.
289 The references preceding Einstein’s E = mc 2 are: S. Tolver Preston, Physics of the Ether,
E. & F.N. Spon, 1875, J.H. Poincaré, La théorie de Lorentz et le principe de réaction,
Archives néerlandaises des sciences exactes et naturelles 5, pp. 252–278, 1900, O. De Pretto,
Ipotesi dell’etere nella vita dell’universo, Reale Istituto Veneto di Scienze, Lettere ed Arti tomo
LXIII, parte 2, pp. 439–500, Febbraio 1904, F. Hasenöhrl, Berichte der Wiener Akademie
113, p. 1039, 1904, F. Hasenöhrl, Zur Theorie der Strahlung in bewegten Körpern, An-
nalen der Physik 15, pp. 344–370, 1904, F. Hasenöhrl, Zur Theorie der Strahlung in be-
wegten Körpern – Berichtigung, Annalen der Physik 16, pp. 589–592, 1905. Hasenöhrl died
in 1915, De Pretto in 1921. All these publications were published before the famous paper
by Albert Einstein, Ist die Trägheit eines Körpers von seinem Energieinhalt abhängig?,
Annalen der Physik 18, pp. 639–641, 1905. Cited on page 319.
290 A jewel among the textbooks on special relativity is the booklet by Ulrich E. Schröder,
Spezielle Relativitätstheorie, Verlag Harri Deutsch, Thun, 1981. Cited on pages 321 and 324.
291 A readable article showing a photocopy of a letter by Einstein making this point is Lev B.
Okun, The concept of mass, Physics Today, pp. 31–36, June 1989. The topic is not without
controversy, as the letters by readers following that article show; they are found in Physics
Today, pp. 13–14 and pp. 115–117, May 1990. The topic is still a source of debates. Cited on
page 323.
292 Christian Møller, The Theory of Relativity, Clarendon Press, 1952, 1972. This standard
text has been translated in several languages. Cited on page 324.
Copyright © Christoph Schiller November 1997–May 2006

293 P. Ehrenfest, Gleichförmige Rotation starrer Körper und Relativitätstheorie, Physikalis-

che Zeitschrift 10, pp. 918–928, 1909. Ehrenfest (incorrectly) suggested that this meant that
relativity cannot be correct. A modern summary of the issue can be found in M.L. Rug-
giero, The relative space: space measurements on a rotating platform, http://www.arxiv.
org/abs/gr-qc/0309020. Cited on page 325.
294 R.J. Low, When moving clocks run fast, European Journal of Physics 16, pp. 228–229, 1995.
Cited on page 329.
295 G. Stephenson & C.W. Kilmister, Special Relativity for Physicists, Longmans, Lon-
don, 1965. See also W.N. Matthews, Relativistic velocity and acceleration transforma-
tions from thought experiments, American Journal of Physics 73, pp. 45–51, 2005. Cited on
page 330.

Dvipsbugw
348 ii special rel ativity

296 The impossibility of defining rigid coordinate frames for non-uniformly accelerating observ-
ers is discussed by Charles Misner, Kip Thorne & John A. Wheeler, Gravitation,
Freeman, p. 168, 1973. Cited on page 332.
297 E.A. Desloge & R.J. Philpott, Uniformly accelerated reference frames in special re-
lativity, American Journal of Physics 55, pp. 252–261, 1987. Cited on page 332.
298 R.H. Good, Uniformly accelerated reference frame and twin paradox, American Journal
of Physics 50, pp. 232–238, 1982. Cited on pages 332, 333, and 336.
299 J. Dwayne Hamilton, The uniformly accelerated reference frame, American Journal of
Dvipsbugw
Physics 46, pp. 83–89, 1978. Cited on page 333.
300 The best and cheapest mathematical formula collection remains the one by K. Rottmann,
Mathematische Formelsammlung, BI Hochschultaschenbücher, 1960. Cited on page 333.
301 C.G. Adler & R.W. Brehme, Relativistic solutions to a falling body in a uniform gravit-
ation field, American Journal of Physics 59, pp. 209–213, 1991. Cited on page 334.
302 See for example the excellent lecture notes by D.J. R aymond, A radically modern ap-
proach to freshman physics, on the http://www.physics.nmt.edu/~raymond/teaching.html

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

website. Cited on pages 334 and 347.
303 L. Mishra, The relativistic acceleration addition theorem, Classical and Quantum Gravity
11, pp. L97–L102, 1994. Cited on page 334.
304 Edward A. Desloge, The gravitational red-shift in a uniform field, American Journal of
Physics 58, pp. 856–858, 1990. Cited on page 336.
305 One of the latest of these debatable experiments is T.P. Krisher, L. Maleki, G.F. Lu tes,
L.E. Primas, R.T. Logan, J.D. Anderson & C.M. Will, Test of the isotropy of the
one-way speed of light using hydrogen-maser frequency standards, Physical Review D 42,
pp. 731–734, 1990. Cited on page 338.
306 Edwin F. Taylor & A.P. French, Limitation on proper length in special relativity,
American Journal of Physics 51, pp. 889–893, 1983. Cited on page 339.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 C h a pter III

GR AV I TATION AND R EL ATI V I T Y

Dvipsbugw

G
eneral relativity is easy. Nowadays, it can be made as intuitive as universal
ravity and its inverse square law – by using the right approach. The main ideas of
eneral relativity, like those of special relativity, are accessible to secondary-school
students. Black holes, gravitational waves, space-time curvature and the limits of the uni-
verse can then be understood as easily as the Doppler effect or the twins paradox.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We will discover that, just as special relativity is based on a maximum speed c, general
relativity is based on a maximum force c 4 ~4G or on a maximum power c 5 ~4G. We first
show that all known experimental data are consistent with these limits. In fact, we find
that the maximum force and the maximum power are achieved only on insurmountable
limit surfaces; these limit surfaces are called horizons. We will then be able to deduce the
field equations of general relativity. In particular, the existence of a maximum for force
or power implies that space-time is curved. It explains why the sky is dark at night, and
it shows that the universe is of finite size.
We also discuss the main counter-arguments and paradoxes arising from the limits.
The resolutions of the paradoxes clarify why the limits have remained dormant for so
long, both in experiments and in teaching.
After this introduction, we will study the effects of relativistic gravity in more detail.
In particular, we will study the consequences of space-time curvature for the motions of
bodies and of light in our everyday environment. For example, the inverse square law will
be modified. (Can you explain why this is necessary in view of what we have learned so
Challenge 679 n far?) Most fascinating of all, we will discover how to move and bend the vacuum. Then we
will study the universe at large; finally, we will explore the most extreme form of gravity:
black holes.
Copyright © Christoph Schiller November 1997–May 2006

6. maximum force – general rel ativit y in one

statement
We just saw that the theory of special relativity appears when we recognize the speed limit
c in nature and take this limit as a basic principle. At the end of the twentieth century it
was shown that general relativity can be approached by using a similar basic principle:*

Ref. 307, Ref. 308 * This principle was published in the year 2000 in this text, and independently in a conference proceedings in
Ref. 309 2002 by Gary Gibbons. The present author discovered the maximum force in 1998 when searching for a way
to derive the results of chapter XI that would be so simple that it would convince even a secondary-school
student.

Dvipsbugw
350 iii gravitation and rel ativity • 6. maximum force

U There is in nature a maximum force:

c4
FD = 3.0 ë 1043 N . (202)
4G
In nature, no force in any muscle, machine or system can exceed this value.
For the curious, the value of the force limit is the energy of a (Schwarzschild) black
hole divided by twice its radius. The force limit can be understood intuitively by noting
that (Schwarzschild) black holes are the densest bodies possible for a given mass. Since Dvipsbugw
there is a limit to how much a body can be compressed, forces – whether gravitational,
electric, centripetal or of any other type – cannot be arbitrary large.
Alternatively, it is possible to use another, equivalent statement as a basic principle:

U There is a maximum power in nature:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

c5
PD = 9.1 ë 1051 W . (203)
4G
No power of any lamp, engine or explosion can exceed this value. The maximum power
is realized when a (Schwarzschild) black hole is radiated away in the time that light takes
to travel along a length corresponding to its diameter. We will see below precisely what
black holes are and why they are connected to these limits.
The existence of a maximum force or power implies the full theory of general relativ-
ity. In order to prove the correctness and usefulness of this approach, a sequence of argu-
ments is required. The sequence is the same as for the establishment of the limit speed in
special relativity. First of all, we have to gather all observational evidence for the claimed
limit. Secondly, in order to establish the limit as a principle of nature, we have to show
that general relativity follows from it. Finally, we have to show that the limit applies in all
possible and imaginable situations. Any apparent paradoxes will need to be resolved.
These three steps structure this introduction to general relativity. We start the story by
explaining the origin of the idea of a limiting value.

The maximum force and power limits

In the nineteenth and twentieth centuries many physicists took pains to avoid the concept
of force. Heinrich Hertz made this a guiding principle of his work, and wrote an influ-
Copyright © Christoph Schiller November 1997–May 2006

ential textbook on classical mechanics without ever using the concept. The fathers of
quantum theory, who all knew this text, then dropped the term ‘force’ completely from
the vocabulary of microscopic physics. Meanwhile, the concept of ‘gravitational force’
was eliminated from general relativity by reducing it to a ‘pseudo-force’. Force fell out of
fashion.
Nevertheless, the maximum force principle does make sense, provided that we visu-
alize it by means of the useful definition: force is the flow of momentum per unit time.
Momentum cannot be created or destroyed. We use the term ‘flow’ to remind us that
momentum, being a conserved quantity, can only change by inflow or outflow. In other
words, change of momentum always takes place through some boundary surface. This
fact is of central importance. Whenever we think about force at a point, we mean the

Dvipsbugw
 a simple principle encompassing general rel ativity 351

momentum ‘flowing’ through a surface at that point. General relativity states this idea
usually as follows: force keeps bodies from following geodesics. The mechanism under-
lying a measured force is not important. In order to have a concrete example to guide
the discussion it can be helpful to imagine force as electromagnetic in origin. In fact, any
type of force is possible.
The maximum force principle thus boils down to the following: if we imagine any
physical surface (and cover it with observers), the integral of momentum flow through the
surface (measured by all those observers) never exceeds a certain value. It does not matter
how the surface is chosen, as long as it is physical, i.e., as long as we can fix observers* Dvipsbugw
onto it.
This principle imposes a limit on muscles, the effect of hammers, the flow of material,
the acceleration of massive bodies, and much more. No system can create, measure or
experience a force above the limit. No particle, no galaxy and no bulldozer can exceed it.
The existence of a force limit has an appealing consequence. In nature, forces can be
measured. Every measurement is a comparison with a standard. The force limit provides a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

natural unit of force which fits into the system of natural units** that Max Planck derived
from c, G and h (or ħ). The maximum force thus provides a standard of force valid in
every place and at every instant of time.
The limit value of c 4 ~4G differs from Planck’s proposed unit in two ways. First, the
numerical factor is different (Planck had in mind the value c 4 ~G). Secondly, the force
unit is a limiting value. In the this respect, the maximum force plays the same role as the
Ref. 308 maximum speed. As we will see later on, this limit property is valid for all other Planck
Page 1068 units as well, once the numerical factors have been properly corrected. The factor 1/4 has
no deeper meaning: it is just the value that leads to the correct form of the field equations
of general relativity. The factor 1/4 in the limit is also required to recover, in everyday
Page 368 situations, the inverse square law of universal gravitation. When the factor is properly
taken into account, the maximum force (or power) is simply given by the (corrected)
Planck energy divided by the (corrected) Planck length or Planck time.
The expression for the maximum force involves the speed of light c and the gravita-
tional constant G; it thus qualifies as a statement on relativistic gravitation. The funda-
mental principle of special relativity states that speed v obeys v D c for all observers. Ana-
logously, the basic principle of general relativity states that in all cases force F and power
P obey F D c 4 ~4G and P D c 5 ~4G. It does not matter whether the observer measures the
force or power while moving with high velocity relative to the system under observation,
during free fall, or while being strongly accelerated. However, we will see that it is essen-
Copyright © Christoph Schiller November 1997–May 2006

tial that the observer records values measured at his own location and that the observer
is realistic, i.e., made of matter and not separated from the system by a horizon. These
conditions are the same that must be obeyed by observers measuring velocity in special
relativity.
Since physical power is force times speed, and since nature provides a speed limit,
the force bound and the power bound are equivalent. We have already seen that force
* Observers in general relativity, like in special relativity, are massive physical systems that are small enough
so that their influence on the system under observation is negligible.
Page 706 ** When Planck discovered the quantum of action, he had also noticed the possibility to define natural units.
On a walk with his seven-year-old son in the forest around Berlin, he told him that he had made a discovery
as important as the discovery of universal gravity.

Dvipsbugw
 352 iii gravitation and rel ativity • 6. maximum force

Page 323 and power appear together in the definition of 4-force; we can thus say that the upper
bound is valid for every component of a force, as well as for its magnitude. The power
bound limits the output of car and motorcycle engines, lamps, lasers, stars, gravitational
radiation sources and galaxies. It is equivalent to 1.2 ë 1049 horsepowers. The maximum
power principle states that there is no way to move or get rid of energy more quickly than
that.
The power limit can be understood intuitively by noting that every engine produces
exhausts, i.e. some matter or energy that is left behind. For a lamp, a star or an evaporating
black hole, the exhausts are the emitted radiation; for a car or jet engine they are hot Dvipsbugw
gases; for a water turbine the exhaust is the slowly moving water leaving the turbine; for
a rocket it is the matter ejected at its back end; for a photon rocket or an electric motor
it is electromagnetic energy. Whenever the power of an engine gets close to the limit
value, the exhausts increase dramatically in mass–energy. For extremely high exhaust
masses, the gravitational attraction from these exhausts – even if they are only radiation
– prevents further acceleration of the engine with respect to them. The maximum power

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

principle thus expresses that there is a built-in braking mechanism in nature; this braking
mechanism is gravity.
Yet another, equivalent limit appears when the maximum power is divided by c 2 .
U There is a maximum rate of mass change in nature:

c3
D = 1.0 ë 1035 kg~s .
dm
(204)
dt 4G
This bound imposes a limit on pumps, jet engines and fast eaters. Indeed, the rate of flow
of water or any other material through tubes is limited. The mass flow limit is obviously
equivalent to either the force or the power limit.
The claim of a maximum force, power or mass change in nature seems almost too
fantastic to be true. Our first task is therefore to check it empirically as thoroughly as we
can.

The experimental evidence

Like the maximum speed principle, the maximum force principle must first of all be
checked experimentally. Michelson spent a large part of his research life looking for pos-
sible changes in the value of the speed of light. No one has yet dedicated so much effort to
Copyright © Christoph Schiller November 1997–May 2006

testing the maximum force or power. However, it is straightforward to confirm that no ex-
periment, whether microscopic, macroscopic or astronomical, has ever measured force
values larger than the stated limit. Many people have claimed to have produced speeds
larger than that of light. So far, nobody has ever claimed to have produced a force larger
than the limit value.
The large accelerations that particles undergo in collisions inside the Sun, in the most
powerful accelerators or in reactions due to cosmic rays correspond to force values much
smaller than the force limit. The same is true for neutrons in neutron stars, for quarks
inside protons, and for all matter that has been observed to fall towards black holes. Fur-
thermore, the search for space-time singularities, which would allow forces to achieve or
exceed the force limit, has been fruitless.

Dvipsbugw
 a simple principle encompassing general rel ativity 353

In the astronomical domain, all forces between stars or galaxies are below the limit
value, as are the forces in their interior. Not even the interactions between any two halves
of the universe exceed the limit, whatever physically sensible division between the two
Page 366 halves is taken. (The meaning of ‘physically sensible division’ will be defined below; for
divisions that are not sensible, exceptions to the maximum force claim can be constructed.
Challenge 680 n You might enjoy searching for such an exception.)
Astronomers have also failed to find any region of space-time whose curvature (a
concept to be introduced below) is large enough to allow forces to exceed the force limit.
Indeed, none of the numerous recent observations of black holes has brought to light Dvipsbugw
forces larger than the limit value or objects smaller than the corresponding black hole
radii. Observations have also failed to find a situation that would allow a rapid observer
to observe a force value that exceeds the limit due to the relativistic boost factor.
The power limit can also be checked experimentally. It turns out that the power – or
luminosity – of stars, quasars, binary pulsars, gamma ray bursters, galaxies or galaxy
clusters can indeed be close to the power limit. However, no violation of the limit has

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 310 ever been observed. Even the sum of all light output from all stars in the universe does
not exceed the limit. Similarly, even the brightest sources of gravitational waves, merging
black holes, do not exceed the power limit. Only the brightness of evaporating black holes
in their final phase could equal the limit. But so far, none has ever been observed.
Similarly, all observed mass flow rates are orders of magnitude below the correspond-
ing limit. Even physical systems that are mathematical analogues of black holes – for
example, silent acoustical black holes or optical black holes – do not invalidate the force
and power limits that hold in the corresponding systems.
The experimental situation is somewhat disappointing. Experiments do not contradict
the limit values. But neither do the data do much to confirm them. The reason is the
lack of horizons in everyday life and in experimentally accessible systems. The maximum
speed at the basis of special relativity is found almost everywhere; maximum force and
maximum power are found almost nowhere. Below we will propose some dedicated tests
Page 371 of the limits that could be performed in the future.

Deducing general relativity*

In order to establish the maximum force and power limits as fundamental physical prin-
ciples, it is not sufficient to show that they are consistent with what we observe in nature.
It is necessary to show that they imply the complete theory of general relativity. (This sec-
Copyright © Christoph Schiller November 1997–May 2006

tion is only for readers who already know the field equations of general relativity. Other
Page 357 readers may skip to the next section.)
In order to derive the theory of relativity we need to study those systems that realize
the limit under scrutiny. In the case of the special theory of relativity, the main system
that realizes the limit speed is light. For this reason, light is central to the exploration of
special relativity. In the case of general relativity, the systems that realize the limit are less
obvious. We note first that a maximum force (or power) cannot be realized throughout
a volume of space. If this were possible, a simple boost** could transform the force (or

* This section can be skipped at first reading. (The mentioned proof dates from December 2003.)
** A boost was defined in special relativity as a change of viewpoint to a second observer moving in relation
to the first.

Dvipsbugw
 354 iii gravitation and rel ativity • 6. maximum force

power) to a higher value. Therefore, nature can realize maximum force and power only
on surfaces, not volumes. In addition, these surfaces must be unattainable. These unat-
Ref. 308 tainable surfaces are basic to general relativity; they are called horizons. Maximum force
Page 336 and power only appear on horizons. We have encountered horizons in special relativity,
where they were defined as surfaces that impose limits to observation. (Note the contrast
with everyday life, where a horizon is only a line, not a surface.) The present definition
of a horizon as a surface of maximum force (or power) is equivalent to the definition as
a surface beyond which no signal may be received. In both cases, a horizon is a surface
beyond which interaction is impossible. Dvipsbugw
The connection between horizons and the maximum force is a central point of relativ-
istic gravity. It is as important as the connection between light and the maximum speed
in special relativity. In special relativity, we showed that the fact that light speed is the
maximum speed in nature implies the Lorentz transformations. In general relativity, we
will now prove that the maximum force in nature, which we can call the horizon force,
implies the field equations of general relativity. To achieve this aim, we start with the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

realization that all horizons have an energy flow across them. The flow depends on the
horizon curvature, as we will see. This connection implies that horizons cannot be planes,
as an infinitely extended plane would imply an infinite energy flow.
The simplest finite horizon is a static sphere, corresponding to a Schwarzschild black
hole. A spherical horizon is characterized by its radius of curvature R, or equivalently, by
its surface gravity a; the two quantities are related by 2aR = c 2 . Now, the energy flowing
through any horizon is always finite in extension, when measured along the propagation
direction. One can thus speak more specifically of an energy pulse. Any energy pulse
through a horizon is thus characterized by an energy E and a proper length L. When the
energy pulse flows perpendicularly through a horizon, the rate of momentum change, or
force, for an observer at the horizon is

F=
E
. (205)
L
Our goal is to show that the existence of a maximum force implies general relativity. Now,
maximum force is realized on horizons. We thus need to insert the maximum possible
values on both sides of equation (205) and to show that general relativity follows.
Using the maximum force value and the area 4πR 2 for a spherical horizon we get Copyright © Christoph Schiller November 1997–May 2006

c4
=
E
4πR 2 . (206)
4G LA

The fraction E~A is the energy per area flowing through any area A that is part of a horizon.
The insertion of the maximum values is complete when one notes that the length L of
the energy pulse is limited by the radius R. The limit L D R follows from geometrical
considerations: seen from the concave side of the horizon, the pulse must be shorter than
the radius of curvature. An independent argument is the following. The length L of an
Ref. 311 object accelerated by a is limited, by special relativity, by L D c 2 ~2a. Special relativity
already shows that this limit is related to the appearance of a horizon. Together with
relation (206), the statement that horizons are surfaces of maximum force leads to the

Dvipsbugw
 a simple principle encompassing general rel ativity 355

following important relation for static, spherical horizons:

c2
E= aA. (207)
8πG
This horizon equation relates the energy flow E through an area A of a spherical horizon
with surface gravity a. It states that the energy flowing through a horizon is limited, that
this energy is proportional to the area of the horizon, and that the energy flow is propor-
tional to the surface gravity. (The horizon equation is also called the first law of black hole Dvipsbugw
Ref. 312 mechanics or the first law of horizon mechanics.)
The above derivation also yields the intermediate result

c4 A
ED . (208)
16πG L

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

This form of the horizon equation states more clearly that no surface other than a horizon
can achieve the maximum energy flow, when the area and pulse length (or surface gravity)
are given. No other domain of physics makes comparable statements: they are intrinsic
to the theory of gravitation.
An alternative derivation of the horizon equation starts with the emphasis on power
instead of on force, using P = E~T as the initial equation.
It is important to stress that the horizon equations (207) and (208) follow from only
two assumptions: first, there is a maximum speed in nature, and secondly, there is a max-
imum force (or power) in nature. No specific theory of gravitation is assumed. The hori-
zon equation might even be testable experimentally, as argued below. (We also note that
the horizon equation – or, equivalently, the force or power limit – implies a maximum
mass change rate in nature given by dm~dt D c 3 ~4G.)
Next, we have to generalize the horizon equation from static and spherical horizons
to general horizons. Since the maximum force is assumed to be valid for all observers,
whether inertial or accelerating, the generalization is straightforward. For a horizon that
is irregularly curved or time-varying the horizon equation becomes

c2
δE = a δA . (209)
8πG Copyright © Christoph Schiller November 1997–May 2006

This differential relation – it might be called the general horizon equation – is valid for any
horizon. It can be applied separately for every piece δA of a dynamic or spatially changing
horizon. The general horizon equation (209) has been known to be equivalent to general
Ref. 313 relativity at least since 1995, when this equivalence was (implicitly) shown by Jacobson.
We will show that the differential horizon equation has the same role for general relativity
as the equation dx = c dt has for special relativity. From now on, when we speak of the
horizon equation, we mean the general, differential form (209) of the relation.
It is instructive to restate the behaviour of energy pulses of length L in a way that holds

Dvipsbugw
 356 iii gravitation and rel ativity • 6. maximum force

for any surface, even one that is not a horizon. Repeating the above derivation, one gets

c4 1
D
δE
. (210)
δA 16πG L
Equality is only realized when the surface A is a horizon. In other words, whenever the
value δE~δA in a physical system approaches the right-hand side, a horizon starts to form.
This connection will be essential in our discussion of apparent counter-examples to the
limit principles. Dvipsbugw
If one keeps in mind that on a horizon the pulse length L obeys L D c 2 ~2a, it becomes
clear that the general horizon equation is a consequence of the maximum force c 4 ~4G
or the maximum power c 5 ~4G. In addition, the horizon equation takes also into account
maximum speed, which is at the origin of the relation L D c 2 ~2a. The horizon equation
thus follows purely from these two limits of nature.
The remaining part of the argument is simply the derivation of general relativity from

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 313 the general horizon equation. This derivation was implicitly provided by Jacobson, and
the essential steps are given in the following paragraphs. (Jacobson did not stress that his
derivation was valid also for continuous space-time, or that his argument could also be
used in classical general relativity.) To see the connection between the general horizon
equation (209) and the field equations, one only needs to generalize the general horizon
equation to general coordinate systems and to general directions of energy–momentum
flow. This is achieved by introducing tensor notation that is adapted to curved space-time.
To generalize the general horizon equation, one introduces the general surface element
dΣ and the local boost Killing vector field k that generates the horizon (with suitable
norm). Jacobson uses these two quantities to rewrite the left-hand side of the general
horizon equation (209) as
δE = ∫T ab k
a
dΣ b , (211)

where Tab is the energy–momentum tensor. This expression obviously gives the energy
at the horizon for arbitrary coordinate systems and arbitrary energy flow directions.
Jacobson’s main result is that the factor a δA in the right hand side of the general hori-
zon equation (209) can be rewritten, making use of the (purely geometric) Raychaudhuri
equation, as
a δA = c 2 ∫R ab k
a
dΣ b , (212)
Copyright © Christoph Schiller November 1997–May 2006

where R ab is the Ricci tensor describing space-time curvature. This relation describes
how the local properties of the horizon depend on the local curvature.
Combining these two steps, the general horizon equation (209) becomes

c4
∫ Tab k a dΣ b = ∫R ab k
a
dΣ b . (213)
8πG
Jacobson then shows that this equation, together with local conservation of energy (i.e.,

Dvipsbugw
 a simple principle encompassing general rel ativity 357

vanishing divergence of the energy–momentum tensor) can only be satisfied if

c4
Tab = ‹R ab − ( + Λ)дab  ,
R
(214)
8πG 2

where R is the Ricci scalar and Λ is a constant of integration the value of which is not
determined by the problem. The above equations are the full field equations of general
relativity, including the cosmological constant Λ. The field equations thus follow from
the horizon equation. They are therefore shown to be valid at horizons. Dvipsbugw
Since it is possible, by choosing a suitable coordinate transformation, to position a
horizon at any desired space-time point, the field equations must be valid over the whole
of space-time. This observation completes Jacobson’s argument. Since the field equations
follow, via the horizon equation, from the maximum force principle, we have also shown
that at every space-time point in nature the same maximum force holds: the value of the
maximum force is an invariant and a constant of nature.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In other words, the field equations of general relativity are a direct consequence of the
limit on energy flow at horizons, which in turn is due to the existence of a maximum
force (or power). In fact, as Jacobson showed, the argument works in both directions.
Maximum force (or power), the horizon equation, and general relativity are equivalent.
In short, the maximum force principle is a simple way to state that, on horizons, energy
flow is proportional to area and surface gravity. This connection makes it possible to de-
duce the full theory of general relativity. In particular, a maximum force value is sufficient
to tell space-time how to curve. We will explore the details of this relation shortly. Note
that if no force limit existed in nature, it would be possible to ‘pump’ any desired amount
of energy through a given surface, including any horizon. In this case, the energy flow
would not be proportional to area, horizons would not have the properties they have, and
general relativity would not hold. We thus get an idea how the maximum flow of energy,
the maximum flow of momentum and the maximum flow of mass are all connected to
horizons. The connection is most obvious for black holes, where the energy, momentum
Page 478 or mass are those falling into the black hole.
By the way, since the derivation of general relativity from the maximum force principle
or from the maximum power principle is now established, we can rightly call these limits
horizon force and horizon power. Every experimental or theoretical confirmation of the
field equations indirectly confirms their existence. Copyright © Christoph Schiller November 1997–May 2006

Space-time is curved
Imagine two observers who start moving parallel to each other and who continue straight
ahead. If after a while they discover that they are not moving parallel to each other any
Challenge 681 n more, then they can deduce that they have moved on a curved surface (try it!) or in a
curved space. In particular, this happens near a horizon. The derivation above showed
that a finite maximum force implies that all horizons are curved; the curvature of horizons
in turn implies the curvature of space-time. If nature had only flat horizons, there would
be no space-time curvature. The existence of a maximum force implies that space-time
is curved.
A horizon so strongly curved that it forms a closed boundary, like the surface of a

Dvipsbugw
 358 iii gravitation and rel ativity • 6. maximum force

Page 476 sphere, is called a black hole. We will study black holes in detail below. The main property
of a black hole, like that of any horizon, is that it is impossible to detect what is ‘behind’
the boundary.*
The analogy between special and general relativity can thus be carried further. In spe-
cial relativity, maximum speed implies dx = c dt, and the change of time depends on the
observer. In general relativity, maximum force (or power) implies the horizon equation
δE = 8πGc2
a δA and the observation that space-time is curved.
The maximum force (or power) thus has the same double role in general relativity as
Dvipsbugw
the maximum speed has in special relativity. In special relativity, the speed of light is the
maximum speed; it is also the proportionality constant that connects space and time, as
the equation dx = c dt makes apparent. In general relativity, the horizon force is the max-
imum force; it also appears (with a factor 2π) in the field equations as the proportionality
constant connecting energy and curvature. The maximum force thus describes both the
elasticity of space-time and – if we use the simple image of space-time as a medium – the
maximum tension to which space-time can be subjected. This double role of a material

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

constant as proportionality factor and as limit value is well known in materials science.
Does this analogy make you think about aether? Do not worry: physics has no need
Page 581 for the concept of aether, because it is indistinguishable from vacuum. General relativity
does describe the vacuum as a sort of material that can be deformed and move.
Why is the maximum force also the proportionality factor between curvature and en-
ergy? Imagine space-time as an elastic material. The elasticity of a material is described
by a numerical material constant. The simplest definition of this material constant is the
ratio of stress (force per area) to strain (the proportional change of length). An exact
definition has to take into account the geometry of the situation. For example, the shear
modulus G (or µ) describes how difficult it is to move two parallel surfaces of a material
against each other. If the force F is needed to move two parallel surfaces of area A and
length l against each other by a distance ∆l, one defines the shear modulus G by

=G
F ∆l
. (215)
A l
The shear modulus for metals and alloys ranges between 25 and 80 GPa. The continuum
theory of solids shows that for any crystalline solid without any defect (a ‘perfect’ solid)
there is a so-called theoretical shear stress: when stresses higher than this value are ap-
plied, the material breaks. The theoretical shear stress, in other words, the maximum
Copyright © Christoph Schiller November 1997–May 2006

stress in a material, is given by

G tss =
G
. (216)
2π
The maximum stress is thus essentially given by the shear modulus. This connection is
similar to the one we found for the vacuum. Indeed, imagining the vacuum as a material
Ref. 314 that can be bent is a helpful way to understand general relativity. We will use it regularly
in the following.
What happens when the vacuum is stressed with the maximum force? Is it also torn
apart like a solid? Yes: in fact, when vacuum is torn apart, particles appear. We will find
* Analogously, in special relativity it is impossible to detect what moves faster than the light barrier.

Dvipsbugw
 a simple principle encompassing general rel ativity 359

out more about this connection later on: since particles are quantum entities, we need
to study quantum theory first, before we can describe the effect in the last part of our
mountain ascent.

Conditions of validity of the force and power limits

The maximum force value is valid only under certain assumptions. To clarify this point,
we can compare it to the maximum speed. The speed of light (in vacuum) is an upper limit
for motion of systems with momentum or energy only. It can, however, be exceeded for Dvipsbugw
motions of non-material points. Indeed, the cutting point of a pair of scissors, a laser light
spot on the Moon, or the group velocity or phase velocity of wave groups can exceed the
Page 306 speed of light. In addition, the speed of light is a limit only if measured near the moving
mass or energy: the Moon moves faster than light if one turns around one’s axis in a
second; distant points in a Friedmann universe move apart from each other with speeds
larger than the speed of light. Finally, the observer must be realistic: the observer must be

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

made of matter and energy, and thus move more slowly than light, and must be able to
Ref. 315 observe the system. No system moving at or above the speed of light can be an observer.
The same three conditions apply in general relativity. In particular, relativistic gravity
forbids point-like observers and test masses: they are not realistic. Surfaces moving faster
than light are also not realistic. In such cases, counter-examples to the maximum force
Challenge 682 ny claim can be found. Try and find one – many are possible, and all are fascinating. We will
explore some of the most important ones.

Gedanken experiments and paradoxes about the force limit

Wenn eine Idee am Horizonte eben aufgeht, ist

“ gewöhnlich die Temperatur der Seele dabei sehr

kalt. Erst allmählich entwickelt die Idee ihre
Wärme, und am heissesten ist diese (das heisst
sie tut ihre grössten Wirkungen), wenn der
Glaube an die Idee schon wieder im Sinken ist.
Friedrich Nietzsche*

The last, but central, step in our discussion of the force limit is the same as in the discus-
sion of the speed limit. We need to show that any imaginable experiment – not only any
real one – satisfies the hypothesis. Following a tradition dating back to the early twen-
”
tieth century, such an imagined experiment is called a Gedanken experiment, from the
Copyright © Christoph Schiller November 1997–May 2006

German Gedankenexperiment, meaning ‘thought experiment’.

In order to dismiss all imaginable attempts to exceed the maximum speed, it is suf-
ficient to study the properties of velocity addition and the divergence of kinetic energy
near the speed of light. In the case of maximum force, the task is much more involved.
Indeed, stating a maximum force, a maximum power and a maximum mass change easily
provokes numerous attempts to contradict them. We will now discuss some of these.

* ‘When an idea is just rising on the horizon, the soul’s temperature with respect to it is usually very cold.
Only gradually does the idea develop its warmth, and it is hottest (which is to say, exerting its greatest in-
fluence) when belief in the idea is already once again in decline.’ Friedrich Nietzsche (1844–1900), German
philosopher and scholar. This is aphorism 207 – Sonnenbahn der Idee – from his text Menschliches Allzu-
menschliches – Der Wanderer und sein Schatten.

Dvipsbugw
 360 iii gravitation and rel ativity • 6. maximum force

**
The brute force approach. The simplest attempt to exceed the force limit is to try to ac-
celerate an object with a force larger than the maximum value. Now, acceleration implies
the transfer of energy. This transfer is limited by the horizon equation (209) or the limit
(210). For any attempt to exceed the force limit, the flowing energy results in the appear-
ance of a horizon. But a horizon prevents the force from exceeding the limit, because it
imposes a limit on interaction.
Page 339 We can explore this limit directly. In special relativity we found that the acceleration
of an object is limited by its length. Indeed, at a distance given by c 2 ~2a in the direction Dvipsbugw
opposite to the acceleration a, a horizon appears. In other words, an accelerated body
breaks, at the latest, at that point. The force F on a body of mass M and radius R is thus
limited by
FD
M 2
c . (217)
2R

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

It is straightforward to add the (usually small) effects of gravity. To be observable, an
accelerated body must remain larger than a black hole; inserting the corresponding radius
R = 2GM~c 2 we get the force limit (202). Dynamic attempts to exceed the force limit thus
fail.
**
The rope attempt. We can also try to generate a higher force in a static situation, for ex-
ample by pulling two ends of a rope in opposite directions. We assume for simplicity that
an unbreakable rope exists. To produce a force exceeding the limit value, we need to store
large (elastic) energy in the rope. This energy must enter from the ends. When we increase
the tension in the rope to higher and higher values, more and more (elastic) energy must
be stored in smaller and smaller distances. To exceed the force limit, we would need to
add more energy per distance and area than is allowed by the horizon equation. A hori-
zon thus inevitably appears. But there is no way to stretch a rope across a horizon, even if
it is unbreakable. A horizon leads either to the breaking of the rope or to its detachment
from the pulling system. Horizons thus make it impossible to generate forces larger than
the force limit. In fact, the assumption of infinite wire strength is unnecessary: the force
limit cannot be exceeded even if the strength of the wire is only finite.
We note that it is not important whether an applied force pulls – as for ropes or wires
– or pushes. In the case of pushing two objects against each other, an attempt to increase
Copyright © Christoph Schiller November 1997–May 2006

the force value without end will equally lead to the formation of a horizon, due to the
limit provided by the horizon equation. By definition, this happens precisely at the force
limit. As there is no way to use a horizon to push (or pull) on something, the attempt
to achieve a higher force ends once a horizon is formed. Static forces cannot exceed the
limit value.
**
The braking attempt. A force limit provides a maximum momentum change per time. We
can thus search for a way to stop a moving physical system so abruptly that the maximum
force might be exceeded. The non-existence of rigid bodies in nature, already known from
Page 339 special relativity, makes a completely sudden stop impossible; but special relativity on its

Dvipsbugw
 a simple principle encompassing general rel ativity 361

own provides no lower limit to the stopping time. However, the inclusion of gravity does.
Stopping a moving system implies a transfer of energy. The energy flow per area cannot
exceed the value given by the horizon equation. Thus one cannot exceed the force limit
by stopping an object.
Similarly, if a rapid system is reflected instead of stopped, a certain amount of energy
needs to be transferred and stored for a short time. For example, when a tennis ball is
reflected from a large wall its momentum changes and a force is applied. If many such
balls are reflected at the same time, surely a force larger than the limit can be realized?
It turns out that this is impossible. If one attempted it, the energy flow at the wall would Dvipsbugw
reach the limit given by the horizon equation and thus create a horizon. In that case, no
reflection is possible any more. So the limit cannot be exceeded.
**
The classical radiation attempt. Instead of systems that pull, push, stop or reflect mat-
ter, we can explore systems where radiation is involved. However, the arguments hold

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

in exactly the same way, whether photons, gravitons or other particles are involved. In
particular, mirrors, like walls, are limited in their capabilities.
It is also impossible to create a force larger than the maximum force by concentrating
a large amount of light onto a surface. The same situation as for tennis balls arises: when
the limit value E~A given by the horizon equation (210) is reached, a horizon appears
that prevents the limit from being broken.
**
The brick attempt. The force and power limits can also be tested with more concrete
Gedanken experiments. We can try to exceed the force limit by stacking weight. But even
building an infinitely high brick tower does not generate a sufficiently strong force on its
foundations: integrating the weight, taking into account its decrease with height, yields
a finite value that cannot reach the force limit. If we continually increase the mass dens-
ity of the bricks, we need to take into account that the tower and the Earth will change
into a black hole. And black holes, as mentioned above, do not allow the force limit to be
exceeded.
**
The boost attempt. A boost can apparently be chosen in such a way that a force value F in
Ref. 316 one frame is transformed into any desired value F ′ in another frame. However, this result
Copyright © Christoph Schiller November 1997–May 2006

is not physical. To be more concrete, imagine a massive observer, measuring the value F,
at rest with respect to a large mass, and a second observer moving towards the charged
mass with relativistic speed, measuring the value F ′ . Both observers can be thought as
being as small as desired. If one transforms the force field at rest F applying the Lorentz
transformations, the force F ′ for the moving observer can reach extremely high values,
as long as the speed is high enough. However, a force must be measured by an observer
located at the specific point. One has thus to check what happens when the rapid ob-
server moves towards the region where the force is supposed to exceed the force limit.
Suppose the observer has a mass m and a radius r. To be an observer, it must be larger
than a black hole; in other words, its radius must obey r A 2Gm~c 2 , implying that the ob-
server has a non-vanishing size. When the observer dives into the force field surrounding

Dvipsbugw
 362 iii gravitation and rel ativity • 6. maximum force

the sphere, there will be an energy flow E towards the observer determined by the trans-
formed field value and the crossing area of the observer. This interaction energy can be
made as small as desired, by choosing a sufficiently small observer, but the energy is never
zero. When the moving observer approaches the large massive charge, the interaction en-
ergy increases. Before the observer arrives at the point where the force was supposed to
be higher than the force limit, the interaction energy will reach the horizon limits (209) or
(210) for the observer. Therefore, a horizon appears and the moving observer is prevented
from observing anything at all, in particular any value above the horizon force.
The same limitation appears when electrical or other interactions are studied using a Dvipsbugw
test observer that is charged. In summary, boosts cannot beat the force limit.
**
The divergence attempt. The force on a test mass m at a radial distance d from a Schwarz-
Ref. 310 schild black hole (for Λ = 0) is given by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F= ¼
GMm
. (218)
d 2 1 − 2G M
d c2

In addition, the inverse square law of universal gravitation states that the force between
two masses m and M is
F=
GMm
. (219)
d2
Both expressions can take any value; this suggest that no maximum force limit exists.
A detailed investigation shows that the maximum force still holds. Indeed, the force
in the two situations diverges only for non-physical point-like masses. However, the max-
imum force implies a minimum approach distance to a mass m given by

d min =
2Gm
. (220)
c2
The minimum approach distance – in simple terms, this would be the corresponding
black hole radius – makes it impossible to achieve zero distance between two masses or
between a horizon and a mass. This implies that a mass can never be point-like, and that
there is a (real) minimum approach distance, proportional to the mass. If this minimum
Copyright © Christoph Schiller November 1997–May 2006

approach distance is introduced in equations (218) and (219), one gets

c4 c4
F= D
1
¼
Mm
4G (M + m)
(221)
2
1− M 4G
M+m

and
c4 c4
F= D
Mm
4G (M + m)2 4G
. (222)

The maximum force value is thus never exceeded, as long as we take into account the size

Dvipsbugw
 a simple principle encompassing general rel ativity 363

of observers.

**
The consistency problem. If observers cannot be point-like, one might question whether
it is still correct to apply the original definition of momentum change or energy change
as the integral of values measured by observers attached to a given surface. In general
relativity, observers cannot be point-like, but they can be as small as desired. The original
definition thus remains applicable when taken as a limit procedure for ever-decreasing
observer size. Obviously, if quantum theory is taken into account, this limit procedure Dvipsbugw
comes to an end at the Planck length. This is not an issue for general relativity, as long as
the typical dimensions in the situation are much larger than this value.
**
The quantum problem. If quantum effects are neglected, it is possible to construct surfaces
Challenge 683 ny with sharp angles or even fractal shapes that overcome the force limit. However, such

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

surfaces are not physical, as they assume that lengths smaller than the Planck length can
be realized or measured. The condition that a surface be physical implies that it must have
Ref. 308 an intrinsic uncertainty given by the Planck length. A detailed study shows that quantum
effects do not allow the horizon force to be exceeded.
**
The relativistically extreme observer attempt. Any extreme observer, whether in rapid
inertial or in accelerated motion, has no chance to beat the limit. In classical physics
we are used to thinking that the interaction necessary for a measurement can be made
as small as desired. This statement, however, is not valid for all observers; in particular,
extreme observers cannot fulfil it. For them, the measurement interaction is large. As a
result, a horizon forms that prevents the limit from being exceeded.
**
The microscopic attempt. We can attempt to exceed the force limit by accelerating a small
particle as strongly as possible or by colliding it with other particles. High forces do in-
deed appear when two high energy particles are smashed against each other. However,
if the combined energy of the two particles became high enough to challenge the force
limit, a horizon would appear before they could get sufficiently close.
In fact, quantum theory gives exactly the same result. Quantum theory by itself already
Copyright © Christoph Schiller November 1997–May 2006

Ref. 317 provides a limit to acceleration. For a particle of mass m it is given by

2mc 3
aD . (223)
ħ

Here, ħ = 1.1 ë 10−34 Js is the quantum of action, a fundamental constant of nature. In

particular, this acceleration limit is satisfied in particle accelerators, in particle collisions
and in pair creation. For example, the spontaneous generation of electron–positron pairs
in intense electromagnetic fields or near black hole horizons does respect the limit (223).
Inserting the maximum possible mass for an elementary particle, namely the (corrected)
Page 1078 Planck mass, we find that equation (223) then states that the horizon force is the upper

Dvipsbugw
 364 iii gravitation and rel ativity • 6. maximum force

bound for elementary particles.

**
The compaction attempt. Are black holes really the most dense form of matter or energy?
The study of black hole thermodynamics shows that mass concentrations with higher
Ref. 310 density than black holes would contradict the principles of thermodynamics. In black
hole thermodynamics, surface and entropy are related: reversible processes that reduce
entropy could be realized if physical systems could be compressed to smaller values than
the black hole radius. As a result, the size of a black hole is the limit size for a mass in Dvipsbugw
nature. Equivalently, the force limit cannot be exceeded in nature.
**
The force addition attempt. In special relativity, composing velocities by a simple vector
addition is not possible. Similarly, in the case of forces such a naive sum is incorrect; any
attempt to add forces in this way would generate a horizon. If textbooks on relativity had

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

explored the behaviour of force vectors under addition with the same care with which
they explored that of velocity vectors, the force bound would have appeared much earlier
in the literature. (Obviously, general relativity is required for a proper treatment.)
**
Challenge 684 r Can you propose and resolve another attempt to exceed the force or power limit?

Gedanken experiments with the power limit and the mass flow
limit
Like the force bound, the power bound must be valid for all imaginable systems. Here are
some attempts to refute it.

**
The cable-car attempt. Imagine an engine that accelerates a mass with an unbreakable
and massless wire (assuming that such a wire could exist). As soon as the engine reached
the power bound, either the engine or the exhausts would reach the horizon equation.
When a horizon appears, the engine cannot continue to pull the wire, as a wire, even
an infinitely strong one, cannot pass a horizon. The power limit thus holds whether the
engine is mounted inside the accelerating body or outside, at the end of the wire pulling
it.
Copyright © Christoph Schiller November 1997–May 2006

**
The mountain attempt. It is possible to define a surface that is so strangely bent that
it passes just below every nucleus of every atom of a mountain, like the surface A in
Figure 175. All atoms of the mountain above sea level are then just above the surface,
barely touching it. In addition, imagine that this surface is moving upwards with almost
the speed of light. It is not difficult to show that the mass flow through this surface is
higher than the mass flow limit. Indeed, the mass flow limit c 3 ~4G has a value of about
1035 kg~s. In a time of 10−22 s, the diameter of a nucleus divided by the speed of light, only
1013 kg need to flow through the surface: that is the mass of a mountain.

Dvipsbugw
a simple principle encompassing general rel ativity 365

6000 m

mountain

Dvipsbugw
nuclei

surface A

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

0m
surface B

F I G U R E 175 The mountain attempt to exceed the maximum

mass flow value

This surface seems to provide a counter-example to the limit. However, a closer look
shows that this is not the case. The problem is the expression ‘just below’. Nuclei are
quantum particles and have an indeterminacy in their position; this indeterminacy is
essentially the nucleus–nucleus distance. As a result, in order to be sure that the surface
of interest has all atoms above it, the shape cannot be that of surface A in Figure 175. It
must be a flat plane that remains below the whole mountain, like surface B in the figure.
However, a flat surface beneath a mountain does not allow the mass change limit to be
exceeded.

**
The multiple atom attempt. One can imagine a number of atoms equal to the number of
the atoms of a mountain that all lie with large spacing (roughly) in a single plane. Again,
the plane is moving upwards with the speed of light. But also in this case the uncertainty
Copyright © Christoph Schiller November 1997–May 2006

in the atomic positions makes it impossible to say that the mass flow limit has been ex-
ceeded.

**
The multiple black hole attempt. Black holes are typically large and the uncertainty in their
position is thus negligible. The mass limit c 3 ~4G, or power limit c 5 ~4G, corresponds to
the flow of a single black hole moving through a plane at the speed of light. Several black
holes crossing a plane together at just under the speed of light thus seem to beat the limit.
However, the surface has to be physical: an observer must be possible on each of its points.
But no observer can cross a black hole. A black hole thus effectively punctures the plane
surface. No black hole can ever be said to cross a plane surface; even less so a multiplicity

Dvipsbugw
 366 iii gravitation and rel ativity • 6. maximum force

of black holes. The limit remains valid.

**
The multiple neutron star attempt. The mass limit seems to be in reach when several
neutron stars (which are slightly less dense than a black hole of the same mass) cross a
plane surface at the same time, at high speed. However, when the speed approaches the
speed of light, the crossing time for points far from the neutron stars and for those that
actually cross the stars differ by large amounts. Neutron stars that are almost black holes
cannot be crossed in a short time in units of a coordinate clock that is located far from Dvipsbugw
the stars. Again, the limit is not exceeded.
**
The luminosity attempt. The existence of a maximum luminosity bound has been dis-
Ref. 310 cussed by astrophysicists. In its full generality, the maximum bound on power, i.e. on
energy per time, is valid for any energy flow through any physical surface whatsoever. The

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

physical surface may even run across the whole universe. However, not even bringing
together all lamps, all stars and all galaxies of the universe yields a surface which has a
larger power output than the proposed limit.
The surface must be physical.* A surface is physical if an observer can be placed on each
of its points. In particular, a physical surface may not cross a horizon, or have local detail
Page 1003 finer than a certain minimum length. This minimum length will be introduced later on;
it is given by the corrected Planck length. If a surface is not physical, it may provide a
Challenge 685 n counter-example to the power or force limits. However, these counter-examples make no
statements about nature. (Ex falso quodlibet.**)
**
The many lamp attempt. An absolute power limit imposes a limit on the rate of energy
transport through any imaginable surface. At first sight, it may seem that the combined
power emitted by two radiation sources that each emit 3/4 of the maximum value should
give 3/2 times that value. However, two such lamps would be so massive that they would
form a black hole. No amount of radiation that exceeds the limit can leave. Again, since
the horizon limit (210) is achieved, a horizon appears that swallows the light and prevents
the force or power limit from being exceeded.
**
Copyright © Christoph Schiller November 1997–May 2006

The light concentration attempt. Another approach is to shine a powerful, short and spher-
ical flash of light onto a spherical mass. At first sight it seems that the force and power
limits can be exceeded, because light energy can be concentrated into small volumes.
However, a high concentration of light energy forms a black hole or induces the mass to
form one. There is no way to pump energy into a mass at a faster rate than that dictated
by the power limit. In fact, it is impossible to group light sources in such a way that their
total output is larger than the power limit. Every time the force limit is approached, a
horizon appears that prevents the limit from being exceeded.

* It can also be called physically sensible.

** Anything can be deduced from a falsehood.

Dvipsbugw
 a simple principle encompassing general rel ativity 367

**
The black hole attempt. One possible system in nature that actually achieves the power
limit is the final stage of black hole evaporation. However, even in this case the power
limit is not exceeded, but only equalled.
**
The water flow attempt. One could try to pump water as rapidly as possible through a
large tube of cross-section A. However, when a tube of length L filled with water flowing
at speed v gets near to the mass flow limit, the gravity of the water waiting to be pumped Dvipsbugw
through the area A will slow down the water that is being pumped through the area. The
limit is again reached when the cross-section A turns into a horizon.

Checking that no system – from microscopic to astrophysical – ever exceeds the max-
imum power or maximum mass flow is a further test of general relativity. It may seem easy
to find a counter-example, as the surface may run across the whole universe or envelop

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

any number of elementary particle reactions. However, no such attempt succeeds.
In summary, in all situations where the force, power or mass-flow limit is challenged,
whenever the energy flow reaches the black hole mass–energy density in space or the
corresponding momentum flow in time, an event horizon appears; this horizon makes it
impossible to exceed the limits. All three limits are confirmed both in observation and
in theory. Values exceeding the limits can neither be generated nor measured. Gedanken
experiments also show that the three bounds are the tightest ones possible. Obviously, all
three limits are open to future tests and to further Gedanken experiments. (If you can
Challenge 686 ny think of a good one, let me know.)

Hide and seek

The absence of horizons in everyday life is the first reason why the maximum force prin-
ciple remained undiscovered for so long. Experiments in everyday life do not highlight
the force or power limits. The second reason why the principle remained hidden is the
erroneous belief in point particles. This is a theoretical reason. (Prejudices against the
concept of force in general relativity have also been a factor.) The principle of maximum
force – or of maximum power – has thus remained hidden for so long because of a ‘con-
spiracy’ of nature that hid it both from theorists and from experimentalists.
For a thorough understanding of general relativity it is essential to remember that
Copyright © Christoph Schiller November 1997–May 2006

point particles, point masses and point-like observers do not exist. They are approxima-
tions only applicable in Galilean physics or in special relativity. In general relativity, hori-
zons prevent their existence. The habit of believing that the size of a system can be made
as small as desired while keeping its mass constant prevents the force or power limit from
being noticed.

Dvipsbugw
 368 iii gravitation and rel ativity • 6. maximum force

An intuitive understanding of general relativity

Wir leben zwar alle unter dem gleichen Himmel,

“ aber wir haben nicht alle den gleichen

Horizont.*
Konrad Adenauer

The concepts of horizon force and horizon power can be used as the basis for a direct,
intuitive approach to general relativity.
” Dvipsbugw
**
What is gravity? Of the many possible answers we will encounter, we now have the first:
gravity is the ‘shadow’ of the maximum force. Whenever we experience gravity as weak,
we can remember that a different observer at the same point and time would experience
the maximum force. Searching for the precise properties of that observer is a good exer-
cise. Another way to put it: if there were no maximum force, gravity would not exist.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
The maximum force implies universal gravity. To see this, we study a simple planetary
system, i.e., one with small velocities and small forces. A simple planetary system of size
L consists of a (small) satellite circling a central mass M at a radial distance R = L~2.
º implies the 2condition aL  c ,
2
Let a be the acceleration of the object. Small velocity
deduced from special relativity; small force implies 4GMa  c , deduced from the
force limit. These conditions are valid for the system as a whole and for all its components.
Both expressions have the dimensions of speed squared.ºSince the system has only one
characteristic speed, the two expressions aL = 2aR and 4GMa must be proportional,
yielding
a= f 2 ,
GM
(224)
R

where the numerical factor f must still be determined. To determine it, we study the
escape velocity necessary to leave the central body. The escape velocity must be smaller
than the speed of light for any body larger than a black hole. The escape velocity, derived
from expression (224), from a body of mass M and radius R is given by v esc 2
= 2 f GM~R.
The minimum radius R of objects, given by R = 2GM~c , then implies that f = 1. There-
2

fore, for low speeds and low forces, the inverse square law describes the orbit of a satellite
Copyright © Christoph Schiller November 1997–May 2006

around a central mass.

**
If empty space-time is elastic, like a piece of metal, it must also be able to oscillate. Any
physical system can show oscillations when a deformation brings about a restoring force.
We saw above that there is such a force in the vacuum: it is called gravitation. In other
words, vacuum must be able to oscillate, and since it is extended, it must also be able to
sustain waves. Indeed, gravitational waves are predicted by general relativity, as we will
Page 402 see below.

* ‘We all live under the same sky, but we do not have the same horizon.’ Konrad Adenauer, German chancel-
lor.

Dvipsbugw
 a simple principle encompassing general rel ativity 369

**
If curvature and energy are linked, the maximum speed must also hold for gravitational
energy. Indeed, we will find that gravity has a finite speed of propagation. The inverse
square law of everyday life cannot be correct, as it is inconsistent with any speed limit.
More about the corrections induced by the maximum speed will become clear shortly.
In addition, since gravitational waves are waves of massless energy, we would expect the
Page 402 maximum speed to be their propagation speed. This is indeed the case, as we will see.
** Dvipsbugw
A body cannot be denser than a (non-rotating) black hole of the same mass. The max-
imum force and power limits that apply to horizons make it impossible to squeeze mass
into smaller horizons. The maximum force limit can therefore be rewritten as a limit for
the size L of physical systems of mass m:

LE
4Gm

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

. (225)
c2
If we call twice the radius of a black hole its ‘size’, we can state that no physical system
of mass m is smaller than this value.* The » size limit plays an important role in general
relativity. The opposite inequality, m E A~16π c ~G, which describes the maximum
2

‘size’ of black holes, is called the Penrose inequality and has been proven for many physic-
Ref. 318, Ref. 319, ally realistic situations. The Penrose inequality can be seen to imply the maximum force
Ref. 320 limit, and vice versa. The maximum force principle, or the equivalent minimum size of
matter–energy systems, thus prevents the formation of naked singularities, and implies
the validity of the so-called cosmic censorship.

**
There is a power limit for all energy sources. In particular, the value c 5 ~4G limits the lu-
minosity of all gravitational sources. Indeed, all formulae for gravitational wave emission
Ref. 310 imply this value as an upper limit. Furthermore, numerical relativity simulations never
exceed it: for example, the power emitted during the simulated merger of two black holes
is below the limit.

**
Perfectly plane waves do not exist in nature. Plane waves are of infinite extension. But
Copyright © Christoph Schiller November 1997–May 2006

neither electrodynamic nor gravitational waves can be infinite, since such waves would
carry more momentum per time through a plane surface than is allowed by the force
limit. The non-existence of plane gravitational waves also precludes the production of
singularities when two such waves collide.

**
In nature, there are no infinite forces. There are thus no naked singularities in nature.
Horizons prevent the appearance of naked singularities. In particular, the big bang was

* The maximum value for the mass to size limit is obviously equivalent to the maximum mass change given
above.

Dvipsbugw
 370 iii gravitation and rel ativity • 6. maximum force

not a singularity. The mathematical theorems by Penrose and Hawking that seem to imply
the existence of singularities tacitly assume the existence of point masses – often in the
form of ‘dust’ – in contrast to what general relativity implies. Careful re-evaluation of
each such proof is necessary.
**
The force limit means that space-time has a limited stability. The limit suggests that space-
time can be torn into pieces. This is indeed the case. However, the way that this happens
is not described general relativity. We will study it in the third part of this text. Dvipsbugw

**
The maximum force is the standard of force. This implies that the gravitational constant
G is constant in space and time – or at least, that its variations across space and time
Ref. 321 cannot be detected. Present data support this claim to a high degree of precision.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
The maximum force principle implies that gravitational energy – as long as it can be
defined – falls in gravitational fields in the same way as other type of energy. As a result,
Ref. 310 the maximum force principle predicts that the Nordtvedt effect vanishes. The Nordtvedt
effect is a hypothetical periodical change in the orbit of the Moon that would appear if the
gravitational energy of the Earth–Moon system did not fall, like other mass–energy, in
the gravitational field of the Sun. Lunar range measurements have confirmed the absence
of this effect.

**
If horizons are surfaces, we can ask what their colour is. This question will be explored
Page 477 later on.
**
Page 1077 Later on we will find that quantum effects cannot be used to exceed the force or power
Challenge 687 e limit. (Can you guess why?) Quantum theory also provides a limit to motion, namely
a lower limit to action; however, this limit is independent of the force or power limit.
(A dimensional analysis already shows this: there is no way to define an action by com-
binations of c and G.) Therefore, even the combination of quantum theory and general
relativity does not help in overcoming the force or power limits.
Copyright © Christoph Schiller November 1997–May 2006

An intuitive understanding of cosmology

A maximum power is the simplest possible explanation of Olbers’ paradox. Power and
luminosity are two names for the same observable. The sum of all luminosities in the
universe is finite; the light and all other energy emitted by all stars, taken together, is finite.
If one assumes that the universe is homogeneous and isotropic, the power limit P D c 5 ~4G
must be valid across any plane that divides the universe into two halves. The part of the
universe’s luminosity that arrives on Earth is then so small that the sky is dark at night. In
fact, the actually measured luminosity is still smaller than this calculation, as a large part
of the power is not visible to the human eye (since most of it is matter anyway). In other

Dvipsbugw
 a simple principle encompassing general rel ativity 371

words, the night is dark because of nature’s power limit. This explanation is not in contrast
to the usual one, which uses the finite lifetime of stars, their finite density, their finite size,
and the finite age and the expansion of the universe. In fact, the combination of all these
usual arguments simply implies and repeats in more complex words that the power limit
cannot be exceeded. However, this more simple explanation seems to be absent in the
literature.
The existence of a maximum force in nature, together with homogeneity and isotropy,
implies that the visible universe is of finite size. The opposite case would be an infinitely
large, homogeneous and isotropic universe. But in that case, any two halves of the uni- Dvipsbugw
verse would attract each other with a force above the limit (provided the universe were
sufficiently old). This result can be made quantitative by imagining a sphere whose centre
lies at the Earth, which encompasses all the universe, and whose radius decreases with
time almost as rapidly as the speed of light. The mass flow dm~dt = ρAv is predicted to
reach the mass flow limit c 3 ~4G; thus one has

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

c3
= ρ 0 4πR 02 c =
dm
, (226)
dt 4G
Ref. 322 a relation also predicted by the Friedmann models. The precision measurements of the
cosmic background radiation by the WMAP satellite confirm that the present-day total
energy density ρ 0 (including dark matter and dark energy) and the horizon radius R 0
just reach the limit value. The maximum force limit thus predicts the observed size of the
universe.
A finite power limit also suggests that a finite age for the universe can be deduced. Can
Challenge 688 ny you find an argument?

Experimental challenges for the third millennium

The lack of direct tests of the horizon force, power or mass flow is obviously due to the
lack of horizons in the environment of all experiments performed so far. Despite the
difficulties in reaching the limits, their values are observable and falsifiable.
In fact, the force limit might be tested with high-precision measurements in binary
pulsars or binary black holes. Such systems allow precise determination of the positions
of the two stars. The maximum force principle implies a relation between the position
Ref. 308 error ∆x and the energy error ∆E. For all systems one has
Copyright © Christoph Schiller November 1997–May 2006

c4
D
∆E
. (227)
∆x 4G

For example, a position error of 1 mm gives a mass error of below 3 ë 1023 kg. In everyday
life, all measurements comply with this relation. Indeed, the left side is so much smaller
than the right side that the relation is rarely mentioned. For a direct check, only systems
which might achieve direct equality are interesting. Dual black holes or dual pulsars are
such systems.
It might be that one day the amount of matter falling into some black hole, such as the
one at the centre of the Milky Way, might be measured. The limit dm~dt D c 3 ~4G could

Dvipsbugw
 372 iii gravitation and rel ativity • 6. maximum force

then be tested directly.

The power limit implies that the highest luminosities are only achieved when systems
emit energy at the speed of light. Indeed, the maximum emitted power is only achieved
when all matter is radiated away as rapidly as possible: the emitted power P = Mc 2 ~(R~v)
cannot reach the maximum value if the body radius R is larger than that of a black hole
(the densest body of a given mass) or the emission speed v is lower than that of light. The
sources with highest luminosity must therefore be of maximum density and emit entities
without rest mass, such as gravitational waves, electromagnetic waves or (maybe) gluons.
Candidates to detect the limit are black holes in formation, in evaporation or undergoing Dvipsbugw
mergers.
A candidate surface that reaches the limit is the night sky. The night sky is a horizon.
Provided that light, neutrino, particle and gravitational wave flows are added together, the
limit c 5 ~4G is predicted to be reached. If the measured power is smaller than the limit
(as it seems to be at present), this might even give a hint about new particles yet to be
discovered. If the limit were exceeded or not reached, general relativity would be shown

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

to be incorrect. This might be an interesting future experimental test.
The power limit implies that a wave whose integrated intensity approaches the force
limit cannot be plane. The power limit thus implies a limit on the product of intensity
I (given as energy per unit time and unit area) and the size (curvature radius) R of the
front of a wave moving with the speed of light c:

c5
4πR 2 I D . (228)
4G
Obviously, this statement is difficult to check experimentally, whatever the frequency
and type of wave might be, because the value appearing on the right-hand side is ex-
tremely large. Possibly, future experiments with gravitational wave detectors, X-ray de-
tectors, gamma ray detectors, radio receivers or particle detectors might allow us to test
relation (228) with precision. (You might want to predict which of these experiments will
Challenge 689 e confirm the limit first.)
The lack of direct experimental tests of the force and power limits implies that indirect
tests become particularly important. All such tests study the motion of matter or energy
and compare it with a famous consequence of the force and power limits: the field equa-
tions of general relativity. This will be our next topic. Copyright © Christoph Schiller November 1997–May 2006

A summary of general relativity

There is a simple axiomatic formulation of general relativity: the horizon force c 4 ~4G
and the horizon power c 5 ~4G are the highest possible force and power values. No contra-
dicting observation is known. No counter-example has been imagined. General relativity
follows from these limits. Moreover, the limits imply the darkness of the night and the
finiteness of the size of the universe.
The principle of maximum force has obvious applications for the teaching of general re-
lativity. The principle brings general relativity to the level of first-year university, and pos-
sibly to well-prepared secondary school, students: only the concepts of maximum force
and horizon are necessary. space-time curvature is a consequence of horizon curvature.

Dvipsbugw
 a simple principle encompassing general rel ativity 373

The concept of a maximum force points to an additional aspect of gravitation. The cos-
Challenge 690 ny mological constant Λ is not fixed by the maximum force principle. (However, the prin-
Page 460 ciple does fix its sign to be positive.) Present measurements give the result Λ  10−52 ~m2 .
A positive cosmological constant implies the existence of a negative energy volume dens-
ity −Λc 4 ~G. This value corresponds to a negative pressure, as pressure and energy dens-
ity have the same dimensions. Multiplication by the (numerically corrected) Planck area
Page 1076 2Għ~c 3 , the smallest area in nature, gives a force value

F = 2Λħc = 0.60 ë 10−77 N . (229) Dvipsbugw

» is also the gravitational force between two (numerically

This º corrected) Planck masses
ħc~8G located at the cosmological distance 1~4 Λ . If we make the somewhat wishful
assumption that expression (229) is the smallest possible force in nature (the numerical
factors are not yet verified), we get the fascinating conjecture that the full theory of general
relativity, including the cosmological constant, may be defined by the combination of a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 691 ny maximum and a minimum force in nature. (Can you find a smaller force?)
Proving the minimum force conjecture is more involved than for the case of the max-
imum force. So far, only some hints are possible. Like the maximum force, the minimum
force must be compatible with gravitation, must not be contradicted by any experiment,
and must withstand any Gedanken experiment. A quick check shows that the minimum
force, as we have just argued, allows us to deduce gravitation, is an invariant, and is not
contradicted by any experiment. There are also hints that there may be no way to generate
or measure a smaller value. For example, the minimum force corresponds to the energy
per length contained by a photon with a wavelength of the size of the universe. It is hard
– but maybe not impossible – to imagine the production of a still smaller force.
We have seen that the maximum force principle and general relativity fail to fix the
value of the cosmological constant. Only a unified theory can do so. We thus get two re-
quirements for such a theory. First, any unified theory must predict the same upper limit
to force. Secondly, a unified theory must fix the cosmological constant. The appearance of
ħ in the conjectured expression for the minimum force suggests that the minimum force
is determined by a combination of general relativity and quantum theory. The proof of
this suggestion and the direct measurement of the minimum force are two important
challenges for our ascent beyond general relativity.
We are now ready to explore the consequences of general relativity and its field equa-
tions in more detail. We start by focusing on the concept of space-time curvature in every-
Copyright © Christoph Schiller November 1997–May 2006

day life, and in particular, on its consequences for the observation of motion.

Acknowledgement
The author thanks Steve Carlip, Corrado Massa, Tom Helmond, Gary Gibbons, Heinrich
Neumaier and Peter Brown for interesting discussions on these topics.

Dvipsbugw
374 iii gravitation and rel ativity

Biblio graphy

307 It seems that the first published statement of the principle was in an earlier edition of this
text, in the chapter on gravitation and relativity. The reference is Christoph Schiller,
Motion Mountain – A Hike Through and Beyond Space and Time Following the Concepts of
Modern Physics, found at http://www.motionmountain.net. The approach was discussed in
various usenet discussion groups in the early twenty-first century. The result is also pub-
lished in C. Schiller, General Relativity and cosmology derived from principle of max-
imum power or force, International Journal of Theoretical Physics 44, pp. 1629–1647, 2005. Dvipsbugw
Cited on page 349.
308 C. Schiller, Maximum force and minimum distance: physics in limit statements, part
of this text and downloadable at http://www.motionmountain.net/C19-LIMI.pdf. Cited on
pages 349, 351, 354, 363, and 371.
309 G.W. Gibbons, The maximum tension principle in general relativity, Foundations of Phys-
ics 32, pp. 1891–1901, 2002, or http://www.arxiv.org/abs/hep-th/0210109. Gary Gibbons ex-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

plains that the maximum force follows from general relativity; he does not make a statement
about the converse. Cited on page 349.
310 H.C. Ohanian & R. Ruffini, Gravitation and Spacetime, W.W. Norton & Co., New York,
1994. Cited on pages 353, 362, 364, 366, 369, and 370.
311 See for example Wolf gang R indler, Relativity – Special, General and Cosmological, Ox-
ford University Press, 2001, p. 70 ff, or R ay d’Inverno Introducing Einstein’s Relativity,
Clarendon Press, 1992, p. 36 ff. Cited on page 354.
312 See for example A. Ashtekar, S. Fairhust & B. Krishnan, Isolated horizons:
Hamiltonian evolution and the first law, http://www.arxiv.org/abs/gr-qc/0005083. Cited
on page 355.
313 T. Jacobson, Thermodynamics of spacetime: the Einstein equation of state, Physical Re-
view Letters 75, pp. 1260–1263, 1995 or http://www.arxiv.org/abs/gr-qc/9504004. Cited on
pages 355 and 356.
314 See for example Ekkehart Kröner, Kontinuumstheorie der Versetzungen und Eigenspan-
nungen, Springer, 1958, volume 5 of the series ‘Ergebnisse der angewandten Mathematik’.
Kröner shows the similarity between the equations, methods and results of solid-state con-
tinuum physics and those of general relativity. Cited on page 358.
315 Edwin F. Taylor & John A. Wheeler, Spacetime Physics – Introduction to Special
Relativity, second edition, Freeman, 1992. Cited on page 359.
316 This counter-example was suggested by Steve Carlip. Cited on page 361.
Copyright © Christoph Schiller November 1997–May 2006

317 E.R. Caianiello, Lettere al Nuovo Cimento 41, p. 370, 1984. Cited on page 363.
318 R. Penrose, Naked singularities, Annals of the New York Academy of Sciences 224, pp. 125–
134, 1973. Cited on page 369.
319 G. Huisken & T. Ilmanen, The Riemannian Penrose inequality, Int. Math. Res. Not. 59,
pp. 1045–1058, 1997. Cited on page 369.
320 S.A. Hay ward, Inequalities relating area, energy, surface gravity and charge of black holes,
Physical Review Letters 81, pp. 4557–4559, 1998. Cited on page 369.
321 C. Will, Was Einstein Right? – Putting General Relativity to the Test, Oxford University
Press, 1993. See also his paper http://www.arxiv.org/abs/gr-qc/9811036. Cited on page 370.
322 The measurement results by the WMAP satellite are summarized on the website http://map.

Dvipsbugw
bibliography 375

gsfc.nasa.gov/m_mm/mr_limits.html; the papers are available at http://lambda.gsfc.nasa.

gov/product/map/current/map_bibliography.cfm. Cited on page 371.

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 the new ideas on space, time and gravity 377

7. the new ideas on space, time and gravit y

Sapere aude.

“
Gravitational influences do transport energy.** Our description of motion must therefore
Horace*

”
be precise enough to imply that this transport can happen at most with the speed of light.
Henri Poincaré stated this requirement as long ago as 1905. The results following from this Dvipsbugw
principle will be fascinating: we will find that empty space can move, that the universe has
a finite age and that objects can be in permanent free fall. It will turn out that empty space
can be bent, although it is much stiffer than steel. Despite these strange consequences, the
theory and all its predictions have been confirmed by all experiments.
The theory of universal gravitation, which describes motion due to gravity using the
relation a = GM~r 2 , allows speeds higher than that of light. Indeed, the speed of a mass
in orbit is not limited. It is also unclear how the values of a and r depend on the observer.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

So this theory cannot be correct. In order to reach the correct description, called general
Ref. 323, Ref. 324 relativity by Albert Einstein, we have to throw quite a few preconceptions overboard.

Rest and free fall

The opposite of motion in daily life is a body at rest, such as a child sleeping or a rock
defying the waves. A body is at rest whenever it is not disturbed by other bodies. In the
Galilean description of the world, rest is the absence of velocity. In special relativity, rest
became inertial motion, since no inertially moving observer can distinguish its own mo-
tion from rest: nothing disturbs him. Both the rock in the waves and the rapid protons
crossing the galaxy as cosmic rays are at rest. The inclusion of gravity leads us to an even
more general definition.
Challenge 692 e If any body moving inertially is to be considered at rest, then any body in free fall
must also be. Nobody knows this better than Joseph Kittinger, the man who in August
Ref. 325 1960 stepped out of a balloon capsule at the record height of 31.3 km. At that altitude, the
air is so thin that during the first minute of his free fall he felt completely at rest, as if
he were floating. Although an experienced parachutist, he was so surprised that he had
to turn upwards in order to convince himself that he was indeed moving away from his
balloon! Despite his lack of any sensation of movement, he was falling at up to 274 m~s
or 988 km~h with respect to the Earth’s surface. He only started feeling something when
Copyright © Christoph Schiller November 1997–May 2006

he encountered the first substantial layers of air. That was when his free fall started to be
disturbed. Later, after four and a half minutes of fall, his special parachute opened; and
nine minutes later he landed in New Mexico.
Kittinger and all other observers in free fall, such as the cosmonauts circling the Earth
or the passengers in parabolic aeroplane flights,*** make the same observation: it is im-
possible to distinguish anything happening in free fall from what would happen at rest.
This impossibility is called the principle of equivalence; it is one of the starting points of

* ‘Venture to be wise.’ Quintus Horatius Flaccus, Ep. 1, 2, 40.

** The details of this statement are far from simple. They are discussed on page 402 and page 433.
*** Nowadays it is possible to book such flights in specialized travel agents.

Dvipsbugw
 378 iii gravitation and rel ativity • 7. the new ideas

general relativity. It leads to the most precise – and final – definition of rest: rest is free
fall. Rest is lack of disturbance; so is free fall.
The set of all free-falling observers at a point in space-time generalizes the special-
relativistic notion of the set of the inertial observers at a point. This means that we must
describe motion in such a way that not only inertial but also freely falling observers can
talk to each other. In addition, a full description of motion must be able to describe gravit-
ation and the motion it produces, and it must be able to describe motion for any observer
imaginable. General relativity realizes this aim.
As a first step, we put the result in simple words: true motion is the opposite of free fall. Dvipsbugw
This statement immediately rises a number of questions: Most trees or mountains are not
Challenge 693 n in free fall, thus they are not at rest. What motion are they undergoing? And if free fall is
rest, what is weight? And what then is gravity anyway? Let us start with the last question.

What is gravity? – A second answer

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In the beginning, we described gravity as the shadow of the maximum force. But there
is a second way to describe it, more related to everyday life. As William Unruh likes to
Ref. 326 explain, the constancy of the speed of light for all observers implies a simple conclusion:
gravity is the uneven running of clocks at different places.* Of course, this seemingly ab-
Challenge 695 e surd definition needs to be checked. The definition does not talk about a single situation
seen by different observers, as we often did in special relativity. The definition depends
of the fact that neighbouring, identical clocks, fixed against each other, run differently in
the presence of a gravitational field when watched by the same observer; moreover, this
difference is directly related to what we usually call gravity. There are two ways to check
this connection: by experiment and by reasoning. Let us start with the latter method, as
it is cheaper, faster and more fun.
An observer feels no difference between
gravity and constant acceleration. We can thus v(t)=gt
study constant acceleration and use a way of
reasoning we have encountered already in the
chapter on special relativity. We assume light is B light F
emitted at the back end of a train of length ∆h
that is accelerating forward with acceleration д,
as shown in Figure 176. The light arrives at the F I G U R E 176 Inside an accelerating train or
front after a time t = ∆h~c. However, during bus
Copyright © Christoph Schiller November 1997–May 2006

this time the accelerating train has picked up

some additional velocity, namely ∆v = дt = д∆h~c. As a result, because of the Doppler
Page 305 effect we encountered in our discussion of special relativity, the frequency f of the light
Challenge 696 e arriving at the front has changed. Using the expression of the Doppler effect, we thus get*

* Gravity is also the uneven length of metre bars at different places, as we will see below. Both effects are
needed to describe it completely; but for daily life on Earth, the clock effect is sufficient, since it is much
Challenge 694 n larger than the length effect, which can usually be neglected. Can you see why?

Dvipsbugw
 the new ideas on space, time and gravity 379

= 2 .
∆ f д∆h
(230)
f c

The sign of the frequency change depends on whether the light motion and the train
acceleration are in the same or in opposite directions. For actual trains or buses, the
Challenge 698 n frequency change is quite small; nevertheless, it is measurable. Acceleration induces fre-
quency changes in light. Let us compare this effect of acceleration with the effects of grav-
ity.
To measure time and space, we use light. What happens to light when gravity is in- Dvipsbugw
Ref. 327 volved? The simplest experiment is to let light fall or rise. In order to deduce what must
happen, we add a few details. Imagine a conveyor belt carrying masses around two wheels,
a low and a high one, as shown in Figure 177. The descending, grey masses are slightly lar-
ger. Whenever such a larger mass is near the bottom, some mechanism – not shown in the
figure – converts the mass surplus to light, in accordance with the equation E = mc 2 , and
sends the light up towards the top.** At the top, one of the lighter, white masses passing by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

absorbs the light and, because of its added weight, turns the conveyor belt until it reaches
the bottom. Then the process repeats.***
As the grey masses on the descending side are always heavier, the belt would turn for
ever and this system could continuously generate energy. However, since energy conser-
Page 198 vation is at the basis of our definition of time, as we saw in the beginning of our walk, the
whole process must be impossible. We have to conclude that the light changes its energy
when climbing. The only possibility is that the light arrives at the top with a frequency
different from the one at which it is emitted from the bottom.****
In short, it turns out that rising light is gravitationally red-shifted. Similarly, the light
descending from the top of a tree down to an observer is blue-shifted; this gives a darker
colour to the top in comparison with the bottom of the tree. General relativity thus says
that trees have different shades of green along their height.***** How big is the effect?
Challenge 701 e The result deduced from the drawing is again the one of formula (230). That is what we
would, as light moving in an accelerating train and light moving in gravity are equival-
Challenge 702 n ent situations, as you might want to check yourself. The formula gives a relative change
of frequency of only 1.1 ë 10−16 ~m near the surface of the Earth. For trees, this so-called
gravitational red-shift or gravitational Doppler effect is far too small to be observable, at
least using normal light.
Ref. 328 In 1911, Einstein proposed an experiment to check the change of frequency with height
by measuring the red-shift of light emitted by the Sun, using the famous Fraunhofer lines
Copyright © Christoph Schiller November 1997–May 2006

Page 753 as colour markers. The results of the first experiments, by Schwarzschild and others, were
unclear or even negative, due to a number of other effects that induce colour changes at

* The expression v = дt is valid only for non-relativistic speeds; nevertheless, the conclusion of this section
Challenge 697 e is not affected by this approximation.
** As in special relativity, here and in the rest of our mountain ascent, the term ‘mass’ always refers to rest
mass.
Challenge 699 n *** Can this process be performed with 100% efficiency?
**** The precise relation between energy and frequency of light is described and explained in our discussion
on quantum theory, on page 719. But we know already from classical electrodynamics that the energy of light
depends on its intensity and on its frequency.
Challenge 700 ny ***** How does this argument change if you include the illumination by the Sun?

Dvipsbugw
 380 iii gravitation and rel ativity • 7. the new ideas

high temperatures. But in 1920 and 1921, Grebe and Bachem, and independently Perot,
confirmed the gravitational red-shift with careful experiments. In later years, technolo-
gical advances made the measurements much easier, until it was even possible to measure
the effect on Earth. In 1960, in a classic experiment using the Mössbauer effect, Pound and
Ref. 329 Rebka confirmed the gravitational red-shift in their university tower using γ radiation.
But our two thought experiments tell us much more. Let
us use the same arguments as in the case of special relativity:
a colour change implies that clocks run differently at differ- m
ent heights, just as they run differently in the front and in Dvipsbugw
the back of a train. The time difference ∆τ is predicted to
depend on the height difference ∆h and the acceleration of
m+E/c2
gravity д according to

= = 2 .
∆τ ∆ f д∆h h
(231)
τ f c

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Therefore, in gravity, time is height-dependent. That was ex-
actly what we claimed above. In fact, height makes old. Can
Challenge 703 ny you confirm this conclusion?
In 1972, by flying four precise clocks in an aeroplane
while keeping an identical one on the ground, Hafele and light
Ref. 330 Keating found that clocks indeed run differently at differ-
ent altitudes according to expression (231). Subsequently, in
Ref. 331 1976, the team of Vessot et al. shot a precision clock based
on a maser – a precise microwave generator and oscillator –
upwards on a missile. The team compared the maser inside
the missile with an identical maser on the ground and again F I G U R E 177 The necessity of
confirmed the expression. In 1977, Briatore and Leschiutta blue- and red-shift of light:
Ref. 332 showed that a clock in Torino indeed ticks more slowly why trees are greener at the
bottom
than one on the top of the Monte Rosa. They confirmed the
prediction that on Earth, for every 100 m of height gained,
Challenge 704 ny people age more rapidly by about 1 ns per day. This effect has been confirmed for all sys-
tems for which experiments have been performed, such as several planets, the Sun and
numerous other stars.
Do these experiments show that time changes or are they simply due to clocks that
Copyright © Christoph Schiller November 1997–May 2006

Challenge 705 e function badly? Take some time and try to settle this question. We will give one argument
only: gravity does change the colour of light, and thus really does change time. Clock
precision is not an issue here.
In summary, gravity is indeed the uneven running of clocks at different heights. Note
that an observer at the lower position and another observer at the higher position agree
on the result: both find that the upper clock goes faster. In other words, when gravity is
present, space-time is not described by the Minkowski geometry of special relativity, but
by some more general geometry. To put it mathematically, whenever gravity is present,
the 4-distance ds 2 between events is different from the expression without gravity:

ds 2 x c 2 dt 2 − dx 2 − dy 2 − dz 2 . (232)

Dvipsbugw
 the new ideas on space, time and gravity 381

We will give the correct expression shortly.

Is this view of gravity as height-dependent time really reasonable? No. It turns out that
it is not yet strange enough! Since the speed of light is the same for all observers, we can
say more. If time changes with height, length must also do so! More precisely, if clocks
run differently at different heights, the length of metre bars must also change with height.
Challenge 706 ny Can you confirm this for the case of horizontal bars at different heights?
If length changes with height, the circumference of a circle around the Earth cannot be
given by 2πr. An analogous discrepancy is also found by an ant measuring the radius and
circumference of a circle traced on the surface of a basketball. Indeed, gravity implies that Dvipsbugw
humans are in a situation analogous to that of ants on a basketball, the only difference
being that the circumstances are translated from two to three dimensions. We conclude
that wherever gravity plays a role, space is curved.

What tides tell us about gravity

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

During his free fall, Kittinger was able to specify an inertial frame for himself. Indeed,
he felt completely at rest. Does this mean that it is impossible to distinguish acceleration
from gravitation? No: distinction is possible. We only have to compare two (or more)
falling observers.
Kittinger could not have found a frame which is also inertial for
Challenge 707 e a colleague falling on the opposite side of the Earth. Such a common
before
frame does not exist. In general, it is impossible to find a single iner-
tial reference frame describing different observers freely falling near
after
a mass. In fact, it is impossible to find a common inertial frame even
for nearby observers in a gravitational field. Two nearby observers
Challenge 708 n observe that during their fall, their relative distance changes. (Why?)
The same happens to orbiting observers.
In a closed room in orbit around the Earth, a person or a mass
at the centre of the room would not feel any force, and in particular
no gravity. But if several particles are located in the room, they will F I G U R E 178 Tidal
behave differently depending on their exact positions in the room. effects: what bodies
feel when falling
Only if two particles were on exactly the same orbit would they keep
the same relative position. If one particle is in a lower or higher orbit
than the other, they will depart from each other over time. Even more interestingly, if a
particle in orbit is displaced sideways, it will oscillate around the central position. (Can
Copyright © Christoph Schiller November 1997–May 2006

Challenge 709 ny you confirm this?)

Gravitation leads to changes of relative distance. These changes evince another effect,
shown in Figure 178: an extended body in free fall is slightly squeezed. This effect also
tells us that it is an essential feature of gravity that free fall is different from point to point.
Page 129 That rings a bell. The squeezing of a body is the same effect as that which causes the
tides. Indeed, the bulging oceans can be seen as the squeezed Earth in its fall towards the
Ref. 333 Moon. Using this result of universal gravity we can now affirm: the essence of gravity is
the observation of tidal effects.
In other words, gravity is simple only locally. Only locally does it look like acceleration.
Only locally does a falling observer like Kittinger feel at rest. In fact, only a point-like
observer does so! As soon as we take spatial extension into account, we find tidal effects.

Dvipsbugw
 382 iii gravitation and rel ativity • 7. the new ideas

Gravity is the presence of tidal effects. The absence of tidal effects implies the absence of
gravity. Tidal effects are the everyday consequence of height-dependent time. Isn’t this a
beautiful conclusion?
In principle, Kittinger could have felt gravitation during his free fall, even with his eyes
closed, had he paid attention to himself. Had he measured the distance change between
his two hands, he would have found a tiny decrease which could have told him that he
was falling. This tiny decrease would have forced Kittinger to a strange conclusion. Two
inertially moving hands should move along two parallel lines, always keeping the same
distance. Since the distance changes, he must conclude that in the space around him lines Dvipsbugw
starting out in parallel do not remain so. Kittinger would have concluded that the space
around him was similar to the surface of the Earth, where two lines starting out north,
parallel to each other, also change distance, until they meet at the North Pole. In other
words, Kittinger would have concluded that he was in a curved space.
By studying the change in distance between his hands, Kittinger could even have con-
cluded that the curvature of space changes with height. Physical space differs from a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

sphere, which has constant curvature. Physical space is more involved. The effect is ex-
tremely small, and cannot be felt by human senses. Kittinger had no chance to detect any-
thing. Detection requires special high-sensitivity apparatus. However, the conclusion re-
mains valid. Space-time is not described by Minkowski geometry when gravity is present.
Tidal effects imply space-time curvature. Gravity is the curvature of space-time.

Bent space and mattresses

Wenn ein Käfer über die Oberfläche einer Kugel

“ krabbelt, merkt er wahrscheinlich nicht, daß der

Weg, den er zurücklegt, gekrümmt ist. Ich
dagegen hatte das Glück, es zu merken.*
Albert Einstein’s answer to his son Eduard’s
question about the reason for his fame

On the 7th of November 1919, Albert Einstein became world-famous. On that day, an
article in the Times newspaper in London announced the results of a double expedition
to South America under the heading ‘Revolution in science / new theory of the universe
”
/ Newtonian ideas overthrown’. The expedition had shown unequivocally – though not
for the first time – that the theory of universal gravity, essentially given by a = GM~r 2 ,
was wrong, and that instead space was curved. A worldwide mania started. Einstein was
presented as the greatest of all geniuses. ‘Space warped’ was the most common headline.
Copyright © Christoph Schiller November 1997–May 2006

Einstein’s papers on general relativity were reprinted in full in popular magazines. People
could read the field equations of general relativity, in tensor form and with Greek indices,
in Time magazine. Nothing like this has happened to any other physicist before or since.
What was the reason for this excitement?
The expedition to the southern hemisphere had performed an experiment proposed
Ref. 334 by Einstein himself. Apart from seeking to verify the change of time with height, Einstein
had also thought about a number of experiments to detect the curvature of space. In the
one that eventually made him famous, Einstein proposed to take a picture of the stars

* ‘When an insect walks over the surface of a sphere it probably does not notice that the path it walks is
curved. I, on the other hand, had the luck to notice it.’

Dvipsbugw
 the new ideas on space, time and gravity 383

near the Sun, as is possible during a solar eclipse, and compare it with a picture of the
same stars at night, when the Sun is far away. Einstein predicted a change in position of
1.75′ (1.75 seconds of arc) for star images at the border of the Sun, a value twice as large
Page 133 as that predicted by universal gravity. The prediction, corresponding to about 1~40 mm
on the photographs, was confirmed in 1919, and thus universal gravity was ruled out.
Does this result imply that space is curved? Not by itself. In fact, other explanations
could be given for the result of the eclipse experiment, such as a potential differing from
the inverse square form. However, the eclipse results are not the only data. We already
know about the change of time with height. Experiments show that two observers at dif- Dvipsbugw
ferent heights measure the same value for the speed of light c near themselves. But these
experiments also show that if an observer measures the speed of light at the position of
the other observer, he gets a value differing from c, since his clock runs differently. There
is only one possible solution to this dilemma: metre bars, like clocks, also change with
height, and in such a way as to yield the same speed of light everywhere.
If the speed of light is constant but clocks and metre bars change with height, the con-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 710 ny clusion must be that space is curved near masses. Many physicists in the twentieth century
checked whether metre bars really behave differently in places where gravity is present.
And indeed, curvature has been detected around several planets, around all the hundreds
of stars where it could be measured, and around dozens of galaxies. Many indirect effects
of curvature around masses, to be described in detail below, have also been observed. All
results confirm the curvature of space and space-time around masses, and in addition
confirm the curvature values predicted by general relativity. In other words, metre bars
near masses do indeed change their size from place to place, and even from orientation
to orientation. Figure 179 gives an impression of the situation.

image image
of star
position
star of star

Sun

Sun

Mercury Earth
Earth
Copyright © Christoph Schiller November 1997–May 2006

F I G U R E 179 The mattress model of space: the path of a light beam and of a satellite near a spherical
mass

But beware: the right-hand figure, although found in many textbooks, can be
Ref. 335 misleading. It can easily be mistaken fora reproduction of a potential around a body.
Indeed, it is impossible to draw a graph showing curvature and potential separately.
Challenge 711 n (Why?) We will see that for small curvatures, it is even possible to explain the change in
metre bar length using a potential only. Thus the figure does not really cheat, at least in
the case of weak gravity. But for large and changing values of gravity, a potential cannot
be defined, and thus there is indeed no way to avoid using curved space to describe grav-
ity. In summary, if we imagine space as a sort of generalized mattress in which masses

Dvipsbugw
 384 iii gravitation and rel ativity • 7. the new ideas

produce deformations, we have a reasonable model of space-time. As masses move, the

deformation follows them.
The acceleration of a test particle only depends on the curvature of the mattress. It
does not depend on the mass of the test particle. So the mattress model explains why all
bodies fall in the same way. (In the old days, this was also called the equality of the inertial
and gravitational mass.)
Space thus behaves like a frictionless mattress that pervades everything. We live inside
the mattress, but we do not feel it in everyday life. Massive objects pull the foam of the mat-
tress towards them, thus deforming the shape of the mattress. More force, more energy or Dvipsbugw
more mass imply a larger deformation. (Does the mattress remind you of the aether? Do
Page 581 not worry: physics eliminated the concept of aether because it is indistinguishable from
vacuum.)
If gravity means curved space, then any accelerated observer, such as a man in a de-
parting car, must also observe that space is curved. However, in everyday life we do not
notice any such effect, because for accelerations and sizes of of everyday life the curvature

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

values are too small to be noticed. Could you devise a sensitive experiment to check the
Challenge 712 ny prediction?

Curved space-time
Figure 179 shows the curvature of space only, but in fact space-time is curved. We will
shortly find out how to describe both the shape of space and the shape of space-time, and
how to measure their curvature.
Let us have a first attempt to describe nature with the idea of curved space-time. In the
case of Figure 179, the best description of events is with the use of the time t shown by a
clock located at spatial infinity; that avoids problems with the uneven running of clocks at
different distances from the central mass. For the radial coordinate r, the most practical
choice to avoid problems with the curvature of space is to use the circumference of a circle
around the central body, divided by 2π. The curved shape of space-time is best described
Page 295 by the behaviour of the space-time distance ds, or by the wristwatch time dτ = ds~c,
between two neighbouring points with coordinates (t, r) and (t + dt, r + dr). As we saw
Page 381 above, gravity means that in spherical coordinates we have

ds 2
dτ 2 = x dt 2 − dr 2 ~c 2 − r 2 dφ 2 ~c 2 . (233)
c2
Copyright © Christoph Schiller November 1997–May 2006

The inequality expresses the fact that space-time is curved. Indeed, the experiments on
time change with height confirm that the space-time interval around a spherical mass is
given by
ds 2 dr 2 r2 2
dτ 2 = 2 = ‹1 − 
2GM 2
dt − − dφ . (234)
c rc 2 c 2 − 2GrM c 2

This expression is called the Schwarzschild metric after one of its discoverers.* The metric
(234) describes the curved shape of space-time around a spherical non-rotating mass.
* Karl Schwarzschild (1873–1916), important German astronomer; he was one of the first people to un-
derstand general relativity. He published his formula in December 1915, only a few months after Einstein

Dvipsbugw
 the new ideas on space, time and gravity 385

Challenge 713 n It is well approximated by the Earth or the Sun. (Why can their rotation be neglected?)
Expression (234) also shows that gravity’s strength around a body of mass M and radius
R is measured by a dimensionless number h defined as

h=
2G M
. (235)
c2 R
This ratio expresses the gravitational strain with which lengths and the vacuum are de-
formed from the flat situation of special relativity, and thus also determines how much Dvipsbugw
clocks slow down when gravity is present. (The ratio also reveals how far one is from any
possible horizon.) On the surface of the Earth the ratio h has the small value of 1.4 ë 10−9 ;
on the surface of the Sun is has the somewhat larger value of 4.2 ë 10−6 . The precision
of modern clocks allows one to detect such small effects quite easily. The various con-
sequences and uses of the deformation of space-time will be discussed shortly.
We note that if a mass is highly concentrated, in particular when its radius becomes

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

equal to its so-called Schwarzschild radius

RS =
2GM
, (236)
c2
the Schwarzschild metric behaves strangely: at that location, time disappears (note that
t is time at infinity). At the Schwarzschild radius, the wristwatch time (as shown by a
clock at infinity) stops – and a horizon appears. What happens precisely will be explored
Page 479 below. This situation is not common: the Schwarzschild radius for a mass like the Earth is
8.8 mm, and for the Sun is 3.0 km; you might want to check that the object size for every
Challenge 714 e system in everyday life is larger than its Schwarzschild radius. Bodies which reach this
Ref. 337 limit are called black holes; we will study them in detail shortly. In fact, general relativity
states that no system in nature is smaller than its Schwarzschild size, in other words that
the ratio h defined by expression (235) is never above unity.
In summary, the results mentioned so far make it clear that mass generates curvature.
The mass–energy equivalence we know from special relativity then tells us that as a con-
sequence, space should also be curved by the presence of any type of energy–momentum.
Every type of energy curves space-time. For example, light should also curve space-time.
However, even the highest-energy beams we can create correspond to extremely small
masses, and thus to unmeasurably small curvatures. Even heat curves space-time; but in
most systems, heat is only about a fraction of 10−12 of total mass; its curvature effect is
Copyright © Christoph Schiller November 1997–May 2006

thus unmeasurable and negligible. Nevertheless it is still possible to show experimentally

that energy curves space. In almost all atoms a sizeable fraction of the mass is due to the
electrostatic energy among the positively charged protons. In 1968 Kreuzer confirmed
Ref. 338 that energy curves space with a clever experiment using a floating mass.
It is straightforward to imagine that the uneven running of clock is the temporal equi-
Challenge 715 ny valent of spatial curvature. Taking the two together, we conclude that when gravity is
present, space-time is curved.

had published his field equations. He died prematurely, at the age of 42, much to Einstein’s distress. We
will deduce the form of the metric later on, directly from the field equations of general relativity. The other
Ref. 336 discoverer of the metric, unknown to Einstein, was the Dutch physicist J. Droste.

Dvipsbugw
 386 iii gravitation and rel ativity • 7. the new ideas

Let us sum up our chain of thoughts. Energy is equivalent to mass; mass produces grav-
ity; gravity is equivalent to acceleration; acceleration is position-dependent time. Since
light speed is constant, we deduce that energy–momentum tells space-time to curve. This
statement is the first half of general relativity.
We will soon find out how to measure curvature, how to calculate it from energy–
momentum and what is found when measurement and calculation are compared. We
will also find out that different observers measure different curvature values. The set of
transformations relating one viewpoint to another in general relativity, the diffeomorph-
ism symmetry, will tell us how to relate the measurements of different observers. Dvipsbugw
Since matter moves, we can say even more. Not only is space-time curved near masses,
it also bends back when a mass has passed by. In other words, general relativity states that
Ref. 339 space, as well as space-time, is elastic. However, it is rather stiff: quite a lot stiffer than steel.
To curve a piece of space by 1 % requires an energy density enormously larger than to
Challenge 716 ny curve a simple train rail by 1 %. This and other interesting consequences of the elasticity
of space-time will occupy us for the remainder of this chapter.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The speed of light and the gravitational constant
Si morior, moror.*

“
We continue on the way towards precision in our understanding of gravitation. All our
theoretical and empirical knowledge about gravity can be summed up in just two general
”
statements. The first principle states:

U The speed v of a physical system is bounded above:

vDc (237)

for all observers, where c is the speed of light.

The theory following from this first principle, special relativity, is extended to general
relativity by adding a second principle, characterizing gravitation. There are several equi-
valent ways to state this principle. Here is one.

U For all observers, the force F on a system is limited by

Copyright © Christoph Schiller November 1997–May 2006

c4
FD , (238)
4G
where G is the universal constant of gravitation.

In short, there is a maximum force in nature. Gravitation leads to attraction of masses.

Challenge 717 e However, this force of attraction is limited. An equivalent statement is:

* ‘If I rest, I die.’ This is the motto of the bird of paradise.

Dvipsbugw
 the new ideas on space, time and gravity 387

U For all observers, the size L of a system of mass M is limited by

E 2 .
L 4G
(239)
M c

In other words, a massive system cannot be more concentrated than a non-rotating black
hole of the same mass. Another way to express the principle of gravitation is the following:

U For all systems, the emitted power P is limited by Dvipsbugw

c5
PD . (240)
4G

In short, there is a maximum power in nature.

The three limits given above are all equivalent to each other; and no exception is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

known or indeed possible. The limits include universal gravity in the non-relativistic case.
They tell us what gravity is, namely curvature, and how exactly it behaves. The limits al-
low us to determine the curvature in all situations, at all space-time events. As we have
Page 353 seen above, the speed limit together with any one of the last three principles imply all of
general relativity.*
For example, can you show that the formula describing gravitational red-shift com-
Challenge 718 ny plies with the general limit (239) on length-to-mass ratios?
We note that any formula that contains the speed of light c is based on special relativity,
and if it contains the constant of gravitation G, it relates to universal gravity. If a formula
contains both c and G, it is a statement of general relativity. The present chapter frequently
underlines this connection.
Our mountain ascent so far has taught us that a precise description of motion requires
the specification of all allowed viewpoints, their characteristics, their differences, and the
transformations between them. From now on, all viewpoints are allowed, without excep-
tion: anybody must be able to talk to anybody else. It makes no difference whether an
observer feels gravity, is in free fall, is accelerated or is in inertial motion. Furthermore,
people who exchange left and right, people who exchange up and down or people who
say that the Sun turns around the Earth must be able to talk to each other and to us. This
gives a much larger set of viewpoint transformations than in the case of special relativ-
ity; it makes general relativity both difficult and fascinating. And since all viewpoints are
Copyright © Christoph Schiller November 1997–May 2006

allowed, the resulting description of motion is complete.**

Why does a stone thrown into the air fall back to Earth? –
Geodesics
A genius is somebody who makes all possible

“ mistakes in the shortest possible time.

Anonymous

”
* This didactic approach is unconventional. It is possible that is has been pioneered by the present author.
Ref. 340 The British physicist Gary Gibbons also developed it independently. Earlier references are not known.
** Or it would be, were it not for a small deviation called quantum theory.

Dvipsbugw
 388 iii gravitation and rel ativity • 7. the new ideas

In our discussion of special relativity, we saw that inertial or free-floating motion is the
Page 326 motion which connecting two events that requires the longest proper time. In the absence
of gravity, the motion fulfilling this requirement is straight (rectilinear) motion. On the
Page 56 other hand, we are also used to thinking of light rays as being straight. Indeed, we are all
accustomed to check the straightness of an edge by looking along it. Whenever we draw
the axes of a physical coordinate system, we imagine either drawing paths of light rays or
drawing the motion of freely moving bodies.
In the absence of gravity, object paths and light paths coincide. However, in the pres-
ence of gravity, objects do not move along light paths, as every thrown stone shows. Light Dvipsbugw
does not define spatial straightness any more. In the presence of gravity, both light and
matter paths are bent, though by different amounts. But the original statement remains
valid: even when gravity is present, bodies follow paths of longest possible proper time.
For matter, such paths are called timelike geodesics. For light, such paths are called lightlike
or null geodesics.
We note that in space-time, geodesics are the curves with maximal length. This is in

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

contrast with the case of pure space, such as the surface of a sphere, where geodesics are
the curves of minimal length. In simple words, stones fall because they follow geodesics.
Let us perform a few checks of this statement.
Since stones move by maximizing proper time for inertial observers, they also must
do so for freely falling observers, like Kittinger. In fact, they must do so for all observers.
The equivalence of falling paths and geodesics is at least coherent.
If falling is seen as a consequence of the Earth’s surface approaching – as we will argue
later on – we can deduce directly that falling implies a proper time that is as long as
Challenge 719 ny possible. Free fall indeed is motion along geodesics.
We saw above that gravitation follows from the existence of a maximum force. The
result can be visualized in another way. If the gravitational attraction between a central
body and a satellite were stronger than it is, black holes would be smaller than they are;
in that case the maximum force limit and the maximum speed could be exceeded by
getting close to such a black hole. If, on the other hand, gravitation were weaker than it
is, there would be observers for which the two bodies would not interact, thus for which
they would not form a physical system. In summary, a maximum force of c 4 ~4G implies
universal gravity. There is no difference between stating that all bodies attract through
gravitation and stating that there is a maximum force with the value c 4 ~4G. But at the
same time, the maximum force principle implies that objects move on geodesics. Can
Challenge 720 ny you show this?
Copyright © Christoph Schiller November 1997–May 2006

Let us turn to an experimental check. If falling is a consequence of curvature, then

the paths of all stones thrown or falling near the Earth must have the same curvature in
space-time. Take a stone thrown horizontally, a stone thrown vertically, a stone thrown
rapidly, or a stone thrown slowly: it takes only two lines of argument to show that in space-
Challenge 721 ny time all their paths are approximated to high precision by circle segments, as shown in
Figure 180. All paths have the same curvature radius r, given by

c2
r=  9.2 ë 1015 m . (241)
д

Dvipsbugw
 the new ideas on space, time and gravity 389

height
c · time
slow, steep throw
h
d

rapid, flat throw

throw distance
Dvipsbugw
F I G U R E 180 All paths of flying stones have the same
curvature in space-time

The large value of the radius, corresponding to a low curvature, explains why we do not
notice it in everyday life. The parabolic shape typical of the path of a stone in everyday

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

life is just the projection of the more fundamental path in 4-dimensional space-time into
3-dimensional space. The important point is that the value of the curvature does not de-
pend on the details of the throw. In fact, this simple result could have suggested the ideas
of general relativity to people a full century before Einstein; what was missing was the
recognition of the importance of the speed of light as limit speed. In any case, this simple
calculation confirms that falling and curvature are connected. As expected, and as men-
tioned already above, the curvature diminishes at larger heights, until it vanishes at infin-
ite distance from the Earth. Now, given that the curvature of all paths for free fall is the
same, and given that all such paths are paths of least action, it is straightforward that they
are also geodesics.
If we describe fall as a consequence of the curvature of space-time, we must show that
the description with geodesics reproduces all its features. In particular, we must be able
to explain that stones thrown with small speed fall back, and stones thrwon with high
Challenge 722 ny speed escape. can you deduce this from space curvature?
In summary, the motion of any particle falling freely ‘in a gravitational field’ is de-
scribed by the same variational principle as the motion of a free particle in special relativ-
ity: the path maximizes the proper time ∫ dτ. We rephrase this by saying that any particle
in free fall from point A to point B minimizes the action S given by

S = −mc 2
B
∫
Copyright © Christoph Schiller November 1997–May 2006

dτ . (242)
A

That is all we need to know about the free fall of objects. As a consequence, any deviation
from free fall keeps you young. The larger the deviation, the younger you stay.
Page 498 As we will see below, the minimum action description of free fall has been tested ex-
Ref. 341 tremely precisely, and no difference from experiment has ever been observed. We will
also find out that for free fall, the predictions of general relativity and of universal grav-
ity differ substantially both for particles near the speed of light and for central bodies of
high density. So far, all experiments have shown that whenever the two predictions dif-
fer, general relativity is right, and universal gravity and other alternative descriptions are
wrong.

Dvipsbugw
 390 iii gravitation and rel ativity • 7. the new ideas

All bodies fall along geodesics. This tells us something important. The fall of bodies
does not depend on their mass. The geodesics are like ‘rails’ in space-time that tell bod-
ies how to fall. In other words, space-time can indeed be imagined as a single, giant, de-
formed entity. Space-time is not ‘nothing’; it is an entity of our thinking. The shape of this
entity tells objects how to move. Space-time is thus indeed like an intangible mattress; this
deformed mattress guides falling objects along its networks of geodesics.
Moreover, bound energy falls in the same way as mass, as is proven by comparing the
fall of objects made of different materials. They have different percentages of bound en-
Challenge 723 n ergy. (Why?) For example, on the Moon, where there is no air, cosmonauts dropped steel Dvipsbugw
balls and feathers and found that they fell together, alongside each other. The independ-
Ref. 342 ence on material composition has been checked and confirmed over and over again.

Can light fall?

How does radiation fall? Light, like any radiation, is energy without rest mass. It moves

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

like a stream of extremely fast and light objects. Therefore deviations from universal grav-
ity become most apparent for light. How does light fall? Light cannot change speed. When
Page 378 light falls vertically, it only changes colour, as we have seen above. But light can also
change direction. Long before the ideas of relativity became current, in 1801, the Prus-
Ref. 343 sian astronomer Johann Soldner understood that universal gravity implies that light is
deflected when passing near a mass. He also calculated how the deflection angle depends
Page 133 on the mass of the body and the distance of passage. However, nobody in the nineteenth
century was able to check the result experimentally.
Obviously, light has energy, and energy has weight; the deflection of light by itself
is thus not a proof of the curvature of space. General relativity also predicts a deflec-
tion angle for light passing masses, but of twice the classical Soldner value, because the
curvature of space around large masses adds to the effect of universal gravity. The de-
flection of light thus only confirms the curvature of space if the value agrees with the one
predicted by general relativity. This is the case: observations do coincide with predictions.
Page 409 More details will be given shortly.
Mass is thus not necessary to feel gravity; energy is sufficient. This result of the mass–
energy equivalence must become second nature when studying general relativity. In par-
ticular, light is not light-weight, but heavy. Can you argue that the curvature of light near
Challenge 724 ny the Earth must be the same as that of stones, given by expression (241)?
In summary, all experiments show that not only mass, but also energy falls along
Copyright © Christoph Schiller November 1997–May 2006

geodesics, whatever its type (bound or free), and whatever the interaction (be it elec-
tromagnetic or nuclear). Moreover, the motion of radiation confirms that space-time is
curved.
Since experiments show that all particles fall in the same way, independently of their
mass, charge or any other property, we can conclude that the system of all possible tra-
jectories forms an independent structure. This structure is what we call space-time.
We thus find that space-time tells matter, energy and radiation how to fall. This state-
ment is the second half of general relativity. It complements the first half, which states
that energy tells space-time how to curve. To complete the description of macroscopic
motion, we only need to add numbers to these statements, so that they become testable.
As usual, we can proceed in two ways: we can deduce the equations of motion directly,

Dvipsbugw
 the new ideas on space, time and gravity 391

or we can first deduce the Lagrangian and then deduce the equations of motion from it.
But before we do that, let’s have some fun.

Curiosities and fun challenges about gravitation

Wenn Sie die Antwort nicht gar zu ernst

“ nehmen und sie nur als eine Art Spaß ansehen,

so kann ich Ihnen das so erklären: Früher hat
man geglaubt, wenn alle Dinge aus der Welt
verschwinden, so bleiben noch Raum und Zeit Dvipsbugw
übrig. Nach der Relativitätstheorie
verschwinden aber auch Zeit und Raum mit den
Dingen.*
Albert Einstein in 1921 in New York

General relativity is a beautiful topic with numerous interesting aspects.

”

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Take a plastic bottle and make some holes in it near the bottom. Fill the bottle with water,
closing the holes with your fingers. If you let the bottle go, no water will leave the bottle
Challenge 725 n during the fall. Can you explain how this experiment confirms the equivalence of rest
and free fall?

**
On his 76th birthday, Einstein received a birthday present specially made for him, shown
in Figure 181. A rather deep cup is mounted on the top of a broom stick. The cup contains
a weak piece of elastic rubber attached to its bottom, to which a ball is attached at the other
end. In the starting position, the ball hangs outside the cup. The rubber is too weak to
pull the ball into the cup against gravity. What is the most elegant way to get the ball into
Challenge 726 n the cup?

**
Gravity has the same properties in the whole universe – except in the US patent office. In
2005, it awarded a patent, Nr 6 960 975, for an antigravity device that works by distorting
space-time in such a way that gravity is ‘compensated’ (see http://patft.uspto.gov). Do you
Challenge 727 n know a simpler device? Copyright © Christoph Schiller November 1997–May 2006

**
The radius of curvature of space-time at the Earth’s surface is 9.2 ë 1015 m. Can you confirm
Challenge 728 e this value?
**
Challenge 729 ny A piece of wood floats on water. Does it stick out more or less in a lift accelerating up-
wards?

* ‘If you do not take the answer too seriously and regard it only for amusement, I can explain it to you in the
following way: in the past it was thought that if all things were to disappear from the world, space and time
would remain. But following relativity theory, space and time would disappear together with the things.’

Dvipsbugw
 392 iii gravitation and rel ativity • 7. the new ideas

**
Page 305 We saw in special relativity that if two twins are identically accelerated in the same dir-
ection, with one twin some distance ahead of the other, then the twin ahead ages more
than the twin behind. Does this happen in a gravitational field as well? And what happens
Challenge 730 ny when the field varies with height, as on Earth?
**
A maximum force and a maximum power also imply a maximum flow of mass. Can you
Challenge 731 ny show that no mass flow can exceed 1.1 ë 1035 kg~s? Dvipsbugw

**
The experiments of Figure 176 and 177 differ in one point: one happens in flat space, the
other in curved space. One seems to be connected with energy conservation, the other
Challenge 732 ny not. Do these differences invalidate the equivalence of the observations?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Challenge 733 n How can cosmonauts weigh themselves to check whether they are eating enough?
**
Is a cosmonaut in orbit really floating freely? No. It turns out that space stations and satel-
lites are accelerated by several small effects. The important ones are the pressure of the
light from the Sun, the friction of the thin air, and the effects of solar wind. (Micrometeor-
ites can usually be neglected.) These three effects all lead to accelerations of the order of
10−6 m~s2 to 10−8 m~s2 , depending on the height of the orbit. Can you estimate how long
it would take an apple floating in space to hit the wall of a space station, starting from the
Challenge 734 n middle? By the way, what is the magnitude of the tidal accelerations in this situation?
**
Page 80 There is no negative mass in nature, as discussed in the beginning of our walk (even
antimatter has positive mass). This means that gravitation cannot be shielded, in contrast
to electromagnetic interactions. Since gravitation cannot be shielded, there is no way to
make a perfectly isolated system. But such systems form the basis of thermodynamics!
Page 868 We will study the fascinating implications of this later on: for example, we will discover
an upper limit for the entropy of physical systems. Copyright © Christoph Schiller November 1997–May 2006

**
Can curved space be used to travel faster than light? Imagine a space-time in which two
points could be connected either by a path leading through a flat portion, or by a second
path leading through a partially curved portion. Could that curved portion be used to
travel between the points faster than through the flat one? Mathematically, this is pos-
sible; however, such a curved space would need to have a negative energy density. Such
a situation is incompatible with the definition of energy and with the non-existence of
Ref. 344 negative mass. The statement that this does not happen in nature is also called the weak
Challenge 735 ny energy condition. Is it implied by the limit on length-to-mass ratios?

**

Dvipsbugw
 the new ideas on space, time and gravity 393

The statement of a length-to-mass limit L~M E 4G~c 2 invites experiments to try to over-
come it. Can you explain what happens when an observer moves so rapidly past a mass
Challenge 736 ny that the body’s length contraction reaches the limit?
**
There is an important mathematical property of R3 which singles out three dimensional
space from all other possibilities. A closed (one-dimensional) curve can form knots only
in R3 : in any higher dimension it can always be unknotted. (The existence of knots also
explains why three is the smallest dimension that allows chaotic particle motion.) How- Dvipsbugw
ever, general relativity does not say why space-time has three plus one dimensions. It is
simply based on the fact. This deep and difficult question will be settled only in the third
part of our mountain ascent.
**
Henri Poincaré, who died in 1912, shortly before the general theory of relativity was fin-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ished, thought for a while that curved space was not a necessity, but only a possibility. He
imagined that one could continue using Euclidean space provided light was permitted to
Challenge 737 ny follow curved paths. Can you explain why such a theory is impossible?
**
Can two hydrogen atoms circle each other, in their mutual gravitational field? What
Challenge 738 n would the size of this ‘molecule’ be?

**
Challenge 739 n Can two light pulses circle each other, in their mutual gravitational field?
**
The various motions of the Earth mentioned in the section on Galilean physics, such as
Page 86 its rotation around its axis or around the Sun, lead to various types of time in physics
and astronomy. The time defined by the best atomic clocks is called terrestrial dynamical
time. By inserting leap seconds every now and then to compensate for the bad definition
Page 1161 of the second (an Earth rotation does not take 86 400, but 86 400.002 seconds) and, in
minor ways, for the slowing of Earth’s rotation, one gets the universal time coordinate or
UTC. Then there is the time derived from this one by taking into account all leap seconds.
One then has the – different – time which would be shown by a non-rotating clock in
Copyright © Christoph Schiller November 1997–May 2006

the centre of the Earth. Finally, there is barycentric dynamical time, which is the time that
Ref. 345 would be shown by a clock in the centre of mass of the solar system. Only using this latter
time can satellites be reliably steered through the solar system. In summary, relativity says
goodbye to Greenwich Mean Time, as does British law, in one of the rare cases were the
law follows science. (Only the BBC continues to use it.)
**
Space agencies thus have to use general relativity if they want to get artificial satellites to
Mars, Venus, or comets. Without its use, orbits would not be calculated correctly, and
satellites would miss their targets and usually even the whole planet. In fact, space agen-
cies play on the safe side: they use a generalization of general relativity, namely the so-

Dvipsbugw
 394 iii gravitation and rel ativity • 7. the new ideas

called parametrized post-Newtonian formalism, which includes a continuous check on

whether general relativity is correct. Within measurement errors, no deviation has been
found so far.*
**
General relativity is also used by space agencies around the world to calculate the exact
Ref. 346 positions of satellites and to tune radios to the frequency of radio emitters on them. In ad-
dition, general relativity is essential for the so-called global positioning system, or GPS. This
modern navigation tool** consists of 24 satellites equipped with clocks that fly around the Dvipsbugw
world. Why does the system need general relativity to operate? Since all the satellites, as
well as any person on the surface of the Earth, travel in circles, we have dr = 0, and we
can rewrite the Schwarzschild metric (234) as

dτ 2 2GM r 2 dφ 2 2GM v 2
‹  =1− − ( ) = 1 − − 2 . (244)
dt rc 2 c 2 dt rc 2 c

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 740 e For the relation between satellite time and Earth time we then get
2
v sat
dt sat 2 1 − r2G M
2 −
‹  = sat c c2
2 . (245)
dt Earth 1 − r Earth
2G M
−
v Earth
c2 c2

Can you deduce how many microseconds a satellite clock gains every day, given that the
Challenge 741 n GPS satellites orbit the Earth once every twelve hours? Since only three microseconds
would give a position error of one kilometre after a single day, the clocks in the satellites
Ref. 347 must be adjusted to run slow by the calculated amount. The necessary adjustments are
monitored, and so far have confirmed general relativity every single day, within experi-
mental errors, since the system began operation.
**
Ref. 348 The gravitational constant G does not seem to change with time. The latest experiments
limit its rate of change to less than 1 part in 1012 per year. Can you imagine how this can

* To give an idea of what this means, the unparametrized post-Newtonian formalism, based on general
relativity, writes the equation of motion of a body of mass m near a large mass M as a deviation from the
Copyright © Christoph Schiller November 1997–May 2006

inverse square expression for the acceleration a:

GM GM v 2 GM v 4 Gm v 5
a= 2
+ f2 2 2 + f4 2 4 + f5 2 5 + ë ë ë (243)
r r c r c r c
Here the numerical factors f n are calculated from general relativity and are of order one. The first two odd
terms are missing because of the (approximate) reversibility of general relativistic motion: gravity wave
emission, which is irreversible, accounts for the small term f 5 ; note that it contains the small mass m instead
of the large mass M. All factors f n up to f 7 have now been calculated. However, in the solar system, only the
term f 2 has ever been detected. This situation might change with future high-precision satellite experiments.
Page 407 Higher-order effects, up to f 5 , have been measured in the binary pulsars, as discussed below.
In a parametrized post-Newtonian formalism, all factors f n , including the uneven ones, are fitted through
the data coming in; so far all these fits agree with the values predicted by general relativity.
** For more information, see the http://www.gpsworld.com website.

Dvipsbugw
 the new ideas on space, time and gravity 395

Challenge 742 d be checked?

**
Could our experience that we live in only three spatial dimensions be due to a limitation
Challenge 743 n of our senses? How?
**
Challenge 744 ny Can you estimate the effect of the tides on the colour of the light emitted by an atom?
Dvipsbugw
**
The strongest possible gravitational field is that of a small black hole. The strongest grav-
Ref. 349 itational field ever observed is somewhat less though. In 1998, Zhang and Lamb used the
X-ray data from a double star system to determine that space-time near the 10 km sized
neutron star is curved by up to 30 % of the maximum possible value. What is the corres-
ponding gravitational acceleration, assuming that the neutron star has the same mass as

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 745 ny the Sun?

**
Ref. 350 Light deflection changes the angular size δ of a mass M with radius r when observed at
Challenge 746 e distance d. The effect leads to the pretty expression
»
r 1 − R S ~d
δ = arcsin( » ) RS =
2GM
where . (246)
d 1 − R S ~r c2

Challenge 747 ny What percentage of the surface of the Sun can an observer at infinity see? We will examine
Page 486 this issue in more detail shortly.

What is weight?
There is no way for a single (and point-like) observer to distinguish the effects of gravity
from those of acceleration. This property of nature allows one to make a strange state-
ment: things fall because the surface of the Earth accelerates towards them. Therefore,
the weight of an object results from the surface of the Earth accelerating upwards and
pushing against the object. That is the principle of equivalence applied to everyday life.
For the same reason, objects in free fall have no weight.
Copyright © Christoph Schiller November 1997–May 2006

Let us check the numbers. Obviously, an accelerating surface of the Earth produces a
weight for each body resting on it. This weight is proportional to the inertial mass. In other
words, the inertial mass of a body is identical to the gravitational mass. This is indeed
Ref. 351 observed in experiments, and to the highest precision achievable. Roland von Eőtvős*
performed many such high-precision experiments throughout his life, without finding
any discrepancy. In these experiments, he used the fact that the inertial mass determines
centrifugal effects and the gravitational mass determines free fall. (Can you imagine how
Challenge 748 ny he tested the equality?) Recent experiments showed that the two masses agree to one part

* Roland von Eőtvős (b. 1848 Budapest, d. 1919 Budapest), Hungarian physicist. He performed many high-
precision gravity experiments; among other discoveries, he discovered the effect named after him. The uni-

Dvipsbugw
 396 iii gravitation and rel ativity • 7. the new ideas

Ref. 351 in 10−12 .

However, the mass equality is not a surprise. Remembering the definition
Page 77 of mass ratio as negative inverse acceleration ratio, independently of the ori-
gin of the acceleration, we are reminded that mass measurements cannot be
used to distinguish between inertial and gravitational mass. As we have seen,
the two masses are equal by definition in Galilean physics, and the whole dis-
Page 133 cussion is a red herring. Weight is an intrinsic effect of mass.
The equality of acceleration and gravity allows us to imagine the following.
Imagine stepping into a lift in order to move down a few stories. You push Dvipsbugw
the button. The lift is pushed upwards by the accelerating surface of the Earth
somewhat less than is the building; the building overtakes the lift, which there-
fore remains behind. Moreover, because of the weaker push, at the beginning
everybody inside the lift feels a bit lighter. When the contact with the build-
ing is restored, the lift is accelerated to catch up with the accelerating surface
of the Earth. Therefore we all feel as if we were in a strongly accelerating car,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pushed in the direction opposite to the acceleration: for a short while, we feel
heavier, until the lift arrives at its destination.

Why do apples fall? FIG-

Vires acquirit eundo. U R E 181

“
An accelerating car will soon catch up with an object thrown forward from
Vergilius*

”
A puzzle

it. For the same reason, the surface of the Earth soon catches up with a stone
thrown upwards, because it is continually accelerating upwards. If you enjoy this way of
seeing things, imagine an apple falling from a tree. At the moment when it detaches, it
stops being accelerated upwards by the branch. The apple can now enjoy the calmness of
real rest. Because of our limited human perception, we call this state of rest free fall. Un-
fortunately, the accelerating surface of the Earth approaches mercilessly and, depending
on the time for which the apple stayed at rest, the Earth hits it with a greater or lesser
velocity, leading to more or less severe shape deformation.
Falling apples also teach us not to be disturbed any more by the statement that gravity
is the uneven running of clocks with height. In fact, this statement is equivalent to saying
that the surface of the Earth is accelerating upwards, as the discussion above shows.
Copyright © Christoph Schiller November 1997–May 2006

Can this reasoning be continued indefinitely? We can go on for quite a while. It is

fun to show how the Earth can be of constant radius even though its surface is acceler-
Challenge 749 ny ating upwards everywhere. We can thus play with the equivalence of acceleration and
gravity. However, this equivalence is only useful in situations involving only one acceler-
ating body. The equivalence between acceleration and gravity ends as soon as two falling
objects are studied. Any study of several bodies inevitably leads to the conclusion that
gravity is not acceleration; gravity is curved space-time.

versity of Budapest is named after him.

* ‘Going it acquires strength.’ Publius Vergilius Maro (b. 70 bce Andes, d. 19 bce Brundisium), Aeneis 4,
175.

Dvipsbugw
 motion in general rel ativity – bent light and wobbling vacuum 397

Many aspects of gravity and curvature can be understood with no or only a little math-
ematics. The next section will highlight some of the differences between universal gravity
and general relativity, showing that only the latter description agrees with experiment.
After that, a few concepts relating to the measurement of curvature are introduced and
applied to the motion of objects and space-time. If the reasoning gets too involved for a
first reading, skip ahead. In any case, the section on the stars, cosmology and black holes
again uses little mathematics.
Dvipsbugw
8. motion in general rel ativit y – bent light and
wobbling vacuum
I have the impression that Einstein understands

“ relativity theory very well.

Chaim Weitzmann, first president of Israel

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Before we tackle the details of general relativity, we will explore how the motion of objects
and light differs from that predicted by universal gravity, and how these differences can
be measured.

Weak fields
Gravity is strong near horizons. This happens when the mass M and the distance scale R
obey
1.
2GM
(247)
Rc 2
Therefore, gravity is strong mainly in three situations: near black holes, near the horizon
of the universe, and at extremely high particle energies. The first two cases are explored
later on, while the last will be explored in the third part of our mountain ascent. In con-
trast, in most parts of the universe there are no nearby horizons; in these cases, gravity
is a weak effect. Despite the violence of avalanches or of falling asteroids, in everyday life
gravity is much weaker than the maximum force. On the Earth the ratio just mentioned
is only about 10−9 . In this and all cases of everyday life, gravitation can still be approx-
imated by a field, despite what was said above. These weak field situations are interesting
because they are simple to understand; they mainly require for their explanation the dif-
Copyright © Christoph Schiller November 1997–May 2006

ferent running of clocks at different heights. Weak field situations allow us to mention
space-time curvature only in passing, and allow us to continue to think of gravity as a
source of acceleration. However, the change of time with height already induces many
new and interesting effects. The only thing we need is a consistent relativistic treatment.

The Thirring effects

In 1918, the Austrian physicist Hans Thirring published two simple and beautiful predic-
tions of motions, one of them with his collaborator Josef Lense. Neither motion appears
Ref. 352 in universal gravity, but they both appear in general relativity. Figure 182 shows these
predictions.

Dvipsbugw
 398 iii gravitation and rel ativity • 8. motion of light and vacuum

THIRRING EFFECT
universal gravity prediction relativistic prediction

Moon a
m

Earth M

universe or mass shell Dvipsbugw

THIRRING-LENSE EFFECT
universal gravity prediction relativistic prediction

Foucault's pendulum
or

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

orbiting satellite

Earth
Earth
universe or mass shell

F I G U R E 182 The Thirring and the Thirring–Lense effects

In the first example, nowadays called the Thirring effect, centrifugal accelerations as
well as Coriolis accelerations for masses in the interior of a rotating mass shell are pre-
dicted. Thirring showed that if an enclosing mass shell rotates, masses inside it are attrac-
ted towards the shell. The effect is very small; however, this prediction is in stark contrast
to that of universal gravity, where a spherical mass shell – rotating or not – has no effect
on masses in its interior. Can you explain this effect using the figure and the mattress
Challenge 750 ny analogy?
The second effect, the Thirring–Lense effect,* is more famous. General relativity pre-
dicts that an oscillating Foucault pendulum, or a satellite circling the Earth in a polar
orbit, does not stay precisely in a fixed plane relative to the rest of the universe, but that
the rotation of the Earth drags the plane along a tiny bit. This frame-dragging, as the ef-
fect is also called, appears because the Earth in vacuum behaves like a rotating ball in a
Copyright © Christoph Schiller November 1997–May 2006

foamy mattress. When a ball or a shell rotates inside the foam, it partly drags the foam
along with it. Similarly, the Earth drags some vacuum with it, and thus turns the plane
of the pendulum. For the same reason, the Earth’s rotation turns the plane of an orbiting
satellite.
The Thirring–Lense or frame-dragging effect is extremely small. It was measured for
the first time in 1998 by an Italian group led by Ignazio Ciufolini, and then again by the
same group in the years up to 2004. They followed the motion of two special artificial
satellites – shown in Figure 183 – consisting only of a body of steel and some Cat’s-eyes.

* Even though the order of the authors is Lense and Thirring, it is customary (but not universal) to stress
the idea of Hans Thirring by placing him first.

Dvipsbugw
 weak fields 399

The group measured the satellite’s motion around the Earth with extremely high precision,
making use of reflected laser pulses. This method allowed this low-budget experiment to
Ref. 353 beat by many years the efforts of much larger but much more sluggish groups.* The results
confirm the predictions of general relativity with an error of about 25 %.
Frame dragging effects have also been measured in bin-
ary star systems. This is possible if one of the stars is a pulsar,
because such stars send out regular radio pulses, e.g. every
millisecond, with extremely high precision. By measuring
the exact times when the pulses arrive on Earth, one can de- Dvipsbugw
duce the way these stars move and confirm that such subtle
Ref. 354 effects as frame dragging do take place.

Gravitomagnetism**
Frame-dragging and the Thirring–Lense effect can be F I G U R E 183 The LAGEOS

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

seen as special cases of gravitomagnetism. (We will show satellites: metal spheres with a
the connection below.) This approach to gravity, already diameter of 60 cm, a mass of
studied in the nineteenth century by Holzmüller and by 407 kg, and covered with 426
retroreflectors
Ref. 355 Tisserand, has become popular again in recent years, espe-
cially for its didactic advantages. As mentioned above, talk-
ing about a gravitational field is always an approximation.
In the case of weak gravity, such as occurs in everyday life, the approximation is very good.
Many relativistic effects can be described in terms of the gravitational field, without using
the concept of space curvature or the metric tensor. Instead of describing the complete
space-time mattress, the gravitational-field model only describes the deviation of the mat-
tress from the flat state, by pretending that the deviation is a separate entity, called the
gravitational field. But what is the relativistically correct way to describe the gravitational
field?
We can compare the situation to electromagnetism. In a relativistic description of elec-
Page 536 trodynamics, the electromagnetic field has an electric and a magnetic component. The
electric field is responsible for the inverse-square Coulomb force. In the same way, in a
relativistic description of (weak) gravity,*** the gravitational field has an gravitoelectric
and a gravitomagnetic component. The gravitoelectric field is responsible for the inverse
square acceleration of gravity; what we call the gravitational field in everyday life is the
gravitoelectric part of the full relativistic gravitational field.
Copyright © Christoph Schiller November 1997–May 2006

In nature, all components of energy–momentum tensor produce gravity effects. In

other words, it is not only mass and energy that produce a field, but also mass or energy
currents. This latter case is called gravitomagnetism (or frame dragging). The name is due
to the analogy with electrodynamics, where it is not only charge density that produces a
field (the electric field), but also charge current (the magnetic field).
In the case of electromagnetism, the distinction between magnetic and electric field

* One is the so-called Gravity Probe B satellite experiment, which should significantly increase the measure-
ment precision; the satellite was put in orbit in 2005, after 30 years of planning.
** This section can be skipped at first reading.
*** The approximation requires low velocities, weak fields, and localized and stationary mass–energy distri-
butions.

Dvipsbugw
 400 iii gravitation and rel ativity • 8. motion of light and vacuum

depends on the observer; each of the two can (partly) be transformed into the other. Grav-
Ref. 356 itation is exactly analogous. Electromagnetism provides a good indication as to how the
two types of gravitational fields behave; this intuition can be directly transferred to grav-
ity. In electrodynamics, the motion x(t) of a charged particle is described by the Lorentz
Page 523 equation
mẍ = qE − qẋ  B . (248)

In other words, the change of speed is due to electric fields E, whereas magnetic fields B
give a velocity-dependent change of the direction of velocity, without changing the speed Dvipsbugw
itself. Both changes depend on the value of the charge q. In the case of gravity this expres-
sion becomes
mẍ = mG − mẋ  H . (249)

The role of charge is taken by mass. In this expression we already know the field G, given
by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

G = ∇φ = ∇ =− 3 .
GM GMx
(250)
r r
As usual, the quantity φ is the (scalar) potential. The field G is the usual gravitational
field of universal gravity, produced by every mass, and in this context is called the grav-
itoelectric field; it has the dimension of an acceleration. Masses are the sources of the
gravitoelectric field. The gravitoelectric field obeys ∆G = −4πGρ, where ρ is the mass
density. A static field G has no vortices; it obeys ∆  G = 0.
It is not hard to show that if gravitoelectric fields exist,
Ref. 357 gravitomagnetic fields must exist as well; the latter appear
whenever one changes from an observer at rest to a moving one. M rod
Page 536 (We will use the same argument in electrodynamics.) A particle
falling perpendicularly towards an infinitely long rod illustrates v
the point, as shown in Figure 184. An observer at rest with respect
to the rod can describe the whole situation with gravitoelectric m particle
forces alone. A second observer, moving along the rod with con- F I G U R E 184 The reality
stant speed, observes that the momentum of the particle along of gravitomagnetism
the rod also increases. This observer will thus not only measure
a gravitoelectric field; he also measures a gravitomagnetic field. Indeed, a mass moving
Challenge 751 ny with velocity v produces a gravitomagnetic (3-) acceleration on a test mass m given by
Copyright © Christoph Schiller November 1997–May 2006

ma = −mv  H (251)

where, almost as in electrodynamics, the static gravitomagnetic field H obeys

H = ∇  A = 16πN ρv (252)

where ρ is mass density of the source of the field and N is a proportionality constant. The
quantity A is called the gravitomagnetic vector potential. In nature, there are no sources for
the gravitomagnetic field; it thus obeys ∇H = 0. The gravitomagnetic field has dimension
of inverse time, like an angular velocity.

Dvipsbugw
 weak fields 401

Challenge 752 ny When the situation in Figure 184 is evaluated, we find that the proportionality constant
N is given by
N = 2 = 7.4 ë 10−28 m~kg ,
G
(253)
c
an extremely small value. We thus find that as in the electrodynamic case, the gravito-
magnetic field is weaker than the gravitoelectric field by a factor of c 2 . It is thus hard to
observe. In addition, a second aspect renders the observation of gravitomagnetism even
more difficult. In contrast to electromagnetism, in the case of gravity there is no way to Dvipsbugw
Challenge 753 n observe pure gravitomagnetic fields (why?); they are always mixed with the usual, gravito-
electric ones. For these reasons, gravitomagnetic effects were measured for the first time
only in the 1990s. We see that universal gravity is the approximation of general relativity
that arises when all gravitomagnetic effects are neglected.
In summary, if a mass moves, it also produces a gravitomagnetic field. How can one
imagine gravitomagnetism? Let’s have a look at its effects. The experiment of Figure 184

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

showed that a moving rod has the effect to slightly accelerate a test mass in the same dir-
ection. In our metaphor of the vacuum as a mattress, it looks as if a moving rod drags the
vacuum along with it, as well as any test mass that happens to be in that region. Gravito-
magnetism can thus be seen as vacuum dragging. Because of a widespread reluctance to
think of the vacuum as a mattress, the expression frame dragging is used instead.
In this description, all frame dragging effects are gravitomagnetic effects. In particular,
a gravitomagnetic field also appears when a large mass rotates, as in the Thirring–Lense
effect of Figure 182. For an angular momentum J the gravitomagnetic field H is a dipole
field; it is given by
Jx
H = ∇  h = ∇  ‹−2 3  (254)
r

exactly as in the electrodynamic case. The gravitomagnetic field around a spinning mass
has three main effects.
First of all, as in electromagnetism, a spinning test particle with angular momentum
S feels a torque if it is near a large spinning mass with angular momentum J. This torque
T is given by
T= = SH .
dS 1
(255)
dt 2
The torque leads to the precession of gyroscopes. For the Earth, this effect is extremely
Copyright © Christoph Schiller November 1997–May 2006

small: at the North Pole, the precession has a conic angle of 0.6 milli-arcseconds and a
rotation rate of the order of 10−10 times that of the Earth.
Since for a torque one has T = Ω̇  S, the dipole field of a large rotating mass with
angular momentum J yields a second effect. An orbiting mass will experience precession
of its orbital plane. Seen from infinity one gets, for an orbit with semimajor axis a and
Challenge 754 ny eccentricity e,

Ω̇ = − =− 2 3 + 2 = 2 3
H G J G 3(Jx)x G 2J
c SxS c SxS5 c a (1 − e 2 )3~2
(256)
2

Dvipsbugw
 402 iii gravitation and rel ativity • 8. motion of light and vacuum

which is the prediction of Lense and Thirring.* The effect is extremely small, giving a
change of only 8 ′′ per orbit for a satellite near the surface of the Earth. Despite this small-
ness and a number of larger effects disturbing it, Ciufolini’s team have managed to con-
Ref. 353 firm the result.
As a third effect of gravitomagnetism, a rotating mass leads to the precession of the
periastron. This is a similar effect to the one produced by space curvature on orbiting
masses even if the central body does not rotate. The rotation just reduces the precession
due to space-time curvature. This effect has been fully confirmed for the famous binary
pulsar PSR B1913+16, as well as for the ‘real’ double pulsar PSR J0737-3039, discovered in 2003. Dvipsbugw
This latter system shows a periastron precession of 16.9°~a, the largest value observed so
far.
The split into gravitoelectric and gravitomagnetic effects is thus a useful approxim-
ation to the description of gravity. It also helps to answer questions such as: How can
gravity keep the Earth orbiting around the Sun, if gravity needs 8 minutes to get from
Challenge 756 ny the Sun to us? To find the answer, thinking about the electromagnetic analogy can help.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In addition, the split of the gravitational field into gravitoelectric and gravitomagnetic
components allows a simple description of gravitational waves.

Gravitational waves
One of the most fantastic predictions of physics is the existence of gravitational waves.
Gravity waves** prove that empty space itself has the ability to move and vibrate. The
basic idea is simple. Since space is elastic, like a large mattress in which we live, space
should be able to oscillate in the form of propagating waves, like a mattress or any other
elastic medium.
TA B L E 36 The expected spectrum of gravitational waves

Frequency Wa v e l e n g t h N a m e Expected
appearance

< 10−4 Hz A 3 Tm extremely low slow binary star systems,

frequencies supermassive black holes
10−4 Hz–10−1 Hz 3 Tm–3 Gm very low frequencies fast binary star systems,
massive black holes, white
dwarf vibrations
10−1 Hz–102 Hz 3 Gm–3 Mm low frequencies binary pulsars, medium and
Copyright © Christoph Schiller November 1997–May 2006

light black holes

102 Hz–105 Hz 3 Mm–3 km medium frequencies supernovae, pulsar
vibrations
105 Hz–108 Hz 3 km–3 m high frequencies unknown; maybe future
human-made sources
A 108 Hz < 3m maybe unknown
cosmological sources

Challenge 755 ny * A homogeneous spinning sphere has an angular momentum given by J = 25 MωR 2 .
** To be strict, the term ‘gravity wave’ has a special meaning: gravity waves are the surface waves of the sea,
where gravity is the restoring force. However, in general relativity, the term is used interchangeably with
‘gravitational wave’.

Dvipsbugw
 weak fields 403

Jørgen Kalckar and Ole Ulfbeck

Ref. 358 have given a simple argument for
the necessity of gravitational waves
based on the existence of a max-
imum speed. They studied two
equal masses falling towards each
other under the effect of gravita- F I G U R E 185 A Gedanken experiment showing the
necessity of gravity waves
tional attraction, and imagined a
spring between them. Such a spring will make the masses bounce towards each other Dvipsbugw
again and again. The central spring stores the kinetic energy from the falling masses. The
energy value can be measured by determining the length by which the spring is com-
pressed. When the spring expands again and hurls the masses back into space, the gravit-
ational attraction will gradually slow down the masses, until they again fall towards each
other, thus starting the same cycle again.
However, the energy stored in the spring must get smaller with each cycle. Whenever

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

a sphere detaches from the spring, it is decelerated by the gravitational pull of the other
sphere. Now, the value of this deceleration depends on the distance to the other mass;
but since there is a maximal propagation velocity, the effective deceleration is given by
the distance the other mass had when its gravity effect started out towards the second
mass. For two masses departing from each other, the effective distance is thus somewhat
smaller than the actual distance. In short, while departing, the real deceleration is larger
than the one calculated without taking the time delay into account.
Similarly, when one mass falls back towards the other, it is accelerated by the other
mass according to the distance it had when the gravity effect started moving towards it.
Therefore, while approaching, the acceleration is smaller than the one calculated without
time delay.
Therefore, the masses arrive with a smaller energy than they departed with. At every
bounce, the spring is compressed a little less. The difference between these two energies
is lost by each mass: it is taken away by space-time, in other words, it is radiated away as
gravitational radiation. The same thing happens with mattresses. Remember that a mass
deforms the space around it as a metal ball on a mattress deforms the surface around it.*
If two metal balls repeatedly bang against each other and then depart again, until they
come back together, they will send out surface waves on the mattress. Over time, this
effect will reduce the distance that the two balls depart from each other after each bang.
As we will see shortly, a similar effect has already been measured, where the two masses,
Copyright © Christoph Schiller November 1997–May 2006

instead of being repelled by a spring, were orbiting each other.

A simple mathematical description of gravity waves follows from the split into gravito-
Ref. 359 magnetic and gravitoelectric effects. It does not take much effort to extend gravitomag-
netostatics and gravitoelectrostatics to gravitodynamics. Just as electrodynamics can be
deduced from Coulomb’s attraction when one switches to other inertial observers, grav-
Challenge 757 ny itodynamics can be deduced from universal gravity. One gets the four equations

* However, in contrast to actual mattresses, there is no friction between the ball and the mattress.

Dvipsbugw
 404 iii gravitation and rel ativity • 8. motion of light and vacuum

∇ G = −4πGρ , ∇G=−
∂H
∂t
∇H = 0 , ∇  H = −16πGρv +
N ∂G
. (257)
G ∂t
We have met two of these equations already. The two other equations are expanded ver-
sions of what we have encountered, taking time-dependence into account. Except for a
factor of 16 instead of 4 in the last equation, the equations for gravtitodynamics are the
same as Maxwell’s equations for electrodynamics.* These equations have a simple prop- Dvipsbugw
erty: in vacuum, one can deduce from them a wave equation for the gravitoelectric and
Challenge 758 ny the gravitomagnetic fields G and H. (It is not hard: try!) In other words, gravity can be-
have like a wave: gravity can radiate. All this follows from the expression of universal
gravity when applied to moving observers, with the requirement that neither observers
nor energy can move faster than c. Both the above argument involving the spring and
the present mathematical argument use the same assumptions and arrive at the same

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

conclusion.
Challenge 759 e A few manipulations show that the speed of these waves is given by
¾
c=
G
. (258)
N

Page 563 This result corresponds to the electromagnetic expression

c=º
1
. (259)
ε0 µ0

The same letter has been used for the two speeds, as they are identical. Both influences
travel with the speed common to all energy with vanishing rest mass. (We note that this is,
strictly speaking, a prediction: the speed of gravitational waves has not yet been measured.
Ref. 361 A claim from 2003 to have done so has turned out to be false.)
Ref. 363 How should one imagine these waves? We sloppily said above that a gravitational wave
corresponds to a surface wave of a mattress; now we have to do better and imagine that
we live inside the mattress. Gravitational waves are thus moving and oscillating deform-
ations of the mattress, i.e., of space. Like mattress waves, it turns out that gravity waves
are transverse. Thus they can be polarized. (Surface waves on mattresses cannot, because
Copyright © Christoph Schiller November 1997–May 2006

in two dimensions there is no polarization.) Gravity waves can be polarized in two inde-
pendent ways. The effects of a gravitational wave are shown in Figure 186, for both linear
* The additional factor reflects the fact that the ratio between angular momentum and energy (the ‘spin’)
of gravity waves is different from that of electromagnetic waves. Gravity waves have spin 2, whereas electro-
magnetic waves have spin 1. Note that since gravity is universal, there can exist only a single kind of spin 2
radiation particle in nature. This is in strong contrast to the spin 1 case, of which there are several examples
in nature.
By the way, the spin of radiation is a classical property. The spin of a wave is the ratio E~Lω, where E is
the energy, L the angular momentum, and ω is the angular frequency. For electromagnetic waves, this ratio
is equal to 1; for gravitational waves, it is 2.
Ref. 360 Note that due to the approximation at the basis of the equations of gravitodynamics, the equations are
neither gauge-invariant nor generally covariant.

Dvipsbugw
weak fields 405

Wave, moving perpendicularly to page

t1 t2 t3 t4 t5

linear polarization in + direction

Dvipsbugw
No wave
(all times)

linear polarization in x direction

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

circular polarization in R sense

circular polarization in L sense

F I G U R E 186 Effects on a circular or spherical body due to a plane gravitational wave

moving in a direction perpendicular to the page

and circular polarization.* We note that the waves are invariant under a rotation by π
and that the two linear polarizations differ by an angle π~4; this shows that the particles
corresponding to the waves, the gravitons, are of spin 2. (In general, the classical radi-

* A (small amplitude) plane gravity wave travelling in the z-direction is described by a metric д given by

’ 1 0 0 0 “
Copyright © Christoph Schiller November 1997–May 2006

– —
д=– —
0 −1 + h x x hx y 0
– —
(260)
0 hx y −1 + h x x 0
” 0 0 0 −1 •

where its two components, whose amplitude ratio determine the polarization, are given by

h ab = B ab sin(kz − ωt + φ ab ) (261)

as in all plane harmonic waves. The amplitudes B ab , the frequency ω and the phase φ are determined by
the specific physical system. The general dispersion relation for the wave number k resulting from the wave
equation is
ω
=c (262)
k
and shows that the waves move with the speed of light.

Dvipsbugw
 406 iii gravitation and rel ativity • 8. motion of light and vacuum

ation field for a spin S particle is invariant under a rotation by 2π~S. In addition, the two
orthogonal linear polarizations of a spin S particle form an angle π~2S. For the photon,
for example, the spin is 1; indeed, its invariant rotation angle is 2π and the angle formed
by the two polarizations is π~2.)
If we image empty space as a mattress that fills space, gravitational waves are wobbling
deformations of the mattress. More precisely, Figure 186 shows that a wave of circular po-
larization has the same properties as a corkscrew advancing through the mattress. We will
discover later on why the analogy between a corkscrew and a gravity wave with circular
polarization works so well. Indeed, in the third part we will find a specific model of the Dvipsbugw
space-time mattress material that automatically incorporates corkscrew waves (instead
of the spin 1 waves shown by ordinary latex mattresses).
How does one produce gravitational waves? Obviously, masses must be accelerated.
But how exactly? The conservation of energy forbids mass monopoles from varying in
strength. We also know from universal gravity that a spherical mass whose radius oscil-
lates would not emit gravitational waves. In addition, the conservation of momentum

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 760 ny forbids mass dipoles from changing.
As a result, only changing quadrupoles can emit waves. For example, two masses in
orbit around each other will emit gravitational waves. Also, any rotating object that is not
cylindrically symmetric around its rotation axis will do so. As a result, rotating an arm
leads to gravitational wave emission. Most of these statements also apply to masses in
Challenge 761 ny mattresses. Can you point out the differences?
Einstein found that the amplitude h of waves at a distance r from a source is given, to
Ref. 362 a good approximation, by the second derivative of the retarded quadrupole moment Q:

h ab = = 4 d tt Q ab (t − r~c) .
2G 1 ret 2G 1
d tt Q ab (264)
c r
4 c r

This expression shows that the amplitude of gravity waves decreases only with 1~r, in
contrast to naive expectations. However, this feature is the same as for electromagnetic
waves. In addition, the small value of the prefactor, 1.6 ë 10−44 Wm~s, shows that truly
gigantic systems are needed to produce quadrupole moment changes that yield any de-
tectable length variations in bodies. To be convinced, just insert a few numbers, keep-
Challenge 762 ny ing in mind that the best present detectors are able to measure length changes down to
h = δl~l = 10−19 . The production of detectable gravitational waves by humans is probably
impossible.
Copyright © Christoph Schiller November 1997–May 2006

Gravitational waves, like all other waves, transport energy.* If we apply the general
formula for the emitted power P to the case of two masses m 1 and m 2 in circular orbits

In another gauge, a plane wave can be written as

’ c (1 + 2φ) “
2
A1 A2 A3
– —
д=– —
A1 −1 + 2φ hx y 0
– —
(263)
A2 hx y −1 + h x x 0
” A3 0 0 −1 •

∂A
where φ and A are the potentials such that G = ∇φ − c∂t and H = ∇  A.
Page 552 * Gravitomagnetism and gravitoelectricity allow one to define a gravitational Poynting vector. It is as easy
Ref. 357 to define and use as in the case of electrodynamics.

Dvipsbugw
 weak fields 407

Ref. 324 around each other at distance l and get

m1 m2 2 4 6
P=− = = ‹  l ω
dE G ... ret ... ret 32 G
Q Q (265)
dt 45c 5 ab ab 5 c 5 m1 + m2

which, using Kepler’s relation 4π 2 r 3 ~T 2 = G(m 1 + m 2 ), becomes

32 G 4 (m 1 m 2 )2 (m 1 + m 2 )
P= . (266) Dvipsbugw
5 c5 l5
Ref. 324 For elliptical orbits, the rate increases with the ellipticity, as explained by Goenner. Insert-
ing the values for the case of the Earth and the Sun, we get a power of about 200 W, and
a value of 400 W for the Jupiter–Sun system. These values are so small that their effect
cannot be detected at all.
For all orbiting systems, the frequency of the waves is twice the orbital frequency, as

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 763 ny you might want to check. These low frequencies make it even more difficult to detect
them.
As a result, the only observation of effects
of gravitational waves to date is in binary time delay
pulsars. Pulsars are small but extremely dense
stars; even with a mass equal to that of the Sun, 1975 1980 1990 2000
their diameter is only about 10 km. Therefore
they can orbit each other at small distances year

and high speeds. Indeed, in the most famous

binary pulsar system, PSR 1913+16, the two stars
orbit each other in an amazing 7.8 h, even
though their semimajor axis is about 700 Mm,
just less than twice the Earth–Moon distance.
Since their orbital speed is up to 400 km~s, the
system is noticeably relativistic.
Pulsars have a useful property: because of
preliminary figure
their rotation, they emit extremely regular ra-
dio pulses (hence their name), often in milli- F I G U R E 187 Comparison between measured
second periods. Therefore it is easy to follow time delay for the periastron of the binary
their orbit by measuring the change of pulse pulsar PSR 1913+16 and the prediction due to
Copyright © Christoph Schiller November 1997–May 2006

energy loss by gravitational radiation

arrival time. In a famous experiment, a team
of astrophysicists led by Joseph Taylor** meas-
Ref. 364 ured the speed decrease of the binary pulsar system just mentioned. Eliminating all other
effects and collecting data for 20 years, they found a decrease in the orbital frequency,
shown in Figure 187. The slowdown is due to gravity wave emission. The results exactly
fit the prediction by general relativity, without any adjustable parameter. (You might want
Challenge 764 ny to check that the effect must be quadratic in time.) This is the only case so far in which
Page 394 general relativity has been tested up to (v~c)5 precision. To get an idea of the precision,
consider that this experiment detected a reduction of the orbital diameter of 3.1 mm per

** In 1993 he shared the Nobel Prize in physics for his life’s work.

Dvipsbugw
 408 iii gravitation and rel ativity • 8. motion of light and vacuum

mirror
L1

mirror
light L2
source

F I G U R E 188 Detection of Dvipsbugw

gravitational waves

Ref. 364 orbit, or 3.5 m per year! The measurements were possible only because the two stars in
this system are neutron stars with small size, large velocities and purely gravitational in-
teractions. The pulsar rotation period around its axis, about 59 ms, is known to eleven
digits of precision, the orbital time of 7.8 h is known to ten digits and the eccentricity of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 324 the orbit to six digits.
The direct detection of gravitational waves is one of the aims of experimental general
relativity. The race has been on since the 1990s. The basic idea is simple, as shown in
Figure 188: take four bodies, usually four mirrors, for which the line connecting one pair
is perpendicular to the line connecting the other pair. Then measure the distance changes
of each pair. If a gravitational wave comes by, one pair will increase in distance and the
other will decrease, at the same time.
Since detectable gravitational waves cannot be produced by humans, wave detection
first of all requires the patience to wait for a strong enough wave to come by. Secondly, a
system able to detect length changes of the order of 10−22 or better is needed – in other
words, a lot of money. Any detection is guaranteed to make the news on television.*
It turns out that even for a body around a black hole, only about 6 % of the rest mass
can be radiated away as gravitational waves; furthermore, most of the energy is radiated
during the final fall into the black hole, so that only quite violent processes, such as black
hole collisions, are good candidates for detectable gravity wave sources.
Gravitational waves are a fascinating area of study. They still provide many topics to
Challenge 766 r explore. For example: can you find a method to measure their speed? A well-publicized
Ref. 361 but false claim appeared in 2003. Indeed, any correct measurement that does not simply
use two spaced detectors of the type of Figure 188 would be a scientific sensation.
For the time being, another question on gravitational waves remains open: If all change
Copyright © Christoph Schiller November 1997–May 2006

is due to motion of particles, as the Greeks maintained, how do gravity waves fit into the
Challenge 767 ny picture? If gravitational waves were made of particles, space-time would also have to be.
We have to wait until the beginning of the third part of our ascent to say more.

Bending of light and radio waves

As we know from above, gravity also influences the motion of light. A distant observer
measures a changing value for the light speed v near a mass. (Measured at his own loca-
tion, the speed of light is of course always c.) It turns out that a distant observer measures

Ref. 365 * The topic of gravity waves is full of interesting sidelines. For example, can gravity waves be used to power
Challenge 765 ny a rocket? Yes, say Bonnor and Piper. You might ponder the possibility yourself.

Dvipsbugw
 weak fields 409

a lower speed, so that for him, gravity has the same effects as a dense optical medium. It
takes only a little bit of imagination to see that this effect will thus increase the bending
of light near masses already deduced in 1801 by Soldner for universal gravity.
The following is a simple way to calculate the effect. As usual, we
use the coordinate system of flat space-time at infinity. The idea is to
do all calculations to first order, as the value of the bending is very
Ref. 366 small. The angle of deflection α, to first order, is simply α
Dvipsbugw
α=
ª ∂v
∫−ª ∂x
dy , (267)
M
b

where v is the speed of light measured by a distant observer. (Can you

y
Challenge 768 ny confirm this?) The next step is to use the Schwarzschild metric

dr 2 r2 2
dτ 2 = (1 − )dt
2GM 2
− − dφ
(c 2 − r ) c 2
(268)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2G M x
rc 2
F I G U R E 189
Challenge 769 ny and transform it into (x, y) coordinates to first order. That gives Calculating the
bending of light by
a mass
dτ 2 = (1 − )dt 2 − (1 + ) (dx 2 + dy 2 )
2GM 2GM 1
(269)
rc 2 rc 2 c 2
which again to first order leads to

= (1 − )c .
∂v 2GM
(270)
∂x rc 2
This confirms what we know already, namely that distant observers see light slowed down
when passing near a mass. Thus we can also speak of a height-dependent index of refrac-
tion. In other words, constant local light speed leads to a global slowdown.
Challenge 770 ny Inserting the last result in (267) and using a clever substitution, we get a deviation
angle α given by
α= 2
4GM 1
(271)
c b

where the distance b is the so-called impact parameter of the approaching light beam.
Copyright © Christoph Schiller November 1997–May 2006

Page 133 The resulting deviation angle α is twice the result we found for universal gravity. For a
beam just above the surface of the Sun, the result is the famous value of 1.75 ′′ which was
confirmed by the measurement expedition of 1919. (How did they measure the deviation
Challenge 771 ny angle?) This was the experiment that made Einstein famous, as it showed that universal
gravity is wrong. In fact, Einstein was lucky. Two earlier expeditions organized to meas-
ure the value had failed. In 1912, it was impossible to take data because of rain, and in 1914
in Crimea, scientists were arrested (by mistake) as spies, because the world war had just
Ref. 367 begun. But in 1911, Einstein had already published an incorrect calculation, giving only the
Soldner value with half the correct size; only in 1915, when he completed general relativ-
Page 133 ity, did he find the correct result. Therefore Einstein became famous only because of the

Dvipsbugw
 410 iii gravitation and rel ativity • 8. motion of light and vacuum

failure of the two expeditions that took place before he published his correct calculation.
For high-precision experiments around the Sun, it is more effective to measure the
bending of radio waves, as they encounter fewer problems when they propagate through
the solar corona. So far, over a dozen independent experiments have done so, using ra-
dio sources in the sky which lie on the path of the Sun. They have confirmed general
Ref. 346, Ref. 323, relativity’s prediction within a few per cent.
Ref. 324 So far, bending of radiation has also been observed near Jupiter, near certain stars,
Page 468 near several galaxies and near galaxy clusters. For the Earth, the angle is at most 3 nrad,
too small to be measured yet, even though this may be feasible in the near future. There Dvipsbugw
is a chance to detect this value if, as Andrew Gould proposes, the data of the satellite
Hipparcos, which is taking precision pictures of the night sky, are analysed properly in
the future.
Page 417 Of course, the bending of light also confirms that in a triangle, the sum of the angles
does not add up to π, as is predicted later for curved space. (What is the sign of the
Challenge 772 ny curvature?)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Time delay
The above calculation of the bending of light near masses shows that for a distant observer,
light is slowed down near a mass. Constant local light speed leads to a global light speed
slowdown. If light were not slowed down near a mass, it would have to go faster than c
Ref. 368 for an observer near the mass!* In 1964, Irwin Shapiro had the idea to measure this effect.
He proposed two methods. The first was to send radar pulses to Venus, and measure the
time taken for the reflection to get back to Earth. If the signals pass near the Sun, they
will be delayed. The second was to use an artificial satellite communicating with Earth.
Ref. 369 The first measurement was published in 1968, and directly confirmed the prediction of
general relativity within experimental errors. All subsequent tests of the same type, such
as the one shown in Figure 190, have also confirmed the prediction within experimental
errors, which nowadays are of the order of one part in a thousand. The delay has also
Ref. 370 been measured in binary pulsars, as there are a few such systems in the sky for which the
line of sight lies almost precisely in the orbital plane.
The simple calculations presented here suggest a challenge: Is it also possible to de-
scribe full general relativity – thus gravitation in strong fields – as a change of the speed
Challenge 774 ny of light with position and time induced by mass and energy? Copyright © Christoph Schiller November 1997–May 2006

Effects on orbits
Astronomy allows precise measurements of motions. So, Einstein first of all tried to ap-
ply his results to the motion of planets. He looked for deviations of their motions from
the predictions of universal gravity. Einstein found such a deviation: the precession of
the perihelion of Mercury. The effect is shown in Figure 191. Einstein said later that the
moment he found out that his calculation for the precession of Mercury matched obser-
vations was one of the happiest moments of his life.
* A nice exercise is to show that the bending of a slow particle gives the Soldner value, whereas with increas-
Challenge 773 e ing speed, the value of the bending approaches twice that value. In all these considerations, the rotation
of the mass has been neglected. As the effect of frame dragging shows, rotation also changes the deviation
angle; however, in all cases studied so far, the influence is below the detection threshold.

Dvipsbugw
 weak fields 411

10 May 1970
Earth orbit
31 March 1970

Sun Mariner 6
orbit

Dvipsbugw
a: semimajor periastron
axis
240 a
Time delay (µs)

180
M
120

60

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

0
Jan Feb Mar Apr May Jun
1970
F I G U R E 190 Time delay in radio signals – one F I G U R E 191 The orbit around a
of the experiments by Irwin Shapiro central body in general relativity

The calculation is not difficult. In universal gravity, orbits are calculated by setting
a grav = a centri , in other words, by setting GM~r 2 = ω 2 r and fixing energy and angular
momentum. The mass of the orbiting satellite does not appear explicitly.
In general relativity, the mass of the orbiting satellite is made to disappear by rescaling
Ref. 323, Ref. 324 energy and angular momentum as e = E~mc 2 and j = J~m. Next, the space curvature
Page 384 needs to be included. We use the Schwarzschild metric (268) mentioned above to deduce
that the initial condition for the energy e, together with its conservation, leads to a rela-
Challenge 775 e tion between proper time τ and time t at infinity:

=
dt e
, (272)
Copyright © Christoph Schiller November 1997–May 2006

dτ 1 − 2GM~rc 2

whereas the initial condition on the angular momentum j and its conservation imply that

= 2 .
dφ j
(273)
dτ r
These relations are valid for any particle, whatever its mass m. Inserting all this into the
Schwarzschild metric, we find that the motion of a particle follows

( ) + V 2 ( j, r) = e 2
dr 2
(274)
cdτ

Dvipsbugw
 412 iii gravitation and rel ativity • 8. motion of light and vacuum

where the effective potential V is given by

j2
V 2 (J, r) = (1 − )(1 ).
2GM
+ (275)
rc 2 r2 c2
Challenge 776 ny The expression differs slightly from the one in universal gravity, as you might want to
Challenge 777 e check. We now need to solve for r(φ). For circular orbits we get two possibilities

6GM~c 2 Dvipsbugw
r = ¼ (276)
1  1 − 12( GcMj )2
º
where the minus sign gives a stable and the plus sign an unstable orbit. If c j~GM < 2 3 ,
º
no stable orbit exists; the object will impact the surface or, for a black hole, be swallowed.
There is a stable circular orbit only if the angular momentum j is larger than 2 3 GM~c.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We thus find that in general relativity, in contrast to universal gravity, there is a smallest
stable circular orbit. The radius of this smallest stable circular orbit is 6GM~c 2 = 3R S .
What is the situation for elliptical orbits? Setting u = 1~r in (274) and differentiating,
the equation for u(φ) becomes

u′ + u =
GM 3GM 2
+ 2 u . (277)
j2 c

Without the nonlinear correction due to general relativity on the far right, the solutions
Challenge 778 e are the famous conic sections

u 0 (φ) = (1 + ε cos φ) ,
GM
(278)
j2

i.e. ellipses, parabolas or hyperbolas. The type of conic section depends on the value of
the parameter ε, the so-called eccentricity. We know the shapes of these curves from uni-
Page 127 versal gravity. Now, general relativity introduces the nonlinear term on the right-hand
side of equation (277). Thus the solutions are not conic sections any more; however, as
Challenge 779 e the correction is small, a good approximation is given by

3G 2 M 2
u 1 (φ) = [1
GM
Copyright © Christoph Schiller November 1997–May 2006

+ ε cos(φ − φ)] . (279)

j2 j2 c 2

The hyperbolas and parabolas of universal gravity are thus slightly deformed. Instead of
elliptical orbits we get the famous rosetta path shown in Figure 191. Such a path is above
all characterized by a periastron shift. The periastron, or perihelion in the case of the Sun,
is the nearest point to the central body reached by an orbiting body. The periastron turns
Challenge 780 e around the central body by an angle

α  6π
GM
a(1 − ε 2 )c 2
(280)

Dvipsbugw
 weak fields 413

geodesic
precession
Earth
start
Lense– S
Dvipsbugw
after one
Thirring
orbit
precession

F I G U R E 192 The geodesic effect

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

for every orbit, where a is the semimajor axis. For Mercury, the value is 43 ′′ per century.
Around 1900, this was the only known effect that was unexplained by universal gravity;
when Einstein’s calculation led him to exactly that value, he was overflowing with joy for
many days.
To be sure about the equality between calculation and experiment, all other effects
leading to rosetta paths must be eliminated. For some time, it was thought that the quad-
rupole moment of the Sun could be an alternative source of this effect; later measurements
ruled out this possibility.
In the meantime, the perihelion shift has been measured also for the orbits of Icarus,
Venus and Mars around the Sun, as well as for several binary star systems. In binary
Ref. 370 pulsars, the periastron shift can be as large as several degrees per year. In all cases, expres-
sion (280) describes the motion within experimental errors.
We note that even the rosetta orbit itself is not really stable, due to the emission of
gravitational waves. But in the solar system, the power lost this way is completely negli-
Page 407 gible even over thousands of millions of years, as we saw above, so that the rosetta path
remains a good description of observations.

The geodesic effect

When a pointed body orbits a central mass m at distance r, the direction of the tip will
Copyright © Christoph Schiller November 1997–May 2006

not be the same after a full orbit. This effect exists only in general relativity. The angle α
describing the direction change is given by
¾
’ 3Gm “ 3πGm
α = 2π 1 − 1−  .
” rc 2 •
(281)
rc 2

The angle change is called the geodesic effect – ‘geodetic’ in other languages. It is a further
consequence of the split into gravitoelectric and gravitomagnetic fields, as you may want
Challenge 781 e to show. Obviously, it does not exist in universal gravity.
In cases where the pointing of the orbiting body is realized by an intrinsic rotation,
such as a spinning satellite, the geodesic effect produces a precession of the axis. Thus the

Dvipsbugw
 414 iii gravitation and rel ativity • 8. motion of light and vacuum

effect is comparable to spin–orbit coupling in atomic theory. (The Thirring–Lense effect

mentioned above is analogous to spin–spin coupling.)
Ref. 371 The geodesic effect, or geodesic precession, was predicted by Willem de Sitter in 1916;
in particular, he proposed detecting that the Earth–Moon system would change its point-
ing direction in its fall around the Sun. The effect is tiny; for the axis of the Moon the
precession angle is about 0.019 arcsec per year. The effect was first detected in 1987 by an
Ref. 372 Italian team for the Earth–Moon system, through a combination of radio-interferometry
and lunar ranging, making use of the Cat’s-eyes deposited by Lunakhod and Apollo on
the Moon. Experiments to detect it in artificial satellites are also under way. Dvipsbugw
At first sight, geodesic precession is similar to the Thomas precession found in special
Page 309 relativity. In both cases, a transport along a closed line results in the loss of the original
direction. However, a careful investigation shows that Thomas precession can be added
to geodesic precession by applying some additional, non-gravitational interaction, so the
analogy is shaky.
This completes our discussion of weak gravity effects. We now turn to strong gravity,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where curvature cannot be neglected and where the fun is even more intense.

Curiosities and fun challenges about weak fields

Here are some questions to think about.
**
Is there a static, oscillating gravitational field?

**
Are beams of gravitational waves, analogous to beams of light, possible?
**
Would two parallel beams of gravitational waves attract each other?

How is curvature measured?

We have seen that in the precise description of gravity, motion depends on space-time
curvature. In order quantify this idea, we first of all need to describe curvature itself as
accurately as possible. To clarify the issue, we will start the discussion in two dimensions,
and then move to three and four dimensions.
Copyright © Christoph Schiller November 1997–May 2006

Obviously, a flat sheet of paper has no curvature. If we roll it into a cone or a cylinder,
it gets what is called extrinsic curvature; however, the sheet of paper still looks flat for any
two-dimensional animal living on it – as approximated by an ant walking over it. In other
words, the intrinsic curvature of the sheet of paper is zero even if the sheet as a whole is
extrinsically curved. (Can a one-dimensional space have intrinsic curvature? Is a torus
Challenge 782 n intrinsically curved?)
Intrinsic curvature is thus the stronger concept, measuring the curvature which can be
observed even by an ant. The surface of the Earth, the surface of an island, or the slopes
of a mountain* are intrinsically curved. Whenever we talk about curvature in general

Challenge 783 e * Unless the mountain has the shape of a perfect cone. Can you confirm this?

Dvipsbugw
 how is curvature measured? 415

Dvipsbugw

F I G U R E 193 Positive, vanishing and negative curvature in two

dimensions

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

relativity, we always mean intrinsic curvature, since any observer in nature is by definition
in the same situation as an ant on a surface: their experience, their actions and plans
always only concern their closest neighbourhood in space and time.
But how can an ant determine whether it lives on an intrinsically curved surface?**
One way is shown in Figure 193. The ant can check whether either the circumference of
a circle or its area bears a Euclidean relation to the measured radius. She can even use
the difference between the measured and the Euclidean values as a measure for the local
intrinsic curvature, if she takes the limit for vanishingly small circles and if she normalizes
the values correctly. In other words, the ant can imagine to cut out a little disc around the
point she is on, to iron it flat and to check whether the disc would tear or produce folds.
Any two-dimensional surface is intrinsically curved whenever ironing is not able to make
a flat street map out of it. The ‘density’ of folds or tears is related to the curvature.
This means that we can recognize intrinsic curvature also by checking whether two
parallel lines stay parallel, approach each other, or depart from each other. In the first case,
such as lines on a paper cylinder, the surface is said to have vanishing intrinsic curvature;
a surface with approaching parallels, such as the Earth, is said to have positive curvature,
and a surface with diverging parallels, such as a saddle, is said to have negative curvature.
In short, positive curvature means that we are more restricted in our movements, negative
that we are less restricted. A constant curvature even implies being locked in a finite space.
You might want to check this with Figure 193.
Copyright © Christoph Schiller November 1997–May 2006

The third way to measure curvature uses triangles. On curved surfaces the sum of
angles in a triangle is either larger or smaller than π.
Ref. 373 Let us see how we can quantify curvature. First a question of vocabulary: a sphere with
radius a is said, by definition, to have an intrinsic curvature K = 1~a 2 . Therefore a plane
has zero curvature. You might check that for a circle on a sphere, the measured radius r,
Challenge 784 e circumference C, and area A are related by

C = 2πr(1 − A = πr 2 (1 −
K 2 K 2
r + ...) and r + ...) (282)
6 12
** Note that the answer to this question also tells us how to distinguish real curvature from curved coordinate
systems on a flat space. This question is often asked by those approaching general relativity for the first time.

Dvipsbugw
 416 iii gravitation and rel ativity • 8. motion of light and vacuum

direction of point of interest

minimal curvature

right
angle direction of
maximal curvature

F I G U R E 194 The maximum and minimum curvature Dvipsbugw

of a curved surface

where the dots imply higher-order terms. This allows one to define the intrinsic curvature
K, also called the Gaussian curvature, for a general point on a two-dimensional surface
in either of the following two equivalent ways:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

K = 6 lim(1 − ) K = 12 lim(1 − ) .
C 1 A 1
or (283)
r 0 2πr r 2 r 0 πr 2 r 2
These expressions allow an ant to measure the intrinsic curvature at each point for any
smooth surface.* From now on in this text, curvature will always mean intrinsic curvature.
Note that the curvature can be different from place to place, and that it can be positive, as
for an egg, or negative, as for the part of a torus nearest to the hole. A saddle is another ex-
ample of the latter case, but, unlike the torus, its curvature changes along all directions. In
fact, it is not possible at all to fit a two-dimensional surface of constant negative curvature
inside three-dimensional space; one needs at least four dimensions, as you can find out
Challenge 786 e if you try to imagine the situation.
For any surface, at every point, the direction of maximum curvature and the direc-
tion of minimum curvature are perpendicular to each other. This relationship, shown in
Figure 194, was discovered by Leonhard Euler in the eighteenth century. You might want
to check this with a tea cup, with a sculpture by Henry Moore, or with any other curved
Challenge 787 e object from your surroundings, such as a Volkswagen Beetle. The Gaussian curvature K
defined in (283) is in fact the product of the two corresponding inverse curvature radii.
Thus, even though line curvature is not an intrinsic property, this special product is. Gaus-
sian curvature is a measure of the intrinsic curvature. Intrinsic measures of curvature are
needed if one is forced to stay inside the surface or space one is exploring. Physicists are
Copyright © Christoph Schiller November 1997–May 2006

thus particularly interested in Gaussian curvature and its higher-dimensional analogues.

For three-dimensional ‘surfaces’, the issue is a bit more involved. First of all, we have
difficulties imagining the situation. But we can still accept that the curvature of a small
disc around a point will depend on its orientation. Let us first look at the simplest case. If

* If the n-dimensional volume of a sphere is written as Vn = C n r n and its (n − 1)-dimensional ‘surface’ as

Ref. 374 O n = nC n r n−1 , one can generalize the expressions for curvature to

K = 3(n + 2) lim(1 − ) K = 3n lim(1 − ) ,

Vn 1 On 1
or (284)
r 0 Cn r n r2 r 0 nC n r n−1 r 2

Challenge 785 ny as shown by Vermeil. A famous riddle is to determine the number C n .

Dvipsbugw
 how is curvature measured? 417

the curvature at a point is the same in all directions, the point is called isotropic. We can
imagine a small sphere around that point. In this special case, in three dimensions, the
relation between the measured radius r and the measured surface area A and volume V
Challenge 788 ny of the sphere lead to

A = 4πr 2 (1 − V= r (1 − r 2 + ...) ,
K 2 4π 3 K
r + ...) and (285)
3 3 5
where K is the curvature for an isotropic point. This leads to Dvipsbugw
»
r − A~4π
K = 3 lim(1 − ) 2 = 6 lim = 6 lim 3 ,
A 1 r excess
(286)
r 0 4πr r2 r 0 r 3 r 0 r
»
defining the excess radius as r excess = r − A~4π . We thus find that for a three-
dimensional space, the average curvature is six times the excess radius of a small sphere

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

divided by the cube of the radius. A positive curvature is equivalent to a positive excess
radius, and similarly for vanishing and negative cases.
Of course, a curvature value defined in this way is only an average over all possible
directions. The precise definition of curvature involves a disc. For points that are not
isotropic, the value yielded by a disc will differ from the value calculated by using a sphere,
as it will depend on the orientation of the disc. In fact, there is a relationship between all
Challenge 789 ny possible disc curvatures at a given point; taken together, they must form a tensor. (Why?)
For a full description of curvature, we thus have to specify, as for any tensor in three
dimensions, the main curvature values in three orthogonal directions.*
What are the curvature values in the space around us? Already in 1827, the mathem-
atician and physicist Carl-Friedrich Gauss* is said to have checked whether the three
angles formed by three mountain peaks near his place of residence added up to π.
Nowadays we know that the deviation δ from the angle π on the surface of a body of
mass M and radius r is given by

δ = π − (α + β + γ)  A triangle K = A triangle
GM
. (287)
r3 c2
* These three disc values are not independent however, since together, they must yield the just-mentioned
average volume curvature K. In total, there are thus three independent scalars describing the curvature in
three dimensions (at each point). With the metric tensor дab and the Ricci tensor R ab to be introduced below,
Copyright © Christoph Schiller November 1997–May 2006

one possibility is to take for the three independent numbers the values R = −2K, R ab R ab and detR~detд.
* Carl-Friedrich Gauß (b. 1777 Braunschweig, d. 1855 Göttingen), German mathematician. Together with
the Leonhard Euler, the most important mathematician of all times. A famous enfant prodige, when he was
19 years old, he constructed the regular heptadecagon with compass and ruler (see http://www.mathworld.
wolfram.com/Heptadecagon.html). He was so proud of this result that he put a drawing of the figure on his
tomb. Gauss produced many results in number theory, topology, statistics, algebra, complex numbers and
differential geometry which are part of modern mathematics and bear his name. Among his many accom-
plishments, he produced a theory of curvature and developed non-Euclidean geometry. He also worked on
electromagnetism and astronomy.
Gauss was a difficult character, worked always for himself, and did not found a school. He published
little, as his motto was: pauca sed matura. As a consequence, when another mathematician published a new
result, he regularly produced a notebook in which he had noted the very same result already years before.
His notebooks are now available online at http://www.sub.uni-goettingen.de.

Dvipsbugw
 418 iii gravitation and rel ativity • 8. motion of light and vacuum

F I G U R E 195 Curvature (in two dimensions) and geodesic behaviour

Dvipsbugw

This expression is typical for hyperbolic geometries. For the case of mathematical negative
curvature K, the first equality was deduced by Johann Lambert (1728–1777). However, it
was Einstein who discovered that the negative curvature K is related to the mass and
gravitation of a body. For the case of the Earth and typical mountain distances, the angle
δ is of the order of 10−14 rad. Gauss had no chance to detect any deviation, and in fact he
detected none. Even today, studies with lasers and high-precision apparatus have detected

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

no deviation yet – on Earth. The right-hand factor, which measures the curvature of space-
time on the surface of the Earth, is simply too small. But Gauss did not know, as we do
today, that gravity and curvature go hand in hand.

Curvature and space-time

Notre tête est ronde pour permettre à la pensée

“ de changer de direction.**

In nature, with four space-time dimensions, specifying curvature requires a more in-
Francis Picabia

”
volved approach. First of all, the use of space-time coordinates automatically introduces
the speed of light c as limit speed, which is a central requirement in general relativity.
Furthermore, the number of dimensions being four, we expect a value for an average
curvature at a point, defined by comparing the 4-volume of a 4-sphere in space-time
with the one deduced from the measured radius; then we expect a set of ‘almost aver-
age’ curvatures defined by 3-volumes of 3-spheres in various orientations, plus a set of
‘low-level’ curvatures defined by usual 2-areas of usual 2-discs in even more orientations.
Obviously, we need to bring some order to bear on this set, and we need to avoid the
double counting we encountered in the case of three dimensions.
Fortunately, physics can help to make the mathematics easier. We start by defining
Copyright © Christoph Schiller November 1997–May 2006

what we mean by curvature of space-time. Then we will define curvatures for discs of
various orientations. To achieve this, we interpret the definition of curvature in another
way, which allows us to generalize it to time as well. Figure 195 illustrates the fact that the
curvature K also describes how geodesics diverge. Geodesics are the straightest paths on
a surface, i.e. those paths that a tiny car or tricycle would follow if it drove on the surface
keeping the steering wheel straight.
Challenge 790 e If a space is curved, the separation s will increase along the geodesics as

** ‘Our head is round in order to allow our thougths to change direction.’ Francis Picabia (b. 1879 Paris, d.
1953 Paris) French dadaist and surrealist painter.

Dvipsbugw
 how is curvature measured? 419

d2 s
= −Ks + higher orders (288)
dl 2

where l measures the length along the geodesic, and K is the curvature, in other words,
the inverse squared curvature radius. In space-time, this relation is extended by sub-
stituting proper time (times the speed of light) for proper length. Thus separation and
curvature are related by
d2 s
= −Kc 2 s + higher orders . (289) Dvipsbugw
dτ 2
But this is the definition of an acceleration. In other words, what in the purely spatial
case is described by curvature, in the case of space-time becomes the relative acceleration
of two particles freely falling from nearby points. Indeed, we have encountered these
Page 129 accelerations already: they describe tidal effects. In short, space-time curvature and tidal
effects are precisely the same.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Obviously, the magnitude of tidal effects, and thus of curvature, will depend on the
orientation – more precisely on the orientation of the space-time plane formed by the
two particle velocities. The definition also shows that K is a tensor, so that later on we
Challenge 791 ny will have to add indices to it. (How many?) The fun is that we can avoid indices for a
Ref. 375 while by looking at a special combination of spatial curvatures. If we take three planes in
space, all orthogonal to each other and intersecting at a given point, the sum of the three
so-called sectional curvature values does not depend on the observer. (This corresponds
to the tensor trace.) Can you confirm this, by using the definition of the curvature just
Challenge 792 ny given?
The sum of the three sectional curvatures defined for mutually orthogonal planes K(12),
Challenge 793 ny K(23) and K(31) , is related to the excess radius defined above. Can you find out how?
If a surface has constant (intrinsic) curvature, i.e. the same curvature at all locations,
Challenge 794 e geometrical objects can be moved around without deforming them. Can you picture this?
In summary, curvature is not such a difficult concept. It describes the deformation
Ref. 376 of space-time. If we imagine space (-time) as a big blob of rubber in which we live, the
curvature at a point describes how this blob is squeezed at that point. Since we live inside
the rubber, we need to use ‘insider’ methods, such as excess radii and sectional curvatures,
to describe the deformation. Relativity often seems difficult to learn because people do
not like to think about the vacuum in this way, and even less to explain it in this way. (For
a hundred years it was an article of faith for every physicist to say that the vacuum was
Copyright © Christoph Schiller November 1997–May 2006

empty.) Picturing vacuum as a substance can help us in many ways to understand general
relativity.

Curvature and motion in general relativity

As mentioned above, one half of general relativity is the statement that any object moves
along paths of maximum proper time, i.e. along geodesics. The other half is contained
in a single expression: for every observer, the sum of all three proper sectional spatial
curvatures at a point is given by

K(12) + K(23) + K(31) =

8πG (0)
W (290)
c4

Dvipsbugw
 420 iii gravitation and rel ativity • 8. motion of light and vacuum

where W (0) is the proper energy density at the point. The lower indices indicate the mixed
curvatures defined by the three orthogonal directions 1, 2 and 3. This is all of general
relativity in one paragraph.
Challenge 795 e An equivalent expression is easily found using the excess radius defined above, by
introducing the mass M = V W (0) ~c 2 . For the surface area A of the volume V containing
the mass, we get
»
r excess = r − A~4π = 2 M .
G
(291)
3c
Dvipsbugw
In short, general relativity affirms that for every observer, the excess radius of a small
sphere is given by the mass inside the sphere.*
Note that the above expression implies that the average space curvature at a point in
empty space vanishes. As we will see shortly, this means that near a spherical mass the
negative of the curvature towards the mass is equal to twice the curvature around the
mass; the total sum is thus zero.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Curvature will also differ from point to point. In particular, the expression implies that
if energy moves, curvature will move with it. In short, both space curvature and, as we
will see shortly, space-time curvature change over space and time.
We note in passing that curvature has an annoying effect: the relative velocity of distant
Challenge 796 ny observers is undefined. Can you provide the argument? In curved space, relative velocity
is defined only for nearby objects – in fact only for objects at no distance at all. Only in
flat space are relative velocities of distant objects well defined.
The quantities appearing in expression (290) are independent of the observer. But often
people want to use observer-dependent quantities. The relation then gets more involved;
the single equation (290) must be expanded to ten equations, called Einstein’s field equa-
tions. They will be introduced below. But before we do that, we will check that general
relativity makes sense. We will skip the check that it contains special relativity as a limit-
ing case, and go directly to the main test.

Universal gravity
The only reason which keeps me here is gravity.

“
For small velocities and low curvature values, the temporal curvatures K(0 j) turn out to
Anonymous

”
have a special property. In this case, they can be defined as the second spatial derivatives
Copyright © Christoph Schiller November 1997–May 2006

Challenge 797 e of a single scalar function φ. In other words, we can write

∂2 φ
K(0 j) = .
∂(x j )2
(293)

Ref. 377 * Another, equivalent formulation is that for small radii the area A is given by

A = 4πr (1 + r R)
2 1 2
(292)
9
where R is the Ricci scalar, to be introduced later on.

Dvipsbugw
 how is curvature measured? 421

In everyday situations, the function φ turns out to be the gravitational potential. Indeed,
universal gravity is the limiting case of general relativity for small speeds and small spatial
curvature. These two limits imply, making use of W (0) = ρc 2 and c ª, that

K(i j) = 0 and K(01) + K(02) + K(03) = 4πGρ . (294)

In other words, for small speeds, space is flat and the potential obeys Poisson’s equation.
Universal gravity is thus indeed the low speed and low curvature limit of general relativity.
Can you show that relation (290) between curvature and energy density indeed means Dvipsbugw
Challenge 798 ny that time near a mass depends on the height, as stated in the beginning of this chapter?

The Schwarzschild metric

Ref. 375 What is the curvature of space-time near a spherical mass?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

– CS – more to be inserted – CS –

Challenge 799 ny The curvature of the Schwarzschild metric is given by

K rφ = K rθ = − and K θ φ = 2 2 3
GM GM
c2 r3 c r
K tφ = K tθ = 2 3 and K tr = −2 2 3
GM GM
(295)
c r c r

Ref. 375 everywhere. The dependence on 1~r 3 follows from the general dependence of all tidal
Page 129 effects; we have already calculated them in the chapter on universal gravity. The factors
G~c 2 are due to the maximum force of gravity; only the numerical prefactors need to be
calculated from general relativity. The average curvature obviously vanishes, as it does
Challenge 800 ny for all vacuum. As expected, the values of the curvatures near the surface of the Earth are
exceedingly small.

Curiosities and fun challenges about curvature

Every physicist should have an intuitive understanding of curvature.

**
Copyright © Christoph Schiller November 1997–May 2006

A fly has landed on the outside of a cylindrical glass, 1 cm below its rim. A drop of honey
is located halfway around the glass, also on the outside, 2 cm below the rim. What is the
Challenge 801 e shortest distance from the fly to the drop? What is the shortest distance if the drop is on
the inside of the glass?
**
Challenge 802 e Where are the points of highest and lowest Gaussian curvature on an egg?

Dvipsbugw
422 iii gravitation and rel ativity • 8. motion of light and vacuum

All observers – heavier mathematics*

Jeder Straßenjunge in unserem mathematischen

“ Göttingen versteht mehr von vierdimensionaler

Geometrie als Einstein. Aber trotzdem hat
Einstein die Sache gemacht, und nicht die
großen Mathematiker.
David Hilbert**

Now that we have a feeling for curvature, we want to describe it in a way that allows any
observer to talk to any other observer. Unfortunately, this means using formulae with
tensors. These formulae look daunting. The challenge is to see in each of the expressions
” Dvipsbugw

the essential point (e.g. by forgetting all indices for a while) and not to be distracted by
those small letters sprinkled all over them.

The curvature of space-time

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Il faut suivre sa pente, surtout si elle monte.***

“
We mentioned above that a 4-dimensional space-time is described by 2-curvature, 3-
curvature and 4-curvature. Many texts on general relativity start with 3-curvature. These
André Gide

”
curvatures describing the distinction between the 3-volume calculated from a radius and
the actual 3-volume. They are described by the Ricci tensor.**** With an argument we
encountered already for the case of geodesic deviation, it turns out that the Ricci tensor
describes how the shape of a spherical cloud of freely falling particles is deformed on its
path.

– CS – to be expanded – CS –

In short, the Ricci tensor is the general-relativistic version of ∆φ, or better, of jφ.
The most global, but least detailed, definition of curvature is the one describing the
distinction between the 4-volume calculated from a measured radius and the actual 4-
volume. This is the average curvature at a space-time point and is represented by the so-
called Ricci scalar R, defined as

R = −2K = −
2
2
. (296)
r curvature
Copyright © Christoph Schiller November 1997–May 2006

It turns out that the Ricci scalar can be derived from the Ricci tensor by a so-called con-
traction, which is a precise averaging procedure. For tensors of rank two, contraction is
the same as taking the trace:
R = Rλ λ = дλµ Rλµ . (297)

** ‘Every street urchin in our mathematical Göttingen knows more about four-dimensional geometry than
Einstein. Nevertheless, it was Einstein who did the work, not the great mathematicians.’
*** ‘One has to follow one’s inclination, especially if it climbs upwards.’
**** Gregorio Ricci-Cubastro (b. 1853 Lugo , d. 1925 Bologna), Italian mathematician. He is the father of
absolute differential calculus, once also called ‘Ricci calculus’. Tullio Levi-Civita was his pupil.

Dvipsbugw
 all observers – heavier mathematics 423

The Ricci scalar describes the curvature averaged over space and time. In the image of
a falling spherical cloud, the Ricci scalar describes the volume change of the cloud. The
Ricci scalar always vanishes in vacuum. This result allows one, on the surface of the Earth,
Challenge 803 ny to relate the spatial curvature to the change of time with height.
Now comes an idea discovered by Einstein after two years of hard work. The import-
ant quantity for the description of curvature in nature is not the Ricci tensor R ab , but
a tensor built from it. This Einstein tensor G ab is defined mathematically (for vanishing
cosmological constant) as
Dvipsbugw
G ab = R ab − дab R .
1
(298)
2
It is not difficult to understand its meaning. The value G 00 is the sum of sectional
curvatures in the planes orthogonal to the 0 direction and thus the sum of all spatial
sectional curvatures:
G 00 = K(12) + K(23) + K(31) . (299)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Similarly, for each dimension i the diagonal element G i i is the sum (taking into consid-
eration the minus signs of the metric) of sectional curvatures in the planes orthogonal to
the i direction. For example, we have

G 11 = K(02) + K(03) − K(23) . (300)

The distinction between the Ricci tensor and the Einstein tensor thus lies the way in which
the sectional curvatures are combined: discs containing the coordinate in question in one
case, discs orthogonal to the coordinate in the other case. Both describe the curvature of
space-time equally well, and fixing one means fixing the other. (What are the trace and
Challenge 804 d the determinant of the Einstein tensor?)
The Einstein tensor is symmetric, which means that it has ten independent compon-
ents. Most importantly, its divergence vanishes; it therefore describes a conserved quant-
ity. This was the essential property which allowed Einstein to relate it to mass and energy
in mathematical language.

The description of momentum, mass and energy

Obviously, for a complete description of gravity, also motions of momentum and energy
also need to be quantified in such a way that any observer can talk to any other. We have
Copyright © Christoph Schiller November 1997–May 2006

seen that momentum and energy always appear together in relativistic descriptions; the
next step is thus to find out how their motions can be quantified for general observers.
First of all, the quantity describing energy, let us call it T, must be defined using the
energy–momentum vector p = mu = (γmc, γmv) of special relativity. Furthermore, T
does not describe a single particle, but the way energy–momentum is distributed over
space and time. As a consequence, it is most practical to use T to describe a density of
energy and momentum. T will thus be a field, and depend on time and space, a fact usually
indicated by the notation T = T(t, x).
Since the energy–momentum density T describes a density over space and time, it
defines, at every space-time point and for every infinitesimal surface dA around that

Dvipsbugw
 424 iii gravitation and rel ativity • 8. motion of light and vacuum

point, the flow of energy–momentum dp through that surface. In other words, T is

defined by the relation
dp = T dA . (301)

The surface is assumed to be characterized by its normal vector dA. Since the energy–
momentum density is a proportionality factor between two vectors, T is a tensor.
Of course, we are talking about 4-flows and 4-surfaces here. Therefore the energy–
momentum density tensor can be split in the following way:
Dvipsbugw
’ w S1 S2 S3 “ ’ energy energy flow or “
– — – —
T =– —=– —
S1 t 11 t 12 t 13 density momentum density
– — – —
(302)
S2 t 21 t 22 t 23 energy flow or momentum
” S3 t 31 t 32 t 33 • ” momentum density flow density •

where w = T00 is a 3-scalar, S a 3-vector and t a 3-tensor. The total quantity T is called the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

energy–momentum (density) tensor. It has two essential properties: it is symmetric and
its divergence vanishes.
The vanishing divergence of the tensor T, often written as

∂ a T ab = 0 or abbreviated T ab , a = 0 , (303)

expresses the fact that the tensor describes a conserved quantity. In every volume, en-
ergy can change only via flow through its boundary surface. Can you confirm that the
description of energy–momentum with this tensor satisfies the requirement that any two
observers, differing in position, orientation, speed and acceleration, can communicate
Challenge 805 ny their results to each other?
The energy–momentum density tensor gives a full description of the distribution of
energy, momentum and mass over space and time. As an example, let us determine the
energy–momentum density for a moving liquid. For a liquid of density ρ, a pressure p
and a 4-velocity u, we have

T ab = (ρ 0 + p)u a u b − pд ab (304)

where ρ 0 is the density measured in the comoving frame, the so-called proper density.*
Obviously, ρ, ρ 0 and p depend on space and time.
Copyright © Christoph Schiller November 1997–May 2006

Of course, for a particular material fluid, we need to know how pressure p and density
ρ are related. A full material characterization thus requires the knowledge of the relation

p = p(ρ) . (306)

* In the comoving frame we thus have

’ ρ0 c “
2
0 0 0
– 0 —
=– — .
p 0 0
– 0 —
ab
T (305)
0 p 0
” 0 0 0 p •

Dvipsbugw
 all observers – heavier mathematics 425

This relation is a material property and thus cannot be determined from relativity. It has
to be derived from the constituents of matter or radiation and their interactions. The
simplest possible case is dust, i.e. matter made of point particles** with no interactions at
all. Its energy–momentum tensor is given by

T ab = ρ 0 u a u b . (307)

Challenge 806 ny Can you explain the difference from the liquid case?
Dvipsbugw
The divergence of the energy–momentum tensor vanishes for all times and positions,
Challenge 807 ny as you may want to check. This property is the same as for the Einstein tensor presented
above. But before we elaborate on this issue, a short remark. We did not take into account
gravitational energy. It turns out that gravitational energy cannot be defined in general.
Gravity is not an interaction and does not have an associated energy.***

Hilbert’s action – how things fall?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

When Einstein discussed his work with David Hilbert, Hilbert found a way to do in a
few weeks what had taken years for Einstein. Hilbert understood that general relativity
in empty space could be described by an action integral, like all other physical systems.
Thus Hilbert set out to find the measure of change, as this is what an action describes,
for motion due to gravity. Obviously, the measure must be observer-invariant; in particu-
lar, it must be invariant under all possible changes of viewpoints.
Motion due to gravity is determined by curvature. Any curvature measure independ-
ent of the observer must be a combination of the Ricci scalar R and the cosmological
constant Λ. It thus makes sense to expect that the change of space-time is described by
an action S given by
c4
S= (R + 2Λ) dV .
∫ (308)
16πG
The volume element dV must be specified to use this expression in calculations. The cos-
mological constant Λ (added some years after Hilbert’s work) appears as a mathematical
possibility to describe the most general action that is diffeomorphism-invariant. We will
see below that its value in nature, though small, seems to be different from zero.
The Hilbert action of a chunk of space-time is thus the integral of the Ricci scalar plus
twice the cosmological constant over that chunk. The principle of least action states that
space-time moves in such a way that this integral changes as little as possible.
Copyright © Christoph Schiller November 1997–May 2006

– CS – to be finished – CS –

** Even though general relativity expressly forbids the existence of point particles, the approximation is
useful in cases when the particle distances are large compared to their own size.
*** In certain special circumstances, such as weak fields, slow motion, or an asymptotically flat space-time,
we can define the integral of the G 00 component of the Einstein tensor as negative gravitational energy. Grav-
itational energy is thus only defined approximately, and only for our everyday environment. Nevertheless,
Page 471 this approximation leads to the famous speculation that the total energy of the universe is zero. Do you
Challenge 808 ny agree?

Dvipsbugw
 426 iii gravitation and rel ativity • 8. motion of light and vacuum

In summary, the question ‘how do things move?’ is answered by general relativity in

the same way as by special relativity: things follow the path of maximal ageing.
Challenge 809 ny Can you show that the Hilbert action follows from the maximum force?

The symmetries of general relativity

The main symmetry of the Lagrangian of general relativity is called diffeomorphism in-
variance.
Dvipsbugw
– CS – to be written – CS –

The field equations for empty space-time also show scale symmetry. This is the in-
variance of the equations after multiplication of all coordinates by a common numerical
factor. In 1993, Torre and Anderson showed that diffeomorphism symmetry and trivial
Ref. 378 scale symmetry are the only symmetries of the vacuum field equations.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Apart from diffeomorphism symmetry, full general relativity, including mass–energy,
has an additional symmetry which is not yet fully elucidated. This symmetry connects
the various possible initial conditions of the field equations; the symmetry is extremely
Ref. 379 complex and is still a topic of research. These fascinating investigations should give new
insights into the classical description of the big bang.

Einstein’s field equations

[Einstein’s general theory of relativity] cloaked

“ the ghastly appearance of atheism.

A witch hunter from Boston, around 1935
Do you believe in god? Prepaid reply 50 words.
Subsequent telegram by another to his hero
Albert Einstein
I believe in Spinoza’s god, who reveals himself in
the orderly harmony of what exists, not in a god
who concerns himself with fates and actions of
human beings.
Albert Einstein’s answer

Page 353
Einstein’s famous field equations were the basis of many religious worries. They contain
the full description of general relativity. As explained above, they follow from the max-
imum force – or equivalently, from Hilbert’s action – and are given by
” Copyright © Christoph Schiller November 1997–May 2006

G ab = −κ Tab
or

R ab − дab R − Λдab = −κ T ab .
1
(309)
2
The constant κ, called the gravitational coupling constant, has been measured to be

κ= = 2.1 ë 10−43 ~N
8πG
(310)
c4

Dvipsbugw
 all observers – heavier mathematics 427

and its small value – 2π divided by the maximum force c 4 ~4G – reflects the weakness of
gravity in everyday life, or better, the difficulty of bending space-time. The constant Λ,
the so-called cosmological constant, corresponds to a vacuum energy volume density, or
Page 460 pressure Λ~κ. Its low value is quite hard to measure. The currently favoured value is

Λ  10−52 ~m2 or Λ~κ  0.5 nJ~m3 = 0.5 nPa . (311)

Ref. 380 Current measurements and simulations suggest that this parameter, even though it is
numerically near to the square of the present radius of the universe, is a constant of nature Dvipsbugw
that does not vary with time.
In summary, the field equations state that the curvature at a point is equal to the flow of
energy–momentum through that point, taking into account the vacuum energy density.
In other words, energy–momentum tells space-time how to curve.*
The field equations of general relativity can be simplified for the case in which speeds
are small. In that case T00 = ρc 2 and all other components of T vanish. Using the defini-
tion of the constant κ and setting φ = (c 2 ~2)h 00 in дab = η ab + h ab , we find

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 810 ny

d2 x
∇2 φ = 4πρ and = −∇φ (312)
dt 2
which we know well, since it can be restated as follows: a body of mass m near a body of
mass M is accelerated by
a=G 2,
M
(313)
r
a value which is independent of the mass m of the falling body. And indeed, as noted
already by Galileo, all bodies fall with the same acceleration, independently of their size,

* Einstein arrived at his field equations using a number of intellectual guidelines that are called principles in
the literature. Today, many of them are not seen as central any more. Nevertheless, we give a short overview.
- Principle of general relativity: all observers are equivalent; this principle, even though often stated, is
probably empty of any physical content.
- Principle of general covariance: the equations of physics must be stated in tensorial form; even though
Ref. 381 it is known today that all equations can be written with tensors, even universal gravity, in many cases they
require unphysical ‘absolute’ elements, i.e. quantities which affect others but are not affected themselves.
Page 676 This unphysical idea is in contrast with the idea of interaction, as explained above.
Copyright © Christoph Schiller November 1997–May 2006

- Principle of minimal coupling: the field equations of gravity are found from those of special relativity
by taking the simplest possible generalization. Of course, now that the equations are known and tested
experimentally, this principle is only of historical interest.
- Equivalence principle: acceleration is locally indistinguishable from gravitation; we used it to argue that
space-time is semi-Riemannian, and that gravity is its curvature.
- Mach’s principle: inertia is due to the interaction with the rest of the universe; this principle is correct,
even though it is often maintained that it is not fulfilled in general relativity. In any case, it is not the essence
Page 474 of general relativity.
- Identity of gravitational and inertial mass: this is included in the definition of mass from the outset, but
restated ad nauseam in general relativity texts; it is implicitly used in the definition of the Riemann tensor.
- Correspondence principle: a new, more general theory, such as general relativity, must reduce to previous
theories, in this case universal gravity or special relativity, when restricted to the domains in which those
are valid.

Dvipsbugw
 428 iii gravitation and rel ativity • 8. motion of light and vacuum

their mass, their colour, etc. In general relativity also, gravitation is completely demo-
cratic.* The independence of free fall from the mass of the falling body follows from the
description of space-time as a bent mattress. Objects moving on a mattress also move in
the same way, independently of the mass value.
To get a feeling for the complete field equations, we will take a short walk through
their main properties. First of all, all motion due to space-time curvature is reversible,
Challenge 811 e differentiable and thus deterministic. Note that only the complete motion, of space-time
and matter and energy, has these properties. For particle motion only, motion is in fact
irreversible, since some gravitational radiation is usually emitted. Dvipsbugw
By contracting the field equations we find, for vanishing cosmological constant, the
following expression for the Ricci scalar:

R = −κT . (318)

This result also implies the relation between the excess radius and the mass inside a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 812 ny sphere.
The field equations are nonlinear in the metric д, meaning that sums of solutions usu-
ally are not solutions. That makes the search for solutions rather difficult. For a complete
solution of the field equations, initial and boundary conditions should be specified. The
Ref. 382 ways to do this form a specialized part of mathematical physics; it is not explored here.
Albert Einstein used to say that general relativity only provides the understanding of
one side of the field equations (309), but not of the other. Can you see which side he
Challenge 813 ny meant?
What can we do of interest with these equations? In fact, to be honest, not much that
we have not done already. Very few processes require the use of the full equations. Many
textbooks on relativity even stop after writing them down! However, studying them is
worthwhile. For example, one can show that the Schwarzschild solution is the only spher-
ically symmetric solution. Similarly, in 1923, Birkhoff showed that every rotationally sym-
metric vacuum solution is static. This is the case even if masses themselves move, as for
example during the collapse of a star.
Maybe the most beautiful applications of the field equations are the various films made
of relativistic processes. The worldwide web hosts several of these; they allow one to see

* Here is another way to show that general relativity fits with universal gravity. From the definition of the
Riemann tensor we know that relative acceleration b a and speed of nearby particles are related by
Copyright © Christoph Schiller November 1997–May 2006

c d
∇ e b a = R ced a v v . (314)

From the symmetries of R we know there is a φ such that b a = −∇ a φ. That means that

∇ e b a = ∇ e ∇ a φ = R ced
a
vc vd (315)

which implies that

∆φ = ∇ a ∇ a φ = R cad
a
v c v d = R cd v c v d = κ(Tcd v c v d − T~2) (316)

Introducing Tab = ρv a v b we get

∆φ = 4πGρ (317)
as we wanted to show.

Dvipsbugw
 all observers – heavier mathematics 429

what happens when two black holes collide, what happens when an observer falls into
a black hole, etc. To generate these films, the field equations usually need to be solved
directly, without approximations.*

– CS – more to be added – CS –

Another area of application concerns gravitational waves. The full field equations show
that waves are not harmonic, but nonlinear. Sine waves exist only approximately, for small
amplitudes. Even more interestingly, if two waves collide, in many cases singularities are Dvipsbugw
predicted to appear. This whole theme is still a research topic and might provide new
insights for the quantization of general relativity in the coming years.
We end this section with a side note. Usually, the field equations are read in one sense
only, as stating that energy–momentum produces curvature. One can also read them in
the other way, calculating the energy–momentum needed to produces a given curvature.
When one does this, one discovers that not all curved space-times are possible, as some

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

would lead to negative energy (or mass) densities. Such solutions would contradict the
mentioned limit on length-to-mass ratios for physical systems.

More on the force limit

When the cosmological constant is taken into the picture, the maximum force principle
requires a second look. In the case of a non-vanishing cosmological constant, the force
Ref. 383 limit makes sense only if the constant Λ is positive; this is the case for the currently meas-
Ref. 323, Ref. 324 ured value, which is Λ  10−52 ~m2 . Indeed, the radius–mass relation of black holes

2GM = Rc 2 (1 − R )
Λ 2
(319)
3
implies that a radius-independent maximum force is valid only for positive or zero cos-
mological constant. For a negative cosmological constant the force limit would only be
valid for infinitely small black holes. In the following, we take a pragmatic approach and
note that a maximum force limit can be seen to imply a vanishing or positive cosmolo-
gical constant. Obviously, the force limit does not specify the value of the constant; to
achieve this, a second principle needs to be added. A straightforward formulation, using
Page 373 the additional principle of a minimum force in nature, was proposed above.
One might ask also whether rotating or charged black holes change the argument that
Copyright © Christoph Schiller November 1997–May 2006

leads from maximum force to the derivation of general relativity. However, the deriva-
tion using the Raychaudhuri equation does not change. In fact, the only change of the
argument appears with the inclusion of torsion, which changes the Raychaudhuri equa-
tion itself. As long as torsion plays no role, the derivation given above remains valid. The
inclusion of torsion is still an open research issue.
Another question is how maximum force relates to scalar–tensor theories of gravity,
such as the proposal by Brans and Dicke or its generalizations. If a particular scalar-tensor
theory obeys the general horizon equation (209) then it must also imply a maximum force.
The general horizon equation must be obeyed both for static and for dynamic horizons.

* See for example the http://www.photon.at/~werner/black-earth website.

Dvipsbugw
 430 iii gravitation and rel ativity • 8. motion of light and vacuum

If that were the case, the specific scalar–tensor theory would be equivalent to general
relativity, as it would allow one, using the argument of Jacobson, to deduce the usual
field equations. This case can appear if the scalar field behaves like matter, i.e., if it has
mass–energy like matter and curves space-time like matter. On the other hand, if in the
particular scalar–tensor theory the general horizon equation (209) is not obeyed for all
moving horizons – which is the general case, as scalar–tensor theories have more defin-
ing constants than general relativity – then the maximum force does not appear and the
theory is not equivalent to general relativity. This connection also shows that an experi-
mental test of the horizon equation for static horizons only is not sufficient to confirm Dvipsbugw
general relativity; such a test rules out only some, but not all, scalar–tensor theories.

Deducing universal gravity

For the maximum force limit to be considered a basic physical principle, all properties
of gravitation, including the full theory of general relativity, must be deduced from it. To

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

make this argument easier to follow, we will split it into several steps. First of all, we show
that the force limit implies that in everyday life the inverse square law of universal gravity
holds. Then we show that it implies the main ideas of general relativity. Finally, we show
that the full theory of general relativity follows.
In other words, from this point onwards the force limit is assumed to be valid. We
explore its consequences and compare them with the known properties of nature. The
maximum force limit can also be visualized in another way. If the gravitational attraction
between a central body and a satellite were stronger than it is, black holes would be smaller
than they are; in that case the maximum force limit and the maximum speed could be
exceeded. If, on the other hand, gravitation were weaker than it is, a fast and accelerating
observer would not be able to determine that the two bodies interact. In summary, a
maximum force of c 4 ~4G implies universal gravity. There is no difference between stating
that all bodies attract through gravitation and stating that there is a maximum force with
the value c 4 ~4G.

Deducing linearized general relativity

The next logical step is to show that a maximum force also implies general relativity. The
naive approach is to repeat, step by step, the standard approach to general relativity.
Space-time curvature is a consequence of the fact that the speed of light is the max-
Ref. 323, Ref. 324 imum speed for all observers, even if they are located in a gravitational field. The gravita-
Copyright © Christoph Schiller November 1997–May 2006

tional red shift shows that in gravitational fields, clocks change their rate with height; that
change, together with the constancy of the speed of light, implies space-time curvature.
Gravity thus implies space-time curvature. The value of the curvature in the case of weak
gravitational fields is completely fixed by the inverse square law of gravity. Since universal
gravity follows from the maximum force, we deduce that maximum force implies space-
time curvature.
Apart from curvature, we must also check the other basic ideas of general relativity.
The principle of general relativity states that all observers are equivalent; since the max-
imum force principle applies to all observers, the principle of general relativity is con-
tained in it. The equivalence principle states that, locally, gravitation can be transformed
away by changing to a suitable observer. This is also the case for the maximum force

Dvipsbugw
 all observers – heavier mathematics 431

principle, which is claimed for all observers, thus also for observers that locally eliminate
gravitation. Mach’s principle, whose precise formulation varies, states that only relative
quantities should play a role in the description of nature. Since the maximum force is
a relative quantity – in particular, the relation of mass and curvature remains – Mach’s
principle is also satisfied.
Free bodies in flat space move with constant speed. By the equivalence principle, this
statement generalizes to the statement that freely falling bodies move along geodesics.
The maximum force principle keeps intact the statement that space-time tells matter how
to move. Dvipsbugw
The curvature of space-time for weak gravitational fields is fixed by the inverse square
law of gravity. Space curvature is thus present in the right amount around each mass.
Ref. 384 As Richard Feynman explains, by extending this result to all possible observers, we can
deduce all low-curvature effects of gravitation. In particular, this implies the existence of
linear (low-amplitude) gravitational waves and of the Thirring–Lense effect. Linearized
general relativity thus follows from the maximum force principle.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

How to calculate the shape of geodesics
The other half of general relativity states that bodies fall along geodesics. All orbits are
geodesics, thus curves with the longest proper time. It is thus useful to be able to calcu-
late these trajectories.* To start, one needs to know the shape of space-time, the notion
of ‘shape’ being generalized from its familiar two-dimensional meaning. For a being liv-
ing on the surface, it is usually described by the metric дab , which defines the distances
between neighbouring points through

ds 2 = dx a dx a = дab (x) dx a dx b . (320)

It is a famous exercise of calculus to show from this expression that a curve x a (s) de-
pending on a well behaved (affine) parameter s is a timelike or spacelike (metric) geodesic,
Challenge 814 ny i.e. the longest possible path between the two events,** only if

dx d 1 ∂дbc dx b dx c
(дad )=
d
, (321)
ds ds 2 ∂x a ds ds
as long as ds is different from zero along the path.*** All bodies in free fall follow such
Copyright © Christoph Schiller November 1997–May 2006

Page 388 geodesics. We showed above that the geodesic property implies that a stone thrown in the
air falls back, unless if it is thrown with a speed larger than the escape velocity. Expression

* This is a short section for the more curious; it can be skipped at first reading.
** We remember that in space in everyday life, geodesics are the shortest possible paths; however, in space-
time in general relativity, geodesics are the longest possible paths. In both cases, they are the ‘straightest’
possible paths.
*** This is often written as
d2 x a b
a dx dx
c
+ Γbc =0 (322)
ds 2 ds ds
where the condition
dx a dx b
дab =1 (323)
ds ds

Dvipsbugw
 432 iii gravitation and rel ativity • 8. motion of light and vacuum

(321) thus replaces both the expression d2 x~dt 2 = −∇φ valid for falling bodies and the
expression d2 x~dt 2 = 0 valid for freely floating bodies in special relativity.
The path does not depend on the mass or on the material of the body. Therefore an-
Ref. 385 timatter also falls along geodesics. In other words, antimatter and matter do not repel;
they also attract each other. Interestingly, even experiments performed with normal mat-
Challenge 815 ny ter can show this, if they are carefully evaluated. Can you find out how?
For completeness, we mention that light follows lightlike or null geodesics. In other
words, there is an affine parameter u such that the geodesics follow
Dvipsbugw
2 a b c
=0
d x dx dx
+ Γ a bc (325)
du 2 du du
with the different condition
dx a dx b
дab =0. (326)
du du

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Given all these definitions of various types of geodesics, what are the lines drawn in
Challenge 816 ny Figure 179 on page 383?

Mass in general relativity

The diffeomorphism-invariance of general relativity makes life quite interesting. We will
see that it allows us to say that we live on the inside of a hollow sphere, and that it does
not allow us to say where energy is actually located. If energy cannot be located, what
about mass? It soon becomes clear that mass, like energy, can be localized only if distant
space-time is known to be flat. It is then possible to define a localized mass value by
making precise an intuitive idea: the mass is measured by the time a probe takes to orbit
the unknown body.*
The intuitive mass definition requires flat space-time at infinity; it cannot be extended
Challenge 817 ny to other situations. In short, mass can only be localized if total mass can be defined. And

must be fulfilled, thus simply requiring that all the tangent vectors are unit vectors, and that ds x 0 all along
the path. The symbols Γ appearing above are given by

Γ a bc = ˜  = д ad (∂ b дd c + ∂ c дdb − ∂ d дbc ) ,
a 1
(324)
bc 2
Copyright © Christoph Schiller November 1997–May 2006

and are called Christoffel symbols of the second kind or simply the metric connection.
* This definition was formalized by Arnowitt, Deser and Misner, and since then has often been called the
ADM mass. The idea is to use the metric дi j and to take the integral

(дi j, i ν j − дi i , j ν j )dA
1
m=
16π ∫ SR
(327)

where S R is the coordinate sphere of radius R, ν is the unit vector normal to the sphere and dA is the area
element on the sphere. The limit exists if space-time is asymptotically flat and if the mass distribution is suffi-
Ref. 386 ciently concentrated. Mathematical physicists have also shown that for any manifold whose metric changes
at infinity as
дi j = (1 + f ~r + O(1~r ))δ i j
2
(328)
the total mass is given by M = 2 f .

Dvipsbugw
 all observers – heavier mathematics 433

total mass is defined only for asymptotically flat space-time. The only other notion of
mass that is precise in general relativity is the local mass density at a point. In contrast, it
is not well understood how to define the mass contained in a region larger than a point
but smaller than the entirety of space-time.
Now that we can go on talking about mass without (too much of) a bad conscience,
we turn to the equations of motion.

Is gravity an interaction?
Dvipsbugw
We tend to answer this question affirmatively, as in Galilean physics gravity was seen
as an influence on the motion of bodies. In fact, we described it by a potential, imply-
ing that gravity produces motion. But let us be careful. A force or an interaction is what
changes the motion of objects. However, we just saw that when two bodies attract each
other through gravitation, both always remain in free fall. For example, the Moon circles
the Earth because it continuously falls around it. Since any freely falling observer con-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tinuously remains at rest, the statement that gravity changes the motion of bodies is not
correct for all observers. Indeed, we will soon discover that in a sense to be discussed
shortly, the Moon and the Earth both follow ‘straight’ paths.
Is this correction of our idea of gravity only a question of words? Not at all. Since
gravity is not an interaction, it is not due to a field and there is no potential.
Page 676 Let us check this strange result in yet another way. The most fundamental definition
of ‘interaction’ is as the difference between the whole and the sum of its parts. In the case
of gravity, an observer in free fall could claim that nothing special is going on, independ-
ently of whether the other body is present or not, and could claim that gravity is not an
interaction.
However, that is going too far. An interaction transports energy between systems. We
have indeed seen that gravity can be said to transport energy only approximately. Grav-
itation is thus an interaction only approximately. But that is not a sufficient reason to
abandon this characterization. The concept of energy is not useful for gravity beyond the
domain of everyday life. For the general case, namely for a general observer, gravity is
thus fundamentally different from electricity or magnetism.
Another way to look at the issue is the following. Take a satellite orbiting Jupiter with
energy–momentum p = mu. If we calculate the energy–momentum change along its path
Challenge 818 ny s, we get

du a de a a du a
=m = m(e a u ) = me a ( + Γ a bd ub uc ) = 0
dp du
Copyright © Christoph Schiller November 1997–May 2006

+ (329)
ds ds ds ds ds
where e describes the unit vector along a coordinate axis. The energy–momentum change
Challenge 819 ny vanishes along any geodesic, as you might check. Therefore, the energy–momentum of
this motion is conserved. In other words, no force is acting on the satellite. One could
Ref. 387 reply that in equation (329) the second term alone is the gravitational force. But this
Challenge 820 ny term can be made to vanish along the entirety of any given world line. In short, nothing
changes between two bodies in free fall around each other: gravity could be said not to
Challenge 821 n be an interaction. The properties of energy confirm this argument.
Of course, the conclusion that gravity is not an interaction is somewhat academic, as

Dvipsbugw
 434 iii gravitation and rel ativity • 8. motion of light and vacuum

it contradicts our experience of daily life. But we will need it for the full understanding of
motion later on. The behaviour of radiation confirms the deduction. In vacuum, radiation
is always moving freely. In a sense, we can say that radiation always is in free fall. Strangely,
since we called free fall the same as rest, we should conclude that radiation always is at
rest. This is not wrong! We have already seen that light cannot be accelerated.* We even
saw that gravitational bending is not an acceleration, since light follows straight paths in
space-time in this case as well. Even though light seems to slow down near masses for
distant observers, it always moves at the speed of light locally. In short, even gravitation
doesn’t manage to move light. Dvipsbugw
There is another way to show that light is always at rest. A clock for an observer trying
to reach the speed of light goes slower and slower. For light, in a sense, time stops: or, if
you prefer, light does not move.

The essence of general relativity

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

If a maximum power or force appearing on horizons is the basis for general relativity, one
can ask whether physical systems other than space-time can also be described in this way.
For special relativity, we found that all its main effects – such as a limit speed, Lorentz
contraction or energy–mass equivalence – are also found for dislocations in solids. Do sys-
tems analogous to general relativity exist? Attempts to find such systems have only been
partially successful. Several equations and ideas of general relativity are applicable to de-
formations of solids, since general relativity describes the deformation of the space-time
Ref. 388 mattress. Kröner has studied this analogy in great detail. Other systems with horizons,
and thus with observables analogous to curvature, are found in certain liquids – where
vortices play the role of black holes – and in certain quantum fluids for the propagation of
Ref. 389 light. Exploring such systems has become a research topic in its own right. A full analogy
of general relativity in a macroscopic system was discovered only a few years ago. This
analogy will be presented in the third part of our adventure; we will need an additional
ingredient that is not visible at this point of our ascent.

Riemann gymnastics
Most books introduce curvature the hard way, namely historically,* using the Riemann
curvature tensor. This is a short summary, so that you can understand that old stuff when
you come across it.
We saw above that curvature is best described by a tensor. In 4 dimensions, this
Copyright © Christoph Schiller November 1997–May 2006

curvature tensor, usually called R, must be a quantity which allows us to calculate, among
other things, the area for any orientation of a 2-disc in space-time. Now, in 4 dimensions,
Challenge 822 e orientations of a disc are defined in terms of two 4-vectors; let us call them p and q. And
instead of a disc, we take the parallelogram spanned by p and q. There are several possible
definitions.

* Refraction, the slowdown of light inside matter, is not a counter-example. Strictly speaking, light inside
matter is constantly being absorbed and re-emitted. In between these processes, light still propagates with
the speed of light in vacuum. The whole process only looks like a slowdown in the macroscopic limit. The
same applies to diffraction and to reflection. A list of apparent ways to bend light can be found on page 567;
details of the quantum-mechanical processes at their basis can be found on page 728.
* This is a short section for the more curious; it can be skipped at first reading.

Dvipsbugw
 all observers – heavier mathematics 435

The Riemann-Christoffel curvature tensor R is then defined as a quantity which allows

us to calculate the curvature K(p, q) for the surface spanned by p and q, with area A,
through
R abcd p a q b p c q d
K(p, q) = 2 =
R pqpq
A (p, q) (дα δ дβγ − дαγ дβδ )p α q β pγ q δ
(330)

where, as usual, Latin indices a, b, c, d, etc. run from 0 to 3, as do Greek indices here,
and a summation is implied when an index name appears twice. Obviously R is a tensor, Dvipsbugw
of rank 4. This tensor thus describes only the intrinsic curvature of a space-time. In con-
trast, the metric д describes the complete shape of the surface, not only the curvature.
The curvature is thus the physical quantity of relevance locally, and physical descriptions
therefore use only the Riemann** tensor R or quantities derived from it.***
But we can forget the just-mentioned definition of curvature. There is a second, more
physical way to look at the Riemann tensor. We know that curvature means gravity. As
we said above, gravity means that when two nearby particles move freely with the same

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 823 e velocity and the same direction, the distance between them changes. In other words, the
local effect of gravity is relative acceleration of nearby particles.
It turns out that the tensor R describes precisely this relative acceleration, i.e. what we
called the tidal effects earlier on. Obviously, the relative acceleration b increases with the
Challenge 824 ny separation d and the square (why?) of the speed u of the two particles. Therefore we can
also define R as a (generalized) proportionality factor among these quantities:

b = R u u d or, more clearly, b a = R a bcd u b u c d d . (333)

The components of the Riemann curvature tensor have the dimensions of inverse square
length. Since it contains all information about intrinsic curvature, we conclude that if R
Challenge 825 ny vanishes in a region, space-time in that region is flat. This connection is easily deduced
from this second definition.*

** Bernhard Riemann (b. 1826 Breselenz, d. 1866 Selasca), important German mathematician.
*** We showed above that space-time is curved by noting changes in clock rates, in metre bar lengths and
in light propagation. Such experiments are the easiest way to determine the metric д. We know that space-
time is described by a 4-dimensional manifold M with a metric дab that locally, at each space-time point,
is a Minkowski metric. Such a manifold is called a Riemannian manifold. Only such a metric allows one to
define a local inertial system, i.e. a local Minkowski space-time at every space-time point. In particular, we
have
Copyright © Christoph Schiller November 1997–May 2006

дab = 1~д ab and дa b = д a b = δ ba . (331)

How are curvature and metric related? The solution to this question usually occupies a large number of
pages in relativity books; just for information, the relation is

∂Γ a bd ∂Γ a bc
R a bcd = − + Γ a ec Γ e bd − Γ a f d Γ f bc . (332)
∂x c ∂x d
The curvature tensor is built from the second derivatives of the metric. On the other hand, we can also
determine the metric if the curvature is known. An approximate relation is given below.
* This second definition is also called the definition through geodesic deviation. It is of course not evident
Ref. 390 that it coincides with the first. For an explicit proof, see the literature. There is also a third way to picture
the tensor R, a more mathematical one, namely the original way Riemann introduced it. If one parallel-
transports a vector w around a parallelogram formed by two vectors u and v, each of length ε, the vector w

Dvipsbugw
 436 iii gravitation and rel ativity • 8. motion of light and vacuum

A final way to define the tensor R is the following. For a free-falling observer, the metric
дab is given by the metric η ab from special relativity. In its neighbourhood, we have

дab = η ab + R acbd x c x d + O(x 3 )

1
3
= (∂ c ∂ d дab )x c x d + O(x 3 ) .
1
(335)
2
The curvature term thus describes the departure of the space-time metric from that of flat Dvipsbugw
space-time. The curvature tensor R is a large beast; it has 44 = 256 components at each
point of space-time; however, its symmetry properties reduce them to 20 independent
numbers.** The actual number of importance in physical problems is still smaller, namely
only 10. These are the components of the Ricci tensor, which can be defined with the help
of the Riemann tensor by contraction, i.e. by setting

R bc = R a bac .

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(338)

Its components, like those of the Riemann tensor, are inverse square lengths. The values
a
of the tensor R bc , or those of R bcd , are independent of the sign convention used in the
Challenge 828 e Minkowski metric, in contrast to R abcd .
Challenge 829 ny Can you confirm the relation R abcd R abcd = 48m 2 ~r 6 for the Schwarzschild solution?

Curiosities and fun challenges about general relativity

There are new results in general relativity every year.

**
For a long time, people have speculated why the Pioneer 10 and 11 artificial satellites,
which are now over 70 astronomical units away from the Sun, are subject to a constant
deceleration of 8 ë 10−10 m~s2 (towards the Sun) since thy passed the orbit of Saturn. This
effect is called the Pioneer anomaly. The origin is not clear and still a subject of research.

is changed to w + δw. One then has

2
δw = −ε R u v w + higher-order terms . (334)

More can be learned about the geodesic deviation by studying the behaviour of the famous south-pointing
Copyright © Christoph Schiller November 1997–May 2006

Page 174 carriage. This device, common in China before the compass was discovered, only works if the world is flat.
Indeed, on a curved surface, after following a large closed path, it will show a different direction than at the
Challenge 826 ny start of the trip. Can you explain why?
** The free-fall definition shows that the Riemann tensor is symmetric in certain indices and antisymmetric
Challenge 827 ny in others:
R abcd = R cd ab , R abcd = −R b acd = −R abd c . (336)
These relations also imply that many components vanish. Of importance also is the relation

R abcd + R adbc + R acdb = 0 . (337)

Note that the order of the indices is not standardized in the literature. The list of invariants which can be
constructed from R is long. We mention that 21 ε abcd R cd e f R ab e f , namely the product ‡ R R of the Riemann
tensor with its dual, is the invariant characterizing the Thirring–Lense effect.

Dvipsbugw
 why can we see the stars? – motion in the universe 437

But several investigations have shown that the reason is not a deviation from the inverse
Ref. 391 square dependence of gravitation, as is sometimes proposed. In other words, the effect
Challenge 830 r must be electromagnetic. Finding it is one of the challenges of modern astrophysics.

9. why can we see the stars? – motion in the

universe
Zwei Dinge erfüllen das Gemüt mit immer
Dvipsbugw

“ neuer und zunehmender Bewunderung und

Ehrfurcht, je öfter und anhaltender sich das
Nachdenken damit beschäftigt: der bestirnte
Himmel über mir und das moralische Gesetz in
mir.*
Immanuel Kant (1724–1804)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

On clear nights, between two and five thousand stars are visible with the naked eye. Sev-
eral hundred of them have names. Indeed, in all parts of the world, the stars and the con-
stellations they form are seen as memories of ancient events, and stories are told about
them.* But the simple fact that we can see the stars is the basis for a story much more
fantastic than all myths. It touches almost all aspects of modern physics.

Which stars do we see?

Democritus says [about the Milky Way] that it is

“ a region of light emanating from numerous

stars small and near to each other, of which the
grouping produces the brightness of the whole.
Aetius, Opinions.

”
Ref. 393

The stars we see on a clear night are mainly the brightest of our nearest neighbours in
the surrounding region of the Milky Way. They lie at distances between four and a few
thousand light years from us. Roughly speaking, in our environment there is a star about
every 400 cubic light years.
Almost all visible stars are from our own galaxy. The only extragalactic object con-
stantly visible to the naked eye in the northern hemisphere is the so-called Andromeda
nebula, shown enlarged in Figure 196. It is a whole galaxy like our own, as Immanuel Kant
had already conjectured in 1755. Several extragalactic objects are visible with the naked
Copyright © Christoph Schiller November 1997–May 2006

eye in the southern hemisphere: the Tarantula nebula, as well as the large and the small
Magellanic clouds. The Magellanic clouds are neighbour galaxies to our own. Other, tem-
porary exceptions are the rare novae, exploding stars which can be seen if they appear
in nearby galaxies, or the still rarer supernovae, which can often be seen even in faraway
galaxies.

* ‘Two things fill the mind with ever new and increasing admiration and awe, the more often and persistently
Ref. 392 thought considers them: the starred sky above me and the moral law inside me.’
* About the myths around the stars and the constellations, see e.g. the text by G. Fasching, Sternbilder und
ihre Mythen, Springer Verlag, 1993. On the internet there are also the beautiful http://www.astro.wisc.edu/
~dolan/constellations/constellations.html and http://www.astro.uiuc.edu/~kaler/sow/sow.html websites.

Dvipsbugw
 438 iii gravitation and rel ativity • 9. why can we see the stars?

Dvipsbugw

F I G U R E 196 The Andromeda nebula M31, our

neighbour galaxy (and the 31st member of the
Messier object listing) (NASA)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 197 How our galaxy looks in the infrared (NASA)

In fact, the visible stars are special in other respects also. For example, telescopes show
that about half of them are in fact double: they consist of two stars circling around each
other, as in the case of Sirius. Measuring the orbits they follow around each other allows
Challenge 831 ny one to determine their masses. Can you explain how?
Is the universe different from our Milky Way? Yes, it is. There are several arguments to
demonstrate this. First of all, our galaxy – th word galaxy is just the original Greek term
for ‘Milky Way’ – is flattened, because of its rotation. If the galaxy rotates, there must be
Copyright © Christoph Schiller November 1997–May 2006

other masses which determine the background with respect to which this rotation takes
place. In fact, there is a huge number of other galaxies – about 1011 – in the universe, a
discovery dating only from the twentieth century.
Why did our understanding of the place of our galaxy in the universe happen so late?
Well, people had the same difficulty as they had when trying to determine the shape
of the Earth. They had to understand that the galaxy is not only a milky strip seen on
clear nights, but an actual physical system, made of about 1011 stars gravitating around
each other.* Like the Earth, the galaxy was found to have a three-dimensional shape; it is

* The Milky Way, or galaxy in Greek, was said to have originated when Zeus, the main Greek god, tried to let
his son Heracles feed at Hera’s breast in order to make him immortal; the young Heracles, in a sign showing

Dvipsbugw
 why can we see the stars? – Motion in the universe 439

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 198 The elliptical galaxy NGC 205 (the 205th
member of the New Galactic Catalogue) (NASA)

shown in Figure 197. Our galaxy is a flat and circular structure, with a diameter of 100 000
light years; in the centre, it has a spherical bulge. It rotates about once every 200 to 250
Challenge 832 ny million years. (Can you guess how this is measured?) The rotation is quite slow: since the
Sun was formed, it has made only about 20 to 25 full turns around the centre.
It is even possible to measure the mass of our galaxy. The trick is to use a binary pulsar
on its outskirts. If it is observed for many years, one can deduce its acceleration around
the galactic centre, as the pulsar reacts with a frequency shift which can be measured
on Earth. Many decades of observation are needed and many spurious effects have to
Ref. 394 be eliminated. Nevertheless, such measurements are ongoing. Present estimates put the
mass of our galaxy at 10411 kg.

What do we see at night?

Copyright © Christoph Schiller November 1997–May 2006

Astrophysics leads to a strange conclusion about matter, quite different from how we are
used to thinking in classical physics: the matter observed in the sky is found in clouds.
Clouds are systems in which the matter density diminishes with the distance from the
centre, with no definite border and with no definite size. Most astrophysical objects are
best described as clouds.
The Earth is also a cloud, if we take its atmosphere, its magnetosphere and the dust
ring around it as part of it. The Sun is a cloud. It is a gas ball to start with, but is even
more a cloud if we take into consideration its protuberances, its heliosphere, the solar
wind it generates and its magnetosphere. The solar system is a cloud if we consider its
his future strength, sucked so forcefully that the milk splashed all over the sky.

Dvipsbugw
 440 iii gravitation and rel ativity • 9. why can we see the stars?

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 199 The colliding galaxies M51 and M110 (NASA)

F I G U R E 200 The X-rays in the night sky, between 1 and 30 MeV (NASA)
Copyright © Christoph Schiller November 1997–May 2006

comet cloud, its asteroid belt and its local interstellar gas cloud. The galaxy is a cloud if
we remember its matter distribution and the cloud of cosmic radiation it is surrounded
by. In fact, even people can be seen as clouds, as every person is surrounded by gases,
little dust particles from skin, vapour, etc.
Ref. 395 In the universe, almost all clouds are plasma clouds. A plasma is an ionized gas, such
as fire, lightning, the inside of neon tubes, or the Sun. At least 99.9 % of all matter in the
universe is in the form of plasma clouds. Only a very small percentage exists in solid or
liquid form, such as toasters, subways or their users.
Clouds in the universe have certain common properties. First, clouds seen in the uni-
verse, when undisturbed by collisions or other interactions from neighbouring objects,

Dvipsbugw
 why can we see the stars? – Motion in the universe 441

Dvipsbugw

F I G U R E 201 Rotating clouds emitting jets along their axis; top row: a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

composite image (visible and infrared) of the galaxy 0313-192, the galaxy
3C296, and the Vela pulsar; bottom row: the star in formation HH30, the
star in formation DG Tauri B, and a black hole jet from the galaxy M87
(NASA)

are rotating. Most clouds are therefore flattened and are in shape of discs. Secondly, in
many rotating clouds, matter is falling towards the centre: most clouds are accretion discs.
Finally, undisturbed accretion discs usually emit something along the rotation axis: they
possess jets. This basic cloud structure has been observed for young stars, for pulsars, for
galaxies, for quasars and for many other systems. Figure 201 gives some examples. (Does
the Sun have a jet? Does the Milky Way have a jet? So far, none has been detected – there
Challenge 833 r is still room for discovery.)
In summary, at night we see mostly rotating, flattened plasma clouds emitting jets
along their axes. A large part of astronomy and astrophysics collects information about
Ref. 396 them. An overview about the observations is given in Table 37.*

TA B L E 37 Some observations about the universe

A sp e c t Main Va l u e
properties
Copyright © Christoph Schiller November 1997–May 2006

Phenomena
galaxy formation observed by Hubble several times
trigger event unknown
galactic collisions momentum 1045 to 1047 kg m~s
star formation cloud collapse form stars between 0.04 and 200 solar
masses

* An overview of optical observations is given by the Sloan Digital Sky Survey at http://skyserver.sdss.org.
More details about the universe can be found in the beautiful text by W.J. Kaufmann & R.A. Fredman,
Universe, fifth edition, W.H. Freeman & Co., 1999. The most recent discoveries are best followed on the
http://sci.esa.int and http://hubble.nasa.gov websites.

Dvipsbugw
442 iii gravitation and rel ativity • 9. why can we see the stars?

A sp e c t Main Va l u e
properties
frequency between 0 and 1000 solar masses per
year per galaxy; around 1 solar mass in
the Milky Way
novae new luminous stars, L < 1031 W
ejecting bubble R  t ë c~100
supernovae new bright stars, L < 1036 W Dvipsbugw
rate 1 to 5 per galaxy per 1000 a
hypernovae optical bursts L A 1037 W
gamma-ray bursts luminosity L up to 1045 W, about one per cent of
the whole visible universe’s luminosity
energy c. 1046 J
duration c. 0.015 to 1000 s

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

observed number c. 2 per day
3
radio sources radio emission 10 3 to 1038 W
2
X-ray sources X-ray emission 10 3 to 1034 W
cosmic rays energy from 1 eV to 1022 eV
gravitational lensing light bending angles down to 10−4 ′′
comets recurrence, evaporation typ. period 50 a, typ. visibility lifetime
2 ka, typ. lifetime 100 ka
meteorites age up to 4.57 ë 109 a
Observed components
intergalactic space mass density c. 10−26 kg~m3
quasars red-shift up to z = 6
luminosity L = 1040 W, about the same as one
galaxy
galaxy superclusters number of galaxies c. 108 inside our horizon
our own local supercluster number of galaxies about 4000
galaxy groups size 100 Zm
number of galaxies between a dozen and 1000
our local group number of galaxies 30
Copyright © Christoph Schiller November 1997–May 2006

galaxies size 0.5 to 2 Zm

number c. 1011 inside horizon
containing 10 to 400 globular clusters
containing typically 1011 stars each
containing typically one supermassive and several
intermediate-mass black holes
our galaxy diameter 1.0(0.1) Zm
mass 1042 kg or 5 ë 1011 solar masses Ref. 397
containing 100 globular clusters each with 1
million stars

Dvipsbugw
why can we see the stars? – Motion in the universe 443

A sp e c t Main Va l u e
properties
speed 600 km~s towards Hydra-Centaurus
globular clusters (e.g. M15) containing thousands of stars, one
intermediate-mass black hole
age up to 12 Ga (oldest known objects)
nebulae, clouds composition dust, oxygen, hydrogen
our local interstellar cloud size 20 light years Dvipsbugw
composition atomic hydrogen at 7500 K
star systems types orbiting double stars, over 70 stars
orbited by brown dwarfs, several
planetary systems
our solar system size 2 light years (Oort cloud)
our solar system speed 368 km~s from Aquarius towards Leo

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

stars mass up to 130 solar masses (except when
stars fuse) Ref. 398
giants and supergiants large size up to 1 Tm
main sequence stars
brown dwarfs low mass below 0.072 solar masses
low temperature below 2800 K Ref. 399
L dwarfs low temperature 1200 to 2800 K
T dwarfs low temperature 900 to 1100 K
white dwarfs small radius r  5000 km
high temperature cools from 100 000 to 5000 K
neutron stars nuclear mass density ρ  1017 kg~m3
small size r  10 km
jet sources
central compact
objects
emitters of X-ray X-ray emission
bursts
pulsars periodic radio emission
mass up to around 25 solar masses
Copyright © Christoph Schiller November 1997–May 2006

magnetars high magnetic fields up to 1011 T and higher Ref. 400

(soft gamma repeaters, anomalous X-ray pulsars)
mass above 25 solar masses Ref. 401
black holes horizon radius r = 2GM~c 2 , observed mass range
from 1 to 100 million solar masses
General properties
cosmic horizon distance c. 1026 m = 100 Ym
expansion Hubble’s constant 71(4) km s−1 Mpc−1 or 2.3(2) ë 10−18 s−1
‘age’ of the universe 13.7(2) Ga

Dvipsbugw
444 iii gravitation and rel ativity • 9. why can we see the stars?

A sp e c t Main Va l u e
properties
vacuum energy density 0.5 nJ~m3 or Ω Λ = 0.73 for k = 0
no evidence for time-dependence
large-scale shape space curvature k  Ω K = 0Page 455
topology simple in our galactic environment,
unknown at large scales
dimensions number 3 for space, 1 for time, at low and Dvipsbugw
moderate energies
matter density 2 to 11 ë 10−27 kg~m3 or 1 to 6 hydrogen
atoms per cubic metre
Ω M = 0.25
baryons density Ω b = 0.04, one sixth of the previous
(included in Ω M )
Ω DM = 0.21 (included in Ω M ),

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

dark matter density
unknown
dark energy density Ω DM = 0.75, unknown
photons number density 4 to 5 ë 108 ~m3
= 1.7 to 2.1 ë 10−31 kg~m3
energy density Ω R = 4.6 ë 10−5
neutrinos energy density Ω ν unknown
average temperature photons 2.725(2) K
neutrinos not measured, predicted value is 2 K
perturbations photon anisotropy ∆T~T = 1 ë 10−5
density amplitude A = 0.8(1)
spectral index n = 0.97(3)
tensor-to-scalar ratio r < 0.53 with 95% confidence
ionization optical depth τ = 0.15(7)
decoupling z = 1100

But while we are speaking of what we see in the sky, we need to clarify a general issue.

What is the universe?

Copyright © Christoph Schiller November 1997–May 2006

I’m astounded by people who want to ‘know’

“ the universe when it’s hard enough to find your

way around Chinatown.
Woody Allen

The term universe implies turning. The universe is what turns around us at night. For a
physicist, at least three definitions are possible for the term ‘universe’:
”
— The (visible) universe is the totality of all observable mass and energy. This includes
everything inside the cosmological horizon. Since the horizon is moving away from
us, the amount of observable mass and energy is constantly increasing. The content of
the term ‘visible universe’ is thus not fixed in time. (What is the origin of this increase?

Dvipsbugw
 why can we see the stars? – Motion in the universe 445

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 202 The universe is full of galaxies – this photograph shows the Perseus
cluster (NASA)

We will come back to this issue later on.)

— The (believed) universe is the totality of all mass and energy, including any that is not
visible. Numerous books on general relativity state that there definitely exists matter
or energy beyond the observation boundaries. We will explain the origin of this belief
below.
— The (full) universe is the sum of matter and energy as well as space-time itself.
These definitions are often mixed up in physical and philosophical discussions. There
is no generally accepted consensus on the terms, so one has to be careful. In this text,
when we use the term ‘universe’, we imply the last definition only. We will discover re-
Copyright © Christoph Schiller November 1997–May 2006

peatedly that without clear distinction between the definitions the complete ascent of
Motion Mountain becomes impossible. (For example: Is the amount of matter and en-
Challenge 834 ny ergy in the full universe the same as in the visible universe?)
Note that the ‘size’ of the visible universe, or better, the distance to its horizon, is a
quantity which can be imagined. The value of 1026 m is not beyond imagination. If one
took all the iron from the Earth’s core and made it into a wire reaching to the edge of the
Challenge 835 ny visible universe, how thick would it be? The answer might surprise you. Also, the content
of the universe is clearly finite. There are about as many visible galaxies in the universe as
there are grains in a cubic metre of sand. To expand on the comparison, can you deduce
how much space you would need to contain all the flour you would get if every little speck
Challenge 836 ny represented one star?

Dvipsbugw
446 iii gravitation and rel ativity • 9. why can we see the stars?

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

F I G U R E 203 An atlas of our cosmic environment: illustrations at scales up to 12.5, 50, 250, 5 000, 50 000,
500 000, 5 million, 100 million, 1 000 million and 14 000 million light years (© Richard Powell,
http://www.anzwers.org/free/universe)

Dvipsbugw
 why can we see the stars? – Motion in the universe 447

The colour and the motion of the stars

᾽Η τοι µὲν πρώτιστα Ξάος γένετ΄ ... *

“ Hesiod, Theogony.

Obviously, the universe is full of motion. To get to know the universe a bit, it is useful to ”
measure the speed and position of as many objects in it as possible. In the twentieth cen-
tury, a large number of such observations were obtained from stars and galaxies. (Can you
Challenge 837 ny imagine how distance and velocity are determined?) This wealth of data can be summed Dvipsbugw
up in two points.
First of all, on large scales, i.e. averaged over about five hundred million light years,
the matter density in the universe is homogeneous and isotropic. Obviously, at smaller
scales inhomogeneities exist, such as galaxies or cheesecakes. Our galaxy for example is
Ref. 402 neither isotropic nor homogeneous. But at large scales the differences average out. This
large-scale homogeneity of matter distribution is often called the cosmological principle.
The second point about the universe is even more important. In the 1920s, independ-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 403 ently, Carl Wirtz, Knut Lundmark and Gustaf Stromberg showed that on the whole, galax-
ies move away from the Earth, and the more so, the more they were distant. There are a
few exceptions for nearby galaxies, such as the Andromeda nebula itself; but in general,
the speed of flight v of an object increases with distance d. In 1929, the US-American as-
tronomer Edwin Hubble** published the first measurement of the relation between speed
and distance. Despite his use of incorrect length scales he found a relation

v=Hd, (339)

where the proportionality constant H is today called the Hubble constant. A graph of
the relation is given in Figure 204. The Hubble constant is known today to have a value
around 71 km s−1 Mpc−1 . (Hubble’s own value was so far from this value that it is not cited
any more.) For example, a star at a distance of 2 Mpc* is moving away from Earth with a
speed between of around 142 km~s, and proportionally more for stars further away.
In fact, the discovery by Wirtz, Lundmark and Stromberg implies that every galaxy
Challenge 838 ny moves away from all the others. (Why?) In other words, the matter in the universe is ex-
panding. The scale of this expansion and the enormous dimensions involved are amazing.
The motion of all the thousand million galaxy groups in the sky is described by the single
equation (339)! Some deviations are observed for nearby galaxies, as mentioned above,
and for faraway galaxies, as we will see.
Copyright © Christoph Schiller November 1997–May 2006

The cosmological principle and the expansion taken together imply that the universe
cannot have existed before time when it was of vanishing size; the universe thus has a
finite age. Together with the evolution equations, as explained in more detail below, the

* ‘Verily, at first chaos came to be ...’. The Theogony, attributed to the probably mythical Hesiodos, was final-
ized around 700 bce. It can be read in English and Greek on the http://www.perseus.tufts.edu website. The
famous quotation here is from verse 117.
** Edwin Powell Hubble (1889–1953), important US-American astronomer. After being an athlete and tak-
ing a law degree, he returned to his childhood passion of the stars; he finally proved Immanuel Kant’s 1755
conjecture that the Andromeda nebula was a galaxy like our own. He thus showed that the Milky Way is
only a tiny part of the universe.
Page 1167 * A megaparsec or Mpc is a distance of 30.8 Zm.

Dvipsbugw
 448 iii gravitation and rel ativity • 9. why can we see the stars?

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 204 The relation between star distance and star velocity

Hubble constant points to an age value of around 13 700 million years. The expansion also
means that the universe has a horizon, i.e. a finite maximum distance for sources whose
signals can arrive on Earth. Signals from sources beyond the horizon cannot reach us.
Since the universe is expanding, in the past it has been much smaller and thus much
denser than it is now. It turns out that it has also been hotter. George Gamow* predicted
Ref. 404 in 1948 that since hot objects radiate light, the sky cannot be completely black at night,
but must be filled with black-body radiation emitted when it was ‘in heat’. That radiation,
called the background radiation, must have cooled down due to the expansion of the uni-
Challenge 839 ny verse. (Can you confirm this?) Despite various similar predictions by other authors, in
one of the most famous cases of missed scientific communication, the radiation was
found only much later, by two researchers completely unaware of all this work. A fam-
Ref. 405 ous paper in 1964 by Doroshkevich and Novikov had even stated that the antenna used
by the (unaware) later discoverers was the best device to search for the radiation! In any
case, only in 1965 did Arno Penzias and Robert Wilson discover the radiation. It was in
one of the most beautiful discoveries of science, for which both later received the Nobel
Ref. 406 Prize for physics. The radiation turns out to be described by the black-body radiation for
Copyright © Christoph Schiller November 1997–May 2006

a body with a temperature of 2.7 K; it follows the black-body dependence to a precision

of about 1 part in 104 .
But apart from expansion and cooling, the past fourteen thousand million years have
also produced a few other memorable events.

* George Gamow (b. 1904 Odessa, d. 1968 St. Boulder), Russian-American physicist; he explained alpha
decay as a tunnelling effect and predicted the microwave background. He wrote the first successful popular
science texts, such as 1, 2, 3, infinity and the Mr. Thompkins series, which were later imitated by many others.

Dvipsbugw
why can we see the stars? – Motion in the universe 449

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 205 The Hertzsprung–Russell diagram (© Richard Powell)

Do stars shine every night?

Don’t the stars shine beautifully? I am the only

“ person in the world who knows why they do.

Friedrich (Fritz) Houtermans (1903–1966)

Stars seem to be there for ever. In fact, every now and then a new star appears in the sky: ” Copyright © Christoph Schiller November 1997–May 2006

a nova. The name is Latin and means ‘new’. Especially bright novae are called supernovae.
Novae and similar phenomena remind us that stars usually live much longer than humans,
but that like us, they are born and die.
It turns out that one can plot all stars on the so-called Hertzsprung–Russell diagram.
This diagram, central to every book on astronomy, is shown in Figure 205. It is a beautiful
example of a standard method used by astrophysicists: collecting statistics over many
examples of a type of object, one can deduce the life cycle of the object, even though their
lifetime is much longer than that of a human. For example, it is possible, by clever use of
the diagram, to estimate the age of stellar clusters, and thus arrive at a minimum age of
the universe. The result is around thirteen thousand million years.
One conclusion is basic: since stars shine, they also die. Stars can only be seen if they

Dvipsbugw
 450 iii gravitation and rel ativity • 9. why can we see the stars?

are born but not yet dead at the moment of light emission. This leads to restrictions on
their visibility, especially for high red-shifts. Indeed, modern telescope can look at places
(and times) so far from now that they contained no stars at all. At those distances one
only observers quasars; these are not stars, but much more massive and bright systems.
Their precise structure are still being studied by astrophysicists.
On the other hand, since the stars shine, they were also formed somehow. The fascin-
ating details of their birth from dust clouds are a central part of astrophysics but will not
be explored here.
Yet we do not have the full answer to our question. Why do stars shine at all? Clearly, Dvipsbugw
they shine because they are hot. They are hot because of nuclear reactions in their interior.
Page 922 We will discuss these processes in more detail in the chapter on the nucleus.

A short history of the universe

Anima scintilla stellaris essentiae.*

“ Heraclitus of Ephesus (c. 540 to c. 480 bce )

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 407

The adventures the universe has experienced, or better, the adventures the matter and radi-
Ref. 408 ation inside it have experienced, are summarized in Table 38. The steps not yet discussed
will be studied in quantum theory. This history table has applications no theoretical phys-
icist would have imagined. The sequence is so beautiful and impressive that nowadays it
is used in certain psychotherapies to point out to people the story behind their existence
and to remind them of their own worth. Enjoy.

TA B L E 38 A short history of the universe

Ti m e Ti m e Event Te m pe r -
from from big at u r e
n o wa b a n gb
 14 ë 109 a  t Pl b
Time, space, matter and initial conditions 1032 K  TPl
indeterminate
13 ë 109 a c. 800 t Pl Distinction of space-time from matter and radiation, 1030 K
 10 s
−42
initial conditions determinate
10 s to
−35
Inflation & GUT epoch starts; strong and 5 ë 1026 K
10 s
−32
electroweak interactions diverge
10−12 s Antiquarks annihilate; electromagnetic and weak 1015 K
Copyright © Christoph Schiller November 1997–May 2006

interaction separate
2 ë 10 s
−6
Quarks get confined into hadrons; universe is a 1013 K
plasma
Positrons annihilate
0.3 s Universe becomes transparent for neutrinos 1010 K
a few seconds Nucleosynthesis: D, He, He and Li nuclei form; 109 K
4 3 7

radiation still dominates

2500 a Matter domination starts; density perturbations 75 000 K
magnify

* ‘The soul is a spark of the substance of the stars.’

Dvipsbugw
why can we see the stars? – Motion in the universe 451

Ti m e Ti m e Event Te m pe r -
from from big at u r e
n o wa b a n gb
z = 1100 380 000 a Recombination: during these latter stages of the big 3000 K
bang, H, He and Li atoms form, and the universe
becomes ‘transparent’ for light, as matter and
radiation decouple, i.e. as they acquire different
temperatures; the ‘night’ sky starts to get darker and
Dvipsbugw
darker
Sky is almost black except for black-body radiation Tγ =
To (1 + z)
z = 10 to 30 Galaxy formation
z=6 Oldest object seen so far
z=5 Galaxy clusters form
z=3 106 a First generation of stars (population II) is formed,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

starting hydrogen fusion; helium fusion produces
carbon, silicon and oxygen
2 ë 109 a First stars explode as supernovae c ; iron is produced
z=1 3 ë 109 a Second generation of stars (population I) appears,
and subsequent supernova explosions of the ageing
stars form the trace elements (Fe, Se, etc.) we are
made of and blow them into the galaxy
4.7 ë 109 a Primitive cloud, made from such explosion
remnants, collapses; Sun forms
4.6 ë 109 a Earth and other planet formation: Azoicum starts
4.3 ë 109 a Craters form on the planets
4.0 ë 109 a Moon forms from material ejected during the
collision of a large asteroid with the still-liquid Earth
4.0 ë 109 a Archean eon (Archaeozoicum) starts: bombardment
from space stops; Earth’s crust solidifies; oldest
minerals form; water condenses
3.5 ë 109 a Unicellular (microscopic) life appears; stromatolites
form
2.5 ë 109 a Proterozoic eon (‘age of first life’) starts: atmosphere
becomes rich in oxygen thanks to the activity of
Copyright © Christoph Schiller November 1997–May 2006

microorganisms Ref. 409

1 ë 109 a Macroscopic, multicellular life appears
800 ë 106 a Earth is completely covered with ice for the first time
(reason still unknown) Ref. 410
600 to Earth is completely covered with ice for the last time
540 ë 106 a
540(5) ë 106 a Paleozoic era (Palaeozoicum, ‘age of old life’) starts,
after a gigantic ice age: animals appear, oldest fossils
(with 540(5) start of Cambrian, 495(5) Ordovician,
440(5) Silurian, 417(5) Devonian, 354(5)
Carboniferous and 292(5) Permian periods)

Dvipsbugw
452 iii gravitation and rel ativity • 9. why can we see the stars?

Ti m e Ti m e Event Te m pe r -
from from big at u r e
n o wa b a n gb
450 ë 106 a Land plants appear
370 ë 106 a Wooden trees appear
250(5) ë 106 a Mesozoic era (Mesozoicum, ‘age of middle life’,
formerly called Secondary) starts: most insects and
other life forms are exterminated; mammals appear Dvipsbugw
(with 250(5) start of Triassic, 205(4) Jurassic and
142(3) Cretaceous periods)
150 ë 106 a Continent Pangaea splits into Laurasia and
Gondwana
The star cluster of the Pleiades forms
150 ë 106 a Birds appear

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

142(3) ë 106 a Golden time of dinosaurs (Cretaceous) starts
100 ë 106 a Start of formation of Alps, Andes and Rocky
Mountains
65.5 ë 106 a Cenozoic era (Caenozoicum, ‘age of new life’) starts:
dinosaurs become extinct after an asteroid hits the
Earth in the Yucatan; primates appear (with 65.5
start of Tertiary, consisting of Paleogene period with
Paleocene, 55.0 Eocene and 33.7 Oligocene epoch,
and of Neogene period, with 23.8 Miocene and 5.32
Pliocene epoch; then 1.81 Quaternary period with
Pleistocene (or Diluvium) and 0.01 Holocene (or
Alluvium) epoch)
50 ë 106 a Large mammals appear
7(1) ë 106 a Hominids appears
3 ë 106 a Supernova explodes, with following consequences:
more intense cosmic radiation, higher formation rate
of clouds, Earth cools down drastically, high
evolutionary pressure on the hominids and as a
result, Homo appears Ref. 411
500 000 a Formation of youngest stars in galaxy
500 000 a Homo sapiens appears
Copyright © Christoph Schiller November 1997–May 2006

100 000 a Beginning of last ice age

90 000 a Homo sapiens sapiens appears
11 800 a End of last ice age, start of Holocene
6 000 a First written texts
2 500 a Physics starts
500 a Use of coffee, pencil and modern physics starts
200 a Electricity use begins
100 a Einstein publishes
10 to 120 a You are a unicellular being

Dvipsbugw
 why can we see the stars? – Motion in the universe 453

Ti m e Ti m e Event Te m pe r -
from from big at u r e
n o wa b a n gb
present c. 14 ë 109 a You are reading this Tγ = 2.73 K,
Tν  1.6 K,
Tb  0 K
future You enjoy life; for details and reasons, see page 616
Dvipsbugw

a. The time coordinate used here is the one given by the coordinate system defined by the microwave back-
ground radiation, as explained on page 456. A year is abbreviated ‘a’ (Latin ‘annus’). Errors in the last digits
are given between parentheses.
b. This quantity is not exactly defined since the big bang is not a space-time event. More on this issue on
page 1034.
c. The history of the atoms on Earth shows that we are made from the leftovers of a supernova. We truly are

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

made of stardust.
The geological time scale is the one of the International Commission on Stratigraphy; the times are
measured through radioactive dating.

Despite its length and its interest, this table has its limitations. For example, what
happened elsewhere in the last few thousand million years? There is still a story to be
written of which next to nothing is known. For obvious reasons, investigations have been
rather Earth-centred.
Research in astrophysics is directed at discovering and understanding all phenomena
observed in the skies. Here we skip most of this fascinating topic, since as usual, we want
to focus on motion. Interestingly, general relativity allows us to explain many of the gen-
eral observations about motion in the universe.

The history of space-time

A number of rabbits run away from a central

Challenge 840 n
“ point in various directions, all with the same
speed. While running, one rabbit turns its head,
and makes a startling observation. What does it
see?

”
The data showing that the universe is sprinkled with stars all over lead to a simple conclu-
sion: the universe cannot be static. Gravity always changes the distances between bodies;
the only exceptions are circular orbits. Gravity also changes the average distances between
Copyright © Christoph Schiller November 1997–May 2006

bodies: gravity always tries to collapse clouds. The biggest cloud of all, the one formed by
all the matter in the universe, must therefore either be collapsing, or still be in expansion.
Ref. 412 The first to dare to draw this conclusion was Aleksander Friedmann.* In 1922 he de-

* Aleksander Aleksandrowitsch Friedmann (1888–1925), Russian physicist who predicted the expansion of
the universe. Following his early death from typhus, his work remained almost unknown until Georges A.
Lemaître (b. 1894 Charleroi, d. 1966 Leuven), Belgian priest and cosmologist, took it up and expanded
it in 1927, focusing, as his job required, on solutions with an initial singularity. Lemaître was one of the
propagators of the (erroneous!) idea that the big bang was an ‘event’ of ‘creation’ and convinced his whole
Page 463, page 464 organization of it. The Friedmann–Lemaître solutions are often erroneously called after two other physicists,
who studied them again much later, in 1935 and 1936, namely H.P. Robertson and A.G. Walker.

Dvipsbugw
 454 iii gravitation and rel ativity • 9. why can we see the stars?

duced the detailed evolution of the universe in the case of homogeneous, isotropic mass
distribution. His calculation is a classic example of simple but powerful reasoning. For a
universe which is homogeneous and isotropic for every point, the line element is given
Challenge 841 ny by
ds 2 = c 2 dt 2 − a 2 (t)(dx 2 + dy 2 + dz 2 ) (340)

and matter is described by a density ρ M and a pressure p M . Inserting all this into the field
equations, we get two equations
Dvipsbugw
2
‹  + 2 =
ȧ k 8πG Λ
ρM + and (341)
a a 3 3
ä = − (ρ M + 3p M ) a + a
4πG Λ
(342)
3 3
which imply

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ρ̇ M = −3 (ρ M + p M ) .
ȧ
(343)
a
At the present time t 0 , the pressure of matter is negligible. (In the following, the index 0
refers to the present time.) In this case, the expression ρ M a 3 is constant in time.
Equations (341) and (342) depend on only two constants of nature: the gravitational
constant G, related to the maximum force or power in nature, and the cosmological con-
stant Λ, describing the energy density of the vacuum, or, if one prefers, the smallest force
in nature.
Before we discuss the equations, first a few points of vocabulary. It is customary to
Challenge 842 ny relate all mass densities to the so-called critical mass density ρ c given by

3H 02
ρc =  (8  2) ë 10−27 kg~m3 (344)
8πG
corresponding to about 8, give or take 2, hydrogen atoms per cubic metre. On Earth, one
would call this value an extremely good vacuum. Such are the differences between every-
day life and the universe as a whole. In any case, the critical density characterizes a matter
distribution leading to an evolution of the universe just between never-ending expansion
and collapse. In fact, this density is the critical one, leading to a so-called marginal evolu-
Copyright © Christoph Schiller November 1997–May 2006

tion, only in the case of vanishing cosmological constant. Despite this restriction, the term
is now used for this expression in all other cases as well. One thus speaks of dimensionless
mass densities Ω M defined as
Ω M = ρ 0 ~ρ c . (345)

The cosmological constant can also be related to this critical density by setting

Λc 2 Λc 2
ΩΛ = = =
ρΛ
. (346)
ρ c 8πGρ c 3H 02

A third dimensionless parameter Ω K describes the curvature of space. It is defined in

Dvipsbugw
 why can we see the stars? – Motion in the universe 455

terms of the present-day radius of the universe R 0 and the curvature constant k =
˜1, −1, 0 as
−k
ΩK = 2 2 (347)
R0 H0

and its sign is opposite to the one of the curvature k; Ω K vanishes for vanishing curvature.
Note that a positively curved universe, when homogeneous and isotropic, is necessarily
closed and of finite volume. A flat or negatively curved universe with the same matter
distribution can be open, i.e. of infinite volume, but does not need to be so. It could be Dvipsbugw
simply or multiply connected. In these cases the topology is not completely fixed by the
curvature.
The present-time Hubble parameter is defined
by H 0 = ȧ 0 ~a 0 . From equation (341) we then get no big
Challenge 843 ny the central relation bang

ΩM + ΩΛ + ΩK = 1 .

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(348) 2 experimental
values
ion
In the past, when data were lacking, physicists x p ans
ed
e ion
were divided into two camps: the claustrophobics 1 rat ans
believing that Ω K A 0 and the agoraphobics believ-
e l e e x p
acc rated
ing that Ω K < 0. More details about the measured
ΩΛ ce le
de
ansion
eternal exp
values of these parameters will be given shortly. 0
collapse
The diagram of Figure 206 shows the most inter-

clo t
fla en
se
esting ranges of parameters together with the cor-
op

d
responding behaviours of the universe.
un
-1 too

ive
For the Hubble parameter, the most modern young

rse
measurements give a value of
0 1 2 3

H 0 = 71  4 km~sMpc = 2.3  2 ë 10−18 ~s (349) ΩM

F I G U R E 206 The ranges for the Ω

which corresponds to an age of the universe of parameters and their consequences
13.7  2 thousand million years. In other words,
the age deduced from the history of space-time agrees with the age, given above, deduced
from the history of stars.
To get a feeling of how the universe evolves, it is customary to use the so-called decel-
Copyright © Christoph Schiller November 1997–May 2006

eration parameter q 0 . It is defined as

q0 = − = ΩM − ΩΛ .
ä 0 1
2
(350)
a0 H0 2

The parameter q 0 is positive if the expansion is slowing down, and negative if the expan-
sion is accelerating. These possibilities are also shown in the diagram.
An even clearer way to picture the expansion of the universe for vanishing pressure is

Dvipsbugw
 456 iii gravitation and rel ativity • 9. why can we see the stars?

scale scale
factor factor
a a

ct

a(t) present
l Pl
Dvipsbugw
experimental
quantum uncertainties
effects
t Pl time t

present time t

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 207 The evolution of the universe’s scale a for different values of its mass density

to rewrite equation (341) using τ = t H 0 and x(τ) = a(t)~a(t 0 ), yielding

( ) + U(x) = Ω K
dx 2
dτ
with U(x) = −Ω Λ x − Ω Λ x 2 (351)

This looks like the evolution equation for the motion of a particle with mass 1, with total
energy Ω K in a potential U(x). The resulting evolutions are easily deduced.
For vanishing Ω Λ , the universe either expands for ever, or recollapses, depending on
the value of the mass–energy density.
For non-vanishing (positive) Ω Λ , the potential has exactly one maximum; if the
particle has enough energy to get over the maximum, it will accelerate continuously. That
is the situation the universe seems to be in today.
For a certain time range, the result is shown in Figure 207. There are two points to be
noted: first the set of possible curves is described by two parameters, not one. In addition,
Copyright © Christoph Schiller November 1997–May 2006

lines cannot be drawn down to zero size. There are two main reasons: we do not yet
understand the behaviour of matter at very high energy, and we do not understand the
behaviour of space-time at very high energy. We return to this important issue later on.
The main conclusion to be drawn from Friedmann’s work is that a homogeneous and
isotropic universe is not static: it either expands or contracts. In either case, it has a finite
age. This profound idea took many years to spread around the cosmology community;
even Einstein took a long time to get accustomed to it.
Note that due to its isotropic expansion, the universe has a preferred reference frame:
the frame defined by average matter. The time measured in that frame is the time listed
in Table 38 and is the one we assume when we talk about the age of the universe.
An overview of the possibilities for the long time evolution is given in Figure 208.

Dvipsbugw
 why can we see the stars? – Motion in the universe 457

Λ>0 Λ=0 Λ<0

scale factor scale factor scale factor

k = –1

time t time t time t

Dvipsbugw
scale factor scale factor scale factor

k=0

time t time t time t

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Λ < Λc Λ = Λc Λ > Λc
scale factor scale factor scale factor scale factor scale factor

k = +1

time t time t time t time t time t

F I G U R E 208 The long-term evolution of the universe’s scale factor a for various parameters

The evolution can have various outcomes. In the early twentieth century, people decided
among them by personal preference. Albert Einstein first preferred the solution k = 1 and
Λ = a −2 = 4πGρ M . It is the unstable solution found when x(τ) remains at the top of the
potential U(x).
In 1917, the Dutch physicist Willem de Sitter had found, much to Einstein’s personal
dismay, that an empty universe with ρ M = p M = 0 and k = 1 is also possible. This type of
Challenge 844 ny universe expands for large times. The de Sitter universe shows that in special cases, matter
is not needed for space-time to exist.
Lemaître had found expanding universes for positive mass, and his results were also
Copyright © Christoph Schiller November 1997–May 2006

contested by Einstein at first. When later the first measurements confirmed the calcu-
lations, the idea of a massive and expanding universe became popular. It became the
standard model in textbooks. However, in a sort of collective blindness that lasted from
around 1950 to 1990, almost everybody believed that Λ = 0.* Only towards the end of
the twentieth century did experimental progress allow one to make statements based on
evidence rather than beliefs or personal preferences, as we will find out shortly. But first
of all we will settle an old issue.

Challenge 845 ny * In this case, for Ω M E 1, the age of the universe follows t 0 D 2~(3H 0 ), where the limits correspond. For
vanishing mass density one has t 0 = 1~H o .

Dvipsbugw
 458 iii gravitation and rel ativity • 9. why can we see the stars?

Dvipsbugw

F I G U R E 209 The fluctuations of the cosmic background radiation (WMAP/NASA)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Why is the sky dark at night?
In der Nacht hat ein Mensch nur ein

“ Nachthemd an, und darunter kommt gleich der

Charakter.**
Rober Musil

First of all, the sky is not black at night. It has the same intrinsic colour as during the day,
as any long-exposure photograph shows. (See, for example, Figure 61.) But that colour,
like the colour of the sky during the day, is not due to the temperature of the sky, but
”
to scattered light from the stars. If we look for the real colour of the sky, we need to
look for its thermal radiation. Indeed, measurements show that even the empty sky is
not completely cold or black at night. It is filled with radiation of around 200 GHz; more
precise measurements show that the radiation corresponds to the thermal emission of a
body at 2.73 K. This background radiation is the thermal radiation left over from the big
bang.
Ref. 413 The universe is indeed colder than the stars. But why is this so? If the universe were
homogeneous on large scales and infinitely large, it would have an infinite number of
stars. Looking in any direction, we would see the surface of a star. The night sky would
Challenge 846 n be as bright as the surface of the Sun! Can you convince your grandmother about this?
In a deep forest, one sees a tree in every direction. Similarly, in a ‘deep’ universe, we
Copyright © Christoph Schiller November 1997–May 2006

would see a star in every direction. Now, the average star has a surface temperature of
about 6000 K. If we lived in a deep and old universe, we would effectively live inside an
oven with a temperature of around 6000 K, making it impossible to enjoy ice cream.
This paradox was most clearly formulated in 1823 by the astronomer Wilhelm Ol-
bers.*** As he extensively discussed the question, it is also called Olbers’ paradox. Today

** ‘At night, a person is dressed only with a nightgown, and directly under it there is the character.’ Robert
Musil (b. 1880 Klagenfurt, d. 1942 Geneva), German writer.
*** Heinrich Wilhelm Matthäus Olbers (b. 1758 Arbergen, d. 1840 Bremen), astronomer. He discovered
two planetoids, Pallas and Vesta, and five comets; he developed the method of calculating parabolic orbits
for comets which is still in use today. Olbers also actively supported the mathematician and astronomer
Page 101 Friedrich Wilhelm Bessel in his career choice. The paradox is named after Olbers, though others had made

Dvipsbugw
 why can we see the stars? – Motion in the universe 459

we know that even if all matter in the universe were converted into radiation, the universe
would still not be as bright as just calculated. In other words, the power and lifetime of
stars are much too low to produce the oven brightness just mentioned. So something is
Ref. 414 wrong.
In fact, two main effects can be invoked to avoid the contradiction. First, since the uni-
verse is finite in age, distant stars are shining for less time. We see them in a younger stage
or even during their formation, when they were darker. As a result, the share of bright-
ness of distant stars is smaller than that of nearby stars, so that the average temperature
of the sky is reduced.* Secondly, we could imagine that the radiation of distant stars is Dvipsbugw
red-shifted and that the volume the radiation must fill is increasing continuously, so that
the average temperature of the sky is also reduced.
Calculations are necessary to decide which effect is the greater one. This issue has been
Ref. 415 studied in great detail by Paul Wesson; he explains that the first effect is larger than the
second by a factor of about three. We may thus state correctly that the sky is dark at night
mostly because the universe has a finite age. We can add that the sky would be somewhat

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

brighter if the universe were not expanding.
Ref. 413 In addition, the darkness of the sky is possible only because the speed of light is finite.
Challenge 848 ny Can you confirm this?
Finally, the darkness of the sky also tells us that the universe has a large (but fi-
nite) age. Indeed, the 2.7 K background radiation is that cold, despite having been emit-
ted at 3000 K, because it is red-shifted, thanks to the Doppler effect. Under reasonable
Ref. 416 assumptions, the temperature T of this radiation changes with the scale factor R(t) of
the universe as
T
1
. (352)
R(t)

In a young universe, we would thus not be able to see the stars, even if they existed.
From the brightness of the sky at night, measured to be about 3 ë 10−13 times that of an
average star like the Sun, we can deduce something interesting: the density of stars in the
universe must be much smaller than in our galaxy. The density of stars in the galaxy can
be deduced by counting the stars we see at night. But the average star density in the galaxy
would lead to much higher values for the night brightness if it were constant throughout
Ref. 414 the universe. We can thus deduce that the galaxy is much smaller than the universe simply
by measuring the brightness of the night sky and by counting the stars in the sky! Can
Challenge 849 ny you make the explicit calculation?
Copyright © Christoph Schiller November 1997–May 2006

In summary, the sky is black at night because space-time and matter are of finite, but
Challenge 850 ny old age. As a side issue, here is a quiz: is there an Olbers’ paradox also for gravitation?

similar points before, such as the Swiss astronomer Jean Philippe Loÿs de Cheseaux in 1744 and Johannes
Kepler in 1610.
* Can you explain that the sky is not black just because it is painted black or made of black chocolate? Or
more generally, that the sky is not made of and does not contain any dark and cold substance, as Olbers
Challenge 847 ny himself suggested, and as John Herschel refuted in 1848?

Dvipsbugw
 460 iii gravitation and rel ativity • 9. why can we see the stars?

Is the universe open, closed or marginal?

- Doesn’t the vastness of the universe make you

“ feel small?
- I can feel small without any help from the
universe.
Anonymous

Sometimes the history of the universe is summed up in two words: bang!...crunch. But
will the universe indeed recollapse, or will it expand for ever? Or is it in an intermediate,
marginal situation? The parameters deciding its fate are the mass density and cosmolo-
” Dvipsbugw

gical constant.
The main news of the last decade of twentieth-century astrophysics are the experi-
mental results allowing one to determine all these parameters. Several methods are being
used. The first method is obvious: determine the speed and distance of distant stars. For
large distances, this is difficult, since the stars are so faint. But it has now become possible

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

to search the sky for supernovae, the bright exploding stars, and to determine their dis-
tance from their brightness. This is presently being done with the help of computerized
Ref. 417 searches of the sky, using the largest available telescopes.
A second method is the measurement of the anisotropy of the cosmic microwave back-
ground. From the observed power spectrum as a function of the angle, the curvature of
space-time can be deduced.
A third method is the determination of the mass density using the gravitational lensing
Page 468 effect for the light of distant quasars bent around galaxies or galaxy clusters.
A fourth method is the determination of the mass density using galaxy clusters. All
these measurements are expected to improve greatly in the years to come.
At present, these four completely independent sets of measurements provide the
Ref. 418 values
(Ω M , Ω Λ , Ω K )  (0.3, 0.7, 0.0) (353)

where the errors are of the order of 0.1 or less. The values imply that the universe is spatially
flat, its expansion is accelerating and there will be no big crunch. However, no definite
Page 469 statement on the topology is possible. We will return to this last issue shortly.
In particular, the data show that the density of matter, including all dark matter, is only
about one third of the critical value.* Two thirds are given by the cosmological term. For
the cosmological constant Λ one gets the value
Copyright © Christoph Schiller November 1997–May 2006

3H 02
Λ = ΩΛ  10−52 ~m2 . (354)
c2
This value has important implications for quantum theory, since it corresponds to a va-

* The difference between the total matter density and the separately measurable baryonic matter density,
only about one sixth of the former value, is also explained yet. It might even be that the universe contains
matter of a type unknown so far. This issue is called the dark matter problem; it is one of the important
unsolved questions of cosmology.

Dvipsbugw
 why can we see the stars? – Motion in the universe 461

cuum energy density

Λc 4 10−46 (GeV)4
ρΛ c2 =  0.5 nJ~m3  .
(ħc)3
(355)
8πG

But the cosmological term also implies a negative vacuum pressure p Λ = −ρ Λ c 2 . Inserting
Page 427 this result into the relation for the potential of universal gravity deduced from relativity

∆φ = 4πG(ρ + 3p~c 2 )
Dvipsbugw
(356)

Ref. 419 we get

∆φ = 4πG(ρ M − 2ρ Λ ) . (357)

Challenge 851 ny Thus the gravitational acceleration is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

a= − c r = 2 − Ω Λ H 02 r ,
GM Λ 2 GM
(358)
r2 3 r
which shows that a positive vacuum energy indeed leads to a repulsive gravitational effect.
Inserting the mentioned value (354) for the cosmological constant Λ we find that the
repulsive effect is small even for the distance between the Earth and the Sun. In fact, the
order of magnitude of the repulsive effect is so much smaller than that of attraction that
one cannot hope for a direct experimental confirmation of this deviation from universal
Challenge 852 ny gravity at all. Probably astrophysical determinations will remain the only possible ones.
A positive gravitational constant manifests itself through a positive component in the
expansion rate, as we will see shortly.
But the situation is puzzling. The origin of this cosmological constant is not explained
by general relativity. This mystery will be solved only with the help of quantum theory.
In any case, the cosmological constant is the first local and quantum aspect of nature
detected by astrophysical means.

Why is the universe transparent?

Could the universe be filled with water, which is transparent, as maintained by some
Ref. 420 popular books in order to explain rain? No. Even if it were filled with air, the total mass
would never have allowed the universe to reach the present size; it would have recollapsed
Copyright © Christoph Schiller November 1997–May 2006

Challenge 853 ny much earlier and we would not exist.

The universe is thus transparent because it is mostly empty. But why is it so empty? First
of all, in the times when the size of the universe was small, all antimatter annihilated with
the corresponding amount of matter. Only a tiny fraction of matter, which originally was
slightly more abundant than antimatter, was left over. This 10−9 fraction is the matter we
Page 939 see now. As a consequence, there are 109 as many photons in the universe as electrons or
quarks.
In addition, 380 000 years after antimatter annihilation, all available nuclei and elec-
trons recombined, forming atoms, and their aggregates, like stars and people. No free
charges interacting with photons were lurking around any more, so that from that period

Dvipsbugw
 462 iii gravitation and rel ativity • 9. why can we see the stars?

onwards light could travel through space as it does today, being affected only when it hits
a star or dust particle.
If we remember that the average density of the universe is 10−26 kg~m3 and that most
of the matter is lumped by gravity in galaxies, we can imagine what an excellent va-
cuum lies in between. As a result, light can travel along large distances without noticeable
hindrance.
But why is the vacuum transparent? That is a deeper question. Vacuum is transparent
because it contains no electric charges and no horizons: charges or horizons are indispens-
able in order to absorb light. In fact, quantum theory shows that vacuum does contain Dvipsbugw
Page 854 so-called virtual charges. However, virtual charges have no effects on the transmission of
light.

The big bang and its consequences

Learn to die.

“
Μελέτη θανάτου.
Plato, Phaedo, 81a.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Above all, the big bang model, which is deduced from the colour of the stars and galaxies,
states that about fourteen thousand million years ago the whole universe was extremely
Page 690 small. This fact gave the big bang its name. The term was created (with a sarcastic un-
Ref. 421 dertone) in 1950 by Fred Hoyle, who by the way never believed that it applies to nature.
Nevertheless, the term caught on. Since the past smallness of the universe be checked
directly, we need to look for other, verifiable consequences. The central ones are the fol-
lowing:
— all matter moves away from all other matter;
— the mass of the universe is made up of about 75% hydrogen and 23% helium;
— there is thermal background radiation of about 2.7 K;
— the maximal age for any system in the universe is around fourteen thousand million
years;
— there are background neutrinos with a temperature of about 2 K;*
— for non-vanishing cosmological constant, Newtonian gravity is slightly reduced.
All predictions except the last two have been confirmed by observations. Technology will
probably not allow us to check the last two in the foreseeable future; however, there is no
evidence against them.
Competing descriptions of the universe have not been so successful in matching
Copyright © Christoph Schiller November 1997–May 2006

Ref. 421 observations. In addition, theoretical arguments state that with matter distributions such
as the observed one, and some rather weak general assumptions, there is no way to avoid
Ref. 422 a period in the finite past in which the universe was extremely small. Therefore it is worth
having a close look at the situation.

Was the big bang a big bang?

Was it a kind of explosion? This description implies that some material transforms in-
ternal energy into motion of its parts. There was no such process in the early history

* The theory states that Tν ~Tγ  (4~11)1~3 . These neutrinos appeared about 0.3 s after the big bang.

Dvipsbugw
 why can we see the stars? – Motion in the universe 463

of the universe. In fact, a better description is that space-time is expanding, rather than
matter moving. The mechanism and the origin of the expansion is unknown at this point
of our mountain ascent. Because of the importance of spatial expansion, the whole phe-
nomenon cannot be called an explosion at all. And obviously there neither was nor is any
sound carrying medium in interstellar space, so that one cannot speak of a ‘bang’ in any
sense of the term.
Was it big? The visible universe was rather small about fourteen thousand million years
ago, much smaller than an atom. In summary, the big bang was neither big nor a bang;
but the rest is correct. Dvipsbugw

Was the big bang an event?

The big bang theory is a description of what happened in the whole of space-time. Des-
pite what is often written in careless newspaper articles, at every moment of the expansion
space has been of non-vanishing size: space was never a single point. People who pretend

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

it was are making ostensibly plausible, but false statements. The big bang theory is a de-
scription of the expansion of space-time, not of its beginning. Following the motion of
matter back in time, general relativity cannot deduce the existence of an initial singularity.
The issue of measurement errors is probably not a hindrance; however, the effect of the
nonlinearities in general relativity at situations of high energy densities is not clear.
Most importantly, quantum theory shows that the big bang was not a true singular-
ity, as no physical observable, neither density nor temperature, ever reaches an infinitely
Page 1042 large (or infinitely small) value. Such values cannot exist in nature.* In any case, there is a
general agreement that arguments based on pure general relativity alone cannot make cor-
rect statements about the big bang. Nevertheless, most statements in newspaper articles
are of this sort.

Was the big bang a beginning?

Asking what was before the big bang is like asking what is north of the North Pole. Just
as nothing is north of the North Pole, so nothing ‘was’ before the big bang. This analogy
could be misinterpreted to imply that the big bang took its start at a single point in time,
which of course is incorrect, as just explained. But the analogy is better than it looks: in
fact, there is no precise North Pole, since quantum theory shows that there is a funda-
mental indeterminacy as to its position. There is also a corresponding indeterminacy for
the big bang.
Copyright © Christoph Schiller November 1997–May 2006

In fact, it does not take more than three lines to show with quantum theory that time
and space are not defined either at or near the big bang. We will give this simple argument
Page 1001 in the first chapter of the third part of our mountain ascent. The big bang therefore cannot
be called a ‘beginning’ of the universe. There never was a time when the scale factor a(t)
of the universe was zero.
This conceptual mistake is frequently encountered. In fact, quantum theory shows that
near the big bang, events can neither be ordered nor even be defined. More bluntly, there
is no beginning; there has never been an initial event or singularity.

* Many physicists are still wary of making such strong statements on this point. The first sections of the third
Page 996 part of our mountain ascent give the precise arguments leading to them.

Dvipsbugw
 464 iii gravitation and rel ativity • 9. why can we see the stars?

Obviously the concept of time is not defined ‘outside’ or ‘before’ the existence of the
Ref. 423 universe; this fact was already clear to thinkers over a thousand years ago. It is then tempt-
ing to conclude that time must have started. But as we saw, that is a logical mistake as well:
first of all, there is no starting event, and secondly, time does not flow, as clarified already
Page 49 in the beginning of our walk.
A similar mistake lies behind the idea that the universe had certain ‘initial condi-
Page 153 tions.’ Initial conditions by definition make sense only for objects or fields, i.e. for entities
which can be observed from the outside, i.e. for entities which have an environment. The
universe does not comply with this requirement; it thus cannot have initial conditions. Dvipsbugw
Nevertheless, many people still insist on thinking about this issue; interestingly, Stephen
Ref. 424 Hawking sold millions of copies of a book explaining that a description without initial
conditions is the most appealing, overlooking the fact that there is no other possibility
anyway.*
In summary, the big bang is not a beginning, nor does it imply one. We will uncover
the correct way to think about it in the third part of our mountain ascent.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Does the big bang imply creation?
[The general theory of relativity produces]

“ universal doubt about god and his creation.

Creation, i.e. the appearance of something out of nothing, needs an existing concept of
A witch hunter

”
Page 683 space and time to make sense. The concept of ‘appearance’ makes no sense otherwise. But
whatever the description of the big bang, be it classical, as in this chapter, or quantum
mechanical, as in later ones, this condition is never fulfilled. Even in the present, classical
description of the big bang, which gave rise to its name, there is no appearance of mat-
ter, nor of energy, nor of anything else. And this situation does not change in any later,
improved description, as time or space are never defined before the appearance of matter.
In fact, all properties of a creation are missing: there is no ‘moment’ of creation, no
appearance from nothing, no possible choice of any ‘initial’ conditions out of some set of
possibilities, and as we will see in more detail later on, not even any choice of particular
physical ‘laws’ from any set of possibilities.
In summary, the big bang does not imply nor harbour a creation process. The big bang
was not an event, not a beginning and not a case of creation. It is impossible to continue
Challenge 854 ny the ascent of Motion Mountain if one cannot accept each of these three conclusions. To
Copyright © Christoph Schiller November 1997–May 2006

deny them is to continue in the domain of beliefs and prejudices, thus effectively giving
up on the mountain ascent.
Note that this requirement is not new. In fact, it was already contained in equation (1)
Page 51 at the start of our walk, as well as in all the following ones. It appears even more clearly at
this point. But what then is the big bang? We’ll find out in the third part. We now return
to the discussion of what the stars can tell us about nature.

* This statement will still provoke strong reactions among physicists; it will be discussed in more detail in
the section on quantum theory.

Dvipsbugw
 why can we see the stars? – Motion in the universe 465

Dvipsbugw

F I G U R E 210 The absorption of the atmosphere (NASA)

Why can we see the Sun?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

First of all, the Sun is visible because air is transparent. It is not self-evident that air is
transparent; in fact it is transparent only to visible light and to a few selected other fre-
quencies. Infrared and ultraviolet radiation are mostly absorbed. The reasons lie in the
behaviour of the molecules the air consists of, namely mainly nitrogen, oxygen and a
few other transparent gases. Several moons and planets in the solar system have opaque
atmospheres: we are indeed lucky to be able to see the stars at all.
In fact, even air is not completely transparent; air molecules scatter light a little bit. That
is why the sky and distant mountains appear blue and sunsets red,* and stars are invisible
during daylight. At many frequencies far from the visible spectrum the atmosphere is
even opaque, as Figure 210 shows.
Secondly, we can see the Sun because the Sun, like all hot bodies, emits light. We de-
Page 611 scribe the details of incandescence, as this effect is called, below.
Thirdly, we can see the Sun because we and our environment and the Sun’s environ-
ment are colder than the Sun. In fact, incandescent bodies can be distinguished from their
background only if the background is colder. This is a consequence of the properties of
incandescent light emission, usually called black-body radiation. The radiation is material-
independent, so that for an environment with the same temperature as the body, nothing
can be seen at all. Just have a look at the photograph on page 613 as a proof.
Finally, we can see the Sun because it is not a black hole. If it were, it would emit
(almost) no light.
Copyright © Christoph Schiller November 1997–May 2006

Obviously, each of these conditions applies to stars as well. For example, we can only
see them because the night sky is black. But then, how to explain the multicoloured sky?

Why are the colours of the stars different?

Stars are visible because they emit visible light. We have encountered several important
effects which determine colours: the diverse temperatures among the stars, the Doppler
shift due to a relative speed with respect to the observer, and the gravitational red-shift.
* Air scattering makes the sky blue also at night, as can be proven by long-exposure photographs. (See, for
example, Figure 61.) However, our eyes are not able to perceive this, and the low levels of light make it appear
black to us.

Dvipsbugw
 466 iii gravitation and rel ativity • 9. why can we see the stars?

TA B L E 39 The colour of the stars

C l as s Te m pe r - Example L o c at i o n Colour
at u r e
O 30 kK Mintaka δ Orionis blue-violet
O 31(10) kK Alnitak ζ Orionis blue-violet
B 22(6) kK Bellatrix γ Orionis blue
B 26 kK Saiph κ Orionis blue-white
B 12 kK Rigel β Orionis blue-white Dvipsbugw
B 25 kK Alnilam ε Orionis blue-white
B 17(5) kK Regulus α Leonis blue-white
A 9.9 kK Sirius α Canis Majoris blue-white
A 8.6 kK Megrez δ Ursae Majoris white
A 7.6(2) kK Altair α Aquilae yellow-white
F 7.4(7) kK Canopus α Carinae yellow-white

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F 6.6 kK Procyon α Canis Minoris yellow-white
G 5.8 kK Sun ecliptic yellow
K 3.5(4) kK Aldebaran α Tauri orange
M 2.8(5) kK Betelgeuse α Orionis red
D <80 kK – – all

Note. White dwarfs, or class-D stars, are remnants of imploded stars, with a size of only a few tens of
kilometres. Not all are white; they can be yellow or red. They comprise 5% of all stars. None is visible with
the naked eye. Temperature uncertainties in the last digit are given between parentheses.
The size of all other stars is an independent variable and is sometimes added as roman numerals at the end
of the spectral type. (Sirius is an A1V star, Arcturus a K2III star.) Giants and supergiants exist in all classes
from O to M.
To accommodate brown dwarfs, two new star classes, L and T, have been proposed.

Not all stars are good approximations to black bodies, so that the black-body radiation
Page 612 law does not always accurately describe their colour. However, most stars are reasonable
approximations of black bodies. The temperature of a star depends mainly on its size,
Ref. 425 its mass, its composition and its age, as astrophysicists are happy to explain. Orion is a
good example of a coloured constellation: each star has a different colour. Long-exposure
Copyright © Christoph Schiller November 1997–May 2006

Page 72 photographs beautifully show this.

The basic colour determined by temperature is changed by two effects. The first, the
Challenge 855 ny Doppler red-shift z, depends on the speed v between source and observer as
½
c+v
z= = −1=
∆λ fS
−1 . (359)
λ fO c−v

Such shifts play a significant role only for remote, and thus faint, stars visible through the
telescope. With the naked eye, Doppler shifts cannot be seen. But Doppler shifts can make
distant stars shine in the infrared instead of in the visible domain. Indeed, the highest
Doppler shifts observed for luminous objects are larger than 5.0, corresponding to a re-

Dvipsbugw
 why can we see the stars? – Motion in the universe 467

Challenge 856 ny cessional speed of more than 94 % of the speed of light. Note that in the universe, the
red-shift is also related to the scale factor R(t) by

R(t 0 )
z= −1 .
R(t emission )
(360)

Light at a red-shift of 5.0 was thus emitted when the universe was one sixth of its present
age.
The other colour-changing effect, the gravitational red-shift z g , depends on the matter Dvipsbugw
density of the source and is given by

zg = = −1= ¼
∆λ f S 1
−1 . (361)
λ f0 1 − 2G M
c2 R

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 857 ny It is usually quite a bit smaller than the Doppler shift. Can you confirm this?
No other red-shift processes are known; moreover, such processes would contradict
Page 476 all the known properties of nature. But the colour issue leads to the next question.

Are there dark stars?

It could be that some stars are not seen because they are dark. This could be one explan-
ation for the large amount of dark matter seen in the recent measurements of the back-
ground radiation. This issue is currently of great interest and hotly debated. It is known
that objects more massive than Jupiter but less massive than the Sun can exist in states
which emit hardly any light. They are called brown dwarfs. It is unclear at present how
many such objects exist. Many of the so-called extrasolar ‘planets’ are probably brown
dwarfs. The issue is not yet settled.
Page 476 Another possibility for dark stars are black holes. These are discussed in detail below.

Are all stars different? – Gravitational lenses

Per aspera ad astra.*

“
Are we sure that at night, two stars are really different? The answer is no. Recently, it
was shown that two ‘stars’ were actually two images of the same object. This was found
” Copyright © Christoph Schiller November 1997–May 2006

by comparing the flicker of the two images. It was found that the flicker of one image
was exactly the same as the other, just shifted by 423 days. This result was found by the
Estonian astrophysicist Jaan Pelt and his research group while observing two images of
Ref. 426 quasars in the system Q0957+561.
The two images are the result of gravitational lensing, as shown in Figure 211. Indeed, a
large galaxy can be seen between the two images, much nearer to the Earth. This effect had
been already considered by Einstein; however he did not believe that it was observable.
Ref. 427 The real father of gravitational lensing is Fritz Zwicky, who predicted in 1937 that the effect

* ‘Through hardship to the stars.’ A famous Latin motto. Often incorrectly given as ‘per ardua at astra’.

Dvipsbugw
 468 iii gravitation and rel ativity • 9. why can we see the stars?

first image GRAVITATIONAL LENSING first image TOPOLOGICAL EFFECT

star star
Earth
galaxy
Earth

Dvipsbugw
second image
second image

F I G U R E 211 How one star can lead to several images

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 212 The Zwicky-Einstein ring B1938+666,
seen in the radio spectrum (left) and in the optical
domain (right) (NASA)

would be quite common and easy to observe, if lined-up galaxies instead of lined-up stars
were considered, as indeed turned out to be the case.
Interestingly, when the time delay is known, astronomers are able to determine the
Challenge 858 ny size of the universe from this observation. Can you imagine how?
In fact, if the two observed objects are lined up exactly behind each other, the more dis-
tant one is seen as ring around the nearer one. Such rings have indeed been observed, and
Copyright © Christoph Schiller November 1997–May 2006

the galaxy image around a central foreground galaxy at B1938+666, shown in Figure 212,
is one of the most beautiful examples. In 2005, several cases of gravitational lensing by
stars were also discovered. More interestingly, three events where one of the two stars has
a Earth-mass planet have also been observed. The coming years will surely lead to many
additional observations, helped by the sky observation programme in the southern hemi-
sphere that checks the brightness of about 100 million stars every night.
Generally speaking, images of nearby stars are truly unique, but for the distant stars the
problem is tricky. For single stars, the issue is not so important, seen overall. Reassuringly,
only about 80 multiple star images have been identified so far. But when whole galaxies
are seen as several images at once (and several dozens are known so far) we might start
to get nervous. In the case of the galaxy cluster CL0024+1654, shown in Figure 213, seven

Dvipsbugw
 why can we see the stars? – Motion in the universe 469

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 213 Multiple blue images of a galaxy
formed by the yellow cluster CL0024+1654 (NASA)

thin, elongated, blue images of the same distant galaxy are seen around the yellow, nearer,
elliptical galaxies.
But multiple images can be created not only by gravitational lenses; the shape of the
universe could also play some tricks.

What is the shape of the universe?

There is a popular analogy which avoids some of the just-mentioned problems. The uni-
verse in its evolution is similar to the surface of an ever-expanding sphere: the surface is
finite, but it has no boundary. The universe simply has an additional dimension; therefore
its volume is also ever increasing, finite, but without boundary. This statement presup-
poses that the universe has the same topology, the same ‘shape’ as that of a sphere with
an additional dimension.
Ref. 428 But what is the experimental evidence for this statement? There is none. Nothing is
yet known about the shape of the universe. It is extremely hard to determine it, simply
because of its sheer size.
What do experiments say? In the nearby region of the universe, say within a few mil-
Copyright © Christoph Schiller November 1997–May 2006

lion light years, the topology is simply connected. But for large distances, almost nothing
is certain. Maybe research into gamma-ray bursts will tell us something about the topo-
logy, as these bursts often originate from the dawn of time.* Maybe even the study of
fluctuations of the cosmic background radiation can tell us something. All this research
is still in its infancy.
Since little is known, we can ask about the range of possible answers. As just mentioned,
in the standard model with k = 1, space-time is usually assumed to be a product of linear
time, with the topology R of the real line, and a sphere S 3 for space. That is the simplest
possible shape, corresponding to a simply-connected universe. For k = 0, the simplest

* The story is told from the mathematical point of view by B ob Osserman, Poetry of the Universe, 1996.

Dvipsbugw
 470 iii gravitation and rel ativity • 9. why can we see the stars?

topology of space is three-dimensional real space R3 , and for k = −1 it is a hyperbolic

manifold H3 .
Page 456 In addition, Figure 206 showed that depending on the value of the cosmological con-
stant, space could be finite and bounded, or infinite and unbounded. In Friedmann–
Lemaître calculations, simple-connectedness is usually tacitly assumed, even though it
is not at all required.
It could well be that space-time is multiply connected, like a higher-dimensional ver-
sion of a torus. It could also have even more complex topologies.** In these cases, it could
even be that the actual number of galaxies is much smaller than the observed number. Dvipsbugw
This situation would correspond to a kaleidoscope, where a few beads produce a large
number of images. In addition, topological surprises could also be hidden behind the
horizon.
In fact, the range of possibilities is not limited to the simply and multiply connec-
ted cases suggested by classical physics. An additional and completely unexpected twist
will appear in the third part of our walk, when quantum theory is included in the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 1041 investigations.

What is behind the horizon?

The universe is a big place; perhaps the biggest.

“ Kilgore Trout, Venus on the Half Shell.

The horizon is a tricky entity. In fact, all cosmological models show that it moves rapidly ”
Ref. 430 away from us. A detailed investigation shows that for a matter-dominated universe the
Challenge 859 ny horizon moves away from us with a velocity

v horizon = 3c . (362)

A pretty result, isn’t it? Obviously, since the horizon does not transport any signal, this is
not a contradiction of relativity. But what is behind the horizon?
If the universe is open or marginal, the matter we see at night is predicted by naively
applied general relativity to be a – literally – infinitely small part of all matter existing.
Indeed, an open or marginal universe implies that there is an infinite amount of matter
Challenge 860 ny behind the horizon. Is such a statement verifiable?
In a closed universe, matter is still predicted to exist behind the horizon; however, in
this case it is only a finite amount.
Copyright © Christoph Schiller November 1997–May 2006

In short, the standard model of cosmology states that there is a lot of matter behind
the horizon. Like most cosmologists, we sweep the issue under the rug and take it up
only later in our walk. A precise description of the topic is provided by the hypothesis of
inflation.

** The Friedmann–Lemaître metric is also valid for any quotient of the just-mentioned simple topologies by
a group of isometries, leading to dihedral spaces and lens spaces in the case k = 1, to tori in the case k = 0,
Ref. 429 and to any hyperbolic manifold in the case k = −1.

Dvipsbugw
 why can we see the stars? – Motion in the universe 471

Why are there stars all over the place? – Inflation

What were the initial conditions of matter? Matter was distributed in a constant density
over space expanding with great speed. How could this happen? The person who has ex-
plored this question most thoroughly is Alan Guth. So far, we have based our studies of
the night sky, cosmology, on two observational principles: the isotropy and the homo-
geneity of the universe. In addition, the universe is (almost) flat. Inflation is an attempt
to understand the origin of these observations. Flatness at the present instant of time is
strange: the flat state is an unstable solution of the Friedmann equations. Since the uni- Dvipsbugw
verse is still flat after fourteen thousand million years, it must have been even flatter near
the big bang.
Ref. 431 Guth argued that the precise flatness, the homogeneity and the isotropy could follow
if in the first second of its history, the universe had gone through a short phase of expo-
nential size increase, which he called inflation. This exponential size increase, by a factor
of about 1026 , would homogenize the universe. This extremely short evolution would be

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

driven by a still-unknown field, the inflaton field. Inflation also seems to describe correctly
the growth of inhomogeneities in the cosmic background radiation.
However, so far, inflation poses as many questions as it solves. Twenty years after his
initial proposal, Guth himself is sceptical on whether it is a conceptual step forward. The
final word on the issue has not been said yet.

Why are there so few stars? – The energy and entropy content of
the universe
Die Energie der Welt ist constant. Die Entropie

“ der Welt strebt einem Maximum zu.*

Rudolph Clausius
The matter–energy density of the universe is near the critical one. Inflation, described in
”
the previous section, is the favourite explanation for this connection. This implies that the
actual number of stars is given by the behaviour of matter at extremely high temperatures,
and by the energy density left over at lower temperature. The precise connection is still
the topic of intense research. But this issue also raises a question about the quotation
above. Was the creator of the term ‘entropy’, Rudolph Clausius, right when he made this
famous statement? Let us have a look at what general relativity has to say about all this.
In general relativity, a total energy can indeed be defined, in contrast to localized energy,
which cannot. The total energy of all matter and radiation is indeed a constant of motion.
Copyright © Christoph Schiller November 1997–May 2006

It is given by the sum of the baryonic, luminous and neutrino parts:

c 2 M0 c2
E = Eb + Eγ + E ν  + ... + ...  + ... . (363)
T0 G

This value is constant only when integrated over the whole universe, not when just the
inside of the horizon is taken.*
Many people also add a gravitational energy term. If one tries to do so, one is obliged
to define it in such a way that it is exactly the negative of the previous term. This value
* ‘The energy of the universe is constant. Its entropy tends towards a maximum.’
* Except for the case when pressure can be neglected.

Dvipsbugw
 472 iii gravitation and rel ativity • 9. why can we see the stars?

for the gravitational energy leads to the popular speculation that the total energy of the
universe might be zero. In other words, the number of stars could also be limited by this
relation.
However, the discussion of entropy puts a strong question mark behind all these seem-
ingly obvious statements. Many people have tried to give values for the entropy of the
Ref. 432 universe. Some have checked whether the relation

kc 3 A kG
S= = 4πM 2 , (364) Dvipsbugw
Għ 4 ħc
Challenge 861 ny which is correct for black holes, also applies to the universe. This assumes that all the
matter and all the radiation of the universe can be described by some average temperature.
They argue that the entropy of the universe is surprisingly low, so that there must be some
ordering principle behind it. Others even speculate over where the entropy of the universe
comes from, and whether the horizon is the source for it.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

But let us be careful. Clausius assumes, without the slightest doubt, that the universe is
a closed system, and thus deduces the statement quoted above. Let us check this assump-
tion. Entropy describes the maximum energy that can be extracted from a hot object.
After the discovery of the particle structure of matter, it became clear that entropy is also
given by the number of microstates that can make up a specific macrostate. But neither
definition makes any sense if applied to the universe as a whole. There is no way to extract
energy from it, and no way to say how many microstates of the universe would look like
the macrostate.
The basic reason is the impossibility of applying the concept of state to the universe.
Page 39 We first defined the state as all those properties of a system which allow one to distin-
guish it from other systems with the same intrinsic properties, or which differ from one
observer to another. You might want to check for yourself that for the universe, such state
Challenge 862 ny properties do not exist at all.
We can speak of the state of space-time and we can speak of the state of matter and
energy. But we cannot speak of the state of the universe, because the concept makes no
sense. If there is no state of the universe, there is no entropy for it. And neither is there
an energy value. This is in fact the only correct conclusion one can draw about the issue.

Why is matter lumped?

We are able to see the stars because the universe consists mainly of empty space, in other
Copyright © Christoph Schiller November 1997–May 2006

words, because stars are small and far apart. But why is this the case? Cosmic expansion
was deduced and calculated using a homogeneous mass distribution. So why did matter
lump together?
It turns out that homogeneous mass distributions are unstable. If for any reason the
density fluctuates, regions of higher density will attract more matter than regions of lower
density. Gravitation will thus cause the denser regions to increase in density and the re-
gions of lower density to be depleted. Can you confirm the instability, simply by assuming
Challenge 863 ny a space filled with dust and a = GM~r 2 ? In summary, even a tiny quantum fluctuation in
the mass density will lead, after a certain time, to lumped matter.
But how did the first inhomogeneities form? That is one of the big problems of mod-

Dvipsbugw
 why can we see the stars? – Motion in the universe 473

ern physics and astrophysics, and there is no accepted answer yet. Several modern ex-
periments are measuring the variations of the cosmic background radiation spectrum
with angular position and with polarization; these results, which will be available in the
Ref. 433 coming years, might provide some information on the way to settle the issue.

Why are stars so small compared with the universe?

Given that the matter density is around the critical one, the size of stars, which contain
most of the matter, is a result of the interaction of the elementary particles composing Dvipsbugw
Page 492 them. Below we will show that general relativity (alone) cannot explain any size appearing
in nature. The discussion of this issue is a theme of quantum theory.

Are stars and galaxies moving apart or is the universe

expanding?
Can we distinguish between space expanding and galaxies moving apart? Yes, we can.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 864 ny Can you find an argument or devise an experiment to do so?
The expansion of the universe does not apply to the space on the Earth. The expansion
is calculated for a homogeneous and isotropic mass distribution. Matter is neither homo-
geneous nor isotropic inside the galaxy; the approximation of the cosmological principle
is not valid down here. It has even been checked experimentally, by studying atomic spec-
Ref. 434 tra in various places in the solar system, that there is no Hubble expansion taking place
around us.

Is there more than one universe?

The existence of ‘several’ universes might be an option when we study the question
whether we see all the stars. But you can check that neither definition of ‘universe’ given
above, be it ‘all matter-energy’ or ‘all matter–energy and all space-time’, allows us to an-
Challenge 865 ny swer the question positively.
There is no way to define a plural for universe: either the universe is everything, and
then it is unique, or it is not everything, and then it is not the universe. We will discover
Page 812 that quantum theory does not change this conclusion, despite recurring reports to the
contrary.
Whoever speaks of many universes is talking ghibberish.

Why are the stars fixed? – Arms, stars and Mach’s principle
Copyright © Christoph Schiller November 1997–May 2006

Si les astres étaient immobiles, le temps et

“ l’espace n’existeraient plus.

Maurice Maeterlink.

The two arms possessed by humans have played an important role in discussions about
”
motion, and especially in the development of relativity. Looking at the stars at night, we
can make a simple observation, if we keep our arms relaxed. Standing still, our arms hang
down. Then we turn rapidly. Our arms lift up. In fact they do so whenever we see the stars
turning. Some people have spent a large part of their lives studying this phenomenon.
Why?
Ref. 435 Stars and arms prove that motion is relative, not absolute.* This observation leads to

Dvipsbugw
 474 iii gravitation and rel ativity • 9. why can we see the stars?

two possible formulations of what Einstein called Mach’s principle.

— Inertial frames are determined by the rest of the matter in the universe.
This idea is indeed realized in general relativity. No question about it.
— Inertia is due to the interaction with the rest of the universe.
This formulation is more controversial. Many interpret it as meaning that the mass of an
object depends on the distribution of mass in the rest of the universe. That would mean
Dvipsbugw
that one needs to investigate whether mass is anisotropic when a large body is nearby.
Of course, this question has been studied experimentally; one simply needs to measure
whether a particle has the same mass values when accelerated in different directions. Un-
Ref. 436 surprisingly, to a high degree of precision, no such anisotropy has been found. Many
therefore conclude that Mach’s principle is wrong. Others conclude with some pain in
Ref. 437 their stomach that the whole topic is not yet settled.
But in fact it is easy to see that Mach cannot have meant a mass variation at all: one then

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

would also have to conclude that mass is distance-dependent, even in Galilean physics.
Challenge 866 ny But this is known to be false; nobody in his right mind has ever had any doubts about it.
The whole debate is due to a misunderstanding of what is meant by ‘inertia’: one
can interpret it as inertial mass or as inertial motion (like the moving arms under the
stars). There is no evidence that Mach believed either in anisotropic mass or in distance-
dependent mass; the whole discussion is an example people taking pride in not making
a mistake which is incorrectly imputed to another, supposedly more stupid, person.**
Obviously, inertial effects do depend on the distribution of mass in the rest of the
universe. Mach’s principle is correct. Mach made some blunders in his life (he is infam-
ous for opposing the idea of atoms until he died, against experimental evidence) but his
principle is not one of them. Unfortunately it is to be expected that the myth about the
Ref. 437 incorrectness of Mach’s principle will persist, like that of the derision of Columbus.
In fact, Mach’s principle is valuable. As an example, take our galaxy. Experiments show
that it is flattened and rotating. The Sun turns around its centre in about 250 million years.
Indeed, if the Sun did not turn around the galaxy’s centre, we would fall into it in about 20
million years. As the physicist Dennis Sciama pointed out, from the shape of our galaxy
we can draw a powerful conclusion: there must be a lot of other matter, i.e. a lot of other
Challenge 867 n stars and galaxies in the universe. Can you confirm his reasoning?

At rest in the universe

Copyright © Christoph Schiller November 1997–May 2006

There is no preferred frame in special relativity, no absolute space. Is the same true in the
actual universe? No; there is a preferred frame. Indeed, in the standard big-bang cosmo-
logy, the average galaxy is at rest. Even though we talk about the big bang, any average

* The original reasoning by Newton and many others used a bucket and the surface of the water in it; but
the arguments are the same.
** For another example, at school one usually hears that Columbus was derided because he thought the
Earth to be spherical. But he was not derided at all for this reason; there were only disagreements on the size
of the Earth, and in fact it turned out that his critics were right, and that he was wrong in his own, much too
small, estimate of the radius.

Dvipsbugw
 why can we see the stars? – Motion in the universe 475

galaxy can rightly maintain that it is at rest. Each one is in free fall. An even better realiz-
ation of this privileged frame of reference is provided by the background radiation.
In other words, the night sky is black because we move with almost no speed through
background radiation. If the Earth had a large velocity relative to the background radi-
ation, the sky would be bright even at night, thanks to the Doppler effect for the back-
ground radiation. In other words, the fact that the night sky is dark in all directions is a
consequence of our slow motion against the background radiation.
This ‘slow’ motion has a speed of 368 km~s. (This is the value of the motion of the
Sun; there are variations due to addition of the motion of the Earth.) the value is large Dvipsbugw
in comparison to everyday life, but small compared to the speed of light. More detailed
studies do not change this conclusion. Even the motion of the Milky Way and that of the
local group against the cosmic background radiation is of the order of 600 km~s; that is
still much slower than the speed of light. The reasons why the galaxy and the solar system
move with these ‘low’ speeds across the universe have already been studied in our walk.
Challenge 868 ny Can you give a summary?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

By the way, is the term ‘universe’ correct? Does the universe rotate, as its name implies?
If by universe one means the whole of experience, the question does not make sense,
because rotation is only defined for bodies, i.e. for parts of the universe. However, if by
Ref. 438 universe one only means ‘all matter’, the answer can be determined by experiments. It
turns out that the rotation is extremely small, if there is any: measurements of the cosmic
background radiation show that in the lifetime of the universe, it cannot have rotated by
more than a hundredth of a millionth of a turn! In short, ‘universe’ is a misnomer.

Does light attract light?

Another reason why we can see stars is that their light reaches us. But why are travelling
light rays not disturbed by each other’s gravitation? We know that light is energy and that
any energy attracts other energy through gravitation. In particular, light is electromag-
netic energy, and experiments have shown that all electromagnetic energy is subject to
gravitation. Could two light beams that are advancing with a small angle between them
converge, because of mutual gravitational attraction? That could have measurable and
possibly interesting effects on the light observed from distant stars.
The simplest way to explore the issue is to study the following question: Do parallel
light beams remain parallel? Interestingly, a precise calculation shows that mutual gravit-
Ref. 439 ation does not alter the path of two parallel light beams, even though it does alter the path
Copyright © Christoph Schiller November 1997–May 2006

of antiparallel light beams.* The reason is that for parallel beams moving at light speed,
Challenge 869 ny the gravitomagnetic component exactly cancels the gravitoelectric component.
Since light does not attract light moving along, light is not disturbed by its own gravity
during the millions of years that it takes to reach us from distant stars. Light does not
attract or disturb light moving alongside. So far, all known quantum-mechanical effects
also confirm this conclusion.

* Antiparallel beams are parallel beams travelling in opposite directions.

Dvipsbugw
 476 iii gravitation and rel ativity • 10. bl ack holes – falling forever

Does light decay?

In the section on quantum theory we will encounter experiments showing that light is
made of particles. It is plausible that these photons might decay into some other particle,
as yet unknown, or into lower-frequency photons. If that actually happened, we would
not be able to see distant stars.
Challenge 870 ny But any decay would also mean that light would change its direction (why?) and thus
produce blurred images for remote objects. However, no blurring is observed. In addition,
the Soviet physicist Matvey Bronstein demonstrated in the 1930s that any light decay pro- Dvipsbugw
Ref. 440 cess would have a larger rate for smaller frequencies. When people checked the shift of
radio waves, in particular the famous 21 cm line, and compared it with the shift of light
from the same source, no difference was found for any of the galaxies tested.
People even checked that Sommerfeld’s fine-structure constant, which determines the
Ref. 441 colour of objects, does not change over time. Despite an erroneous claim in recent years,
no change could be detected over thousands of millions of years.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Of course, instead of decaying, light could also be hit by some hitherto unknown entity.
Challenge 871 ny But this possibility is excluded by the same arguments. These investigations also show
that there is no additional red-shift mechanism in nature apart from the Doppler and
Page 466 gravitational red-shifts.
The visibility of the stars at night has indeed shed light on numerous properties of
nature. We now continue our mountain ascent with a more general issue, nearer to our
quest for the fundamentals of motion.

10. bl ack holes – falling forever

Why study black holes?
Qui iacet in terra non habet unde cadat.**

“
Black holes are the most extreme gravitational phenomena. They realize nature’s limit
Alanus de Insulis

of length-to-mass ratios. They produce the highest force value possible in nature; as a
”
result, they produce high space-time curvatures. Therefore, black holes cannot be studied
without general relativity. In addition, the study of black holes is a major stepping stone
towards unification and the final description of motion.
Copyright © Christoph Schiller November 1997–May 2006

Ref. 337 ‘Black hole’ is shorthand for ‘gravitationally completely collapsed object’. For many
years it was unclear whether or not they exist. But the available experimental data have
now led most experts to conclude that there is a black hole at the centre of most galaxies,
Ref. 442 including our own. Black holes are also suspected at the heart of quasars and of gamma
ray bursters. It seems that the evolution of galaxies is strongly tied to the evolution of
black holes. In addition, half a dozen smaller black holes have been identified elsewhere
in our galaxy. For these and many other reasons, black holes, the most impressive, the
Ref. 443 most powerful and the most relativistic systems in nature, are a fascinating subject of
study.
** ‘He who lies on the ground cannot fall down from it.’ The author’s original name is Alain de Lille (c. 1128–
1203).

Dvipsbugw
 bl ack holes – falling forever 477

Horizons
The escape velocity is the speed needed to launch an projectile in such a way that it never
falls back down. The escape velocity depends on the mass and the size of the planet from
which the launch takes place. What happens when a planet or star has an escape velocity
that is larger than the speed of light c? Such objects were first imagined by the British
geologist John Michell in 1784, and independently by the French mathematician Pierre
Ref. 444 Laplace in 1795, long before general relativity was developed. Michell and Laplace realized
something fundamental: even if an object with such a high escape velocity were a hot star, Dvipsbugw
it would appear to be completely black. The object would not allow any light to leave it; in
addition, it would block all light coming from behind it. In 1967, John Wheeler* coined
Ref. 337 the now standard term black hole.
It only takes a short calculation to show that light cannot escape from a body of mass
Challenge 872 ny M whenever the radius is smaller than a critical value given by

RS =
2GM

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(365)
c2
called the Schwarzschild radius. The formula is valid both in universal gravity and in gen-
eral relativity, provided that in general relativity we take the radius as meaning the cir-
cumference divided by 2π. Such a body realizes the limit value for length-to-mass ratios
in nature. For this and other reasons to be given shortly, we will call R S also the size of
the black hole of mass M. (But note that it is only half the diameter. In addition, the term
‘size’ has to be taken with some grain of salt.) In principle, it is possible to imagine an ob-
ject with a smaller length-to-mass ratio; but nobody has yet observed one. In fact, we will
discover that there is no way to observe an object smaller than the Schwarzschild radius,
just as an object moving faster than the speed of light cannot be observed. However, we
can observe black holes – the limit case – just as we can observe entities moving at the
speed of light.
When a test mass approaches the critical radius R S , two things happen. First, the local
proper acceleration for (imaginary) point masses increases without bound. For realistic
objects of finite size, the black hole realizes the highest force possible in nature. Something
that falls into a black hole cannot be pulled back out. A black hole thus swallows all matter
that falls into it. It acts like a cosmic vacuum cleaner.
At the surface of a black hole, the red-shift factor for a distant observer also increases
without bound. The ratio between the two quantities is called the surface gravity of a black
Copyright © Christoph Schiller November 1997–May 2006

Challenge 873 ny hole. It is given by

c4 c2
дsurf = 2 = =
GM
. (366)
RS 4GM 2R S

A black hole thus does not allow any light to leave it.

* John Archibald Wheeler (1911–), US-American physicist, important expert on general relativity and au-
thor of several excellent textbooks, among them the beautiful John A. Wheeler, A Journey into Gravity
and Spacetime, Scientific American Library & Freeman, 1990, in which he explains general relativity with
passion and in detail, but without any mathematics.

Dvipsbugw
 478 iii gravitation and rel ativity • 10. bl ack holes – falling forever

A surface that realizes the force limit and an in-

finite red-shift makes it is impossible to send light,
matter, energy or signals of any kind to the outside
world. A black hole is thus surrounded by a hori-
zon. We know that a horizon is a limit surface. In event horizon
fact, a horizon is a limit in two ways. First, a hori-
zon is a limit to communication: nothing can com-
municate across it. Secondly, a horizon is a surface
of maximum force and power. These properties are Dvipsbugw
sufficient to answer all questions about the effects
of horizons. For example: What happens when a
Challenge 874 ny light beam is sent upwards from the horizon? And
from slightly above the horizon?
F I G U R E 214 The light cones in the
Black holes, regarded as astronomical objects, equatorial plane around a non-rotating
are thus different from planets. During the form- black hole, seen from above

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ation of planets, matter clumps together; as soon
as it cannot be compressed any further, an equilibrium is reached, which determines the
radius of the planet. That is the same mechanism as when a stone is thrown towards the
Earth: it stops falling when it hits the ground. A ‘ground’ is formed whenever matter hits
other matter. In the case of a black hole, there is no ground; everything continues falling.
That is why, in Russian, black holes used to be called collapsars.
This continuous falling takes place when the concentration of matter is so high that
it overcomes all those interactions which make matter impenetrable in daily life. In 1939,
Ref. 445 Robert Oppenheimer* and Hartland Snyder showed theoretically that a black hole forms
whenever a star of sufficient mass stops burning. When a star of sufficient mass stops
burning, the interactions that form the ‘floor’ disappear, and everything continues falling
without end.
A black hole is matter in permanent free fall. Nevertheless, its radius for an outside
observer remains constant! But that is not all. Furthermore, because of this permanent
free fall, black holes are the only state of matter in thermodynamic equilibrium! In a sense,
floors and all other every-day states of matter are metastable: these forms are not as stable
as black holes.
The characterizing property of a black hole is thus its horizon. The first time we en-
Page 335 countered horizons was in special relativity, in the section on accelerated observers. The
horizons due to gravitation are similar in all their properties; the section on the maximum
Copyright © Christoph Schiller November 1997–May 2006

force and power gave a first impression. The only difference we have found is due to the
neglect of gravitation in special relativity. As a result, horizons in nature cannot be planar,
in contrast to what is suggested by the observations of the imagined point-like observers
assumed to exist in special relativity.
Both the maximum force principle and the field equations imply that the space-time
around a rotationally symmetric (thus non-rotating) and electrically neutral mass is de-
* Robert Oppenheimer (1904–1967), important US-American physicist. He can be called the father of the-
oretical physics in the USA. He worked on quantum theory and atomic physics. He then headed the team
that developed the nuclear bomb during the Second World War. He was also the most prominent (innocent)
victim of one of the greatest witch-hunts ever organized in his home country. See also the http://www.nap.
edu/readingroom/books/biomems/joppenheimer.html website.

Dvipsbugw
 bl ack holes – falling forever 479

Page 385 scribed by

dr 2
di 2 = (1 − )dt − r 2 dφ 2 ~c 2 .
2GM 2
− (367)
rc 2 1 − 2G M
rc 2

This is the so-called Schwarzschild metric. As mentioned above, r is the circumference

divided by 2π; t is the time measured at infinity. No outside observer will ever receive any
signal emitted from a radius value r = 2GM~c 2 or smaller. Indeed, as the proper time i
of an observer at radius r is related to the time t of an observer at infinity through Dvipsbugw
¾
di =
2GM
1− dt , (368)
rc 2

we find that an observer at the horizon would have vanishing proper time. In other words,
at the horizon the red-shift is infinite. (In fact, the surface of infinite red-shift and the hori-
zon coincide only for non-rotating black holes. For rotating black holes, the two surfaces

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

are distinct.) Everything happening at the horizon goes on infinitely slowly, as observed
by a distant observer. In other words, for a distant observer observing what is going on
at the horizon itself, nothing at all ever happens.
In the same way that observers cannot reach the speed of light, observers cannot reach
a horizon. For a second observer, it can only happen that the first is moving almost as fast
as light; in the same way, for a second observer, it can only happen that the first has almost
reached the horizon. In addition, a traveller cannot feel how much he is near the speed of
light for another, and experiences light speed as unattainable; in the same way, a traveller
(into a large black hole) cannot feel how much he is near a horizon and experiences the
horizon as unattainable.
In general relativity, horizons of any kind are predicted to be black. Since light cannot
escape from them, classical horizons are completely dark surfaces. In fact, horizons are
the darkest entities imaginable: nothing in nature is darker. Nonetheless, we will later
discover that physical horizons are not completely black.

Orbits
Ref. 440 Since black holes curve space-time strongly, a body moving near a black hole behaves in
more complicated ways than predicted by universal gravity. In universal gravity, paths
are either ellipses, parabolas, or hyperbolas; all these are plane curves. It turns out that
Copyright © Christoph Schiller November 1997–May 2006

paths lie in a plane only near non-rotating black holes.*

Around non-rotating black holes, also called Schwarzschild black holes, circular paths
Challenge 876 ny are impossible for radii less than 3R S ~2 (can you show why?) and are unstable to per-
turbations from there up to a radius of 3R S . Only at larger radii are circular orbits stable.
Around black holes, there are no elliptic paths; the corresponding rosetta path is shown

* For such paths, Kepler’s rule connecting the average distance and the time of orbit

GMt 3
= r3
(2π)2
(369)

Challenge 875 ny still holds, provided the proper time and the radius measured by a distant observer are used.

Dvipsbugw
 480 iii gravitation and rel ativity • 10. bl ack holes – falling forever

Dvipsbugw
impact
parameter

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 215 Motions of uncharged objects around a non-rotating black hole – for
different impact parameters and initial velocities

in Figure 215. Such a path shows the famous periastron shift in all its glory.
Note that the potential around a black hole is not appreciably different from 1~r for
Challenge 877 ny distances above about fifteen Schwarzschild radii. For a black hole of the mass of the Sun,
that would be 42 km from its centre; therefore, we would not be able to note any difference
for the path of the Earth around the Sun.
We have mentioned several times in our adventure that gravitation is characterized
by its tidal effects. Black holes show extreme properties in this respect. If a cloud of dust
falls into a black hole, the size of the cloud increases as it falls, until the cloud envelops the
whole horizon. In fact, the result is valid for any extended body. This property of black
holes will be of importance later on, when we will discuss the size of elementary particles.
For falling bodies coming from infinity, the situation near black holes is even more in-
teresting. Of course there are no hyperbolic paths, only trajectories similar to hyperbolas
for bodies passing far enough away. But for small, but not too small impact parameters,
Copyright © Christoph Schiller November 1997–May 2006

a body will make a number of turns around the black hole, before leaving again. The
number of turns increases beyond all bounds with decreasing impact parameter, until a
value is reached at which the body is captured into an orbit at a radius of 2R, as shown in
Figure 215. In other words, this orbit captures incoming bodies if they approach it below
a certain critical angle. For comparison, remember that in universal gravity, capture is
never possible. At still smaller impact parameters, the black hole swallows the incoming
mass. In both cases, capture and deflection, a body can make several turns around the
black hole, whereas in universal gravity it is impossible to make more than half a turn
around a body.
The most absurd-looking orbits, though, are those corresponding to the parabolic case
Challenge 878 ny of universal gravity. (These are of purely academic interest, as they occur with probability

Dvipsbugw
 bl ack holes – falling forever 481

limit orbit

limit orbit

the photon sphere the photon sphere

Dvipsbugw

F I G U R E 216 Motions of light passing near a non-rotating black hole

zero.) In summary, relativity changes the motions due to gravity quite drastically.
Around rotating black holes, the orbits of point masses are even more complex than

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

those shown in Figure 215; for bound motion, for example, the ellipses do not stay in
one plane – thanks to the Thirring–Lense effect – leading to extremely involved orbits in
three dimensions filling the space around the black hole.
For light passing a black hole, the paths are equally interesting, as shown in Figure 216.
There are no qualitative differences with the case of rapid particles. For a non-rotating
black hole, the path obviously lies in a single plane. Of course, if light passes sufficiently
nearby, it can be strongly bent, as well as captured. Again, light can also make one or
several turns around the black hole before leaving or being captured. The limit between
the two cases is the path in which light moves in a circle around a black hole, at 3R~2.
If we were located on that orbit, we would see the back of our head by looking forward!
Challenge 879 ny However, this orbit is unstable. The surface containing all orbits inside the circular one
is called the photon sphere. The photon sphere thus divides paths leading to capture from
those leading to infinity. Note that there is no stable orbit for light around a black hole.
Challenge 880 ny Are there any rosetta paths for light around a black hole?
For light around a rotating black hole, paths are much more complex. Already in the
equatorial plane there are two possible circular light paths: a smaller one in the direction
Challenge 881 ny of the rotation, and a larger one in the opposite direction.
For charged black holes, the orbits for falling charged particles are even more complex.
The electrical field lines need to be taken into account. Several fascinating effects appear
which have no correspondence in usual electromagnetism, such as effects similar to elec-
Copyright © Christoph Schiller November 1997–May 2006

trical versions of the Meissner effect. The behaviour of such orbits is still an active area of
research in general relativity.

Hair and entropy

How is a black hole characterized? It turns out that all properties of black holes follow
from a few basic quantities characterizing them, namely their mass M, their angular mo-
mentum J, and their electric charge Q.* All other properties – such as size, shape, colour,

* The existence of three basic characteristics is reminiscent of particles. We will find out more about the
connection between black holes and particles in the third part of our mountain ascent.

Dvipsbugw
 482 iii gravitation and rel ativity • 10. bl ack holes – falling forever

magnetic field – are uniquely determined by these.** It is as though, to use Wheeler’s

colourful analogy, one could deduce every characteristic of a woman from her size, her
waist and her height. Physicists also say that black holes ‘have no hair,’ meaning that (clas-
sical) black holes have no other degrees of freedom. This expression was also introduced
Ref. 448 by Wheeler.*** This fact was proved by Israel, Carter, Robinson and Mazur; they showed
that for a given mass, angular momentum and charge, there is only one possible black
Ref. 449 hole. (However, the uniqueness theorem is not valid any more if the black hole carries
nuclear quantum numbers, such as weak or strong charges.)
In other words, a black hole is independent of how it has formed, and of the materials Dvipsbugw
used when forming it. Black holes all have the same composition, or better, they have no
composition at all (at least classically).
The mass M of a black hole is not restricted by general relativity. It may be as small
as that of a microscopic particle and as large as many million solar masses. But for their
angular momentum J and electric charge Q, the situation is different. A rotating black
hole has a maximum possible angular momentum and a maximum possible electric (and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

magnetic) charge.**** The limit on the angular momentum appears because its perimeter
Challenge 882 ny may not move faster than light. The electric charge is also limited. The two limits are not
independent: they are related by

J 2 GQ 2 GM 2
‹  + D ‹  . (370)
cM 4πε 0 c 4 c2

Challenge 883 ny This follows from the limit on length-to-mass ratios at the basis of general relativity. Ro-
tating black holes realizing the limit (370) are called extremal black holes. The limit (370)
implies that the horizon radius of a general black hole is given by
¾
GM ’ J2 c2 Q2 “
rh = 2 1 + 1− −
c ” 4πε 0 GM •
(371)
M G
4 2 2

For example, for a black hole with the mass and half the angular momentum of the Sun,
namely 2 ë 1030 kg and 0.45 ë 1042 kg m2 ~s, the charge limit is about 1.4 ë 1020 C.
How does one distinguish rotating from non-rotating black holes? First of all by the
shape. Non-rotating black holes must be spherical (any non-sphericity is radiated away as
Ref. 450 gravitational waves) and rotating black holes have a slightly flattened shape, uniquely de-
Copyright © Christoph Schiller November 1997–May 2006

termined by their angular momentum. Because of their rotation, their surface of infinite

** Mainly for marketing reasons, non-rotating and electrically neutral black holes are often called Schwarz-
Ref. 446 schild black holes; uncharged and rotating ones are often called Kerr black holes, after Roy Kerr, who
discovered the corresponding solution of Einstein’s field equations in 1963. Electrically charged but non-
rotating black holes are often called Reissner–Nordström black holes, after the German physicist Hans Re-
issner and the Finnish physicist Gunnar Nordström. The general case, charged and rotating, is sometimes
Ref. 447 named after Kerr and Newman.
Ref. 337 *** Wheeler claims that he was inspired by the difficulty of distinguishing between bald men; however, Feyn-
man, Ruffini and others had a clear anatomical image in mind when they stated that ‘black holes, in contrast
to their surroundings, have no hair.’
**** More about the still hypothetical magnetic charge later on. In black holes, it enters like an additional
type of charge into all expressions in which electric charge appears.

Dvipsbugw
 bl ack holes – falling forever 483

gravity or infinite red-shift, called the static limit, is different from their (outer) horizon.
The region in between is called the ergosphere; this is a misnomer as it is not a sphere.
(It is so called because, as we will see shortly, it can be used to extract energy from the
black hole.) The motion of bodies within the ergosphere can be quite complex. It suffices
to mention that rotating black holes drag any in-falling body into an orbit around them;
this is in contrast to non-rotating black holes, which swallow in-falling bodies. In other
words, rotating black holes are not really ‘holes’ at all, but rather vortices.
The distinction between rotating and non-rotating
black holes also appears in the horizon surface area. The Dvipsbugw
rotation axis
(horizon) surface area A of a non-rotating and uncharged
Challenge 884 ny black hole is obviously related to its mass M by event horizon

16πG 2 2
A= M . (372) ergosphere
c4
static limit

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The relation between surface area and mass for a rotating
and charged black hole is more complex: it is given by
F I G U R E 217 The ergosphere of
¾
’ “
a rotating black hole
8πG 2 J2 c2 Q2
A = 4 M2 1 + 1− 4 2 −
” 4πε 0 GM 2 •
(373)
c M G

where J is the angular momentum and Q the charge. In fact, the relation

A=
8πG
Mr h (374)
c2
is valid for all black holes. Obviously, in the case of an electrically charged black hole, the
rotation also produces a magnetic field around it. This is in contrast with non-rotating
black holes, which cannot have a magnetic field.

Black holes as energy sources

Can one extract energy from a black hole? Roger Penrose has discovered that this is pos-
Ref. 451 sible for rotating black holes. A rocket orbiting a rotating black hole in its ergosphere
could switch its engines on and would then get hurled into outer space at tremendous
Copyright © Christoph Schiller November 1997–May 2006

velocity, much greater than what the engines could have produced by themselves. In fact,
the same effect is used by rockets on the Earth, and is the reason why all satellites orbit
the Earth in the same direction; it would require much more fuel to make them turn the
other way.*
The energy gained by the rocket would be lost by the black hole, which would thus
slow down and lose some mass; on the other hand, there is a mass increases due to the

* And it would be much more dangerous, since any small object would hit such an against-the-stream satel-
Challenge 885 ny lite at about 15.8 km~s, thus transforming the object into a dangerous projectile. In fact, any power wanting
to destroy satellites of the enemy would simply have to load a satellite with nuts or bolts, send it into space
the wrong way, and distribute the bolts into a cloud. It would make satellites impossible for many decades
to come.

Dvipsbugw
 484 iii gravitation and rel ativity • 10. bl ack holes – falling forever

exhaust gases falling into the black hole. This increase always is larger than, or at best
equal to, the loss due to rotation slowdown. The best one can do is to turn the engines on
exactly at the horizon; then the horizon area of the black hole stays constant, and only its
rotation is slowed down.**
º a neutral black hole rotating with its maximum possible angular mo-
As a result, for
mentum, 1 − 1~ 2 = 29.3 % of its total energy can be extracted through the Penrose
Challenge 886 ny process. For black holes rotating more slowly, the percentage is obviously smaller.
For charged black holes, such irreversible energy extraction processes are also possible.
Challenge 887 ny Can you think of a way? Using expression (370), we find that up to 50 % of the mass Dvipsbugw
Challenge 888 ny of a non-rotating black hole can be due to its charge. In fact, in the second part of our
mountain ascent we will encounter an energy extraction process which nature seems to
Page 875 use quite frequently.
The Penrose process allows one to determine how angular momentum and charge
Ref. 452 increase the mass of a black hole. The result is the famous mass–energy relation

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

E2 Q2 J2 c2 Q2 J2 1
M2 = = (m irr + ) 2
+ = (m irr + ) 2
+ (375)
c4 16πε 0 Gm irr 4m irr G 2
2 8πε 0 ρ irr 2 c2
ρ irr

which shows how the electrostatic and the rotational energy enter the mass of a black
hole. In the expression, m irr is the irreducible mass defined as

A(M, Q = 0, J = 0) c 4
2
c2
2
m irr = = Œρ irr ‘ (376)
16π G2 2G

and ρ irr is the irreducible radius.

Detailed investigations show that there is no process which decreases the horizon area,
and thus the irreducible mass or radius, of the black hole. People have checked this in all
ways possible and imaginable. For example, when two black holes merge, the total area
increases. One calls processes which keep the area and energy of the black hole constant
reversible, and all others irreversible. In fact, the area of black holes behaves like the en-
tropy of a closed system: it never decreases. That the area in fact is an entropy was first
Ref. 453 stated in 1970 by Jakob Bekenstein. He deduced that only when an entropy is ascribed to
a black hole, is it possible to understand where the entropy of all the material falling into
it is collected.
The black hole entropy is a function only of the mass, the angular momentum and
Copyright © Christoph Schiller November 1997–May 2006

the charge of the black hole. You might want to confirm Bekenstein’s deduction that the
Challenge 889 ny entropy S is proportional to the horizon area. Later it was found, using quantum theory,
that
A kc 3 A k
S= = 2 . (377)
4 ħG 4 l Pl

This famous relation cannot be deduced without quantum theory, as the absolute value
of entropy, as for any other observable, is never fixed by classical physics alone. We will
Page 877 discuss this expression later on in our mountain ascent.

** It is also possible to extract energy from rotational black holes through gravitational radiation.

Dvipsbugw
 bl ack holes – falling forever 485

If black holes have an entropy, they also must have a temperature. If they have a tem-
perature, they must shine. Black holes thus cannot be black! This was proven by Stephen
Hawking in 1974 with extremely involved calculations. However, it could have been de-
duced in the 1930s, with a simple Gedanken experiment which we will present later
Page 872 on. You might want to think about the issue, asking and investigating what strange con-
sequences would appear if black holes had no entropy. Black hole radiation is a further,
though tiny (quantum) mechanism for energy extraction, and is applicable even to non-
Page 872 rotating, uncharged black holes. The interesting connections between black holes, ther-
modynamics, and quantum theory will be presented in the second part of our mountain Dvipsbugw
Challenge 890 ny ascent. Can you imagine other mechanisms that make black holes shine?

Curiosities and fun challenges about black holes

Tiens, les trous noirs. C’est troublant.*

“ Anonymous

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Black holes have many counter-intuitive properties. We will first have a look at the clas-
Page 879 sical effects, leaving the quantum effects for later on.

**
Following universal gravity, light could climb upwards from the surface of a black hole
and then fall back down. In general relativity, a black hole does not allow light to climb
Challenge 891 ny up at all; it can only fall. Can you confirm this?

**
What happens to a person falling into a black hole? An outside observer gives a clear
answer: the falling person never arrives there since she needs an infinite time to reach the
Challenge 892 ny horizon. Can you confirm this result? The falling person, however, reaches the horizon
Challenge 893 ny in a finite amount of her own time. Can you calculate it?
This result is surprising, as it means that for an outside observer in a universe with
finite age, black holes cannot have formed yet! At best, we can only observe systems that
are busy forming black holes. In a sense, it might be correct to say that black holes do not
exist. Black holes could have existed right from the start in the fabric of space-time. On
the other hand, we will find out later why this is impossible. In short, it is important to
keep in mind that the idea of black hole is a limit concept but that usually, limit concepts
(like baths or temperature) are useful descriptions of nature. Independently of this last
Copyright © Christoph Schiller November 1997–May 2006

issue, we can confirm that in nature, the length-to-mass ratio always satisfies

E 2 .
L 4G
(378)
M c
**
Interestingly, the size of a person falling into a black hole is experienced in vastly different
ways by the falling person and a person staying outside. If the black hole is large, the in-
falling observer feels almost nothing, as the tidal effects are small. The outside observer

* No translation possible.

Dvipsbugw
 486 iii gravitation and rel ativity • 10. bl ack holes – falling forever

observer dense
star

F I G U R E 218 Motion of some light rays

from a dense body to an observer

Dvipsbugw
makes a startling observation: he sees the falling person spread all over the horizon of the
black hole. In-falling, extended bodies cover the whole horizon. Can you explain this fact,
Challenge 894 ny for example by using the limit on length-to-mass ratios?
This strange result will be of importance later on in our exploration, and lead to im-
portant results about the size of point particles.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
An observer near a (non-rotating) black hole, or in fact near any object smaller than 7~4
times its gravitational radius, can even see the complete back side of the object, as shown
Challenge 895 ny in Figure 218. Can you imagine what the image looks like? Note that in addition to the
paths shown in Figure 218, light can also turn several times around the black hole before
reaching the observer! Therefore, such an observer sees an infinite number of images of
the black hole. The resulting formula for the angular size of the innermost image was
Page 395 given above.
In fact, the effect of gravity means that it is possible to observe more than half the
surface of any spherical object. In everyday life, however, the effect is small: for example,
light bending allows us to see about 50.0002 % of the surface of the Sun.
**
A mass point inside the smallest circular path of light around a black hole, at 3R~2, cannot
stay in a circle, because in that region, something strange happens. A body which circles
another in everyday life always feels a tendency to be pushed outwards; this centrifugal
effect is due to the inertia of the body. But at values below 3R~2, a circulating body is
Ref. 454 pushed inwards by its inertia. There are several ways to explain this paradoxical effect. The
simplest is to note that near a black hole, the weight increases faster than the centrifugal
Challenge 896 ny force, as you may want to check yourself. Only a rocket with engines switched on and
Copyright © Christoph Schiller November 1997–May 2006

pushing towards the sky can orbit a black hole at 3R~2.

**
By the way, how can gravity, or an electrical field, come out of a black hole, if no signal
Challenge 897 n and no energy can leave it?
**
Do white holes exist, i.e. time-inverted black holes, in which everything flows out of, in-
Challenge 898 ny stead of into, some bounded region?

**

Dvipsbugw
 bl ack holes – falling forever 487

Challenge 899 ny Show that a cosmological constant Λ leads to the following metric for a black hole:

ds 2 dr 2 r2 2
dτ 2 = = ‹1 − − r  dt 2 −
2GM Λ 2
Λc 2 2
− dφ . (379)
c2 rc 2 3 c2 − 2G M
r
− 3
r c2

Note that this metric does not turn into the Minkowski metric for large values of r.
º
However, in the case that Λ is small, the metric is almost flat for values of r that satisfy
1~ Λ Q r Q 2Gm~c 2 . Dvipsbugw
As a result, the inverse square law is also modified:

Gm c 2 Λ
F=− + r. (380)
r2 6
With the known values of the cosmological constant, the second term is negligible inside
the solar system.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
In quantum theory, the gyromagnetic ratio is an important quantity for any rotating
Challenge 900 ny charged system. What is the gyromagnetic ratio for rotating black holes?
**
A large black hole is, as the name implies, black. Still, it can be seen. If we were to travel
towards it in a spaceship, we would note that the black hole is surrounded by a bright
circle, like a thin halo. The ring at the radial distance of the photon sphere is due to those
photons which come from other luminous objects, then circle the hole, and finally, after
Challenge 901 n one or several turns, end up in our eye. Can you confirm this result?
**
Challenge 902 ny Do moving black holes Lorentz-contract? Black holes do shine a little bit. It is true that
the images they form are complex, as light can turn around them a few times before
reaching the observer. In addition, the observer has to be far away, so that the effects of
curvature are small. All these effects can be taken into account; nevertheless, the question
remains subtle. The reason is that the concept of Lorentz contraction makes no sense in
general relativity, as the comparison with the uncontracted situation is difficult to define
precisely.
Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 903 ny Can you confirm that black holes imply a limit to power? Power is energy change over
time. General relativity limits power to P D c 5 ~4G. In other words, no engine in nature
can provide more than 0.92 ë 1052 W or 1.2 ë 1049 horsepower.

Formation of and search for black holes

How might black holes form? At present, at least three possible mechanisms have been
distinguished; the question is still a hot subject of research. First of all, black holes could
Ref. 455 have formed during the early stages of the universe. These primordial black holes might

Dvipsbugw
 488 iii gravitation and rel ativity • 10. bl ack holes – falling forever

grow through accretion, i.e. through the swallowing of nearby matter and radiation, or
Page 874 disappear through one of the mechanisms to be studied later on.
Of the observed black holes, the so-called supermassive black holes are found at the
centre of every galaxy studied so far. They have masses in the range from 106 to 109 solar
masses and contain about 0.5 % of the mass of a galaxy. They are conjectured to exist at
the centre of all galaxies, and seem to be related to the formation of galaxies themselves.
Supermassive black holes are supposed to have formed through the collapse of large dust
clouds, and to have grown through subsequent accretion of matter. The latest ideas imply
that these black holes accrete a lot of matter in their early stage; the matter falling in Dvipsbugw
emits lots of radiation, which would explain the brightness of quasars. Later on, the rate
of accretion slows, and the less spectacular Seyfert galaxies form. It may even be that the
supermassive black hole at the centre of the galaxy triggers the formation of stars. Still
later, these supermassive black holes become almost dormant, like the one at the centre
of the Milky Way.
Ref. 456 On the other hand, black holes can form when old massive stars collapse. It is estimated

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that when stars with at least three solar masses burn out their fuel, part of the matter
remaining will collapse into a black hole. Such stellar black holes have a mass between
one and a hundred solar masses; they can also continue growing through subsequent
accretion. This situation provided the first ever candidate for a black hole, Cygnus X-1,
which was discovered in 1971.
Recent measurements suggest also the existence of intermediate black holes, with
masses around a thousand solar masses or more; the mechanisms and conditions for
their formation are still unknown.
The search for black holes is a popular sport among astrophysicists. Conceptually, the
simplest way to search for them is to look for strong gravitational fields. But only double
stars allow one to measure gravitational fields directly, and the strongest ever measured
Ref. 457 is 30 % of the theoretical maximum value. Another way is to look for strong gravitational
lenses, and try to get a mass-to-size ratio pointing to a black hole. Still another way is
to look at the dynamics of stars near the centre of galaxies. Measuring their motion, one
can deduce the mass of the body they orbit. The most favoured method to search for
black holes is to look for extremely intense X-ray emission from point sources, using
space-based satellites or balloon-based detectors. If the distance to the object is known,
its absolute brightness can be deduced; if it is above a certain limit, it must be a black hole,
since normal matter cannot produce an unlimited amount of light. This method is being
perfected with the aim of directly observing of energy disappearing into a horizon. This
Copyright © Christoph Schiller November 1997–May 2006

Ref. 458 may in fact have been observed.

To sum up the experimental situation, measurements show that in all galaxies studied
so far – more than a dozen – a supermassive black hole seems to be located at their centre.
The masses vary: the black hole at the centre of our own galaxy has about 2.6 million solar
Ref. 442 masses, while the central black hole of the galaxy M87 has 3000 million solar masses.
About a dozen stellar black holes of between 4 and 20 solar masses are known in the
Ref. 442 rest of our own galaxy; all have been discovered since 1971, when Cygnus X-1 was found.
In the year 2000, intermediate-mass black holes were found. Astronomers are also study-
ing how large numbers of black holes in star clusters behave, how often they collide, and
what sort of measurable gravitational waves these collisions produce. The list of discover-
ies is expected to expand dramatically in the coming years.

Dvipsbugw
 bl ack holes – falling forever 489

Singularities
Solving the equations of general relativity for various initial conditions, one finds that
a cloud of dust usually collapses to a singularity, i.e. to a point of infinite density. The
same conclusion appears when one follows the evolution of the universe backwards in
time. In fact, Roger Penrose and Stephen Hawking have proved several mathematical
theorems on the necessity of singularities for many classical matter distributions. These
theorems assume only the continuity of space-time and a few rather weak conditions on
Ref. 459 the matter in it. The theorems state that in expanding systems such as the universe itself, or Dvipsbugw
in collapsing systems such as black holes in formation, events with infinite matter density
should exist somewhere in the past, or in the future, respectively. This result is usually
summarized by saying that there is a mathematical proof that the universe started in a
singularity.
In fact, the derivation of the initial singularities makes a hidden, but strong assumption
about matter: that dust particles have no proper size. In other words, it is assumed that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

dust particles are singularities. Only with this assumption can one deduce the existence
of initial singularities. However, we have seen that the maximum force principle can be
reformulated as a minimum size principle for matter. The argument that there must have
been an initial singularity of the universe is thus flawed. The experimental situation is
clear: there is overwhelming evidence for an early state of the universe that was extremely
hot and dense; but there is no evidence for infinite temperature or density.
Mathematically inclined researchers distinguish two types of singularities: those with
and without a horizon. The latter ones, the so-called naked singularities, are especially
strange: for example, a toothbrush can fall into a naked singularity and disappear without
leaving any trace. Since the field equations are time invariant, we could thus expect
that every now and then, naked singularities emit toothbrushes. (Can you explain why
Challenge 904 ny ‘dressed’ singularities are less dangerous?)
To avoid the spontaneous appearance of toothbrushes, over the years many people
have tried to discover some theoretical principles forbidding the existence of naked sin-
gularities. It turns out that there are two such principles. The first is the maximum force
or maximum power principle we encountered above. The maximum force implies that
no infinite force values appear in nature; in other words, there are no naked singularities
Ref. 460 in nature. This statement is often called cosmic censorship. Obviously, if general relativity
were not the correct description of nature, naked singularities could still appear. Cosmic
censorship is thus still discussed in research articles. The experimental search for naked
Copyright © Christoph Schiller November 1997–May 2006

singularities has not yielded any success; in fact, there is not even a candidate observation
for the less abstruse dressed singularities. But the theoretical case for ‘dressed’ singular-
ities is also weak. Since there is no way to interact with anything behind a horizon, it is
futile to discuss what happens there. There is no way to prove that behind a horizon a
singularity exists. Dressed singularities are articles of faith, not of physics.
In fact, there is another principle preventing singularities, namely quantum theory.
Whenever we encounter a prediction of an infinite value, we have extended our descrip-
tion of nature to a domain for which it was not conceived. To speak about singularities,
one must assume the applicability of pure general relativity to very small distances and
Page 1042 very high energies. As will become clear in the next two parts of this book, nature does
not allow this: the combination of general relativity and quantum theory shows that it

Dvipsbugw
 490 iii gravitation and rel ativity • 11. d oes space differ from time?

makes no sense to talk about ‘singularities’, nor about what happens ‘inside’ a black hole
Page 1001 horizon. The reason is that time and space are not continuous at very small scales. *

A quiz – is the universe a black hole?

Could it be that we live inside a black hole? Both the universe and black holes have hori-
zons. Interestingly, the horizon distance r 0 of the universe is about

r 0  3ct 0  4 ë 1026 m (381) Dvipsbugw

and its matter content is about

m0  = 72πGρ 0 ct 03 = 6 ë 1026 m
4π 2Gm 0
ρ o r 03 whence (382)
3 c2

for a density of 3 ë 10−27 kg~m3 . Thus we have

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

r0 
2Gm 0
, (383)
c2

Challenge 905 ny which is similar to the black hole relation r S = 2Gm~c 2 . Is this a coincidence? No, it is not:
all systems with high curvature more or less obey this relation. But are we nevertheless
Challenge 906 ny falling into a large black hole? You can answer that question by yourself.

11. d oes space differ from time?

Tempori parce.*

“
People in a bad mood say that time is our master. Nobody says that of space. Time and
space are obviously different in everyday life. But what is the precise difference between
Seneca

”
them in general relativity? And do we need them at all? In general relativity it is assumed
that we live in a (pseudo-Riemannian) space-time of variable curvature. The curvature is
an observable and is related to the distribution and motion of matter and energy in the
way described by the field equations.
Copyright © Christoph Schiller November 1997–May 2006

However, there is a fundamental problem. The equations of general relativity are in-
variant under numerous transformations which mix the coordinates x 0 , x 1 , x 2 and x 3 .

* Many physicists are still wary of making such strong statements at this point; and there are still some who
claim that space and time are continuous even down to the smallest distances. Our discussion of quantum
Page 996 theory, and the first sections of the third part of our mountain ascent, will give the precise arguments leading
to the opposite conclusion.
* ‘Care about time.’ Lucius Annaeus Seneca (c. 4 bce–65), Epistolae 88, 39.

Dvipsbugw
 d oes space differ from time? 491

For example, the viewpoint transformation

x 0′ = x 0 + x 1
x 1′ = −x 0 + x 1
x 2′ = x 2
x 3′ = x 3 (384)

is allowed in general relativity, and leaves the field equations invariant. You might want Dvipsbugw
Challenge 907 ny to search for other examples.
This has a consequence that is clearly in sharp contrast with everyday life: diffeomorph-
ism invariance makes it impossible to distinguish space from time inside general relativity.
More explicitly, the coordinate x 0 cannot simply be identified with the physical time t, as
implicitly done up to now. This identification is only possible in special relativity. In spe-
cial relativity the invariance under Lorentz (or Poincaré) transformations of space and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

time singles out energy, linear momentum and angular momentum as the fundamental
observables. In general relativity, there is no (non-trivial) metric isometry group; con-
sequently, there are no basic physical observables singled out by their characteristic of
being conserved. But invariant quantities are necessary for communication! In fact, we
can talk to each other only because we live in an approximately flat space-time: if the
angles of a triangle did not add up to 180 degrees, there would be no invariant quantities
and we would not be able to communicate.
How have we managed to sweep this problem under the rug so far? We have done so in
several ways. The simplest was to always require that in some part of the situation under
consideration space-time was our usual flat Minkowski space-time, where x 0 can be iden-
tified with t. We can fulfil this requirement either at infinity, as we did around spherical
masses, or in zeroth approximation, as we did for gravitational radiation and for all other
perturbation calculations. In this way, we eliminate the free mixing of coordinates and
the otherwise missing invariant quantities appear as expected. This pragmatic approach
is the usual way out of the problem. In fact, it is used in some otherwise excellent texts
Ref. 416 on general relativity that preclude any deeper questioning of the issue.
A common variation of this trick is to let the distinction ‘sneak’ into the calculations
by the introduction of matter and its properties, or by the introduction of radiation. The
material properties of matter, for example their thermodynamic state equations, always
distinguish between space and time. Radiation does the same, by its propagation. Obvi-
Copyright © Christoph Schiller November 1997–May 2006

ously this is true also for those special combinations of matter and radiation called clocks
and metre bars. In fact, the method of introducing matter is the same as the method of
introducing Minkowski space-time, if one looks closely: properties of matter are always
defined using space-time descriptions of special relativity.*
Another variation of the pragmatic approach is the use of the cosmological time co-
ordinate. An isotropic and homogeneous universe does have a preferred time coordinate,
Page 451, page 616 namely the one used in all the tables on the past and the future of the universe. This
method is in fact a combination of the previous two.
* We note something astonishing here: the inclusion of some condition at small distances (matter) has the
Challenge 908 ny same effect as the inclusion of some condition at infinity. Is this just coincidence? We will come back to this
Page 1051 issue in the third part of our mountain ascent.

Dvipsbugw
 492 iii gravitation and rel ativity • 11. d oes space differ from time?

But we are on a special quest here. We want to understand motion in principle, not
only to calculate it in practice. We want a fundamental answer, not a pragmatic one. And
for this we need to know how the positions x i and time t are connected, and how we
can define invariant quantities. The question also prepares us for the task of combining
gravity with quantum theory, which will be the goal of the third part of our mountain
ascent.
A fundamental solution requires a description of clocks together with the system un-
der consideration, and a deduction of how the reading t of a clock relates to the behaviour
of the system in space-time. But we know that any description of a system requires meas- Dvipsbugw
urements: for example, in order to determine the initial conditions. And initial conditions
require space and time. We thus enter a vicious circle: that is precisely what we wanted
to avoid in the first place.
A suspicion arises. Is there in fact a fundamental difference between space and time?
Let us take a tour of various ways to investigate this question.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Can space and time be measured?
In order to distinguish between space and time in general relativity, we must be able to
Page 265 measure them. But already in the section on universal gravity we have mentioned the im-
possibility of measuring lengths, times and masses with gravitational effects alone. Does
this situation change in general relativity? Lengths and times are connected by the speed
of light, and in addition lengths and masses are connected by the gravitational constant.
Despite this additional connection, it takes only a moment to convince oneself that the
problem persists.
In fact, we need electrodynamics to solve it. It is only be using the elementary charge e
Ref. 461 that can we form length scales, of which the simplest one is
º
l emscale = º  1.4 ë 10−36 m .
e G
(385)
4πε 0 c2

Page 523 Here, ε 0 the permittivity of free space. Alternatively, we can argue that quantum mechan-
ics provides a length scale, since one can use the quantum of action ħ to define the length
scale ¾
l qtscale =  1.6 ë 10−35 m ,
ħG
(386)
c3
Copyright © Christoph Schiller November 1997–May 2006

which is called the Planck length or Planck’s natural length unit. However, this does not
change the argument, because one needs electrodynamics to measure the value of ħ. The
equivalence of the two arguments is shown by rewriting the elementary charge e as a
combination of nature’s fundamental constants:
¼
e = 4πε 0 cħα . (387)

Here, α  1~137.06 is the fine-structure constant that characterizes the strength of electro-

Dvipsbugw
 d oes space differ from time? 493

magnetism. In terms of α, expression (385) becomes

¾
αħG º
l scale = = α l Pl . (388)
c3

Summing up, every length measurement is based on the electromagnetic coupling con-
stant α and on the Planck length. Of course, the same is true for time and mass meas-
Challenge 909 e urements. There is thus no way to define or measure lengths, times and masses using Dvipsbugw
gravitation or general relativity only.*
Given this sobering result, we can ask whether in general relativity space and time are
really required at all.

Are space and time necessary?

Ref. 463 Robert Geroch answers this question in a beautiful five-page article. He explains how to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

formulate the general theory of relativity without the use of space and time, by taking as
starting point the physical observables only.
He starts with the set of all observables. Among them there is one, called v, which
stands out. It is the only observable which allows one to say that for any two observables
a 1 , a 2 there is a third one a 3 , for which

(a 3 − v) = (a 1 − v) + (a 2 − v) . (389)

Such an observable is called the vacuum. Geroch shows how to use such an observable to
construct the derivatives of observables. Then the so-called Einstein algebra can be built,
which comprises the whole of general relativity.
Usually one describes motion by deducing space-time from matter observables, by
calculating the evolution of space-time, and then by deducing the motion of matter fol-
lowing from it. Geroch’s description shows that the middle step, and thus the use of space
and time, is not necessary.
Indirectly, the principle of maximum force makes the same statement. General relativ-
ity can be derived from the existence of limit values for force or power. Space and time
are only tools needed to translate this principle into consequences for real-life observers.
Thus, it is possible to formulate general relativity without the use of space and time.
Since both are unnecessary, it seems unlikely that there should be a fundamental differ-
Copyright © Christoph Schiller November 1997–May 2006

ence between them. Nevertheless, on difference is well-known.

Do closed timelike curves exist?

Is it possible that the time coordinate behaves, at least in some regions, like a torus? When
we walk, we can return to the point of departure. Is it possible, to come back in time
to where we have started? The question has been studied in great detail. The standard
Ref. 422 reference is the text by Hawking and Ellis; they list the required properties of space-time,

Ref. 462 * In the past, John Wheeler used to state that his geometrodynamic clock, a device which measures time
by bouncing light back and forth between two parallel mirrors, was a counter-example; that is not correct,
Challenge 910 n however. Can you confirm this?

Dvipsbugw
 494 iii gravitation and rel ativity • 11. d oes space differ from time?

x Dvipsbugw

F I G U R E 219 A ‘hole’ in space

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

explaining which are mutually compatible or exclusive. They find, for example, that space-
times which are smooth, globally hyperbolic, oriented and time-oriented do not contain
any such curves. It is usually assumed that the observed universe has these properties, so
that observation of closed timelike curves is unlikely. Indeed, no candidate has ever been
suggested. Later on, we will show that searches for such curves at the microscopic scale
Page 883 have also failed to find any.
The impossibility of closed timelike curves seems to point to a difference between
space and time. But in fact, this difference is only apparent. All these investigations are
based on the behaviour of matter. Thus these arguments assume a specific distinction
between space and time right from the start. In short, this line of enquiry cannot help us
to decide whether space and time differ. Let us look at the issue in another way.

Is general relativity local? – The hole argument

When Albert Einstein developed general relativity, he had quite some trouble with diffeo-
morphism invariance. Most startling is his famous hole argument, better called the hole
paradox. Take the situation shown in Figure 219, in which a mass deforms the space-time
around it. Einstein imagined a small region of the vacuum, the hole, which is shown as
a small ellipse. What happens if we somehow change the curvature inside the hole while
Ref. 464 leaving the situation outside it unchanged, as shown in the inset of the picture?
On the one hand, the new situation is obviously physically different from the original
Copyright © Christoph Schiller November 1997–May 2006

one, as the curvature inside the hole is different. This difference thus implies that the
curvature outside a region does not determine the curvature inside it. That is extremely
unsatisfactory. Worse, if we generalize this operation to the time domain, we seem to get
the biggest nightmare possible in physics: determinism is lost.
On the other hand, general relativity is diffeomorphism invariant. The deformation
shown in the figure is a diffeomorphism; so the new situation must be physically equival-
ent to the original situation.
Which argument is correct? Einstein first favoured the first point of view, and therefore
dropped the whole idea of diffeomorphism invariance for about a year. Only later did he
understand that the second assessment is correct, and that the first statement makes a
fundamental mistake: it assumes an independent existence of the coordinate axes x and y,

Dvipsbugw
 d oes space differ from time? 495

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 220 A model of the hollow Earth theory (© Helmut Diehl)

as shown in the figure. But during that deformation, the coordinates x and y automatically
change as well, so that there is no physical difference between the two situations.
The moral of the story is that there is no difference between space-time and gravitational
field. Space-time is a quality of the field, as Einstein put it, and not an entity with a sep-
arate existence, as suggested by the graph. Coordinates have no physical meaning; only
distances (intervals) in space and time have one. In particular, diffeomorphism invari-
ance proves that there is no flow of time. Time, like space, is only a relational entity: time
and space are relative; they are not absolute.
The relativity of space and time has practical consequences. For example, it turns out
that many problems in general relativity are equivalent to the Schwarzschild situation,
even though they appear completely different at first sight. As a result, researchers have
‘discovered’ the Schwarzschild solution (of course with different coordinate systems) over
twenty times, often thinking that they had found a new, unknown solution. We will now
Copyright © Christoph Schiller November 1997–May 2006

discuss a startling consequence of this insight.

Is the Earth hollow?

The hollow Earth hypothesis, i.e. the conjecture that we live on the inside of a sphere, was
popular in paranormal circles around the year 1900, and still remains so among certain
Page 56 eccentrics today, especially in Britain, Germany and the US. They maintain, as illustrated
in Figure 220, that the solid Earth encloses the sky, together with the Moon, the Sun and
the stars. Most of us are fooled by education into another description, because we are
brought up to believe that light travels in straight lines. Get rid of this wrong belief, they
say, and the hollow Earth appears in all its glory.
Interestingly, the reasoning is correct. There is no way to disprove this sort of de-

Dvipsbugw
 496 iii gravitation and rel ativity • 12. a summary for the l ayman

scription of the universe. In fact, as the great Austrian physicist Roman Sexl used to
Ref. 465 explain, the diffeomorphism invariance of general relativity even proclaims the equival-
ence between the two views. The fun starts when either of the two camps wants to tell
the other that only its own description can be correct. You might check that any such
argument is wrong; it is fun to slip into the shoes of such an eccentric and to defend the
Challenge 911 e hollow Earth hypothesis against your friends. It is easy to explain the appearance of day
and night, of the horizon, and of the satellite images of the Earth. It is easy to explain what
happened during the flight to the Moon. You can drive many bad physicists crazy in this
way. The usual description and the hollow Earth description are exactly equivalent. Can Dvipsbugw
you confirm that even quantum theory, with its introduction of length scales into nature,
Challenge 912 n does not change this situation?
Such investigations show that diffeomorphism invariance is not an easy symmetry to
swallow. But it is best to get used to it now, as the rest of our adventure will throw up
even more surprises. Indeed, in the third part of our walk we will discover that there is
an even larger symmetry of nature that is similar to the change in viewpoint from the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

hollow Earth view to the standard view. This symmetry, space-time duality, is valid not
only for distances measured from the centre of the Earth, but for distances measured
Page 1051 from any point in nature.

Are space, time and mass independent?

We can conclude from this short discussion that there is no fundamental distinction
between space and time in general relativity. Pragmatic distinctions, using matter, radi-
ation or space-time at infinity, are the only possible ones.
In the third part of our adventure we will discover that even the inclusion of quantum
theory is consistent with this view. We will show explicitly that no distinction is possible
in principle. We will discover that mass and space-time are on an equal footing, and that,
Page 1013 in a sense, particles and vacuum are made of the same substance. Distinctions between
space and time turn out to be possible only at low, everyday energies.
Page 157 In the beginning of our mountain ascent we found that we needed matter to define
space and time. Now we have found that we even need matter to distinguish between
space and time. Similarly, in the beginning we found that space and time are required to
Page 433 define matter; now we have found that we even need flat space-time to define it.
In summary, general relativity does not provide a way out of the circular reasoning
we discovered in Galilean physics. Indeed, it makes the issue even less clear than before.
Copyright © Christoph Schiller November 1997–May 2006

Continuing the mountain ascent is really worth the effort.

12. general rel ativit y in ten p oints – a summary

for the l ayman
Sapientia felicitas.*

“ ”
* ‘Wisdom is happiness.’ This is the motto of Oxford University.

Dvipsbugw
general rel ativity in ten points – a summary for the l ayman 497

General relativity is the final description of paths of motion, or if one prefers, of mac-
roscopic motion. General relativity describes how the observations of motion of any two
observers are related to each other; it also describes motion due to gravity. In fact, general
relativity is based on the following observations:
— All observers agree that there is a ‘perfect’ speed in nature, namely a common max-
imum energy speed relative to matter. This speed is realized by massless radiation,
such as light or radio signals.
— All observers agree that there is a ‘perfect’ force in nature, a common maximum force Dvipsbugw
that can be realized or measured by realistic observers. This force is realized on event
horizons.
These two statements contain the full theory of relativity. From them we deduce:
— Space-time consists of events in 3 + 1 continuous dimensions, with a variable curvature.
The curvature can be deduced from distance measurements among events or from

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tidal effects. We thus live in a pseudo-Riemannian space-time. Measured times,
lengths and curvatures vary from observer to observer.
— Space-time and space are curved near mass and energy. The curvature at a point is de-
termined by the energy–momentum density at that point, and described by the field
equations. When matter and energy move, the space curvature moves along with them.
A built-in delay in this movement renders faster-than-light transport of energy im-
possible. The proportionality constant between energy and curvature is so small that
the curvature is not observed in everyday life; only its indirect manifestation, namely
gravity, is observed.
— Space is also elastic: it prefers being flat. Being elastic, it can oscillate independently of
matter; one then speaks of gravitational radiation or of gravity waves.
— Freely falling matter moves along geodesics, i.e. along paths of maximal length in
curved space-time; in space this means that light bends when it passes near large
masses by twice the amount predicted by universal gravity.
— To describe gravitation one needs curved space-time, i.e. general relativity, at the latest
whenever distances are of the order of the Schwarzschild radius r S = 2Gm~c 2 . When
distances are much larger than this value, the relativistic description with gravity and
gravitomagnetism (frame-dragging) is sufficient. When distances are even larger, the
description by universal gravity, namely a = Gm~r 2 , together with flat Minkowski
space-time, will do as a first approximation.
Copyright © Christoph Schiller November 1997–May 2006

— Space and time are not distinguished globally, but only locally. Matter is required to
make the distinction.
In addition, all the matter and energy we observe in the sky lead us to the following
conclusions:
— On the cosmological scale, everything moves away from everything else: the universe
is expanding. This expansion of space-time is described by the field equations.
— The universe has a finite age; this is the reason for the darkness of the sky at night. A
horizon limits the measurable space-time intervals to about fourteen thousand million
years.

Dvipsbugw
 498 iii gravitation and rel ativity • 12. a summary for the l ayman

The accuracy of the description

Was general relativity worth the effort? The discussion of its accuracy is most conveniently
Ref. 466 split into two sets of experiments. The first set consists of measurements of how matter
moves. Do objects really follow geodesics? As summarized in Table 40, all experiments
Ref. 467 agree with the theory to within measurement errors, i.e. at least within 1 part in 1012 . In
short, the way matter falls is indeed well described by general relativity.
The second set of measurements concerns the dynamics of space-time itself. Does
space-time move following the field equations of general relativity? In other words, is Dvipsbugw
space-time really bent by matter in the way the theory predicts? Many experiments have
been performed, near to and far from Earth, in both weak and strong fields. All agree
Ref. 466, Ref. 467 with the predictions to within measurement errors. However, the best measurements so
far have only about 3 significant digits. Note that even though numerous experiments
have been performed, there are only few types of tests, as Table 40 shows. The discovery
Challenge 913 ny of a new type of experiment almost guarantees fame and riches. Most sought after, of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

course, is the direct detection of gravitational waves.
Another comment on Table 40 is in order. After many decades in which all measured
effects were only of the order v 2 ~c 2 , several so-called strong field effects in pulsars allowed
Ref. 466 us to reach the order v 4 ~c 4 . Soon a few effects of this order should also be detected even
inside the solar system, using high-precision satellite experiments. The present crown of
all measurements, the gravity wave emission delay, is the only v 5 ~c 5 effect measured so
Page 407 far.
The difficulty of achieving high precision for space-time curvature measurements is
the reason why mass is measured with balances, always (indirectly) using the prototype
kilogram in Paris, instead of defining some standard curvature and fixing the value of
G. Indeed, no useful terrestrial curvature experiment has ever been carried out. A break-
through in this domain would make the news. The terrestrial curvature methods cur-
rently available would not even allow one to define a kilogram of gold or of oranges with
a precision of a single kilogram!
Another way to check general relativity is to search for alternative descriptions of
gravitation. Quite a number of alternative theories of gravity have been formulated and
Ref. 467, Ref. 470 studied, but so far, only general relativity is in agreement with all experiments.
In summary, as Thibault Damour likes to explain, general relativity is at least
99.999 999 999 9 % correct concerning the motion of matter and energy, and at least
Ref. 466 99.9 % correct about the way matter and energy curve and move space-time. No excep-
Copyright © Christoph Schiller November 1997–May 2006

tions, no anti-gravity and no unclear experimental data are known. All motion on Earth
and in the skies is described by general relativity. The importance of Albert Einstein’s
achievement cannot be understated.
We note that general relativity has not been tested for microscopic motion. In this
context, microscopic motion is any motion for which the action is around the quantum of
action, namely 10−34 Js. This issue is central to the third and last part of our adventure.

Research in general relativity and cosmology

Ref. 471 Research in general relativity is more intense than ever.*

Dvipsbugw
 general rel ativity in ten points – a summary for the l ayman 499

TA B L E 40 Types of tests of general relativity

Measured effect Con- Ty pe Refer-

firma- ence
tion
Equivalence principle 10−12 motion of matter Ref. 351,
Ref. 466,
Ref. 468
2
1~r dependence (dimensionality of space-time) 10 −10
motion of matter Ref. 469
Dvipsbugw
Time independence of G 10−19 ~s motion of matter Ref. 466
Red-shift (light and microwaves on Sun, Earth, 10−4 space-time curvature Ref. 329,
Sirius) Ref. 328,
Ref. 466
Perihelion shift (four planets, Icarus, pulsars) 10−3 space-time curvature Ref. 466
Light deflection (light, radio waves around Sun, 10−3 space-time curvature Ref. 466

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

stars, galaxies)
Time delay (radio signals near Sun, near pulsars)10−3 space-time curvature Ref. 466
Gravitomagnetism (Earth, pulsar) 10−1 space-time curvature Ref. 353
Geodesic effect (Moon, pulsars) 10−1 space-time curvature Ref. 372,
Ref. 466
Gravity wave emission delay (pulsars) 10−3 space-time curvature Ref. 466

**
The most interesting experimental studies at present are those of double pulsars, the
search for gravitational waves and various dedicated satellites; among others a special
satellite will capture all possible pulsars of the galaxy. All these experiments will allow
experimental tests in domains that have not been accessible up to now.

**
The description of collisions and many-body problems, invloving stars, neutron stars and
black holes helps astrophysicists to improve their understanding of the rich behaviour
Ref. 472 they observe in their telescopes.
**
Copyright © Christoph Schiller November 1997–May 2006

The study of the early universe and of elementary particle properties, with phenomena
such as inflation, a short period of accelerated expansion during the first few seconds, is
Ref. 473 still an important topic of investigation.
**
The study of chaos in the field equations is of fundamental interest in the study of the
early universe, and may be related to the problem of galaxy formation, one of the biggest
Ref. 474 open problems in physics.

* There is even a free and excellent internet-based research journal, called Living Reviews in Relativity, to be
found at the http://www.livingreviews.org website.

Dvipsbugw
 500 iii gravitation and rel ativity • 12. a summary for the l ayman

**
Gathering data about galaxy formation is the main aim of many satellite systems and
purpose-build telescopes. The main focus is the search for localized cosmic microwave
Ref. 475 background anisotropies due to protogalaxies.
**
The determination of the cosmological parameters, such as the matter density, the
Ref. 418 curvature and the vacuum density, is a central effort of modern astrophysics.
Dvipsbugw
**
Astrophysicists regularly discover new phenomena in the skies. For example, the vari-
ous types of gamma-ray bursts, X-ray bursts and optical bursts are still not completely
Ref. 476 understood. Gamma-ray bursts, for example, can be as bright as 1017 sun-like stars com-
bined; however, they last only a few seconds. More details on this research are given later
Page 875 on.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
A computer database of all solutions of the field equations is being built. Among other
Ref. 477 things, researchers are checking whether they really are all different from each other.
**
The inclusion of torsion in the field equations, a possible extension of the theory, is one of
Ref. 478 the more promising attempts to include particle spin in general relativity. The inclusion
of torsion in general relativity does not require new fundamental constants; indeed, the
absence of torsion was assumed in the Raychaudhuri equation used by Jacobson. The
Ref. 479 use of the extended Raychaudhuri equation, which includes torsion, should allow one to
deduce the full Einstein–Cartan theory from the maximum force principle. This issue is
a topic for future research.
**
Solutions with non-trivial topology, such as wormholes and particle-like solutions, con-
Ref. 480 stitue a fascinating field of enquiry, related to string theory.
**
Other formulations of general relativity, describing space-time with quantities other than
Copyright © Christoph Schiller November 1997–May 2006

the metric, are continuously being developed, in the hope of clarifying its relationship to
Ref. 481 the quantum world. The so-called Ashtekar variables are such a modern description.

**
The unification of quantum physics and general relativity, the topic of the third part of
Ref. 482 this mountain ascent, will occupy researchers for many years to come.
**
Finally, the teaching of general relativity, which for many decades has been hidden behind
Greek indices, differential forms and other antididactic methods, will benefit greatly from
Ref. 483 future improvements focusing more on the physics and less on the formalism.

Dvipsbugw
 general rel ativity in ten points – a summary for the l ayman 501

In short, general relativity is still an extremely interesting field of research and import-
ant discoveries are still expected.

Could general relativity be different?

The constant of gravitation provides a limit for the density and the acceleration of objects,
as well as for the power of engines. We based all our deductions on its invariance. Is it
possible that the constant of gravitation G changes from place to place or that it changes
with time? The question is tricky. At first sight, the answer is a loud: ‘Yes, of course! Just Dvipsbugw
see what happens when the value of G is changed in formulae.’ However, this answer is
Page 340 wrong, as it was wrong for the speed of light c.
Since the constant of gravitation enters into our definition of gravity and acceleration,
and thus, even if we do not notice it, into the construction of all rulers, all measurement
standards and all measuring set-ups, there is no way to detect whether its value actu-
Challenge 914 ny ally varies. No imaginable experiment could detect a variation. Every measurement of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

force is, whether we like it or not, a comparison with the limit force. There is no way, in
principle, to check the invariance of a standard. This is even more astonishing because
Page 499 measurements of this type are regularly reported, as in Table 40. But the result of any
such experiment is easy to predict: no change will ever be found.
Could the number of space dimension be different from 3? This issue is quite involved.
For example, three is the smallest number of dimensions for which a vanishing Ricci
tensor is compatible with non-vanishing curvature. On the other hand, more than three
dimensions give deviations from the inverse square ‘law’ of gravitation. So far, there are
no data pointing in this direction.
Could the equations of general relativity be different? Despite their excellent fit with
experiment, there is one issue that still troubles some people. The rotation speed of matter
far from the centre of galaxies does not seem to be consistent with the inverse square
dependence. There could be many reasons for this effect, and a change in the equations
for large distances might be one of them. This issue is still open.
Theoreticians have explored many alternative theories, such as scalar-tensor theories,
theories with torsion, or theories which break Lorentz invariance. However, none of the
alternative theories proposed so far seem to fit experimental data.

The limits of general relativity

Despite its successes, the description of motion presented so far is unsatisfactory; maybe
Copyright © Christoph Schiller November 1997–May 2006

Challenge 915 e you already have some gut feeling about certain unresolved issues.
First of all, even though the speed of light is the starting point of the whole theory, we
still do not know what light actually is. This will be our next topic.
Secondly, we have seen that everything falls along geodesics. But a mountain does not
fall. Somehow the matter below prevents it from falling. How? And where does mass
come from anyway? What is mass? What is matter? General relativity does not provide
an answer; in fact, it does not describe matter at all. Einstein used to say that the left-hand
side of the field equations, describing the curvature of space-time, was granite, while the
right-hand side, describing matter, was sand. Indeed, at this point we still do not know
what matter and mass are. As already remarked, to change the sand into rock we first
need quantum theory and then, in a further step, its unification with relativity. This is the

Dvipsbugw
 502 iii gravitation and rel ativity • 12. a summary for the l ayman

programme for the rest of our adventure.

We have also seen that matter is necessary to clearly distinguish between space and
time, and in particular, to understand the working of clocks, metre bars and balances. In
particular, one question remains: why are there units of mass, length and time in nature
at all? This deep question will also be addressed in the following chapter.
Finally, we know little about the vacuum. We need to understand the magnitude of
the cosmological constant and the number of space-time dimensions. Only then can we
answer the simple question: Why is the sky so far away? General relativity does not help
here. Worse, the smallness of the cosmological constant contradicts the simplest version Dvipsbugw
of quantum theory; this is one of the reasons why we still have quite some height to scale
Page 996 before we reach the top of Motion Mountain.
In short, to describe motion well, we need a more precise description of light, of matter
and of the vacuum. In other words, we need to know more about everything we know.
Page 840 Otherwise we cannot hope to answer questions about mountains, clocks and stars. In a
sense, it seems that we have not achieved much. Fortunately, this is not true. We have

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

learned so much that for the following topic we are forced to go backwards, to situations
without gravity, i.e. back to the framework of special relativity. That is the next, middle
section of our mountain ascent. Despite this simplification to flat space-time, a lot of fun
awaits us there.

It’s a good thing we have gravity, or else when

“ birds died they’d just stay right up there.

Hunters would be all confused.
Steven Wright, comedian.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
bibliography 503

Biblio graphy
A man will turn over half a library to make one

“ book.
Samuel Johnson*

”
323 The simplest historical source is Albert Einstein, Sitzungsberichte der Preussischen
Akademie der Wissenschaften II pp. 844–846, 1915. It is the first explanation of the general
theory of relativity, in only three pages. The theory is then explained in detail in the famous Dvipsbugw
article Albert Einstein, Die Grundlage der allgemeinen Relativitätstheorie, Annalen
der Physik 49, pp. 769–822, 1916. The historic references can be found in German and Eng-
lish in John Stachel, ed., The Collected Papers of Albert Einstein, Volumes 1–9, Princeton
University Press, 1987–2004.
Below is a selection of English-language textbooks for deeper study, in ascending order
of depth and difficulty:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

— An entertaining book without any formulae, but nevertheless accurate and detailed, is
the paperback by Igor Novikov, Black Holes and the Universe, Cambridge University
Press, 1990.
— Almost no formulae, but loads of insight, are found in the enthusiastic text by John A.
Wheeler, A Journey into Gravity and Spacetime, W.H. Freeman, 1990.
— An excellent didactical presentation is Edwin F. Taylor & John A. Wheeler, Ex-
ploring Black Holes: Introduction to General Relativity, Addison Wesley Longman, 2000.
— Beauty, simplicity and shortness are the characteristics of Malcolm Ludvigsen, Gen-
eral Relativity, a Geometric Approach, Cambridge University Press, 1999.
— Good explanation is the strength of Bernard Schu tz, Gravity From the Ground Up,
Cambridge University Press, 2003.
— A good overview of experiments and theory is given in James Foster & J.D. Night-
ingale, A Short Course in General Relativity, Springer Verlag, 2nd edition, 1998.
— A pretty text is Sam Lilley, Discovering Relativity for Yourself, Cambridge University
Press, 1981.
— A modern text is by R ay d’Inverno Introducing Einstein’s Relativity, Clarendon Press,
1992. It includes an extended description of black holes and gravitational radiation, and
regularly refers to present research.
— A beautiful, informative and highly recommended text is Hans C. Ohanian & R emo
Ruffini, Gravitation and Spacetime, W.W. Norton & Co., 1994.
— A well written and modern book, with emphasis on the theory, by one of the great mas-
ters of the field is Wolf gang R indler, Relativity – Special, General and Cosmological,
Copyright © Christoph Schiller November 1997–May 2006

Oxford University Press, 2001.

— A classic is Steven Weinberg, Gravitation and Cosmology, Wiley, 1972.
— The passion of general relativity can be experienced also in John Kl auder, ed., Magic
without Magic: John Archibald Wheeler – A Collection of Essays in Honour of His Sixtieth
Birthday, W.H. Freeman & Co., 1972.
— An extensive text is Kip S. Thorne, Black Holes and Time Warps – Einstein’s Outrageous
Legacy, W.W. Norton, 1994.
— The most mathematical – and toughest – text is Robert M. Wald, General Relativity,
University of Chicago Press, 1984.

* Samuel Johnson (1709–1784), famous English poet and intellectual.

Dvipsbugw
504 iii gravitation and rel ativity

— Much information about general relativity is available on the internet. As a good starting
point for US-American material, see the http://math.ucr.edu/home/baez/relativity.html
website.

There is still a need for a large and modern textbook on general relativity, with colour
material, that combines experimental and theoretical aspects.
For texts in other languages, see the next reference. Cited on pages 377, 410, 411, 430,
and 431.
324 A beautiful German teaching text is the classic G. Falk & W. Ruppel, Mechanik, Relativ- Dvipsbugw
ität, Gravitation – ein Lehrbuch, Springer Verlag, third edition, 1983.
A practical and elegant booklet is Ulrich E. Schröder, Gravitation – Einführung in
die allgemeine Relativitätstheorie, Verlag Harri Deutsch, Frankfurt am Main, 2001.
A modern reference is Torsten Fliessbach, Allgemeine Relativitätstheorie,
Akademischer Spektrum Verlag, 1998.
Excellent is Hubert Goenner, Einführung in die spezielle und allgemeine Relativität-
stheorie, Akademischer Spektrum Verlag, 1996.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In Italian, there is the beautiful, informative, but expensive Hans C. Ohanian &
R emo Ruffini, Gravitazione e spazio-tempo, Zanichelli, 1997. It is highly recommended.
Cited on pages 377, 407, 408, 410, 411, 430, 431, and 507.
325 P. Mohazzabi & J.H. Shea, High altitude free fall, American Journal of Physics 64,
pp. 1242–1246, 1996. As a note, due to a technical failure Kittinger had his hand in (near)
vacuum during his ascent, without incurring any permanent damage. On the consequences
of human exposure to vacuum, see the http://www.sff.net/people/geoffrey.landis/vacuum.
html website. Cited on page 377.
326 This story is told e.g. by W.G. Unruh, Time, gravity, and quantum mechanics, preprint
available at http://www.arxiv.org/abs/gr-qc/9312027. Cited on page 378.
327 H. B ondi, Gravitation, European Journal of Physics 14, pp. 1–6, 1993. Cited on page 379.
328 J.W. Brault, Princeton University Ph.D. thesis, 1962. See also J.L. Snider, Physical Re-
view Letters 28, pp. 853–856, 1972, and for the star Sirius see J.L. Greenstein & al., As-
trophysical Journal 169, p. 563, 1971. Cited on pages 379 and 499.
329 The famous paper is R.V. Pound & G.A. R ebka, Apparent weight of photons, Physical Re-
view Letters 4, pp. 337–341, 1960. A higher-precision version was published by R.V. Pound
& J.L. Snider, Physical Review Letters 13, p. 539, 1964, and R.V. Pound & J.L. Snider,
Physical Review B 140, p. 788, 1965. Cited on pages 380 and 499.
330 J.C. Hafele & R ichard E. Keating, Around-the-world atomic clocks: predicted re-
lativistic time gains, Science 177, pp. 166–167, and Around-the-world atomic clocks: ob-
Copyright © Christoph Schiller November 1997–May 2006

served relativistic time gains, pp. 168–170, 14 July 1972. Cited on page 380.
331 R.F.C. Vessot & al., Test of relativistic gravitation with a space-borne hydrogen maser,
Physical Review Letters 45, pp. 2081–2084, 1980. The experiment was performed in 1976;
there are more than a dozen co-authors involved in this work, which involved shooting a
maser into space with a scout missile to a height of c. 10 000 km. Cited on page 380.
332 L. Briatore & S. Leschiu tta, Evidence for Earth gravitational shift by direct atomic-
time-scale comparison, Il Nuovo Cimento 37B, pp. 219–231, 1977. Cited on page 380.
333 More information about tides can be found in E.P. Cl ancy, The Tides, Doubleday, New
York, 1969. Cited on page 381.
334 The expeditions had gone to two small islands, namely to Sobral, north of Brazil, and to Prin-
cipe, in the gulf of Guinea. The results of the expedition appeared in The Times before they

Dvipsbugw
bibliography 505

appeared in a scientific journal. Today this would be seen as a gross violation of scientific
honesty. The results were published as F.W. Dyson, A.S. Eddington & C. Davidson,
Philosophical Transactions of the Royal Society (London) 220A, p. 291, 1920, and Memoirs of
the Royal Astronomical Society 62, p. 291, 1920. Cited on page 383.
335 A good source for images of space-time is the text by G.F.R. Ellis & R. Williams, Flat
and Curved Space-times, Clarendon Press, Oxford, 1988. Cited on page 383.
336 J. Droste, Het veld van een enkel centrum in Einstein’s theorie der zwaartekracht, en de
beweging van een stoffelijk punt, Verslag gew. Vergad. Wiss. Amsterdam 25, pp. 163–180, 1916.
Cited on page 385. Dvipsbugw
337 The name black hole was introduced in 1967 at a pulsar conference, as described in his autobi-
ography by John A. Wheeler, Geons, Black Holes, and Quantum Foam: A Life in Physics,
W.W. Norton, 1998, pp. 296–297: ‘In my talk, I argued that we should consider the possibil-
ity that at the center of a pulsar is a gravitationally completely collapsed object. I remarked
that one couldn’t keep saying “gravitationally completely collapsed object” over and over.
One needed a shorter descriptive phrase. “How about black hole?” asked someone in the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

audience. I had been searching for just the right term for months, mulling it over in bed, in
the bathtub, in my car, whenever I had quiet moments. Suddenly, this name seemed exactly
right. When I gave a more formal ... lecture ... a few weeks later on, on December 29, 1967,
I used the term, and then included it into the written version of the lecture published in
the spring of 1968 ... I decided to be casual about the term ”black hole”, dropping it into the
lecture and the written version as if it were an old familiar friend. Would it catch on? Indeed
it did. By now every schoolchild has heard the term.’
The widespread use of the term began with the article by R. Ruffini & J.A. Wheeler,
Introducing the black hole, Physics Today 24, pp. 30–41, January 1971.
In his autobiography, Wheeler also writes that the expression ‘black hole has no hair’ was
criticized as ‘obscene’ by Feynman. An interesting comment by a physicist who used to write
his papers in topless bars. Cited on pages 385, 476, 477, and 482.
338 L.B. Kreuzer, Experimental measurement of the equivalence of active and passive gravit-
ational mass, Physical Review 169, pp. 1007–1012, 1968. With a clever experiment, he showed
that the gravitational masses of fluorine and of bromine are equal. Cited on page 386.
339 A good and accessible book on the topic is David Bl air & Geoff McNamara, Ripples
on a cosmic sea, Allen & Unwin, 1997. Cited on page 386.
340 G.W. Gibbons, The maximum tension principle in general relativity, Foundations of Phys-
ics 32, pp. 1891–1901, 2002, or http://www.arxiv.org/abs/hep-th/0210109. Cited on page 387.
341 That bodies fall along geodesics has been checked by ... Cited on page 389.
342 So far, the experiments confirm that electrostatic and (strong) nuclear energy fall like mat-
Copyright © Christoph Schiller November 1997–May 2006

ter to within one part in 108 , and weak (nuclear) energy to within a few per cent. This is
summarized in Ref. 346. Cited on page 390.
343 J. Soldner, Berliner Astronomisches Jahrbuch auf das Jahr 1804, 1801, p. 161. Cited on page
390.
344 See for example K.D. Olum, Superluminal travel requires negative energies, Physical Re-
view Letters 81, pp. 3567–3570, 1998, or M. Alcubierre, The warp drive: hyper-fast travel
within general relativity, Classical and Quantum Gravity 11, pp. L73–L77, 1994. See also
Chris Van Den Broeck, A warp drive with more reasonable total energy requirements,
Classical and Quantum Gravity 16, pp. 3973–3979, 1999. Cited on page 393.
345 See the Astronomical Almanac, and its Explanatory Supplement, H.M. Printing Office, Lon-
don and U.S. Government Printing Office, Washington, 1992. For the information about

Dvipsbugw
506 iii gravitation and rel ativity

various time coordinates used in the world, such as barycentric coordinate time, the time
at the barycentre of the solar system, see also the http://tycho.usno.navy.mil/systime.html
web page. It also contains a good bibliography. Cited on page 393.
346 An overview is given in C. Will, Theory and Experiment in Gravitational Physics, chapter
14.3, Cambridge University Press, revised edition, 1993. (Despite being a standard reference,
his view the role of tides and the role of gravitational energy within the principle of equi-
valence has been criticised by other researchers.) See also C. Will, Was Einstein Right? –
Putting General Relativity to the Test, Oxford University Press, 1993. See also his paper http://
www.arxiv.org/abs/gr-qc/9811036. Cited on pages 394, 410, and 505. Dvipsbugw
347 The calculation omits several smaller effects, such as rotation of the Earth and red-shift. For
the main effect, see Edwin F. Taylor, ‘The boundaries of nature: special and general
relativity and quantum mechanics, a second course in physics’ – Edwin F. Taylor’s accept-
ance speech for the 1998 Oersted Medal presented by the American Association of Physics
Teachers, 6 January 1998, American Journal of Physics 66, pp. 369–376, 1998. Cited on page
394.
A.G. Lindh, Did Popper solve Hume’s problem?, Nature 366, pp. 105–106, 11 November

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

348
1993, Cited on page 395.
349 The measurement is presented in the Astrophysical Journal, in 1998 or 1999. Some beautiful
graphics at the http://www.physics.uiuc.edu/groups/tastro/movies/spm/ website show the
models of this star system. Cited on page 395.
350 R.J. Nemiroff, Visual distortions near a black hole and a neutron star, American Journal
of Physics 61, pp. 619–632, 1993. Cited on page 395.
351 The equality was first tested with precision by Rol and von Eőtvős, Annalen der Physik
& Chemie 59, p. 354, 1896, and by R. von Eőtvős, V. Pekár & E. Fekete, Beiträge
zum Gesetz der Proportionalität von Trägheit und Gravität, Annalen der Physik 4, Leipzig
68, pp. 11–66, 1922. Eőtvős found agreement to 5 parts in 109 . More experiments were per-
formed by P.G. Roll, R. Krotkow & R.H. Dicke, The equivalence of inertial and
passive gravitational mass, Annals of Physics (NY) 26, pp. 442–517, 1964, one of the most
interesting and entertaining research articles in experimental physics, and by V.B. Brag-
insky & V.I. Panov, Soviet Physics – JETP 34, pp. 463–466, 1971. Modern results, with
errors less than one part in 1012 , are by Y. Su & al., New tests of the universality of free fall,
Physical Review D50, pp. 3614–3636, 1994. Several experiments have been proposed to test
the equality in space to less than one part in 1016 . Cited on pages 395, 396, and 499.
352 The Thirring effect was predicted in H. Thirring, Über die Wirkung rotierender ferner
Massen in der Einsteinschen Gravitationstheorie, Physikalische Zeitschrift 19, pp. 33–39,
1918, and in H. Thirring, Berichtigung zu meiner Arbeit: “Über die Wirkung rotierender
Copyright © Christoph Schiller November 1997–May 2006

Massen in der Einsteinschen Gravitationstheorie”, Physikalische Zeitschrift 22, p. 29, 1921.

The Thirring–Lense effect was predicted in J. Lense & H. Thirring, Über den Einfluß
der Eigenrotation der Zentralkörper auf die Bewegung der Planeten und Monde nach der
Einsteinschen Gravitationstheorie, Physikalische Zeitschrift 19, pp. 156–163, 1918. See also
Ref. 371. Cited on page 398.
353 The feat used the LAGEOS and LAGEOS II satellites and is told in Ignazio Ciufolini, The
1995–99 measurements of the Thirring–Lense effect using laser-ranged satellites, Classical
and Quantum Gravity 17, pp. 2369–2380, 2000. See also I. Ciufolini & E.C. Pavlis, A
confirmation of the general relativistic prediction of the Lense–Thirring effect, Nature 431,
pp. 958–960, 2004. Cited on pages 399, 402, and 499.
354 The detection of the Thirring–Lense effect in binary pulsars is presented in R.D. Bl and-
ford, Lense–Thirring precession of radio pulsars, Journal of Astrophysics and Astronomy

Dvipsbugw
bibliography 507

16, pp. 191–206, 1995. Cited on page 399.

355 G. Holzmüller, Zeitschrift für Mathematik und Physik 15, p. 69, 1870, F. Tisserand,
Comptes Rendus 75, p. 760, 1872, and Comptes Rendus 110, p. 313, 1890. Cited on page 399.
356 B. Mashhoon, Gravitoelectromagnetism, http://www.arxiv.org/abs/gr-qc/0011014. See
also its extensive reference list on gravitomagnetism. Cited on page 400.
357 D. Bedford & P. Krumm, On relativistic gravitation, American Journal of Physics 53,
pp. 889–890, 1985, and P. Krumm & D. Bedford, The gravitational Poynting vector and
energy transfer, American Journal of Physics 55, pp. 362–363, 1987. Cited on pages 400
Dvipsbugw
and 407.
358 This is told in John A. Wheeler, A Journey into Gravity and Spacetime, W.H. Freeman,
1990. Cited on page 403.
359 See, for example, K.T. McDonald, Answer to question #49. Why c for gravitational
waves?, American Journal of Physics 65, pp. 591–592, 1997, and section III of V.B. Brag-
insky, C.M. Caves & K.S. Thorne, Laboratory experiments to test relativistic gravity,
Physical Review D 15, pp. 2047–2068, 1992. Cited on page 404.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

360 A. Tartaglia & M.L. Ruggiero, Gravito-electromagnetism versus electromagnetism,
European Journal of Physics 25, pp. 203–210, 2004. Cited on page 404.
361 The original claim is by S.M. Kopeikin, The post-Newtonian treatment of the VLBI ex-
periment on September 8, 2002, Physics Letters A 312, pp. 147–157, 2003, or http://www.
arxiv.org/abs/gr-qc/0212121. An argument against the claim was published, among others,
by Stuart Samuel, On the speed of gravity and the v~c corrections to the Shapiro time
delay, http://www.arxiv.org/abs/astro-ph/0304006. Cited on pages 405 and 408.
362 The quadrupole formula is explained in the text by Goenner. See Ref. 324. Cited on page
406.
363 For an introduction to gravitational waves, see B.F. Schu tz, Gravitational waves on the
back of an envelope, American Journal of Physics 52, pp. 412–419, 1984. Cited on page 405.
364 The beautiful summary by Daniel Kleppner, The gem of general relativity, Physics
Today 46, pp. 9–11, April 1993, appeared half a year before the authors of the cited work,
Joseph Taylor and Russel Hulse, received the Nobel Prize for the discovery of millisecond
pulsars. A more detailed review article is J.H. Taylor, Pulsar timing and relativistic grav-
ity, Philosophical Transactions of the Royal Society, London A 341, pp. 117–134, 1992. The
original paper is J.H. Taylor & J.M. Weisberg, Further experimental tests of relativ-
istic gravity using the binary pulsar PSR 1913+16, Astrophysical Journal 345, pp. 434–450,
1989. See also J.M. Weisberg, J.H. Taylor & L.A. Fowler, Pulsar PSR 1913+16 sen-
det Gravitationswellen, Spektrum der Wissenschaft, pp. 53–61, December 1981. Cited on page
Copyright © Christoph Schiller November 1997–May 2006

407.
365 W.B. B onnor & M.S. Piper, The gravitational wave rocket, Classical and Quantum Grav-
ity 14, pp. 2895–2904, 1997, or http://www.arxiv.org/abs/gr-qc/9702005. Cited on page 408.
366 L. Lerner, A simple calculation of the deflection of light in a Schwarzschild gravitational
field, American Journal of Physics 65, pp. 1194–1196, 1997. Cited on page 409.
367 A. Einstein, Über den Einfluß der Schwerkraft auf die Ausbreitung des Lichtes, Annalen
der Physik 35, p. 898, 1911. Cited on page 410.
368 I.I. Shapiro & al., Fourth test of general relativity, Physical Review Letters 13, pp. 789–
792, 1964. Cited on page 410.
369 I.I. Shapiro & al., Fourth test of general relativity: preliminary results, Physical Review
Letters 20, pp. 1265–1269, 1968. Cited on page 410.

Dvipsbugw
508 iii gravitation and rel ativity

370 J.H. Taylor, Pulsar timing and relativistic gravity, Proceedings of the Royal Society, London
A 341, pp. 117–134, 1992. Cited on pages 410 and 413.
371 W. de Sitter, On Einstein’s theory of gravitation and its astronomical consequences,
Monthly Notes of the Royal Astrononmical Society 77, pp. 155–184, p. 418E, 1916. For a dis-
cussion of de Sitter precession and Thirring–Lense precession, see also B.R. Holstein,
Gyroscope precession in general relativity, American Journal of Physics 69, pp. 1248–1256,
2001. Cited on pages 414 and 506.
372 B. Bertotti, I. Ciufolini & P.L. Bender, New test of general relativity: measurement
of de Sitter geodetic precession rate for lunar perigee, Physical Review Letters 58, pp. 1062– Dvipsbugw
1065, 1987. Later it was confirmed by I.I. Shapiro & al., Measurement of the de Sitter
precession of the moon: a relativistic three body effect, Physical Review Letters 61, pp. 2643–
2646, 1988. Cited on pages 414 and 499.
373 Wolf gang R indler, Essential Relativity, Springer, revised second edition, 1977. Cited
on page 416.
374 This is told (without the riddle solution) on p. 67, in Wolf gang Pauli, Relativitätsthe-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

orie, Springer Verlag, Berlin, 2000, the edited reprint of a famous text originally published in
1921. The reference is H. Vermeil, Notiz über das mittlere Krümmungsmaß einer n-fach
ausgedehnten Riemannschen Mannigfalktigkeit, Göttinger Nachrichten, mathematische–
physikalische Klasse p. 334, 1917. Cited on page 416.
375 M. Santander, L.M. Nieto & N.A. Cordero, A curvature based derivation of the
Schwarzschild metric, American Journal of Physics 65, pp. 1200–1209, 1997. Cited on pages
419 and 421.
376 Michael H. Soffel, Relativity in Astronomy, Celestial Mechanics and Geodesy, Springer
Verlag, 1989. Cited on page 420.
377 R ichard P. Feynman, Fernand o B. Morinigo, William G. Wagner & Brian
Hatfield, Feynman Lectures on Gravitation, Westview Press, 1995. Cited on page 420.
378 C.G. Torre & I.M. Anderson, Symmetries of the Einstein equations, Physical Review
Letters 70, pp. 3525–3529, 1993, or http://www.arxiv.org/abs/gr-qc/9302033. Cited on page
426.
379 H. Nicol ai, Gravitational billiards, dualities and hidden symmetries, http://www.arxiv.
org//abs/gr-qc/0506031. Cited on page 426.
380 Y. Wang & M. Tegmark, New dark energy constraints from supernovae, microwave
background and galaxy clustering, Physical Review Letters 92, p. 241302, 2004, or http://
www.arxiv.org/astro-ph/0403292. Cited on page 427.
381 Arguments for the emptiness of general covariance are given by John D. Norton, Gen-
Copyright © Christoph Schiller November 1997–May 2006

eral covariance and the foundations of general relativity, Reports on Progress in Physics 56,
pp. 791–858, 1993. The opposite point, including the discussion of ‘absolute elements’, is
made in the book by J.L. Anderson, Principles of Relativity Physics, chapter 4, Academic
Press, 1967. Cited on page 427.
382 For a good introduction to mathematical physics, see the famous three-women text in
two volumes by Yvonne Choquet-Bruhat, Cecile DeWitt-Morette & Mar-
garet Dill ard-Bleick, Analysis, Manifolds, and Physics, North-Holland, 1996 and
2001. The first edition of this classic appeared in 1977. Cited on page 429.
383 See for example R.A. Knop & al., New constraints on Ω M , Ω Λ , and w from an independ-
ent set of eleven high-redshift supernovae observed with HST, Astrophysical Journal 598,
pp. 102–137, 2003. Cited on page 430.

Dvipsbugw
bibliography 509

384 R.P. Feynman, R.B. Leighton & M. Sands, The Feynman Lectures on Physics, Addison
Wesley, 1977, volume II, p. 42–14. Cited on page 431.
385 A recent overview on the experimental tests of the universality of free fall is that by R.J.
Hughes, The equivalence principle, Contemporary Physics 4, pp. 177–191, 1993. Cited on
page 432.
386 See for example H.L. Bray, Black holes, geometric flows, and the Penrose inequality in
general relativity, Notices of the AMS 49, pp. 1372–1381, 2002. Cited on page 433.
387 See for example the paper by K. Dalton, Gravity, geometry and equivalence, preprint to
Dvipsbugw
be found at http://www.arxiv.org/abs/gr-qc/9601004, and L. L andau & E. Lif shitz, The
Classical Theory of Fields, Pergamon, 4th edition, 1975, p. 241. Cited on page 434.
388 Ekkehart Kröner, Kontinuumstheorie der Versetzungen und Eigenspannungen,
Springer, 1958. Kröner shows how to use the Ricci formalism in the solid state. Cited
on page 435.
389 Black Cited on page 435.
The equivalence of the various definitions of the Riemann tensor is explained in ...Cited on

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

390
page 436.
391 K. Tangen, Can the Pioneer anomaly have a gravitational origin?, http://www.arxiv.org/
abs/gr-qc/0602089. Cited on page 437.
392 This famous quote is the first sentence of the final chapter, the ‘Beschluß’, of Immanuel
Kant, Kritik der praktischen Vernunft, 1797. Cited on page 437.
393 Aetius, Opinions, III, I, 6. See Jean-Paul Dumont, Les écoles présocratiques, Folio Es-
sais, Gallimard, 1991, p. 445. Cited on page 438.
394 Galaxy mass measurements are described by ... Cited on page 439.
395 A beautiful introduction to modern astronomy is Paolo Maffei, I mostri del cielo,
Mondadori Editore, 1976. Cited on page 441.
396 See for example A.N. Cox, ed., Allen’s Astrophysical Quantities, AIP Press and Springer
Verlag, 2000. Cited on page 442.
397 P. Jetzer, Gravitational microlensing, Naturwissenschaften 86, pp. 201–211, 1999. Cited on
page 443.
398 D. Figer, An upper limit to the masses of stars, Nature 434, pp. 192–194, 2005. Cited on
page 443.
399 G. Basri, The discovery of brown dwarfs, Scientific American 282, pp. 77–83, April 2001.
Cited on page 443.
400 P.M. Woods & C. Thompson, Soft gamma repeaters and anomalous X-ray pulsars: mag-
Copyright © Christoph Schiller November 1997–May 2006

netar candidates, http://www.arxiv.org/abs/astro-ph/0406133. Cited on page 444.

401 B.M. Gaensler, N.M. McClure-Griffiths, M.S. Oey, M. Haverkorn, J.M.
Dickey & A.J. Green, A stellar wind bubble coincident with the anomalous X-ray pulsar
1E 1048.1-5937: are magnetars formed from massive progenitors?, The Astrophysical Journal
(Letters) 620, pp. L95–L98, 2005, or http://www.arxiv.org/abs/astro-ph/0501563. Cited on
page 444.
402 An opposite idea is defended by ... Cited on page 447.
403 C. Wirtz, Scientia 38, p. 303, 1925, and K. Lundmark, The motions and the distances of
the spiral nebulae, Monthly Notices of the Royal Astronomical Society 85, pp. 865–894, 1925.
See also G. Stromberg, Analysis of radial velocities of globular clusters and non-galactic
nebulae, Astrophysical Journal 61, pp. 353–362, 1925. Cited on page 447.

Dvipsbugw
510 iii gravitation and rel ativity

404 G. Gamow, The origin of the elements and the separation of galaxies, Physical Review 74,
p. 505, 1948. Cited on page 449.
405 A.G. Doroshkevich & I.D. Novikov, Dokl. Akad. Nauk. SSSR 154, p. 809, 1964. It
appeared translated into English a few months later. The story of the prediction was told by
Penzias in his Nobel lecture. Cited on page 449.
406 Arno A. Penzias & Robert W. Wilson, A measurement of excess antenna temper-
ature at 4080 Mcs, Astrophysical Journal 142, pp. 419–421, 1965. Cited on page 449.
407 Macrobius, Somnium Scipionis, XIV, 19. See Jean-Paul Dumont, Les écoles préso-
cratiques, Folio Essais, Gallimard, 1991, p. 61. Cited on page 450. Dvipsbugw
408 On the remote history of the universe, see the excellent texts by G. B örner, The Early
Universe – Facts & Fiction, Springer Verlag, 3rd edition, 1993, or Barry Parker, Creation –
The Story of the Origin and the Evolution of the Universe, Plenum Press, 1988. For an excellent
popular text, see M. Longair, Our Evolving Universe, Cambridge University Press, 1996.
Cited on page 450.
409 The first oxygen seems to have appeared in the atmosphere, produced by microorganisms,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2.32 thousand million years ago. See A. Becker & al., Dating the rise of atmospheric
oxygen, Nature 427, pp. 117–120, 2003. Cited on page 452.
410 Gabriele Walker, Snowball Earth – The Story of the Great Global Catastrophe That
Spawned Life as We Know It, Crown Publishing, 2003. Cited on pages 452 and 823.
411 K. Knie, Spuren einer Sternexplosion, Physik in unserer Zeit 36, p. 8, 2005. The first step of
this connection is found in K. Knie, G. Korschinek, T. Faestermann, E.A. Dorfi,
6
G. Rugel & A. Wallner, 0 Fe anomaly in a deep-sea manganese crust and implica-
tions for a nearby supernova source, Physics Review Letters 93, p. 171103, 2004, the second
step in N.D. Marsh & H. Svensmark, Low cloud properties influenced by cosmic rays,
Physics Review Letters 85, pp. 5004–5007, 2000, and the third step in P.B. de Menocal,
Plio–Pleistocene African climate, Science 270, pp. 53–59, 1995. Cited on page 453.
412 A. Friedman, Über die Krümmung des Raumes, Zeitschrift für Physik 10, pp. 377–386,
1922, and A. Friedmann, Über die Möglichkeit einer Welt mit konstanter negativer Krüm-
mung des Raumes, Zeitschrift für Physik 21, pp. 326–332, 1924. (In the Latin transliteration,
the author aquired a second ‘n’ in his second paper.) Cited on page 454.
413 H. Knu tsen, Darkness at night, European Journal of Physics 18, pp. 295–302, 1997. Cited
on pages 458 and 459.
414 See for example P.D. Peşić, Brightness at night, American Journal of Physics 66, pp. 1013–
1015, 1998. Cited on pages 459 and 460.
415 Paul Wesson, Olbers’ paradox and the spectral intensity of extra-galactic background
light, Astrophysical Journal 367, p. 399, 1991. Cited on page 459.
Copyright © Christoph Schiller November 1997–May 2006

416 Steven Weinberg, Gravitation and Cosmology, John Wiley, 1972. An excellent book writ-
ten with a strong personal touch and stressing most of all the relation with experimental
data. It does not develop a strong feeling for space-time curvature, and does not address the
basic problems of space and time in general relativity. Excellent for learning how to actually
calculate things, but less for the aims of our mountain ascent. Cited on pages 459 and 491.
417 Supernova searches are being performed by ... Cited on page 460.
418 The experiments are discussed in detail in the excellent review by D. Giulini & N.
Straumann, Das Rätsel der kosmischen Vakuumenergiedichte und die beschleunigte Ex-
pansion des Universums, Physikalische Blätter 556, pp. 41–48, 2000. See also N. Strau-
mann, The mystery of the cosmic vacuum energy density and the accelerated expansion of
the universe, European Journal of Physics 20, pp. 419–427, 1999. Cited on pages 460 and 500.

Dvipsbugw
bibliography 511

419 A. Harvey & E. Schucking, Einstein’s mistake and the cosmological contant, American
Journal of Physics 68, pp. 723–727, 2000. Cited on page 461.
420 The author of the bible explains rain in this way, as can be deduced from its very first page,
Genesis 1: 6-7. Cited on page 461.
421 Up to his death, Fred Hoyle defended his belief that the universe is static, e.g. in G. Bur-
bid ge, F. Hoyle & J.V. Narlikar, A different approach to cosmology, Physics Today
52, pp. 38–44, 1999. This team has also written a book with the same title, published in 2000
by Cambridge University Press. Cited on pages 462 and 463.
422 Stephen W. Hawking & G.F.R. Ellis, The Large Scale Structure of Space-Time, Cam- Dvipsbugw
bridge University Press, Cambridge, 1973. Among other things, this reference text discusses
the singularities of space-time, and their necessity in the history of the universe. Cited on
pages 463, 494, and 514.
423 Augustine, Confessions, 398, writes: ‘My answer to those who ask ‘What was god doing
before he made Heaven and Earth?’ is not ‘He was preparing Hell for people who pry into
mysteries’. This frivolous retort has been made before now, so we are told, in order to evade

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the point of the question. But it is one thing to make fun of the questioner and another to
find the answer. So I shall refrain from giving this reply. [...] But if before Heaven and Earth
there was no time, why is it demanded what you [god] did then? For there was no “then”
when there was no time.’ (Book XI, chapter 12 and 15). Cited on page 464.
424 Stephen Hawking, A Brief History of Time – From the Big Bang to Black Holes, 1988.
Reading this bestseller is almost a must for any physicist, as it is a frequent topic at dinner
parties. Cited on page 464.
425 Star details are explained in many texts. See for example ... Cited on page 466.
426 J. Pelt, R. Kayser, S. R ef sdal & T. Schramm, The light curve and the time delay of
QSO 0957+561, Astronomy and Astrophysics 305, p. 97, 1996. Cited on page 468.
427 F. Zwicky, Nebulae as gravitational lenses, Physical Review Letters 51, p. 290, and F.
Zwicky, On the probability to detect nebulae which act as gravitational lenses, p. 679,
1937. The negative view by Einstein is found in A. Einstein, Lens-like action of a star
by the deviatioin of light in the gravitational field, Science 84, pp. 506–507, 1936. A review
on gravitational lensing can even be found online, in the paper by J. Wambsganss, Grav-
itational lensing in astronomy, Living Reviews in Relativity 1-12, pp. 1–80, 1998, to be found
on the http://www.livingreviews.org/Articles/Volume1/1998-12wamb/ website.
There is also the book by P. Schneider, J. Ehlers & E.E. Falco, Gravitational
Lenses, Springer Verlag, Berlin, 1992. Cited on page 468.
428 M. L achièze-R ey & J.-P. Luminet, Cosmic topology, Physics Reports 254, pp. 135–
214, 1995. See also B.F. Roukema, The topology of the universe, http://www.arxiv.org/
Copyright © Christoph Schiller November 1997–May 2006

abs/astro-ph/0010185 preprint. Cited on page 469.

429 Thanks to Steve Carlip for clarifying this point. Cited on page 470.
430 G.F.R. Ellis & T. Rothmann, Lost horizons, American Journal of Physics 61, pp. 883–
893, 1993. Cited on page 470.
431 A. Gu th, Die Geburt des Kosmos aus dem Nichts – Die Theorie des inflationären Universums,
Droemer Knaur, 1999. Cited on page 471.
432 Entropy values for the universe have been discussed by Ilya Prigogine, Is Future
Given?, World Scientific, 2003. This was his last book. For a different approach, see G.A.
Mena Marugán & S. Carneiro, Holography and the large number hypothesis, http://
www.arxiv.org/abs/gr-qc/0111034. This paper also repeats the often heard statement that the
universe has an entropy that is much smaller than the theoretical maximum. The maximum

Dvipsbugw
 512 iii gravitation and rel ativity

Challenge 916 ny is ‘generally’ estimated to be 10120 k, whereas the actual value is ‘estimated’ to be 10100 k. How-
ever, other authors give 1084 k.
In 1974, Roger Penrose also made statements about the entropy of the universe. Cited
on page 472.
433 C.L. Bennet, M.S. Turner & M. White, The cosmic rosetta stone, Physics Today 50,
pp. 32–38, November 1997. The cosmic background radiation differs from black hole radi-
ation by less than 0.005 %. Cited on page 473.
434 The lack of expansion in the solar system is shown in ... Cited on page 473.
Dvipsbugw
435 A pretty article explaining how one can make experiments to find out how the human body
senses rotation even when blindfolded and earphoned is described by M.-L. Mittel-
staedt & H. Mittelstaedt, The effect of centrifugal force on the perception of rota-
tion about a vertical axis, Naturwissenschaften 84, pp. 366–369, 1997. Cited on page 474.
436 The independence of inertia has been tested ... Cited on page 474.
437 The present status is given in the conference proceedings by Julian Barbour &
Herbert Pfister, eds., Mach’s Principle: From Newton’s Bucket to Quantum Gravity,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Birkhäuser, 1995. Various formulations of Mach’s principle – in fact, 21 different ones – are
compared on page 530.
In a related development, in 1953, Dennis Sciama published a paper in which he argues
that inertia of a particle is due to the gravitational attraction of all other matter in the uni-
verse. The paper is widely quoted, but makes no new statements on the issue. See D.W.
Sciama, On the origin of inertia, Monthly Notices of the Royal Astronomical Society 113,
pp. 34–42, 1953. Cited on page 474.
438 Information on the rotation of the universe is given in A. Kogu t, G. Hinshaw & A.J.
Banday, Limits to global rotation and shear from the COBE DMR four-year sky maps,
Physical Review D 55, pp. 1901–1905, 1997. Earlier information is found in J.D. Barrow,
R. Juszkiewicz & D.H. Sonoda, Universal rotation: how large can it be?, Monthly No-
tices of the Royal Astronomical Society 213, pp. 917–943, 1985. See also J.D. Barrow, R.
Juszkiewicz & D.H. Sonoda, Structure of the cosmic microwave background, Nature
309, pp. 397–402, 1983, or E.F. Bunn, P.G. Fereira & J. Silk, How anisotropic is the
universe?, Physical Review Letters 77, pp. 2883–2886, 1996. Cited on page 475.
439 The issue has been discussed within linearized gravity by R ichard Tolman, in his text-
book Relativity, Thermodynamics, and Cosmology, Clarendon Press, 1934, on pp. 272–290.
The exact problem has been solved by A. Peres, Null electromagnetic fields in general re-
lativity theory, Physical Review 118, pp. 1105–1110, 1960, and by W.B. B onnor, The gravita-
tional field of light, Commun. Math. Phys. 13, pp. 163–174, 1969. See also N.V. Mitskievic
& K.K. Kumaradtya, The gravitational field of a spinning pencil of light, Journal of Math-
Copyright © Christoph Schiller November 1997–May 2006

ematical Physics 30, pp. 1095–1099, 1989, and P.C. Aichelburg & R.U. Sexl, On the
gravitational field of a spinning particle, General Relativity and Gravitation 2, pp. 303–312,
1971. Cited on page 475.
440 See the delightful popular account by Igor Novikov, Black Holes and the Universe, Cam-
bridge University Press, 1990. The consequences of light decay were studied by M. Bron-
stein, Die Ausdehnung des Weltalls, Physikalische Zeitschrift der Sowjetunion 3, pp. 73–82,
1933. Cited on pages 476 and 479.
441 C.L. Carilli, K.M. Menten, J.T. Stocke, E. Perlman, R. Vermeulen, F.
Briggs, A.G. de Bruyn, J. Conway & C.P. Moore, Astronomical constraints on the
cosmic evolution of the fine structure constant and possible quantum dimensions, Physical
Review Letters 85, pp. 5511–5514, 25 December 2000. Cited on page 476.

Dvipsbugw
bibliography 513

442 The observations of black holes at the centre of galaxies and elsewhere are summarised by
R. Bl andford & N. Gehrels, Revisiting the black hole, Physics Today 52, pp. 40–46,
June 1999. Cited on pages 477 and 489.
443 An excellent and entertaining book on black holes, without any formulae, but nevertheless
accurate and detailed, is the paperback by Igor Novikov, Black Holes and the Universe,
Cambridge University Press, 1990. See also Edwin F. Taylor & John A. Wheeler,
Exploring Black Holes: Introduction to General Relativity, Addison Wesley Longman 2000.
For a historical introduction, see the paper by R. Ruffini, The physics of gravitationally
collapsed objects, pp. 59–118, in Neutron Stars, Black Holes and Binary X-Ray Sources, Pro- Dvipsbugw
ceedings of the Annual Meeting, San Francisco, Calif., February 28, 1974, Reidel Publishing,
1975. Cited on page 477.
444 J. Michell, On the means of discovering the distance, magnitude, etc of the fixed stars,
Philosophical Transactions of the Royal Society London 74, p. 35, 1784, reprinted in S. De-
tweiler, Black Holes – Selected Reprints, American Association of Physics Teachers, 1982.
Cited on page 477.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

445 The beautiful paper is R. Oppenheimer & H. Snyder, On continued gravitational con-
traction, Physical Review 56, pp. 455–459, 1939. Cited on page 478.
446 R.P. Kerr, Gravitational field of a spinning mass as an example of algebraically special
metrics, Physical Review Letters 11, pp. 237–238, 1963. Cited on page 482.
447 E.T. Newman, E. Couch, R. Chinnapared, A. Exton, A. Prakash & R. Tor-
rence, Metric of a rotating, charged mass, Journal of Mathematical Physics 6, pp. 918–919,
1965. Cited on page 482.
448 For a summary, see P.O. Mazur, Black hole uniqueness theorems, pp. 130–157, in M.A.H.
MacCallum, editor, General Relativity and Gravitation, Cambridge University Press,
1987, or the update at http://www.arxiv.org/abs/hep-th/0101012. See also D.C. Robinson,
Four decades of black hole uniqueness theorems,, available at http://www.mth.kcl.ac.uk/
staff/dc_robinson/blackholes.pdf Cited on page 482.
449 H.P. Künzle & A.K.M. Masood-ul-Al am, Spherically symmetric static SU(2)
Einstein-Yang-Mills fields, Journal of Mathematical Physics 31, pp. 928–935, 1990. Cited on
page 482.
450 For information about the tendency of gravitational radiation to produce spherical shapes,
see for example ... Cited on page 483.
451 R. Penrose & R.M. Floyd, Extraction of rotational energy from a black hole, Nature
229, pp. 177–179, 1971. Cited on page 484.
452 The mass–energy relation for a rotating black hole is due to D. Christod oulou, Re-
Copyright © Christoph Schiller November 1997–May 2006

versible and irreversible transformations in black hole physics, Physical Review Letters 25,
pp. 1596–1597, 1970. For a general, charged and rotating black hole it is due to D. Chris-
tod oulou & R. Ruffini, Reversible transformations of a charged black hole, Physical
Review D 4, pp. 3552–3555, 1971. Cited on page 484.
453 J.D. Bekenstein, Black holes and entropy, Physical Review D7, pp. 2333–2346, 1973. Cited
on page 485.
454 The paradox is discussed in M.A. Abramowicz, Black holes and the centrifugal force
paradox, Scientific American 266, pp. 74–81, March 1993, and in the comment by Don N.
Page, Relative alternatives, Scientific American 266, p. 5, August 1993. See also M.A. Ab-
ramowicz & E. Szuszkiewicz, The wall of death, American Journal of Physics 61,
pp. 982–991, 1993, and M.A. Abramowicz & J.P. L asota, On traveling round without

Dvipsbugw
514 iii gravitation and rel ativity

feeling it and uncurving curves, American Journal of Physics 54, pp. 936–939, 1986. Cited
on page 487.
455 For information about black holes in the early universe, see ... Cited on page 488.
456 For information about black holes formation via star collapse, see ... Cited on page 488.
457 Frederick L amb, APS meeting 1998 press conference: Binary star 4U1820-30, 20 000
light years from Earth, Physics News Update, April 27, 1998. Cited on page 488.
458 The first direct evidence for matter falling into a black hole was publicised in early 2001. ...
Cited on page 489. Dvipsbugw
459 For a readable summary of the Penrose–Hawking singularity theorems, see ... Details can
be found in Ref. 422. Cited on page 489.
460 For an overview of cosmic censorship, see T.P. Singh, Gravitational collapse, black holes
and naked singularities, http://www.arxiv.org/abs/gr-qc/9805066, or R.M. Wald, Grav-
itational collapse and cosmic censorship, http://www.arxiv.org/abs/gr-qc/9710068. The ori-
ginal idea is due to R. Penrose, Gravitational collapse: the role of general relativity, Rivista
del Nuovo Cimento 1, pp. 252–276, 1969. Cited on page 490.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

461 G.J. Stoney, On the physical units of nature, Philosophical Magazine 11, pp. 381–391, 1881.
Cited on page 492.
462 The geometrodynamic clock is discussed in D.E. Brahm & R.P. Gruber, Limitations of
the geometrodynamic clock, General Relativity and Gravitation 24, pp. 297–303, 1992. The
clock itself was introduced by R.F. Marzke, in his Ph.D. thesis The theory of measurement
in general relativity, 1959, with John Wheeler as thesis adviser. Cited on page 493.
463 R. Geroch, Einstein algebras, Commun. Math. Phys. 26, pp. 271–275, 1972. Cited on page
493.
464 A. Macd onald, Einstein’s hole argument, American Journal of Physics 69, pp. 223–225,
2001. Cited on page 494.
465 Roman U. Sexl, Die Hohlwelttheorie, Der mathematisch-naturwissenschaftliche Unter-
richt 368, pp. 453–460, 1983. Roman U. Sexl, Universal conventionalism and space-time.,
General Relativity and Gravitation 1, pp. 159–180, 1970. See also Roman U. Sexl, Die
Hohlwelttheorie, in Arthur Scharmann & Herbert Schramm, editors, Physik, The-
orie, Experiment, Geschichte, Didaktik – Festschrift für Wilfried Kuhn zum 60. Geburtstag am
6. Mai 1983, Aulis Verlag Deubner, 1984, pp. 241–258. Cited on page 495.
466 T. Damour, Experimental tests of relativistic gravity, http://www.arxiv.org/abs/gr-qc/
9904057. It is the latest in a series of his papers on the topic; the first was T. Damour,
Was Einstein 100 % right?, http://www.arxiv.org/abs/gr-qc/9412064. Cited on pages 498
and 499.
Copyright © Christoph Schiller November 1997–May 2006

467 H. Dittus, F. Everitt, C. L ämmerzahl & G. Schäfer, Die Gravitation im Test,

Physikalische Blätter 55, pp. 39–46, 1999. Cited on page 498.
468 See S. Bässler & al., Improved test of the equivalence principle for gravitational selfen-
ergy, Physical Review Letters 83, pp. 3585–3588, 1999. See also C.M. Will, Gravitational
radiation and the validity of general relativity, Physics Today 52, p. 38, October 1999. Cited
on page 499.
469 The inverse square dependence has been checked down to 60 µm, as reported by E. Adel-
berger, B. Heckel & C.D. Hoyle, Testing the gravitational inverse-square law, Physics
World 18, pp. 41–45, 2005. Cited on page 499.
470 For theories competing with general relativity, see for example C.M. Will, The con-
frontation between general relativity and experiment, Living Reviews of Relativity 2001,

Dvipsbugw
bibliography 515

http://www.livingreviews.org/lrr-2001-4. For example, the absence of the Nordtvedt effect,

a hypothetical 28-day oscillation in the Earth–Moon distance, which was looked for by
laser ranging experiments without any result, ‘killed’ several competing theories. This ef-
fect, predicted by Kenneth Nordtvedt, would only appear if the gravitational energy in the
Earth–Moon system would fall in a different way than the Earth and the Moon themselves.
For a summary of the measurements, see J. Müller, M. Schneider, M. Soffel & H.
Ruder, Astrophysical Journal Letters 382, p. L101, 1991. Cited on page 498.
471 Almost everything of importance in general relativity is published in the Journal Classical
and Quantum Gravity. Cited on page 499. Dvipsbugw
472 Collisions and many body problems ... Cited on page 499.
473 Inflation and early universe ... Cited on page 500.
474 The study of chaos in Einstein’s field equations is just beginning. See e.g. L. B ombelli,
F. Lombard o & M. Castagnino, Chaos in Robertson-Walker cosmology, http://www.
arxiv.org/abs/gr-qc/9707051. Cited on page 500.
475 The ESA satellite called ‘Planck’ will measure the polarization of the cosmic microwave back-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ground. Cited on page 500.
476 A good introduction to the topic of gamma-ray bursts is S. Klose, J. Greiner & D.
Hartmann, Kosmische Gammastrahlenausbrüche – Beobachtungen und Modelle, Teil I
und II, Sterne und Weltraum March and April 2001. Cited on page 500.
477 The field solution database is built around the work of A. Karlhede, which allows one to
distinguish between solutions with a limited amount of mathematical computation. Cited
on page 500.
478 Torsion is presented in ... Cited on page 500.
479 G.E. Prince & M. Jerie, Generalising Raychaudhuri’s equation, in Differential Geometry
and Its Applications, Proc. Conf., Opava (Czech Republic), August 27-31, 2001, Silesian Uni-
versity, Opava, 2001, pp. 235–242. Cited on page 500.
480 Wormholes and nontrivial topologies ... A basic approach is the one by T. Diemer &
M. Hadley, Charge and the topology of spacetime, Classical and Quantum Gravity 16,
pp. 3567–3577, 1999, or http://www.arxive.org/abs/gr-qc/9905069 and M. Hadley, Spin
half in classical general relativity, Classical and Quantum Gravity 17, pp. 4187–4194, 2000,
or http://www.arxive.org/abs/gr-qc/0004029. Cited on page 500.
481 An important formulation of relativity is A. Ashtekar, New variables for classical and
quantum gravity, Physical Review Letters 57, pp. 2244–2247, 1986. Cited on page 501.
482 A well written text on the connections between the big bang and particle physics is by
I.L. Rozental, Big Bang – Big Bounce, How Particles and Fields Drive Cosmic Evolution,
Copyright © Christoph Schiller November 1997–May 2006

Springer, 1988. For another connection, see M. Nagano & A.A. Watson, Observations
and implications of the ultrahigh energy cosmic rays, Reviews of Modern Physics 72, pp. 689–
732, 2000. Cited on page 501.
483 Teaching will benefit in particular from new formulations, from concentration on principles
and their consequences, as has happened in special relativity, from simpler descriptions at
the weak field level, and from future research in the theory of general relativity. The newer
textbooks cited above are all steps in these directions. Cited on page 501.

Dvipsbugw
 C h a pter IV

C L A SSIC AL ELEC TRODYNAMIC S

Dvipsbugw

W
hat is light? The study of relativity left us completely in the dark, even though
e had embarked in it precisely to find an answer to that question. True,
e have learned how the motion of light compares with that of objects. We also
learned that light is a moving entity that cannot be stopped; but we haven’t learned any-
thing about its nature. The answer to this long-standing question emerges only from the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

study of those types of motion that are not related to gravitation, such as the ways magi-
cians levitate objects.

13. liquid electricit y, invisible fields and maximum

speed
Page 150 Revisiting the list of motors found in this world, we remark that gravitation hardly de-
scribes any of them. Neither the motion of sea waves, fire and earthquakes, nor that of
a gentle breeze are due to gravity. The same applies to the motion of muscles. Have you
Challenge 917 e ever listened to your own heart beat with a stethoscope? Without having done so, you can-
not claim to have experienced the mystery of motion. Your heart has about 3000 million
beats in your lifetime. Then it stops.
It was one of the most astonishing discoveries of science that heart beats, sea waves and
most other cases of everyday motion, as well as the nature of light itself, are connected to
observations made thousands of years ago using two strange stones. These stones show
that all examples of motion, which are called mechanical in everyday life, are, without
exception, of electrical origin.
In particular, the solidity, the softness and the impenetrability of matter are due to
Ref. 484 internal electricity; also the emission of light is an electrical process. As these aspects
Copyright © Christoph Schiller November 1997–May 2006

are part of everyday life, we will leave aside all complications due to gravity and curved
space-time. The most productive way to study electrical motion is to start, as in the case
of gravity, with those types of motion which are generated without any contact between
the bodies involved.

Amber, lodestone and mobile phones

The story of electricity starts with trees. Trees have a special relation to electricity. When
a tree is cut, a viscous resin appears. With time it solidifies and, after millions of years,
it forms amber. When amber is rubbed with a cat fur, it acquires the ability to attract
small objects, such as saw dust or pieces of paper. This was already known to Thales of

Dvipsbugw
 518 iv cl assical electrodynamics • 13. electricity and fields

Dvipsbugw

F I G U R E 221 Objects surrounded by fields: amber, lodestone and mobile phone

Miletus, one of the original seven sages, in the sixth century bce. The same observation
can be made with many other polymer combinations, for example with combs and hair,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

with soles of the shoe on carpets, and with a TV screen and dust. Children are always
surprised by the effect, shown in Figure 222, that a comb rubbed on wool has on running
tap water. Another interesting effect can be observed when a rubbed comb is put near a
Challenge 918 ny burning candle. (Can you imagine what happens?)
Another part of the story of electricity involves an iron mineral
found in certain caves around the world, e.g. in a region (still) water
called Magnesia in the Greek province of Thessalia, and in some pipe

regions in central Asia. When two stones of this mineral are put rubbed
comb
near each other, they attract or repel each other, depending on
their relative orientation. In addition, these stones attract objects
made of cobalt, nickel or iron.
Today we also find various small objects in nature with more
sophisticated properties, as shown in Figure 221. Some objects en-
able you to switch on a television, others unlock car doors, still
others allow you to talk with far away friends.
All these observations show that in nature there are situations F I G U R E 222 How to
amaze kids
where bodies exert influence on others at a distance. The space
surrounding a body exerting such an influence is said to contain a field. A (physical) field
is thus an entity that manifests itself by accelerating other bodies in its region of space. A
field is some ‘stuff ’ taking up space. Experiments show that fields have no mass. The field
Copyright © Christoph Schiller November 1997–May 2006

surrounding the mineral found in Magnesia is called a magnetic field and the stones are
called magnets.* The field around amber – called ἤλεκτρον in Greek, from a root meaning
‘brilliant, shining’ – is called an electric field. The name is due to a proposal by the famous
English part-time physicist William Gilbert (1544–1603) who was physician to Queen
Elizabeth I. Objects surrounded by a permanent electric field are called electrets. They
are much less common than magnets; among others, they are used in certain loudspeaker
systems.**

* A pretty book about the history of magnetism and the excitement it generates is James D. L ivingston,
Driving Force – the Natural Magic of Magnets, Harvard University Press, 1996.
** The Kirlian effect, which allows one to make such intriguingly beautiful photographs, is due to a time-

Dvipsbugw
amber, lodestone and mobile phones 519

TA B L E 41 Searches for magnetic monopoles, i.e., for magnetic charges

Search Magnetic charge

Smallest magnetic charge suggested by quantum theory д = he = e2α

Z0
= 4.1 pWb
Search in minerals none Ref. 485
Search in meteorites none Ref. 485
Search in cosmic rays none Ref. 485
Search with particle accelerators none Ref. 485 Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 223 Lightning: a picture taken with a moving camera,
showing its multiple strokes (© Steven Horsburgh)

The field around a mobile phone is called a radio field or, as we will see later, an elec-
tromagnetic field. In contrast to the previous fields, it oscillates over time. We will find
out later that many other objects are surrounded by such fields, though these are often
very weak. Objects that emit oscillating fields, such as mobile phones, are called radio
Copyright © Christoph Schiller November 1997–May 2006

transmitters or radio emitters.

Fields influence bodies over a distance, without any material support. For a long time,
this was rarely found in everyday life, as most countries have laws to restrict machines
that use and produce such fields. The laws require that for any device that moves, pro-
duces sound, or creates moving pictures, the fields need to remain inside them. For this
reason a magician moving an object on a table via a hidden magnet still surprises and
entertains his audience. To feel the fascination of fields more strongly, a deeper look into
a few experimental results is worthwhile.

varying electric field.

Dvipsbugw
520 iv cl assical electrodynamics • 13. electricity and fields

TA B L E 42 Some observed magnetic fields

O b s e r va t i o n Magnetic field

Lowest measured magnetic field (e.g., fields of the Schumann reson- 1 fT

ances)
Magnetic field produced by brain currents 0.1 pT to 3 pT
Intergalactic magnetic fields 1 pT to 10 pT
Magnetic field in the human chest, due to heart currents 100 pT Dvipsbugw
Magnetic field of our galaxy 0.5 nT
Magnetic field due to solar wind 0.2 to 80 nT
Magnetic field directly below high voltage power line 0.1 to 1 µT
Magnetic field of Earth 20 to 70 µT
Magnetic field inside home with electricity 0.1 to 100 µT
Magnetic field near mobile phone 100 µT

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Magnetic field that influences visual image quality in the dark 100 µT
Magnetic field near iron magnet 100 mT
Solar spots 1T
Magnetic fields near high technology permanent magnet max 1.3 T
Magnetic fields that produces sense of coldness in humans 5 T or more
Magnetic fields in particle accelerator 10 T
Maximum static magnetic field produced with superconducting coils 22 T
Highest static magnetic fields produced in laboratory, using hybrid 45 T
magnets
Highest pulsed magnetic fields produced without coil destruction 76 T
Pulsed magnetic fields produced, lasting about 1 µs, using imploding 1000 T
coils
Field of white dwarf 104 T
Fields in petawatt laser pulses 30 kT
Field of neutron star from 106 T to 1011 T
Quantum critical magnetic field 4.4 GT
Highest field ever measured, on magnetar and soft gamma repeater 0.8 to 1 ë 1011 T
SGR-1806-20
Field near nucleus 1 TT
Copyright © Christoph Schiller November 1997–May 2006

Maximum (Planck) magnetic field 2.2 ë 1053 T

How can one make lightning?

Everybody has seen a lightning flash or has observed the effect it can have on strik-
ing a tree. Obviously lightning is a moving phenomenon. Photographs such as that of
Figure 223 show that the tip of a lightning flash advance with an average speed of around
600 km~s. But what is moving? To find out, we have to find a way of making lightning for
ourselves.
In 1995, the car company General Motors accidentally rediscovered an old and simple

Dvipsbugw
 amber, lodestone and mobile phones 521

water
nylon ropes pipe nylon ropes

on the roof
pendulum
with metal
ball
metal cylinders
in the hall Dvipsbugw
bang!

metal wires

metal cans in the ground

F I G U R E 224 A simple Kelvin generator F I G U R E 225 Franklin’s personal

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

lightning rod

method of achieving this. Their engineers had inadvertently built a spark generating
mechanism into their cars; when filling the petrol tank, sparks were generated, which
Ref. 486 sometimes lead to the explosion of the fuel. They had to recall 2 million vehicles of its
Opel brand.
What had the engineers done wrong? They had unwittingly copied the conditions for
a electrical device which anyone can build at home and which was originally invented by
William Thomson.* Repeating his experiment today, we would take two water taps, four
empty bean or coffee cans, of which two have been opened at both sides, some nylon rope
Ref. 487 and some metal wire.
Putting this all together as shown in Figure 224, and letting the water flow, we find a
strange effect: large sparks periodically jump between the two copper wires at the point
where they are nearest to each other, giving out loud bangs. Can you guess what condition
for the flow has to be realized for this to work? And what did Opel do to repair the cars
Challenge 919 n they recalled?
If we stop the water flowing just before the next spark is due, we find that both buckets
are able to attract sawdust and pieces of paper. The generator thus does the same that
rubbing amber does, just with more bang for the buck(et). Both buckets are surrounded
Copyright © Christoph Schiller November 1997–May 2006

by electric fields. The fields increase with time, until the spark jumps. Just after the spark,

* William Thomson (1824–1907), important Irish Unionist physicist and professor at Glasgow University.
He worked on the determination of the age of the Earth, showing that it was much older than 6000 years, as
several sects believed. He strongly influenced the development of the theory of magnetism and electricity, the
description of the aether and thermodynamics. He propagated the use of the term ‘energy’ as it is used today,
instead of the confusing older terms. He was one of the last scientists to propagate mechanical analogies for
the explanation of phenomena, and thus strongly opposed Maxwell’s description of electromagnetism. It
was mainly for this reason that he failed to receive a Nobel Prize. He was also one of the minds behind the
laying of the first transatlantic telegraphic cable. Victorian to his bones, when he was knighted, he chose the
name of a small brook near his home as his new name; thus he became Lord Kelvin of Largs. Therefore the
unit of temperature obtained its name from a small Scottish river.

Dvipsbugw
 522 iv cl assical electrodynamics • 13. electricity and fields

the buckets are (almost) without electric field. Obviously, the flow of water somehow
builds up an entity on each bucket; today we call this electric charge. Charge can flow in
metals and, when the fields are high enough, through air. We also find that the two buckets
are surrounded by two different types of electric fields: bodies that are attracted by one
bucket are repelled by the other. All other experiments confirm that there are two types
of charges. The US politician and part-time physicist Benjamin Franklin (1706–1790)
called the electricity created on a glass rod rubbed with a dry cloth positive, and that on
a piece of amber negative. (Previously, the two types of charges were called ‘vitreous’ and
‘resinous’.) Bodies with charges of the same sign repel each other, bodies with opposite Dvipsbugw
charges attract each other; charges of opposite sign flowing together cancel each other
out.*
In summary, electric fields start at bodies, provided they are charged. Charging can
be achieved by rubbing and similar processes. Charge can flow: it is then called an elec-
tric current. The worst conductors of current are polymers; they are called insulators or
dielectrics. A charge put on an insulator remains at the place where it was put. In contrast,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

metals are good conductors; a charge placed on a conductor spreads all over its surface.
The best conductors are silver and copper. This is the reason that at present, after a hun-
dred years of use of electricity, the highest concentration of copper in the world is below
the surface of Manhattan.
Of course, one has to check whether natural lightning is actually electrical in origin.
In 1752, experiments performed in France, following a suggestion by Benjamin Franklin,
published in London in 1751, showed that one can indeed draw electricity from a thun-
derstorm via a long rod.** These French experiments made Franklin famous worldwide;
they were also the start of the use of lightning rods all over the world. Later, Franklin had
Ref. 488 a lightning rod built through his own house, but of a somewhat unusual type, as shown
in Figure 225. Can you guess what it did in his hall during bad weather, all parts being
Challenge 920 n made of metal? (Do not repeat this experiment; the device can kill.)

Electric charge and electric fields

If all experiments with charge can be explained by calling the two charges positive and
negative, the implication is that some bodies have more, and some less charge than an un-
charged, neutral body. Electricity thus only flows when two differently charged bodies are
brought into contact. Now, if charge can flow and accumulate, we must be able to some-
how measure its amount. Obviously, the amount of charge on a body, usually abbreviated
Copyright © Christoph Schiller November 1997–May 2006

q, is defined via the influence the body, say a piece of sawdust, feels when subjected to a
field. Charge is thus defined by comparing it to a standard reference charge. For a charged
body of mass m accelerated in a field, its charge q is determined by the relation

=
q ma
, (390)
q ref m ref a ref

* In fact, there are many other ways to produces sparks or even arcs, i.e. sustained sparks; there is even
a complete subculture of people who do this as a hobby at home. Those who have a larger budget do it
professionally, in particle accelerators. See the http://www.kronjaeger.com/hv/ website.
** The details of how lightning is generated and how it propagates are still a topic of research. An introduction
is given on page 597.

Dvipsbugw
 amber, lodestone and mobile phones 523

TA B L E 43 Properties of classical electric charge

Electric Physical M at h e m at i c a l Definition

charges propert y name

Can be distinguished distinguishability element of set Page 646

Can be ordered sequence order Page 1195
Can be compared measurability metricity Page 1205
Can change gradually continuity completeness Page 1214 Dvipsbugw
Can be added accumulability additivity Page 69
Do not change conservation invariance q = const
Can be separated separability positive or negative

i.e., by comparing it with the acceleration and mass of the reference charge. This definition

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

reflects the observation that mass alone is not sufficient for a complete characterization
of a body. For a full description of motion we need to know its electric charge; charge is
therefore the second intrinsic property of bodies that we discover in our walk.
Nowadays the unit of charge, the coulomb, is defined through a standard flow through
metal wires, as explained in Appendix B. This is possible because all experiments show
that charge is conserved, that it flows, that it flows continuously and that it can accumu-
late. Charge thus behaves like a fluid substance. Therefore we are forced to use for its
description a scalar quantity q, which can take positive, vanishing, or negative values.
In everyday life these properties of electric charge, listed also in Table 43, describe ob-
servations with sufficient accuracy. However, as in the case of all previously encountered
classical concepts, these experimental results for electrical charge will turn out to be only
approximate. More precise experiments will require a revision of several properties. How-
ever, no counter-example to charge conservation has as yet been observed.
A charged object brought near a neutral one polarizes it. Electrical polarization is the
separation of the positive and negative charges in a body. For this reason, even neutral
objects, such as hair, can be attracted to a charged body, such as a comb. Generally, both
insulators and conductors can be polarized; this occurs for whole stars down to single
molecules.
Attraction is a form of acceleration. Experiments show that the entity that accelerates
charged bodies, the electric field, behaves like a small arrow fixed at each point x in space;
Copyright © Christoph Schiller November 1997–May 2006

its length and direction do not depend on the observer. In short, the electric field E(x) is
a vector field. Experiments show that it is best defined by the relation

qE(x) = ma(x) (391)

taken at every point in space x. The definition of the electric field is thus based on how it
Challenge 922 e moves charges.* The field is measured in multiples of the unit N~C or V~m.
To describe the motion due to electricity completely, we need a relation explaining how
charges produce electric fields. This relation was established with precision (but not for

Challenge 921 ny * Does the definition of electric field given here assume a charge speed that is much less than that of light?

Dvipsbugw
524 iv cl assical electrodynamics • 13. electricity and fields

TA B L E 44 Values of electrical charge observed in nature

O b s e r va t i o n Charge

Smallest measured non-vanishing charge 1.6 ë 10−19 C

Charge per bit in computer memory 10−13 C
Charge in small capacitor 10−7 C
Dvipsbugw
Charge flow in average lightning stroke 1 C to 100 C
Charge stored in a fully-charge car battery 0.2 MC
Charge of planet Earth 1 MC
Charge separated by modern power station in one year 3 ë 1011 C
Total charge of positive (or negative) sign observed in universe 10622 C
Total charge observed in universe 0C

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

TA B L E 45 Some observed electric fields

O b s e r va t i o n Electric field

Field 1 m away from an electron in vacuum Challenge 923 n

Field values sensed by sharks down to 0.1 µV~m
Cosmic noise 10 µV~m
Field of a 100 W FM radio transmitter at 100 km distance 0.5 mV~m
Field inside conductors, such as copper wire 0.1 V~m
Field just beneath a high power line 0.1 to 1 V~m
Field of a GSM antenna at 90 m 0.5 V~m
Field inside a typical home 1 to 10 V~m
Field of a 100 W bulb at 1 m distance 50 V~m
Ground field in Earth’s atmosphere 100 to 300 V~m
Field inside thunder clouds up to over 100 kV~m
Copyright © Christoph Schiller November 1997–May 2006

Maximum electric field in air before sparks appear 1 to 3 MV~m

Electric fields in biological membranes 10 MV~m
Electric fields inside capacitors up to 1 GV~m
Electric fields in petawatt laser pulses 10 TV~m
Electric fields in U91+ ions, at nucleus 1 EV~m
Maximum practical electric field in vacuum, limited by electron 1.3 EV~m
pair production
Maximum possible electric field in nature (corrected Planck elec- 2.4 ë 1061 V~m
tric field)

Dvipsbugw
 amber, lodestone and mobile phones 525

the first time) by Charles-Augustin de Coulomb on his private estate, during the French
Revolution.* He found that around any small-sized or any spherical charge Q at rest there
is an electric field. At a position r, the electric field E is given by

E(r) = = 9.0 GV m~C .

1 Q r 1
where (392)
4πε 0 r 2 r 4πε 0

Later we will extend the relation for a charge in motion. The bizarre proportionality con-
stant, built around the so-called permittivity of free space ε 0 , is due to the historical way Dvipsbugw
the unit of charge was defined first.** The essential point of the formula is the decrease of
Challenge 924 n the field with the square of the distance; can you imagine the origin of this dependence?
The two previous equations allow one to write the interaction between two charged
bodies as
= =−
dp1 1 q1 q2 r dp2
, (393)
dt 4πε 0 r 2 r dt

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where dp is the momentum change, and r is the vector connecting the two centres of mass.
This famous expression for electrostatic attraction and repulsion, also due to Coulomb,
is valid only for charged bodies that are of small size or spherical, and most of all, that are
at rest.
Electric fields have two main properties: they contain energy and they can polarize
bodies. The energy content is due to the electrostatic interaction between charges. The
strength of the interaction is considerable. For example, it is the basis for the force of our
muscles. Muscular force is a macroscopic effect of equation 393. Another example is the
material strength of steel or diamond. As we will discover, all atoms are held together
by electrostatic attraction. To convince yourself of the strength of electrostatic attraction,
answer the following: What is the force between two boxes with a gram of protons each,
Challenge 925 n located on the two poles of the Earth? Try to guess the result before you calculate the
astonishing value.
Coulomb’s relation for the field around a charge can be rephrased in a way that helps
to generalize it to non-spherical bodies. Take a closed surface, i.e., a surface than encloses
a certain volume. Then the integral of the electric field over this surface, the electric flux,
is the enclosed charge Q divided by ε 0 :

E dA =
Q
∫ . (394)
Copyright © Christoph Schiller November 1997–May 2006

closedsurface ε0

Challenge 926 n This mathematical relation, called Gauss’s ‘law’, follows from the result of Coulomb. (In
the simplified form given here, it is valid only for static situations.) Since inside conduct-
ors the electrical field is zero, Gauss’s ‘law’ implies, for example, that if a charge q is sur-
* Charles-Augustin de Coulomb (b. 1736 Angoulême, d. 1806 Paris), French engineer and physicist. His
careful experiments on electric charges provided a firm basis for the study of electricity.
** Other definitions of this and other proportionality constants to be encountered later are possible,
leading to unit systems different from the SI system used here. The SI system is presented in detail in
Appendix B. Among the older competitors, the Gaussian unit system often used in theoretical calculations,
the Heaviside–Lorentz unit system, the electrostatic unit system and the electromagnetic unit system are
Ref. 489 the most important ones.

Dvipsbugw
 526 iv cl assical electrodynamics • 13. electricity and fields

rounded by a uncharged metal sphere, the outer surface of the metal sphere shows the
Challenge 927 e same charge q.
Owing to the strength of electromagnetic interactions, separating charges is not an
easy task. This is the reason that electrical effects have only been commonly used for
about a hundred years. We had to wait for practical and efficient devices to be invented
for separating charges and putting them into motion. Of course this requires energy. Bat-
teries, as used in mobile phones, use chemical energy to do the trick.* Thermoelectric
elements, as used in some watches, use the temperature difference between the wrist and
the air to separate charges; solar cells use light, and dynamos or Kelvin generators use Dvipsbugw
kinetic energy.
Do uncharged bodies attract one other? In first approximation they do not. But when
the question is investigated more precisely, one finds that they can attract one other. Can
Challenge 929 n you find the conditions for this to happen? In fact, the conditions are quite important, as
our own bodies, which are made of neutral molecules, are held together in this way.
What then is electricity? The answer is simple: electricity is nothing in particular. It is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the name for a field of inquiry, but not the name for any specific observation or effect.
Electricity is neither electric current, nor electric charge, nor electric field. Electricity
is not a specific term; it applies to all of these phenomena. In fact the vocabulary issue
hides a deeper question that remains unanswered at the beginning of the twenty-first
century: what is the nature of electric charge? In order to reach this issue, we start with
the following question.

Can we detect the inertia of electricity?

If electric charge really is something flowing through metals, we should be able to ob-
serve the effects shown in Figure 226. Maxwell has predicted most of these effects: elec-
tric charge should fall, have inertia and be separable from matter. Indeed, each of these
effects has been observed.** For example, when a long metal rod is kept vertically, we can
measure an electrical potential difference, a voltage, between the top and the bottom. In
other words, we can measure the weight of electricity in this way. Similarly, we can meas-
Ref. 490 ure the potential difference between the ends of an accelerated rod. Alternatively, we can
measure the potential difference between the centre and the rim of a rotating metal disc.
The last experiment was, in fact, the way in which the ratio q~m for currents in metals
was first measured with precision. The result is

q~m = 1.8 ë 1011 C~kg

Copyright © Christoph Schiller November 1997–May 2006

(395)

for all metals, with small variations in the second digit. In short, electrical current has
mass. Therefore, whenever we switch on an electrical current, we get a recoil. This simple
Ref. 491 effect can easily be measured and confirms the mass to charge ratio just given. Also, the
emission of current into air or into vacuum is observed; in fact, every television tube
uses this principle to generate the beam producing the picture. It works best for metal
Ref. 492 objects with sharp, pointed tips. The rays created this way – we could say that they are

Challenge 928 n * Incidentally, are batteries sources of charges?

** Maxwell also performed experiments to detect these effects (apart from the last one, which he did not
predict), but his apparatuses where not sensitive enough.

Dvipsbugw
 amber, lodestone and mobile phones 527

If electric charge in metals moves

like a fluid, it should:

fall under gravity

be subject to centrifugation
Dvipsbugw
a resist acceleration

spray when pumped

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

lead to recoil just after
a switching on a currrent
q

prevent free charges

from falling through
a thin hollow tube

F I G U R E 226 Consequences of the flow of electricity

‘free’ electricity – are called cathode rays. Within a few per cent, they show the same mass
to charge ratio as expression (395). This correspondence thus shows that charges move
almost as freely in metals as in air; this is the reason metals are such good conductors.
If electric charge falls inside vertical metal rods, we can make the astonishing deduc-
tion that cathode rays – as we will see later, they consist of free electrons* – should not
be able to fall through a vertical metal tube. This is due to exact compensation of the
acceleration by the electrical field generated by the displaced electricity in the tube and
Challenge 930 e the acceleration of gravity. Thus electrons should not be able to fall through a long thin
cylinder. This would not be the case if electricity in metals did not behave like a fluid. The
Copyright © Christoph Schiller November 1997–May 2006

Ref. 493 experiment has indeed been performed, and a reduction of the acceleration of free fall
for electrons of 90 % has been observed. Can you imagine why the ideal value of 100 % is
Challenge 931 n not achieved?
If electric current behaves like a liquid, one should be able to measure its speed. The
first to do so, in 1834, was Charles Wheatstone. In a famous experiment, he wire of a
* The name ‘electron’ is due to George Stoney. Electrons are the smallest and lightest charges moving in
metals; they are, usually – but not always – the ‘atoms’ of electricity – for example in metals. Their charge
is small, 0.16 aC, so that flows of charge typical of everyday life consist of large numbers of electrons; as a
result, electrical charge behaves like a continuous fluid. The particle itself was discovered and presented in
1897 by the Prussian physicist Johann Emil Wiechert (1861–1928) and, independently, three months later,
by the British physicist Joseph John Thomson (1856–1940).

Dvipsbugw
 528 iv cl assical electrodynamics • 13. electricity and fields

TA B L E 46 Some observed electric current values

O b s e r va t i o n Current

Smallest regularly measured currents 1 fA

Human nerve signals 20 µA
Lethal current for humans as low as 20 mA, typically
100 mA
Current drawn by a train engine 600 A
Current in a lightning bolt 10 to 100 kA Dvipsbugw
Highest current produced by humans 20 MA
Current inside the Earth around 100 MA

quarter of a mile length, to produce three sparks: one at the start, one at the middle, and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

one at the end. He then mounted a rapidly moving mirror on a mechanical watch; by
noting who much the three spark images were shifted against each other on a screen, he
determined the speed to be 450 Mm~s, though with a large error. Latter, more precise
measurements showed that the speed is always below 300 Mm~s, and that it depends on
the metal and the type of insulation of the wire. The high value of the speed convinced
many people to use electricity for transmitting messages. A modern version of the exper-
Ref. 494 iment, for computer fans, uses the ‘ping’ command. The ‘ping’ command measures the
time for a computer signal to reach another computer and return back. If the cable length
Challenge 932 e between two computers is known, the signal speed can be deduced. Just try.

Feeling electric fields

Why is electricity dangerous to humans? The main reason is that the human body is con-
trolled by ‘electric wires’ itself. As a result, outside electricity interferes with the internal
signals. This has been known since 1789. In that year the Italian medical doctor Luigi
Galvani (1737–1798) discovered that electrical current makes the muscles of a dead an-
imal contract. The famous first experiment used frog legs: when electricity was applied to
them, they twitched violently. Subsequent investigations confirmed that all nerves make
use of electrical signals. Nerves are the ‘control wires’ of animals. However, nerves are
not made of metal: metals are not sufficiently flexible. As a result, nerves do not conduct
electricity using electrons but by using ions. The finer details were clarified only in the
Copyright © Christoph Schiller November 1997–May 2006

twentieth century. Nerve signals propagate using the motion of sodium and potassium
ions in the cell membrane of the nerve. The resulting signal speed is between 0.5 m~s and
120 m~s, depending on the type of nerve. This speed is sufficient for the survival of most
species – it signals the body to run away in case of danger.
Being electrically controlled, all mammals can sense strong electric fields. Humans
can sense fields down to around 10 kV~m, when hair stands on end. In contrast, sev-
eral animals can sense weak electric and magnetic fields. Sharks, for example, can detect
fields down to 1 µV~m using special sensors, the Ampullae of Lorenzini, which are found
around their mouth. Sharks use them to detect the field created by prey moving in wa-
ter; this allows them to catch their prey even in the dark. Several freshwater fish are also
able to detect electric fields. The salamander and the platypus, the famous duck-billed

Dvipsbugw
 amber, lodestone and mobile phones 529

mammal, can also sense electric fields. Like sharks, they use them to detect prey in water
which is too muddy to see through. Certain fish, the so-called weakly-electric fish, even
generate a weak field in order to achieve better prey detection.*
No land animal has special sensors for electric fields, because any electric field in air
is strongly damped when it encounters a water-filled animal body. Indeed, the usual at-
mosphere has an electric field of around 100 V~m; inside the human body this field is
damped to the µV~m range, which is much less than an animal’s internal electric fields.
In other words, humans do not have sensors for low electric fields because they are land
animals. (Do humans have the ability to sense electric fields in water? Nobody seems Dvipsbugw
Challenge 933 ny to know.) However, there a few exceptions. You might know that some older people can
sense approaching thunderstorms in their joints. This is due the coincidence between the
Page 564 electromagnetic field frequency emitted by thunderclouds – around 100 kHz – and the
resonant frequency of nerve cell membranes.
The water content of the human body also means that the electric fields in air that
are found in nature are rarely dangerous to humans. Whenever humans do sense electric

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fields, such as when high voltage makes their hair stand on end, the situation is potentially
dangerous.
The high impedance of air also means that, in the case of time-varying electromagnetic
fields, humans are much more prone to be affected by the magnetic component than by
the electric component.

Magnets
The study of magnetism progressed across the world independently of the study of elec-
tricity. Towards the end of the 12th century, the compass came into use in Europe. At that
time, there were heated debates on whether it pointed to the north or the south. In 1269,
the French military engineer Pierre de Maricourt (1219–1292) published his study of mag-
Ref. 495 netic materials. He found that every magnet has two points of highest magnetization, and
he called them poles. He found that even after a magnet is cut, the resulting pieces always
retain two poles: one points to the north and the other to the south when the stone is left
free to rotate. Magnets are dipoles. Atoms are either dipoles or unmagnetic. There are no
magnetic monopoles. Despite the promise of eternal fame, no magnetic monopole has
ever been found, as shown in Table 41. Like poles repel, and unlike poles attract.
Magnets have a second property: magnets transform unmagnetic materials into mag-
netic ones. There is thus also a magnetic polarization, similar to the electric polarization.
Copyright © Christoph Schiller November 1997–May 2006

Unlike the electric case, some magnetic materials retain the induced magnetic polariza-
tion: they become magnetized. This happens when the atoms in the material get aligned
by the external magnet.

* It took until the year 2000 for technology to make use of the same effect. Nowadays, airbag sensors in cars
often use electric fields to sense whether the person sitting in the seat is a child or an adult, thus changing
the way that the bag behaves in an accident.

Dvipsbugw
 530 iv cl assical electrodynamics • 13. electricity and fields

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 227 The magentotactic bacterium Magnetobacterium
bavaricum with its magnetosomes (© Marianne Hanzlik)

Can humans feel magnetic fields?

Any fool can ask more questions than seven

“ sages can answer.

Antiquity

It is known that honey bees, sharks, pigeons, salmon, trout, sea turtles and certain bac- ”
Ref. 496 teria can feel magnetic fields. One speaks of the ability for magnetoreception. All these
life forms use this ability for navigation. The most common detection method is the use
of small magnetic particles inside a cell; the cell then senses how these small built-in mag-
nets move in a magnetic field. The magnets are tiny, typically around 50 nm in size. These
small magnets are used to navigate along the magnetic field of the Earth. For higher an-
imals, the variations of the magnetic field of the Earth, 20 to 70 µT, produce a landscape
that is similar to the visible landscape for humans. They can remember it and use it for
navigation.
Can humans feel magnetic fields? Magnetic material seems to be present in the human
brain, but whether humans can feel magnetic fields is still an open issue. Maybe you can
Copyright © Christoph Schiller November 1997–May 2006

Challenge 934 ny devise a way to check this?

Are magnetism and electricity related? François Arago* found out that they were. He
observed that a ship that had survived a bad thunderstorm and had been struck by light-
ning, needed a new compass. Thus lightning has the ability to demagnetize compasses.
Arago knew, like Franklin, that lightning is an electrical phenomena. In other words, elec-
tricity and magnetism must be related. More precisely, magnetism must be related to the
motion of electricity.

* Dominique-François Arago (1786–1853) French physicist.

Dvipsbugw
amber, lodestone and mobile phones 531

Oersted's motor Modern motor

current-
carrying
metal
wire N S

N S battery
current-carrying compass
magnet
metal wire wire needle
Dvipsbugw
F I G U R E 228 An old and a newer version of F I G U R E 229 An electrical
an electric motor current always produces a
magnetic field

How can one make a motor?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Communism is the power of the local councils

“ plus electricification of the whole country.

The reason for Lenin’s famous statement were two discoveries. One was made in 1820
Lenin.*

”
by the Danish physicist Hans Christian Oersted (1777–1851) and the other in 1831 by the
English physicist Michael Faraday.** The consequences of these experiments changed the
world completely in less than one century.
On the 21st of July of 1821, Oersted published a leaflet, in Latin, which took Europe by
storm. Oersted had found (during a lecture demonstration to his students) that when a
current is sent through a wire, a nearby magnet is put into motion. In other words, he
found that the flow of electricity can move bodies.
Further experiments show that two wires in which charges flow attract or repel each
other, depending on whether the currents are parallel or antiparallel. These and other
experiments show that wires in which electricity flows behave like magnets.*** In other
words, Oersted had found the definite proof that electricity could be turned into magnet-
ism.
Shortly afterwards, Ampère**** found that coils increase these effects dramatically.

* Lenin (b. 1870 Simbirsk, d. 1924 Gorki), founder of the Union of Soviet Socialist Republics, in 1920 stated
this as the centre of his development plan for the country. In Russian, the local councils of that time were
Copyright © Christoph Schiller November 1997–May 2006

called soviets.
** Michael Faraday (b. 1791 Newington Butts, d. 1867 London) was born to a simple family, without school-
ing, and of deep and naive religious ideas. As a boy he became assistant to the most famous chemist of his
time, Humphry Davy (1778–1829). He had no mathematical training, but late in his life he became member
of the Royal Society. A modest man, he refused all other honours in his life. He worked on chemical topics,
the atomic structure of matter and, most of all, developed the idea of (magnetic) fields and field lines through
all his experimental discoveries, such as effect. Fields were later described mathematically by Maxwell, who
at that time was the only person in Europe to take over Faraday’s field concept.
*** In fact, if one imagines tiny currents moving in circles inside magnets, one gets a unique description for
all magnetic fields observed in nature.
**** André-Marie Ampère (b. 1775 Lyon, d. 1836 Marseille), French physicist and mathematician. Autodi-
dact, he read the famous Encyclopédie as a child; in a life full of personal tragedies, he wandered from maths
to chemistry and physics, worked as a school teacher, and published nothing of importance until 1820. Then

Dvipsbugw
 532 iv cl assical electrodynamics • 13. electricity and fields

Coils behave like small magnets. In particular, coils, like magnetic fields, always have
two poles, usually called the north and the south pole. Opposite poles attract, like poles
repel each other. As is well known, the Earth is itself a large magnet, with its magnetic
north pole near the geographic south pole, and vice versa.
Moving electric charge produces magnetic fields. This result explains why magnetic
fields always have two poles. The lack of magnetic monopoles thus becomes clear. But
one topic is strange. If magnetic fields are due to the motion of charges, this must be also
the case for a normal magnet. Can this be shown?
In 1915, two men in the Netherlands found a simple way to prove that Dvipsbugw
even in a magnet, something is moving. They suspended a metal rod
from the ceiling by a thin thread and then put a coil around the rod, as
shown in Figure 230. They predicted that the tiny currents inside the
rod would become aligned by the magnetic field of the coil. As a res-
ult, they expected that a current passing through the coil would make
the rod turn around its axis. Indeed, when they sent a strong current

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

through the coil, the rod rotated. (As a result of the current, the rod was
Ref. 497 magnetized.) Today, this effect is called the Einstein–de Haas effect after
the two physicists who imagined, measured and explained it.* The effect
thus shows that even in the case of a permanent magnet, the magnetic F I G U R E 230
Current makes
field is due to the internal motion of charges. The size of the effect also a metal rods
shows that the moving particles are electrons. (Twelve years later it be- rotate
came clear that the angular momentum of the electrons responsible for
the effect is a mixture of orbital and spin angular momentum; in fact, the electron spin
plays a central role in the effect.)
Since magnetism is due to the alignment of microscopic rotational motions, an even
more surprising effect can be predicted. Simply rotating a ferromagnetic material**
should magnetize it, because the tiny rotating currents would then be aligned along the
Ref. 498 axis of rotation. This effect has indeed been observed; it is called the Barnett effect after its
discoverer. Like the Einstein–de Haas effect, the Barnett effect can also be used to determ-
Page 760 ine the gyromagnetic ratio of the electron; thus it also proves that the spins of electrons
(usually) play a larger role in magnetism than their orbital angular momentum.

Magnetic fields
Experiments show that the magnetic field always has a given direction in space, and a
Copyright © Christoph Schiller November 1997–May 2006

magnitude common to all (resting) observers, whatever their orientation. We are tempted

the discovery of Oersted reached all over Europe: electrical current can deviate magnetic needles. Ampère
worked for years on the problem, and in 1826 published the summary of his findings, which lead Maxwell
to call him the ‘Newton of electricity’. Ampère named and developed many areas of electrodynamics. In
1832, he and his technician also built the first dynamo, or rotative current generator. Of course, the unit of
electrical current is named after him.
* Wander Johannes de Haas (1878–1960), Dutch physicist. De Haas is best known for two additional
magneto-electric effects named after him, the Shubnikov–de Haas effect (the strong increase of the mag-
netic resistance of bismuth at low temperatures and high magnetic fields) and the de Haas–Van Alphen
effect (the diamagnetic susceptibility of bismuth at low temperatures is a periodic function of the magnetic
field).
** A ferromagnetic material is a special kind of paramagnetic material that has a permanent magnetization.

Dvipsbugw
 amber, lodestone and mobile phones 533

magnet magnet

diamagnetic paramagnetic
material material
Dvipsbugw
F I G U R E 231 The two basic types of magnetic material behaviour
(tested in an inhomogeneous field): diamagnetism and
paramagnetism

to describe the magnetic field by a vector. However, this would be wrong, since a magnetic
field does not behave like an arrow when placed before a mirror. Imagine that a system

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

produces a magnetic field directed to the right. You can take any system, a coil, a machine,
etc. Now build or imagine a second system that is the exact mirror version of the first: a
mirror coil, a mirror machine, etc. The magnetic system produced by the mirror system
does not point to the left, as maybe you expected: it still points to the right. (Check by
Challenge 935 e yourself.) In simple words, magnetic fields do not behave like arrows.
In other words, it is not completely correct to describe a magnetic field by a vector
B = (B x , B y , B z ), as vectors behave like arrows. One also speaks of a pseudovector; angular
momentum and torque are also examples of such quantities. The precise way is to describe
the magnetic field by the quantity*

’ 0 −B z B y “
B = – Bz 0 −B x — , (396)
” −B y B x 0 •

called an antisymmetric tensor. In summary, magnetic fields are defined by the accelera-
tion they impart on moving charges. This acceleration turns out to follow

a= vB = v  B
e e
(397)
m m
Copyright © Christoph Schiller November 1997–May 2006

a relation which is often called Lorentz acceleration, after the important Dutch physicist
Hendrik A. Lorentz (b. 1853 Arnhem, d. 1928 Haarlem) who first stated it clearly.** (The
relation is also called the Laplace acceleration.) The unit of the magnetic field is called
tesla and is abbreviated T. One has 1 T = 1 N s~C m = 1 V s~m2 = 1 V s2 ~A m.
The Lorentz acceleration is the effect at the root of any electric motor. An electric motor
is a device that uses a magnetic field as efficiently as possible to accelerate charges flowing

* The quantity B was not called the ‘magnetic field’ until recently. We follow here the modern, logical defin-
ition, which supersedes the traditional one, where B was called the ‘magnetic flux density’ or ‘magnetic
induction’ and another quantity, H, was called – incorrectly, but for over a century – the magnetic field. This
quantity H will not appear in this walk, but it is important for the description of magnetism in materials.
Challenge 936 ny ** Does the definition of magnetic field given here assume a charge speed much lower than that of light?

Dvipsbugw
 534 iv cl assical electrodynamics • 13. electricity and fields

in a wire. Through the motion of the charges, the wire is then also moved. Electricity is
thus transformed into magnetism and then into motion. The first efficient motor was
built back in 1834.
As in the electric case, we need to know how the strength of a magnetic field is de-
termined. Experiments such as Oersted’s show that the magnetic field is due to moving
charges, and that a charge moving with velocity v produces a field B given by

µ0 v  r
B(r) = = 10−7 N~A2 .
µ0
q 3 where (398) Dvipsbugw
4π r 4π

This is called Ampère’s ‘law’. Again, the strange factor µ 0 ~4π is due to the historical way
in which the electrical units were defined. The constant µ 0 is called the permeability of
the vacuum and is defined by the fraction of newtons per ampere squared given in the
formula. It is easy to see that the magnetic field has an intensity given by vE~c 2 , where E
Challenge 937 e is the electric field measured by an observer moving with the charge. This is the first hint

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that magnetism is a relativistic effect.
We note that equation (398) is valid only for small velocities and accelerations. Can
Challenge 938 n you find the general one?
In 1831, Michael Faraday discovered an additional piece of the puzzle, one that even
the great Ampère had overlooked. He found that a moving magnet could cause a current
flow in an electrical circuit. Magnetism can thus be turned into electricity. This import-
ant discovery allowed the production of electrical current flow by generators, so-called dy-
namos, using water power, wind power or steam power. In fact, the first dynamo was built
in 1832 by Ampère and his technician. Dynamos started the use of electricity throughout
the world. Behind every electrical plug there is a dynamo somewhere.
Additional experiments show that magnetic fields also lead to electric fields when
one changes to a moving viewpoint. You might check this on any of the examples of
Figures 228 to 240. Magnetism indeed is relativistic electricity. Electric and magnetic fields
are partly transformed into each other when switching from one inertial reference frame
to the other. Magnetic and electrical fields thus behave like space and time, which are
also mixed up when changing from one inertial frame to the other. The theory of spe-
cial relativity thus tells us that there must be a single concept, an electromagnetic field,
describing them both. Investigating the details, one finds that the electromagnetic field F
surrounding charged bodies has to be described by an antisymmetric 4-tensor

’ 0 −E x ~c −E y ~c −E z ~c “ ’ 0 E x ~c E y ~c E z ~c “
Copyright © Christoph Schiller November 1997–May 2006

– ~c — – ~c By —
=– x — or F µν = – — .
E 0 −B z By −E 0 −B z
– E y ~c — – −E y ~c B z −B x —
x
F µν
Bz 0 −B x 0
” E z ~c −B y Bx 0 • ” −E z ~c −B y Bx 0 •
(399)
Obviously, the electromagnetic field F, and thus every component of these matrices, de-
pends on space and time. The matrices show that electricity and magnetism are two faces
of the same effect.* In addition, since electric fields appear only in the topmost row and

º
* Actually, the expression for the field contains everywhere the expression 1~ µ o ε 0 instead of the speed of
light c. We will explain the reason for this substitution shortly.

Dvipsbugw
 amber, lodestone and mobile phones 535

leftmost column, the expressions show that in everyday life, for small speeds, electricity
Challenge 939 n and magnetism can be separated. (Why?)
The total influence of electric and magnetic fields on fixed or moving charges is then
given by the following expression for the relativistic force-acceleration relation K = mb:

mb = qFu or
du µ
m = qF µ ν u ν or
dτ Dvipsbugw
’ γc “ ’ 0 E x ~c E y ~c E z ~c “’ γc “
d – γv x — – E ~c —– —
m – — = q– x —– —
0 Bz −B y γv x
dτ – γv y — – E y ~c −B z —– —
or
0 Bx γv y
” γv z • ” E z ~c B y −B x 0 •” γv z •
W = qEv and dp~dt = q(E + v  B) , (400)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

which show how the work W and the three-force d p~dt depend on the electric and mag-
netic fields. All four expressions describe the same content; the simplicity of the first one
is the reason for the involved matrices (399) of the electromagnetic field. In fact, the exten-
ded Lorentz relation (400) is the definition of the electromagnetic field, since the field is
defined as that ‘stuff ’ which accelerates charges. In particular, all devices that put charges
into motion, such as batteries and dynamos, as well as all devices that are put into motion
by flowing charges, such as electric motors and muscles, are described by this relation.
That is why this relation is usually studied, in simple form, already in school. The Lorentz
relation describes all cases in which the motion of objects can be seen by the naked eye
or felt by our senses, such as the movement of an electrical motor in a high speed train,
in a lift and in a dental drills, the motion of the picture generating electron beam in a
television tube, or the travelling of an electrical signal in a cable and in the nerves of the
Ref. 499, Ref. 500 body.
In equation (400) it is understood that one sums over indices that appear twice. The
electromagnetic field tensor F is an antisymmetric 4-tensor. (Can you write down the
Challenge 940 ny relation between F µν , F µν and F µ ν ?) Like any such tensor, it has two invariants, i.e., two
deduced properties that are the same for every observer: the expression B 2 − E 2 ~c 2 =
2
1
tr F2 and the product 4EB = −c tr F‡ F. (Can you confirm this, using the definition of
Challenge 941 n trace as the sum of the diagonal elements?)
The first invariant expression turns out to be the Lagrangian of the electromagnetic
Copyright © Christoph Schiller November 1997–May 2006

field. It is a scalar and implies that if E is larger, smaller, or equal to cB for one observer,
it also is for all other observers. The second invariant, a pseudoscalar, describes whether
the angle between the electric and the magnetic field is acute or obtuse for all observers.*
* There is in fact a third Lorentz invariant, much less known. It is specific to the electromagnetic field and
is a combination of the field and its vector potential:
1
κ3 = A µ Aµ F ρν Fνρ − 2A ρ F ρν Fν µ Aµ
2
= (AE) + (AB) − SA  ES − SA  BS + 4 (AE  B) − ( ) (E + B ) .
2 2 2 2 φ φ 2 2 2
(401)
c c
Ref. 501 This expression is Lorentz (but not gauge) invariant; knowing it can help clarify unclear issues, such as the
lack of existence of waves in which the electric and magnetic fields are parallel. Indeed, for plane mono-

Dvipsbugw
 536 iv cl assical electrodynamics • 13. electricity and fields

The application of electromagnetic effects to daily life has opened up a whole new
world that did not exist before. Electrical light, electric motors, radio, telephone, X-rays,
television and computers have changed human life completely in less than one century.
For example, the installation of electric lighting in city streets has almost eliminated the
previously so common night assaults. These and all other electrical devices exploit the
fact that charges can flow in metals and, in particular, that electromagnetic energy can
be transformed
— into mechanical energy – as used in loudspeakers, motors, piezo crystals; Dvipsbugw
— into light – as in lamps and lasers;
— into heat – as in ovens and tea pots;
— into chemical effects – as in hydrolysis, battery charging and electroplating;
— into coldness – as in refrigerators and Peltier elements;
— into radiation signals – as in radio and television;
— into stored information – as in magnetic records and in computers.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

How motors prove relativity to be right
The only mathematical operation I performed

“ in my life was to turn the handle of a calculator.

Michael Faraday

All electric motors are based on the result that electric currents interact with magnetic ”
fields. The simplest example is the attraction of two wires carrying parallel currents. This
observation alone, made in 1820 by Ampère, is sufficient to make motion larger than a
Ref. 502 certain maximal speed impossible.
The argument is beautifully simple. We
change the original experiment and imagine
v
two long, electrically charged rods of mass m,
d charged rods
moving in the same direction with velocity v
and separation d. An observer moving with
v
the rods would see an electrostatic repulsion
Challenge 943 e between the rods given by F I G U R E 232 The relativistic aspect of
magnetism

1 2λ 2
ma e = − (402)
4πε 0 d
Copyright © Christoph Schiller November 1997–May 2006

where λ is the charge per length of the rods. A second, resting observer sees two effects:
the electrostatic repulsion and the attraction discovered by Ampère. The second observer
Challenge 944 e therefore observes
1 2λ 2 µ 0 λ 2 v 2
ma e m = − + . (403)
4πε 0 d 2π d

This expression must be consistent with the observation of the first observer. This is the
chromatic waves all three invariants vanish in the Lorentz gauge. Also the quantities ∂ µ J µ , J µ Aµ and ∂ µ Aµ
Challenge 942 n are Lorentz invariants. (Why?) The latter, the frame independence of the divergence of the four-potential,
reflects the invariance of gauge choice. The gauge in which the expression is set to zero is called the Lorentz
gauge.

Dvipsbugw
 amber, lodestone and mobile phones 537

case only if both observers find repulsions. It is easy to check that the second observer
sees a repulsion, as does the first one, only if

v2 <
1
. (404)
ε0 µ0

This maximum speed, with a value of 0.3 GM~s, is thus valid for any object carrying
charges. But all everyday objects contain charges: there is thus a maximum speed for
matter. Dvipsbugw
Challenge 945 ny Are you able to extend the argument for a maximum speed to neutral particles as well?
We will find out more on this limit velocity, which we know already, in a minute.
Another argument for magnetism as a relativistic effect is the following. In a wire with
electrical current, the charge is zero for an observer at rest with respect to the wire. The
reason is that the charges enter and exit the wire at the same time for that observer. Now
imagine an observer who flies along the wire. The entrance and exit events do not occur

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

simultaneously any more; the wire is charged for a moving observer. (The charge depends
on the direction of the observer’s motion.) In other words, if the observer himself were
charged, he would experience a force. Moving charges experience forces from current-
carrying wires. This is exactly why magnetic fields were introduced: they only produce
forces on moving charges. In short, current carrying wires are surrounded by magnetic
fields.
In summary, electric effects are due to flow of electric charges and to electric fields;
magnetism is due to moving electric charges. It is not due to magnetic charges.* The
strength of magnetism, used in any running electric motor, proves relativity right: there
is a maximum speed in nature. Both electric and magnetic fields carry energy and mo-
mentum. They are two faces of the same coin. However, our description of electromag-
netism is not complete yet: we need the final description of the way charges produce an
electromagnetic field.

Curiosities and fun challenges about things electric and

magnetic
Et facta mirari et intellectua assequi.

“ Augustine of Hippo

Before we study the motion of an electromagnetic field in detail, let’s have some fun with ” Copyright © Christoph Schiller November 1997–May 2006

electricity.
**
Nowadays, having fun with sparks is straightforward. Tesla coils, named after Nikola
Tesla ** are the simplest devices that allow to produce long sparks at home. Attention: this
is dangerous; that is the reason that such devices cannot be bought anywhere. The basic
Page 590 * ‘Electrons move in metal with a speed of about 1 µm~s; thus if I walk with the same speed along a cable
Challenge 946 ny carrying a constant current, I should not be able to sense any magnetic field.’ What is wrong with this argu-
ment?
** Никола Тесла (1856 Smiljan–1943 New York City), Serbian engineer and inventor. He invented and pro-
moted the polyphase alternating current system, the alternating current electric motor, wireless communic-

Dvipsbugw
538 iv cl assical electrodynamics • 13. electricity and fields

TA B L E 47 Voltage values observed in nature

O b s e r va t i o n Vo l t a g e

Smallest measured voltage c. 10 f V

Human nerves 70 mV
Voltaic cell (‘battery’) 1.5 V
Mains in households 230 V or 110 V
Electric eel 100 to 600 V
Sparks when rubbing a polymer pullover 1 kV Dvipsbugw
Electric fence 0.7 to 10 kV
Colour television tube 30 kV
X-ray tube 30 to 200 kV
Electron microscopes 0.5 kV to 3 MV
Stun gun 65 to 600 kV
Lightning stroke 10 to 100 MV

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Record accelerator voltage 1 TV
Planck voltage, highest value possible in nature 1.5 ë 1027 V

capacitive head
(c.10-20 pF to earth)

c.1000 turns large sparks

10-100nF

230 V c.10 kV c.10 turns

50 Hz 50 Hz spark gap
for switching resonance frequencies
100 - 500 kHz
ground

Copyright © Christoph Schiller November 1997–May 2006

F I G U R E 233 The schematics, the realization and the operation of a Tesla coil, including spark
and corona discharges (© Robert Billon)

ation, fluorescent lighting and many other applications of electricity. He is also one of the inventors of radio.
The SI unit of the magnetic field is named after him. A flamboyant character, his ideas were sometimes

Dvipsbugw
 amber, lodestone and mobile phones 539

diagram and an example is shown in Figure 233. Tesla coils look like large metal mush-
rooms (to avoid unwanted discharges) and plans for their construction can be found on
numerous websites or from numerous enthusiast’s clubs, such as http://www.stefan-kluge.
de.
**
If even knocking on a wooden door is an electric effect, we should be able to detect fields
Challenge 947 ny when doing so. Can you devise an experiment to check this?
Dvipsbugw
**
Birds come to no harm when they sit on unprotected electricity lines. Nevertheless, one
almost never observes any birds on tall, high voltage lines of 100 kV or more, which trans-
Challenge 948 n port power across longer distances. Why?
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

How can you distinguish a magnet from an unmagnetized metal bar of the same size and
Challenge 949 n material, using no external means?
**
How do you wire up a light bulb to the mains and three switches so that the light can
Challenge 950 n be switched on at any of the switches and off at any other switch? And for four switches?
Nobody will take a physicist seriously who is able to write Maxwell’s equations but cannot
solve this little problem.

**
The first appliances built to generate electric currents were large rubbing machines. Then,
in 1799 the Italian scientist Alessandro Volta (1745–1827) invented a new device to gen-
erate electricity and called it a pile; today it is called a (voltaic) cell or, less correctly, a
battery. Voltaic cells are based on chemical processes; they provide much more current
and are smaller and easier to handle than electrostatic machines. The invention of the
battery changed the investigation of electricity so profoundly that Volta became world
famous. At last, a simple and reliable source of electricity was available for use in experi-
ments; unlike rubbing machines, piles are compact, work in all weather conditions and
make no noise.
An apple or a potato with a piece of copper and one of zinc inserted is one of the
Copyright © Christoph Schiller November 1997–May 2006

simplest possible voltaic cells. It provides a voltage of about 1 V and can be used to run
digital clocks or to produce clicks in headphones. Volta was also the discoverer of the
charge law q = CU of capacitors (C being the capacity, and U the voltage) and the inventor
of the high sensitivity capacitor electroscope. A modest man, nevertheless, the unit of
electrical potential, or ‘tension’, as Volta used to call it, was deduced from his name. A
‘battery’ is a large number of voltaic cells; the term was taken from an earlier, almost
purely military use.* A battery in a mobile phone is just an elaborated replacement for a

unrealistic; for example he imagined that Tesla coils could be used for wireless power transmission.
* A pile made of sets of a zinc plate, a sheet of blotting paper soaked with salt water and a copper coin is
Challenge 951 ny easily constructed at home.

Dvipsbugw
 540 iv cl assical electrodynamics • 13. electricity and fields

number of apples or potatoes.

**
A PC or a telephone can communicate without wires, by using radio waves. Why are
these and other electrical appliances not able to obtain their power via radio waves, thus
Challenge 952 n eliminating power cables?
**
Objects that are not right–left symmetric are called chiral, from the Greek word for ‘hand’. Dvipsbugw
Challenge 953 n Can you make a mirror that does not exchange left and right? In two different ways?
**
A Scotch tape roll is a dangerous device. Pulling the roll quickly leads to light emission
(through triboluminescence) and to small sparks. It is suspected that several explosions
in mines were triggered when such a spark ignited a combustible gas mixture.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Take an envelope, wet it and seal it. After letting it dry for a day or more, open it in the
dark. At the place where the two sides of paper are being separated from each other, the
Challenge 954 n envelope glows with a blue colour. Why? Is it possible to speed up the test using a hair
dryer?
**
Electromagnetism is full of surprises and offers many effects that can be reproduced at
home. The internet is full of descriptions of how to construct Tesla coils to produce sparks,
coil guns or rail guns to shoot objects, electrostatic machines to make your hair stand on
end, glass spheres with touch-sensitive discharges and much more. If you like experi-
ments, just search for these terms.
**
A high voltage can lead to current flow through air, because air becomes conductive in
high electric fields. In such discharges, air molecules are put in motion. As a result, one
can make objects that are attached to a pulsed high tension source lift up in the air, if one
optimizes this air motion so that it points downwards everywhere. The high tension is
thus effectively used to accelerate ionized air in one direction and, as a result, an object
Copyright © Christoph Schiller November 1997–May 2006

will move in the opposite direction, using the same principle as a rocket. An example is
shown in Figure 234, using the power supply of a PC monitor. (Watch out: danger!) Nu-
merous websites explain how to build these so-called lifters at home; in Figure 234, the
bottle and the candle are used as high voltage insulator to keep one of the two thin high
voltage wires (not visible in the photograph) high enough in the air, in order to avoid dis-
charges to the environment or to interfere with the lifter’s motion. Unfortunately, the ma-
jority of websites – not all – give incorrect or confused explanations of the phenomenon.
These websites thus provide a good challenge for one to learn to distinguish fact from
Challenge 955 e speculation.

**

Dvipsbugw
 amber, lodestone and mobile phones 541

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 234 Lifting a light object – covered with aluminium foil – using high a
tension discharge (© Jean-Louis Naudin at http://www.jlnlabs.org)

The electric effects produced by friction and by liquid flow are usually small. However, in
the 1990s, a number oil tankers disappeared suddenly. The sailors had washed out the oil
tanks by hosing sea water onto the tank walls. The spraying led to charging of the tank;
a discharge then led to the oil fumes in the tank igniting. This led to an explosion and
subsequently the tankers sank. Similar accidents also happen regularly when chemicals
are moved from one tank to another.
**
Rubbing a plastic spoon with a piece of wool charges it. Such a charged spoon can be
used to extract pepper from a salt–pepper mixture by holding the spoon over the mixture.
Challenge 956 n Why?

**
When charges move, they produce a magnetic field. In particular, when ions inside the
Earth move due to heat convection, they produce the Earth’s magnetic field. When the
ions high up in the stratosphere are moved by solar wind, a geomagnetic storm appears;
Copyright © Christoph Schiller November 1997–May 2006

its field strength can be as high as that of the Earth itself. In 2003, an additional mechan-
ism was discovered. When the tides move the water of the oceans, the ions in the salt water
produce a tiny magnetic field; it can be measured by highly sensitive magnetometers in
satellites orbiting the Earth. After two years of measurements from a small satellite it was
Ref. 503 possible to make a beautiful film of the oceanic flows. Figure 235 gives an impression.
**
The names electrode, electrolyte, ion, anode and cathode were suggested by William
Whewell (1794–1866) on demand of Michael Faraday; Faraday had no formal education
and asked his friend Whewell to form two Greek words for him. For anode and cathode,
Whewell took words that literally mean ‘upward street’ and ‘downward street’. Faraday

Dvipsbugw
 542 iv cl assical electrodynamics • 13. electricity and fields

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 235 The magnetic field due to the tides (© Stefan Maus)

then popularized these terms, like the other words mentioned above.
**
The shortest light pulse produced so far had a length of 100 as. To how many wavelengths
Challenge 957 n of green light would that correspond?
**
Why do we often see shadows of houses and shadows of trees, but never shadows of the
Copyright © Christoph Schiller November 1997–May 2006

Challenge 958 n electrical cables hanging over streets?

**
How would you measure the speed of the tip of a lightning bolt? What range of values do
Challenge 959 n you expect?
**
Ref. 504 One of the simplest possible electric motors was discovered by Faraday in 1831. A magnet
suspended in mercury will start to turn around its axis if a current flows through it. (See
Figure 236.) In addition, when the magnet is forced to turn, the device (often also called
Barlow’s wheel) also works as a current generator; people have even tried to generate

Dvipsbugw
 amber, lodestone and mobile phones 543

suspending
battery wire

S Dvipsbugw

mercury

F I G U R E 236 A unipolar motor F I G U R E 237 The simplest

motor (© Stefan Kluge)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 960 n domestic current with such a system! Can you explain how it works?
The modern version of this motor makes use of a battery, a wire, a conductive
samarium–cobalt magnet and a screw. The result is shown in Figure 237.
**
The magnetic field of the Earth is much higher than that of other planets because of the
Ref. 505 Earth’s Moon. The field has a dipole strength of 7.8 ë 1022 A m2 . It shields us from lethal
solar winds and cosmic radiation particles. Today, a lack of magnetic field would lead to
high radiation on sunny days; but in the past, its lack would have prevented the evolution
of the human species. We owe our existence to the magnetic field.
**
The ionosphere around the Earth has a resonant frequency of 7 Hz; for this reason any
apparatus measuring low frequencies always gets a strong signal at this value. Can you
Challenge 961 n give an explanation of the frequency?
**
Challenge 962 ny The Sun is visible to the naked eye only up to a distance of 50 light years. Is this true? Copyright © Christoph Schiller November 1997–May 2006

**
At home, electricity is mostly used as alternating current. In other words, no electrons
actually flow through cables; as the drift speed of electrons in copper wires is of the order
Page 590 of 1 µm~s, electrons just move back and forward by 20 nm. Nothing flows in or out of the
cables! Why do the electricity companies require a real flow of money in return, instead
Challenge 963 e of being satisfied with a back and forth motion of money?
**
Comparing electricity with water is a good way of understanding electronics. Figure 238
shows a few examples that even a teenager can use. Can you fill in the correspondence
Challenge 964 n for the coil, and thus for a transformer?

Dvipsbugw
 544 iv cl assical electrodynamics • 13. electricity and fields

electrical hydraulic
component component
current, mass flow,
voltage pressure

wire tube

resistor restriction

Dvipsbugw
flexible &
capacitor elastic
closure

battery pump

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

one-way
diode valve

activated
transistor valve

inductor challenge

F I G U R E 238 The correspondence of electronics and water flow

**
Do electrons and protons have the same charge? Experiments show that the values are
Challenge 965 ny equal to within at least twenty digits. How would you check this?
**
Copyright © Christoph Schiller November 1997–May 2006

Challenge 966 ny Charge is also velocity-independent. How would you check this?

**
Magnets can be used, even by school children, to climb steel walls. Have a look at the
http://www.physicslessons.com/TPNN.htm website.
**
Extremely high magnetic fields have strange effects. At fields of 1010 T, vacuum becomes
effectively birefringent, photons can split and coalesce, and atoms get squeezed. Hydro-
gen atoms, for example, are estimated to get two hundred times narrower in one direction.
Fortunately, these conditions exist only in specific neutron stars, called magnetars.

Dvipsbugw
 the description of electromagnetic field evolu tion 545

wire with current I

electric field E
current I

speed v N S
object with
charge ρ

magnetic field B

F I G U R E 239 The first of Maxwell’s equations

Dvipsbugw

**
A good way to make money is to produce electricity and sell it. In 1964, a completely new
method was invented by Fletcher Osterle. The method was presented to a larger public in
Ref. 506 a beautiful experiment in 2003. One can take a plate of glass, add a conducting layers on

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

each side, and then etch a few hundred thousand tiny channels through the plate, each
around 15 µm in diameter. When water is made to flow through the channels, a current
is generated. The contacts at the two conducting plates can be used like battery contacts.
This simple device uses the effect that glass, like most insulators, is covered with a
charged layer when it is immersed in a liquid. Can you imagine why a current is gener-
Challenge 967 n ated? Unfortunately, the efficiency of electricity generation is only about 1%, making the
method much less interesting than a simple blade wheel powering a dynamo.

The description of electromagnetic field evolution

In the years between 1861 and 1865, taking in the details of all the experiments known to
him, James Clerk Maxwell produced a description of electromagnetism that forms one
of the pillars of physics.* Maxwell took all the experimental results and extracted their
common basic principles, as shown in Figures 239 and 240. Twenty years later, Heaviside
and Hertz extracted the main points of Maxwell ideas, calling their summary Maxwell’s
Page 1201 theory of the electromagnetic field. It consists of two equations (four in the non-relativistic
case).
The first equation is the precise statement that electromagnetic fields originate at
Copyright © Christoph Schiller November 1997–May 2006

* James Clerk Maxwell (b. 1831 Edinburgh, d. 1879 Cambridge), Scottish physicist. He founded electro-
magnetism by theoretically unifying electricity and magnetism, as described in this chapter. His work on
thermodynamics forms the second pillar of his activity. In addition, he studied the theory of colours and
developed the now standard horseshoe colour diagram; he was one of the first people to make a colour pho-
tograph. He is regarded by many as the greatest physicist ever. Both ‘Clerk’ and ‘Maxwell’ were his family
names.

Dvipsbugw
 546 iv cl assical electrodynamics • 13. electricity and fields

no magnetic charges I1(t) I2(t)

F I G U R E 240 The second of Maxwell’s equations

Dvipsbugw
charges, and nowhere else. The corresponding equation is variously written*
¾
dF = j
µ0
or
ε0
¾
=j
µ0
d ν F µν µ
or (405)
ε0

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

’ 0 E x ~c E y ~c E z ~c “ ¾
– ~c —
(∂ t ~c, −∂ x , −∂ y , −∂ z ) – —= (ρ, j x ~c, j y ~c, j z ~c) or
−E 0 −B z By µ0
– −E y ~c B z —
x
0 −B x ε0
” −E z ~c −B y Bx 0 •

∇E = = µ0 j .
ρ 1 ∂E
and ∇  B − 2
ε0 c ∂t

Each of these four equivalent ways to write the equation makes a simple statement: elec-
trical charge carries the electromagnetic field. This statement, including its equations, are
equivalent to the three basic observations of Figure 239. It describes Coulomb’s relation,
Ampère’s relation, and the way changing electrical fields induce magnetic effects, as you
Challenge 968 ny may want to check for yourself.
The second half of equation (405) contains the right hand rule for magnetic fields
Challenge 969 e around wires, through the vector product. The equation also states that changing electric
fields induce magnetic fields. The effect is essential for the primary side of transformers.
The factor 1~c 2 implies that the effect is small; that is why coils with many windings or
strong electric currents are needed to find it. Due to the vector product, all induced mag-
netic field lines are closed lines.
The second result by Maxwell is the precise description of how changing electric fields
Copyright © Christoph Schiller November 1997–May 2006

create magnetic fields, and vice versa. In particular, an electric field can have vortices only
when there is a changing magnetic field. In addition it expresses the observation that in
nature there are no magnetic charges, i.e. that magnetic fields have no sources. All these

* Maxwell generalized this equation to cases where the charges are not surrounded by vacuum, but located
inside matter. We will not explore these situations in our walk because, as we will see during our mountain
ascent, the apparently special case of vacuum in fact describes all of nature.

Dvipsbugw
 the description of electromagnetic field evolu tion 547

results are described by the relation variously written

d ‡ F = 0 with = ε ρσ µν F µν or
‡ ρσ 1
F
2
ε µνρ ∂ µ Fνρ = ∂ µ Fνρ + ∂ ν F ρ µ + ∂ ρ F µν = 0 or
’ “’ 0
γ c1 ∂ t Bx By Bz “ ’ 0 “
– — – −B x −E z ~c E y ~c — – —
– —– —=– —
γ∂ x 0 0
– — – −B y E z ~c −E x ~c — – —
or (406)
γ∂ y 0 0
” • ” −B z −E y ~c E x ~c • ” •
Dvipsbugw
γ∂ z 0 0

∇B = 0 and ∇  E = −
∂B
.
∂t

The relation expresses the lack of sources for the dual field tensor, usually written ‡ F: there
are no magnetic charges, i.e. no magnetic monopoles in nature. In practice, this equation
is always needed together with the previous one. Can you see why?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 970 ny
Since there are no magnetic charges, magnetic field lines are always closed; they never
start or end. For example, field lines continue inside magnets. This is often expressed
mathematically by stating that the magnetic flux through a closed surface S – such as a
sphere or a cube – always vanishes: ∫S BdA = 0.
The second half of equation (406), also shown in Figure 240, expresses that changes
in magnetic fields produce electric fields: this effect is used in the secondary side of trans-
formers and in dynamos. The cross product in the expression implies that an electric field
generated in this way – also called an electromotive field – has no start and endpoints. The
electromotive field lines thus run in circles: in most practical cases they run along electric
circuits.
Together with Lorentz’ evolution equation (400), which describes how charges move
given the motion of the fields, Maxwell’s evolution equations (405) and (406) describe
all electromagnetic phenomena occurring on everyday scales, from mobile phones, car
batteries, to personal computers, lasers, lightning, holograms and rainbows. We now have
a system as organized as the expression a = GM~r or as Einstein’s field equations for
gravitation.
We will not study many applications of the field equations but will continue directly
towards our aim to understand the connection to everyday motion and to the motion
of light. In fact, the electromagnetic field has an important property that we mentioned
right at the start: the field itself itself can move.
Copyright © Christoph Schiller November 1997–May 2006

Colliding charged particles

A simple experiment clarifies the properties of electromagnetic fields defined above.
When two charged particles collide, their total momentum is not conserved.
Imagine two particles of identical mass and identical charge just after a collision, when
they are moving away from one another. Imagine also that the two masses are large, so
that the acceleration due to their electrical repulsion is small. For an observer at the centre
of gravity of the two, each particle feels an acceleration from the electric field of the other.
Challenge 971 ny The electric field E is given by the so-called Heaviside formula

Dvipsbugw
 548 iv cl assical electrodynamics • 13. electricity and fields

m, q m, q

v v

0 distance r

Dvipsbugw
F I G U R E 241 Charged particles after a collision

q(1 − v 2 ~c 2 )
E= . (407)
4πe 0 r 2

In other words, the total system has a vanishing total momentum.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Take a second observer, moving with respect to the first with velocity v, so that the first
charge will be at rest. Expression (407) leads to two different values for the electric fields,
Ref. 507 one at the position of each particle. In other words, the system of the two particles is not
in inertial motion, as we would expect; the total momentum is not conserved. Where did
Challenge 972 n it go?
This at first surprising effect has been put in the form of a theorem by Van Dam and
Ref. 508 Wigner. They showed that for a system of particles interacting at a distance the total
particle energy–momentum cannot remain constant in all inertial frames.
The total momentum of the system is conserved only because the electromagnetic
field itself also carries momentum. If electromagnetic fields have momentum, they are
able to strike objects and to be struck by them. As we will show below, light is also an
electromagnetic field. Thus we should be able to move objects by shining light on to them.
We should even be able to suspend particles in mid air by shining light on to them from
below. Both predictions are correct, and some experiments will be presented shortly.
We conclude that any sort of field leading to particle interactions must carry energy
and momentum, as the argument applies to all such cases. In particular, it applies to nuc-
lear interactions. Indeed, in the second part of our mountain ascent we will even find
an additional result: all fields are themselves composed of particles. The energy and mo-
mentum of fields then become an obvious state of affairs.
Copyright © Christoph Schiller November 1997–May 2006

The gauge field – the electromagnetic vector potential

The study of moving fields is called field theory and electrodynamics is the prime example.
(The other classical example is fluid dynamics; moving electromagnetic fields and mov-
ing fluids are very similar mathematically.) Field theory is a beautiful topic; field lines,
equipotential lines and vortex lines are some of the concepts introduced in this domain.
They fascinate many.* However, in this mountain ascent we keep the discussion focused

Challenge 973 n * What is the relation, for static fields, between field lines and (equi-) potential surfaces? Can a field line
cross a potential surface twice? For more details on topics such as these, see the free textbook by B o Thidé,
Electromagnetic Field Theory, on his http://www.plasma.uu.se/CED/Book website. And of course, in English,
Ref. 499
484 have a look at the texts by Schwinger and by Jackson.

Dvipsbugw
 the description of electromagnetic field evolu tion 549

current current magnets

vector S
Dvipsbugw
potential

F I G U R E 242 Vector potentials for selected situations

on motion.
We have seen that fields force us to extend our concept of motion. Motion is not only

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the change in state of objects and of space-time, but also the change in state of fields. We
therefore need, also for fields, a complete and precise description of their state. The obser-
vations with amber and magnets have shown us that fields possess energy and momentum.
They can impart it to particles. The experiments with motors have shown that objects can
add energy and momentum to fields. We therefore have to define a state function which
allows us to define energy and momentum for electric and magnetic fields. Since electric
and magnetic fields transport energy, their motion follows the speed limit in nature.
Maxwell defined the state function in two standard steps. The first step is the definition
of the (magnetic) vector potential, which describes the momentum per charge that the
Ref. 509 field provides:
A= .
p
(408)
q

When a charged particle moves through a magnetic potential A(x), its momentum
changes by q∆A; it changes by the difference between the potential values at the start
and end points, multiplied by its charge. Owing to this definition, the vector potential
has the property that
B = ∇  A = curl A (409)

i.e. that the magnetic field is the curl of the magnetic potential. The curl is called the
Copyright © Christoph Schiller November 1997–May 2006

rotation, abbreviated rot in most languages. The curl (or rotation) of a field describes, for
each point of space, the direction of the local, imagined axis of rotation, as well as (twice)
the rotation speed around that axis. For example, the curl for the velocities of a rotating
Challenge 974 ny solid body is everywhere 2ω, or twice the angular velocity.
Ref. 510 The vector potential for a long straight current-carrying wire is parallel to the wire; it
Challenge 975 d has the magnitude
A(r) = −
µ0 I r
ln , (410)
4π r0

which depends on the radial distance r from the wire and an integration constant r 0 . This
expression for the vector potential, pictured in Figure 242, shows how the moving current

Dvipsbugw
 550 iv cl assical electrodynamics • 13. electricity and fields

produces a linear momentum in the (electro-) magnetic field around it. In the case of a
solenoid, the vector potential ‘circulates’ around the solenoid. The magnitude obeys

A(r) = −
Φ 1
, (411)
4π r
where Φ is the magnetic flux inside the solenoid. We see that, in general, the vector poten-
tial is dragged along by moving charges. The dragging effect decreases for larger distances.
This fits well with the image of the vector potential as the momentum of the electromag- Dvipsbugw
netic field.
This behaviour of the vector potential around charges is reminiscent of the way honey
is dragged along by a spoon moving in it. In both cases, the dragging effect decreases
with distance. However, the vector potential, unlike the honey, does not produce any fric-
tion that slows down charge motion. The vector potential thus behaves like a frictionless
liquid.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Inside the solenoid, the magnetic field is constant and uniform. For such a field B we
Challenge 976 ny find the vector potential
A(r) = − B  r .
1
(412)
2
In this case, the magnetic potential thus increases with increasing distance from the ori-
gin.* In the centre of the solenoid, the potential vanishes. The analogy of the dragged
honey gives exactly the same behaviour.
However, there is a catch. The magnetic potential is not defined uniquely. If A(x) is a
vector potential, then the different vector potential

A′ (x) = A(x) + grad Λ , (413)

where Λ(t, x) is some scalar function, is also a vector potential for the same situation.
(The magnetic field B stays the same, though.) Worse, can you confirm that the corres-
Challenge 977 ny ponding (absolute) momentum values also change? This unavoidable ambiguity, called
gauge invariance, is a central property of the electromagnetic field. We will explore it in
more detail below.
Not only the momentum, but also the energy of the electromagnetic field is defined
ambiguously. Indeed, the second step in the specification of a state for the electromagnetic
field is the definition of the electric potential as the energy U per charge:
Copyright © Christoph Schiller November 1997–May 2006

Ref. 509

φ=
U
(414)
q

In other words, the potential φ(x) at a point x is the energy needed to move a unit charge
to the point x starting from a point where the potential vanishes. The potential energy
is thus given by qφ. From this definition, the electric field E is simply the change of the

* This is only possible as long as the field is constant; since all fields drop again at large distances – because
the energy of a field is always finite – also the vector potential drops at large distances.

Dvipsbugw
 the description of electromagnetic field evolu tion 551

potential with position corrected by the time dependence of momentum, i.e.

E = −∇φ −
∂
A, (415)
∂t

Obviously, there is a freedom in the choice of the definition of the potential. If φ(x) is a
possible potential, then
φ′ (x) = φ(x) − Λ
∂
(416) Dvipsbugw
∂t
is also a potential function for the same situation. This freedom is the generalization of
the freedom to define energy up to a constant. Nevertheless, the electric field E remains
the same for all potentials.
Ref. 509 To be convinced that the potentials really are the energy and momentum of the elec-
Challenge 978 ny tromagnetic field, we note that for a moving charge we have

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

( mv + qφ) = q(φ − vA)
d 1 2 ∂
dt 2 ∂t
(mv + qA) = −∇q(φ − vA) ,
d
(417)
dt
which show that the changes of generalized energy and momentum of a particle (on the
left-hand side) are due to the change of the energy and momentum of the electromagnetic
field (on the right-hand side).*
In relativistic 4-vector notation, the energy and the momentum of the field appear
together in one quantity. The state function of the electromagnetic field becomes

Aµ = (φ~c, A) . (418)

It is easy to see that the description of the field is complete, since we have

F = d A or F µν = ∂ µ A ν − ∂ ν A µ , (419)

which means that the electromagnetic field F is completely specified by the 4-potential
A. But as just said, the 4-potential itself is not uniquely defined. Indeed, any other gauge
field A′ is related to A by the gauge transformation
Copyright © Christoph Schiller November 1997–May 2006

A′ µ = Aµ + ∂ µ Λ (420)

where Λ = Λ(t, x) is any arbitrarily chosen scalar field. The new field A′ leads to the
same electromagnetic field, and to the same accelerations and evolutions. The gauge 4-
field A is thus an overdescription of the physical situation as several different gauge fields
correspond to the same physical situation. Therefore we have to check that all measure-
ment results are independent of gauge transformations, i.e. that all observables are gauge

* This connection also shows why the expression P µ − qAµ appears so regularly in formulae; indeed, it plays
a central role in the quantum theory of a particle in the electromagnetic field.

Dvipsbugw
 552 iv cl assical electrodynamics • 13. electricity and fields

invariant quantities. Such gauge invariant quantities are, as we just saw, the fields F and
‡
F, and in general all classical quantities. We add that many theoretical physicists use the
term ‘electromagnetic field’ loosely for both the quantities F µν and A µ .
There is a simple image, due to Maxwell, to help overcoming the conceptual difficulties
of the vector potential. It turns out that the closed line integral over A µ is gauge invariant,
Challenge 979 e because
y A µ dx = y (A µ + ∂ µ Λ)dx = y A µ dx .
∫ ∫ ∫
µ µ ′ µ
(421)
Dvipsbugw
In other words, if we picture the vector potential as a quantity allowing one to associate
a number to a tiny ring at each point in space, we get a good, gauge invariant picture of
the vector potential.*
Now that we have defined a state function that describes the energy and momentum
of the electromagnetic field, let us look at what happens in more detail when electromag-
netic fields move.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Energy and linear and angular momentum of the
electromagnetic field
The description so far allows us to write the total energy E nergy of the electromagnetic
field as
B2
E nergy =
1
8π
ε 0 E2 +∫µ0
dV . (422)

Energy is thus quadratic in the fields.

For the total linear momentum one obtains

P=
ε0
4π ∫ E  B dV . (423)

The expression is also called the Poynting vector.**

For the total angular momentum one has

L=
ε0
4π ∫ E  A dV , (424)

where A is the magnetic vector potential.

Copyright © Christoph Schiller November 1997–May 2006

The L agrangian of electromagnetism

The motion of charged particles and the motion of the electromagnetic field can also
be described using a Lagrangian instead of using the three equations given above. It is
not hard to see that the action S CED for a particle in classical electrodynamics can be
Challenge 980 ny symbolically defined by***

Ref. 511 * In the second part of the text, on quantum mechanics, we will see that the exponent of this expression,
namely exp(iq x∫ A µ dx µ )~ħ, usually called the phase factor, can indeed be directly observed in experiments.
** John Henry Poynting (1852–1914) introduced the concept in 1884.

Dvipsbugw
 the description of electromagnetic field evolu tion 553

S CED = −mc 2 ∫ dτ − 1
4µ 0 ∫ F, ‡ F − ∫ j, A , (425)

which in index notation becomes

¾
dx n (s) dx nν (s)
µ
S CED = −mc Š 1 F µν F µν + j µ Aµ  d4 x .
ª

−ª
∫ η µν
ds ds
ds −
M 4µ 0 ∫ (426)

What is new is the measure of the change produced by the electromagnetic field. Its in- Dvipsbugw
ternal change is given by the term F ‡ F, and the change due to interaction with matter is
given by the term jA.
The least action principle, as usual, states that the change in a system is always as small
as possible. The action S CED leads to the evolution equations by requiring that the action
be stationary under variations δ and δ ′ of the positions and of the fields which vanish at
infinity. In other terms, the principle of least action requires that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

δS = 0 when xµ = xµ + δµ and A µ = A µ + δ ′µ ,
provided δx µ (θ) 0 for SθS ª
and δA µ (x ν ) 0 for Sx ν S ª . (427)

Page 180 In the same way as in the case of mechanics, using the variational method for the two
Challenge 981 ny variables A and x, we recover the evolution equations for particle and fields
¾
bµ = ∂ µ F µν = j ν ε µνρσ ∂ ν F ρσ = 0 ,
q µ ν µ0
Fν u , , and (428)
m ε0

which we know already. Obviously, they are equivalent to the variational principle based
on S CED . Both descriptions have to be completed by specifying initial conditions for the
particles and the fields, as well as boundary conditions for the latter. We need the first
and zeroth derivatives of the position of the particles, and the zeroth derivative for the
electromagnetic field.
Are you able to specify the Lagrangian of the pure electrodynamic field using the fields
Challenge 982 ny E and B instead of F and ‡ F?
The form of the Lagrangian implies that electromagnetism is time reversible. This
means that every example of motion due to electric or magnetic causes can also take
Copyright © Christoph Schiller November 1997–May 2006

place backwards. This is easily deduced from the properties of the Lagrangian. On the
other hand, everyday life shows many electric and magnetic effects which are not time
invariant, such as the breaking of bodies or the burning of electric light bulbs. Can you
Challenge 983 ny explain how this fits together?
In summary, with the Lagrangian (425) all of classical electrodynamics can be de-
scribed and understood. For the rest of this chapter, we look at some specific topics from
this vast field.

*** The product described by the symbol ,, ‘wedge’ or ‘hat’, has a precise mathematical meaning, defined
for this case in equation (426). Its background, the concept of (mathematical) form, carries us too far from
Ref. 512 our walk.

Dvipsbugw
 554 iv cl assical electrodynamics • 13. electricity and fields

Symmetries – the energy–momentum tensor

We know from classical mechanics that we get the definition of energy and momentum
tensor by using Noether’s theorem, if we determine the conserved quantity from the
Lorentz symmetry of the Lagrangian. For example, we found that relativistic particles
have an energy–momentum vector. At the point at which the particle is located, it de-
scribes the energy and momentum.
Since the electromagnetic field is not a localized entity, like a point particle, but an
extended entity, we need to know the flow of energy and momentum at every point in Dvipsbugw
space, separately for each direction. This makes a description with a tensor necessary. The
Page 424 result is the energy–momentum tensor of the electromagnetic field

’ energy energy flow or “

– —
=– —
density momentum density
– energy flow or —
µν
T
momentum
” momentum density •

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

flow density

u S~c = cp ’ 0 ε 0 cE  B “
=Œ ‘ = – ε 0 cë −ε 0 E i E j − B i B j ~µ 0 — (429)
cp T ” E  B 1~2δ i j (ε 0 E 2 + B 2 ~µ 0 ) •

The energy–momentum tensor shows again that electrodynamics is both Lorentz and
gauge invariant.
Both the Lagrangian and the energy–momentum tensor show that electrodynamics is
symmetric under motion inversion. If all charges change direction of motion – a situation
often incorrectly called ‘time inversion’ – they move backwards along the exact paths they
took when moving forward.
We also note that charges and mass destroy a symmetry of the vacuum that we men-
tioned in special relativity: only the vacuum is invariant under conformal symmetries. In
particular, only the vacuum is invariant under the spatial inversion r 1~r.
To sum up, electrodynamic motion, like all other examples of motion that we have
encountered so far, is deterministic, reversible and conserved. This is no big surprise.
Nevertheless, two symmetries of electromagnetism deserve special mention.

What is a mirror?
We will study the strange properties of mirrors several times during our walk. We start
Copyright © Christoph Schiller November 1997–May 2006

with the simplest one first. Everybody can observe, by painting each of their hands in a
different colour, that a mirror does not exchange right and left, as little as it exchanges up
and down; however, a mirror does exchange right and left handedness. In fact, it does so
by exchanging front and back.
Electrodynamics give a second answer: a mirror is a device that switches magnetic
Challenge 984 n north and south poles. Can you confirm this with a diagram?
But is it always possible to distinguish left from right? This seems easy: this text is
quite different from a derorrim version, as are many other objects in our surroundings.
But take a simple landscape. Are you able to say which of the two pictures of Figure 243
is the original?

Dvipsbugw
 the description of electromagnetic field evolu tion 555

Dvipsbugw

F I G U R E 243 Which one is the original landscape? (NOAA)

Astonishingly, it is actually impossible to distinguish an original picture of nature from

its mirror image if it does not contain any human traces. In other words, everyday nature

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is somehow left–right symmetric. This observation is so common that all candidate excep-
Page 939 tions, from the jaw movement of ruminating cows to the helical growth of plants, such as
hops, or the spiral direction of snail shells, have been extensively studied.* Can you name
Challenge 985 n a few more?
The left–right symmetry of nature appears because everyday nature is described by
gravitation and, as we will see, by electromagnetism. Both interactions share an import-
ant property: substituting all coordinates in their equations by the negative of their val-
ues leaves the equations unchanged. This means that for any solution of these equations,
i.e. for any naturally occurring system, a mirror image is a possibility that can also occur
naturally. Everyday nature thus cannot distinguish between right and left. Indeed, there
are right and left handers, people with their heart on the left and others with their heart
on the right side, etc.
To explore further this strange aspect of nature, try the following experiment: imagine
you are exchanging radio messages with a Martian; are you able to explain to him what
right and left are, so that when you meet, you are sure you are talking about the same
Challenge 986 n thing?
Actually, the mirror symmetry of everyday nature – also called its parity invariance
Ref. 514 – is so pervasive that most animals cannot distinguish left from right in a deeper sense.
Most animals react to mirror stimuli with mirror responses. It is hard to teach them dif-
ferent ways to react, and it is possible almost only for mammals. The many experiments
Copyright © Christoph Schiller November 1997–May 2006

performed in this area gave the result that animals have symmetrical nervous systems,

* The most famous is the position of the heart. The mechanisms leading to this disposition are still being
investigated. Recent research suggests that the oriented motion of the cilia on embryos, probably in the
Ref. 513 region called the node, determines the right–left asymmetry. The deep origin of this asymmetry is not yet
elucidated, however.
Most human bodies have more muscles on the right side for right-handers, such as Albert Einstein and
Pablo Picasso, and correspondingly on the left side for left-handers, such as Charlie Chaplin and Peter
Ustinov. This asymmetry reflects an asymmetry of the human brain, called lateralization, which is essen-
tial to human nature.
Another asymmetry of the human body is the hair whirl on the back of the head; the majority of humans
have only one, and in 80 % of the cases it is left turning. But many people have more than one.

Dvipsbugw
 556 iv cl assical electrodynamics • 13. electricity and fields

and possibly only humans show lateralization, i.e. a preferred hand and different uses for
the left and the right parts of the brain.
To sum up this digression, classical electrodynamics is left–right symmetric, or parity
Challenge 987 n invariant. Can you show this using its Lagrangian?
A concave mirror shows an inverted image; so does a plane mirror if it is partly folded
Challenge 988 n along the horizontal. What happens if this mirror is rotated around the line of sight?
Why do metals provide good mirrors? Metals are strong absorbers of light. Any strong
absorber has a metallic shine. This is true for metals, if they are thick enough, but also
for dye or ink crystals. Any material that strongly absorbs a light wavelength also reflects Dvipsbugw
it efficiently. The cause of the strong absorption of a metal is the electrons inside it; they
can move almost freely and thus absorb most visible light frequencies.

What is the difference between electric and magnetic fields?

Obviously, the standard answer is that electric fields have sources, and magnetic fields

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

do not; moreover, magnetic fields are small relativistic effects of importance only when
charge velocities are high or when electrical fields cancel out.
For situations involving matter, this clear distinction is correct. Up to the present day,
no particle with a magnetic charge, called a magnetic monopole, has ever been found, even
Page 942 though its existence is possible in several unified models of nature. If found, the action
(425) would have to be modified by the addition of a fourth term, namely the magnetic
current density. However, no such particle has yet been detected, despite intensive search
efforts.
In empty space, when matter is not around, it is possible to take a completely different
view. In empty space the electric and the magnetic fields can be seen as two faces of the
same quantity, since a transformation such as

E cB
B −E~c (430)

called (electromagnetic) duality transformation, transforms each vacuum Maxwell equa-

tion into the other. The minus sign is necessary for this. (In fact, there are even more
Challenge 989 n such transformations; can you spot them?) Alternatively, the duality transformation trans-
forms F into ‡ F. In other words, in empty space we cannot distinguish electric from mag-
netic fields.
Copyright © Christoph Schiller November 1997–May 2006

Matter would be symmetric under duality only if magnetic charges, also called mag-
netic monopoles, could exist. In that case the transformation (430) could be extended
to

cρ e ρm , ρm −cρ e . (431)

It was one of the great discoveries of theoretical physics that even though classical electro-
dynamics with matter is not symmetric under duality, nature is. In 1977, Claus Montonen
and David Olive showed that quantum theory allows duality transformations even with
the inclusion of matter. It has been known since the 1930s that quantum theory allows
magnetic monopoles. We will discover the important ramifications of this result in the

Dvipsbugw
 electrodynamic challenges and curiosities 557

third part of the text. This duality turns out to be one of the essential stepping stones
that leads to a unified description of motion. (A somewhat difficult question: extending
this duality to quantum theory, can you deduce what transformation is found for the fine
Challenge 990 ny structure constant, and why it is so interesting?)
Duality, by the way, is a symmetry that works only in Minkowski space-time, i.e. in
space-times of 3 + 1 dimensions. Mathematically, duality is closely related to the exist-
ence of quaternions, to the possibility of interpreting Lorentz boosts as rotations in 3 + 1
dimensions, and last, but not least, to the possibility of defining other smooth mathemat-
ical structures than the standard one on the space R 4 . These mathematical connections Dvipsbugw
are mysterious for the time being; they somehow point to the special role that four space-
time dimensions play in nature. More details will become apparent in the third part of
our mountain ascent.

Electrodynamic challenges and curiosities

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Could electrodynamics be different?
Any interaction such as Coulomb’s rule (392), which acts, for one given observer, between
two particles independently of 3-velocity, must depend on 3-velocity for other inertial
observers.* It turns out that such an interaction cannot be independent of the 4-velocity
either. Such an interaction, even though it would indeed be 3-velocity dependent, would
change the rest mass, since the 4-acceleration would not be 4-orthogonal to the 4-velocity.
The next simplest case is the one in which the acceleration is proportional to the 4-
velocity. Together with the request that the interaction leaves the rest mass constant, we
Ref. 515 then recover electrodynamics.
In fact, the requirements of gauge symmetry and of relativity symmetry also make
it impossible to modify electrodynamics. In short, it does not seem possible to have a
behaviour different from 1~r 2 for a classical interaction.
An inverse square dependence implies a vanishing mass of light and light particles,
Ref. 516 the photons. Is the mass really zero? The issue has been extensively studied. A massive
photon would lead to a wavelength dependence of the speed of light in vacuum, to de-
viations from the inverse square ‘law’, to deviations from Ampère’s ‘law’, to the existence
of longitudinal electromagnetic waves and more. No evidence for these effects has ever
been found. A summary of these studies shows that the photon mass is below 10−53 kg,
or maybe 10−63 kg. Some arguments are not universally accepted, thus the limit varies
somewhat from researcher to researcher.
Copyright © Christoph Schiller November 1997–May 2006

A small non-vanishing mass for the photon would change electrodynamics some-
what. The inclusion of a tiny mass poses no special problems, and the corresponding
Ref. 516 Lagrangian, the so-called Proca Lagrangian, has already been studied, just in case.
Strictly speaking, the photon mass cannot be said to vanish. In particular, a photon
with a Compton wavelength of the radius of the visible universe cannot be distinguished
from one with zero mass through any experiment. This gives a mass of 10−69 kg for the
photon. One notes that the experimental limits are still much larger. Photons with such
a small mass value would not invalidate electrodynamics as we know it.

* This can be deduced from special relativity from the reasoning of page 536 or from the formula in the
footnote of page 324.

Dvipsbugw
 558 iv cl assical electrodynamics • 14. what is light?

Interestingly, a non-zero mass of the photon implies the lack of magnetic monopoles,
as the symmetry between electric and magnetic fields is broken. It is therefore important
on the one hand to try to improve the experimental mass limit, and on the other hand
to explore whether the limit due to the universe’s size has any implications for this issue.
This question is still open.

The toughest challenge for electrodynamics

Electrodynamics faces an experimental and theoretical issue that physicist often avoid. Dvipsbugw
The process of thought is electric in nature. Physics faces two challenges in this domain.
First, physicists must find ways of modelling the thought process. Second, measurement
technology must be extended to allow one to measure the currents in the brain.
Even though important research has been carried out in these domains, researchers
are still far from a full understanding. Research using computer tomography has shown,
for example, that the distinction between the conscious and the unconscious can be meas-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ured and that it has a biological basis. Psychological concepts such as repression can be ob-
served in actual brain scans. Modellers of the brain mechanisms must thus learn to have
the courage to take some of the concepts of psychology as descriptions for actual physical
processes. This approach requires one to translate psychology into physical models, an
approach that is still in its infancy.
Similarly, research into magnetoencephalography devices is making steady progress.
The magnetic fields produced by brain currents are as low as 10 fT, which require sensors
at liquid helium temperature and a good shielding of background noise. Also the spatial
resolution of these systems needs to be improved.
The whole programme would be considered complete as soon as, in a distant future,
it was possible to use sensitive measuring apparatus to detect what is going on inside the
brain and to deduce or ‘read’ the thoughts of a person from these measurements. In fact,
this challenge might be the most complex of all challenges that science is facing. Clearly,
the experiment will require involved and expensive machinery, so that there is no danger
for a misuse of the technique. It could also be that the spatial resolution required is beyond
the abilities of technology. However, the understanding and modelling of the brain will
be a useful technology in other aspects of daily life as well.*

14. what is light?

Copyright © Christoph Schiller November 1997–May 2006

The nature of light has fascinated explorers of nature since at least the time of the ancient
Ref. 517 Greeks. In 1865, Maxwell summarized all data collected in the 2500 years before him by
deducing a basic consequence of the equations of electrodynamics. He found that in the
case of empty space, the equations of the electrodynamic field could be written as

∂2 φ ∂2 A x ∂2 A y ∂2 A z
j A=0 or, equivalently ε0 µ0 + + + =0. (432)
∂t 2 ∂x 2 ∂y 2 ∂z 2

* This vision, formulated here in 2005, is so far from realization that it is unclear whether it will come true
in the twenty-first or in any subsequent century.

Dvipsbugw
 what is light? 559

Figure to be added in future Dvipsbugw

F I G U R E 244 A plane, monochromatic and linearly polarized

electromagnetic wave, with the fields as described by the field equations
of electrodynamics

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 991 e This is called a wave equation, because it admits solutions of the type

A(t, x) = A0 sin(ωt − kx + δ) = A0 sin(2π f t − 2πx~λ + δ) , (433)

which are commonly called plane waves. Such a wave satisfies equation (432) for any
value of amplitude A 0 , of phase δ, and of angular frequency ω, provided the wave vector
k satisfies the relation
º
ω(k) = º ω(k) = º
1 1
k or k2 . (434)
ε0 µ0 ε0 µ0

The relation ω(k) between the angular frequency and the wave vector, the so-called dis-
persion relation, is the main property of any type of wave, be it a sound wave, a water
wave, an electromagnetic wave, or any other kind. Relation (434) specifically character-
izes electromagnetic waves in empty space, and distinguishes them from all other types
of waves.*
Equation (432) for the electromagnetic field is linear in the field; this means that the
sum of two situations allowed by it is itself an allowed situation. Mathematically speaking,
any superposition of two solutions is also a solution. For example, this means that two
waves can cross each other without disturbing each other, and that waves can travel across
Copyright © Christoph Schiller November 1997–May 2006

static electromagnetic fields. Linearity also means that any electromagnetic wave can be
described as a superposition of pure sine waves, each of which is described by expression
(433). The simplest possible electromagnetic wave, a harmonic plane wave with linear
Page 562 polarization, is illustrated in Figure 244.
After Maxwell predicted the existence of electromagnetic waves, in the years between
1885 and 1889 Heinrich Hertz** discovered and studied them. He fabricated a very simple

Page 205 * For completeness, we remember that a wave in physics is any propagating imbalance.
** Heinrich Rudolf Hertz (b. 1857 Hamburg, d. 1894 Bonn), important Hamburger theoretical and exper-
imental physicist. The unit of frequency is named after him. Despite his early death, Hertz was a central
figure in the development of electromagnetism, in the explanation of Maxwell’s theory and in the unfolding
of radio communication technology. More about him on page 152.

Dvipsbugw
 560 iv cl assical electrodynamics • 14. what is light?

Dvipsbugw

F I G U R E 245 The first transmitter (left) and receiver (right) of

electromagnetic (micro-) waves

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

transmitter and receiver for 2 GHz waves. Waves around this frequency are used in the
last generation of mobile phones. These waves are now called radio waves, since physicists
tend to call all moving force fields radiation, recycling somewhat incorrectly a Greek term
that originally meant ‘light emission.’
Hertz also measured the speed of these waves. In fact, you can also
measure the speed at home, with a chocolate bar and a kitchen mi-
crowave oven. A microwave oven emits radio waves at 2.5 GHz – not
far from Hertz’s value. Inside the oven, these waves form standing
waves. Just put the chocolate bar (or a piece of cheese) in the oven and
switch the power off as soon as melting begins. You will notice that the
bar melts at regularly spaced spots. These spots are half a wavelength
apart. From the measured wavelength value and the frequency, the
speed of light and radio waves simply follows as the product of the
two. Heinrich Hertz
If you are not convinced, you can measure the speed directly, by
telephoning a friend on another continent, if you can make sure of using a satellite line
(choose a low cost provider). There is about half a second additional delay between the
end of a sentence and the answer of the friend, compared with normal conversation. In
this half second, the signal goes up to the geostationary satellite, down again and returns
the same way. This half second gives a speed of c  4 ë 36 000 km~0.5 s  3 ë 105 km~s,
Copyright © Christoph Schiller November 1997–May 2006

which is close to the precise value. Radio amateurs who reflect their signals from the
Moon can perform the same experiment and achieve higher precision.
But Maxwell did more. He strengthened earlier predictions that light itself is a solu-
tion of equation (433) and therefore an electromagnetic wave, albeit with a much higher
frequency. Let us see how we can check this.
It is easy to confirm the wave properties of light; indeed they were known already
long before Maxwell. In fact, the first to suggest that light is a wave was, around the year
Page 78 1678, the important physicist Christiaan Huygens. You can confirm that light is a wave
with your own fingers. Simply place your hand one or two centimetres in front of your
eye, look towards the sky through the gap between the middle and the index finger and

Dvipsbugw
 what is light? 561

Dvipsbugw
F I G U R E 246 The primary and secondary rainbow, and the supernumerary bows below the primary bow
(© Antonio Martos and Wolfgang Hinz)

let the two fingers almost touch. You will see a number of dark lines crossing the gap.
These lines are the interference pattern formed by the light behind the slit created by the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fingers. Interference is the name given to the amplitude patterns that appear when several
waves superpose.* The interference patterns depend on the spacing between the fingers.
This experiment therefore allows you to estimate the wavelength of light, and thus, if you
Challenge 993 n know its speed, its frequency. Can you do this?
Historically, another effect was central in convincing everybody that light was a wave:
supernumerary rainbows, the additional bows below the main or primary rainbow. If
we look carefully at a rainbow, below the main red–yellow–green–blue–violet bow, we
Ref. 518 observe weaker, additional green, blue and violet bows. Depending on the intensity of
the rainbow, several of these supernumerary rainbows can be observed. They are due to
an interference effect, as Thomas Young showed around 1803.** Indeed, the repetition
distance of the supernumerary bows depends on the radius of the average water droplets
Page 574 that form them. (Details about the normal rainbows are given below.) Supernumerary
rainbows were central in convincing people that light is a wave. It seems that in those
times scientists either did not trust their own fingers, or did not have any.
There are many other ways in which the wave character of light can be made apparent.
Maybe the most beautiful is an experiment carried out by a team of Dutch physicists in
Ref. 519 1990. They simply measured the light transmitted through a slit in a metal plate. It turns
out that the transmitted intensity depends on the width of the slit. Their surprising result
is shown in Figure 247. Can you explain the origin of the unexpected intensity steps in
Challenge 994 ny the curve?
Copyright © Christoph Schiller November 1997–May 2006

Challenge 992 n * Where does the energy go in an interference pattern?

** Thomas Young (1773 Milverton–1829), read the bible at two, spoke Latin at four; a doctor of medicine, he
became a professor of physics. He introduced the concept of interference into optics, explaining Newtonian
rings and supernumerary rainbows; he was the first person to determine light’s wavelength, a concept that he
also introduced, and its dependence on colour. He was the first to deduce the three-colour vision explanation
of the eye and, after reading of the discovery of polarization, explained light as a transverse wave. In short,
Young discovered most of what people learn at secondary school about light. He was a universal talent:
he also worked on the deciphering of hieroglyphs, on ship building and on engineering problems. Young
collaborated with Fraunhofer and Fresnel. In Britain his ideas on light were not accepted, since Newton and
his followers crushed all opposing views. Towards the end of his life, his results were finally made known to
the physics community by Fresnel and Helmholtz.

Dvipsbugw
 562 iv cl assical electrodynamics • 14. what is light?

Numerous other experiments on the cre- transmitted light power

ation, detection and measurement of elec-
tromagnetic waves were performed in the
nineteenth and twentieth centuries. For
example, in 1800, William Herschel dis-
covered infrared light using a prism and a
Challenge 995 n thermometer. (Can you guess how?) In 1801,
Johann Wilhelm Ritter (1776–1810) a col-
ourful figure of natural Romanticism, dis- (preliminary figure) Dvipsbugw
covered ultraviolet light using silver chlor-
ide, AgCl, and again a prism. The result
of all these experiments is that electromag-
netic waves can be primarily distinguished slit width

by their wavelength or frequency. The main F I G U R E 247 The light power transmitted
Page 564 categories are listed in Table 48. For visible through a slit as function of its width

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

light, the wavelength lies between 0.4 µm
(pure violet) and 0.8 µm (pure red).
At the end of the twentieth century the final confirmation of the wave character of
light became possible. Using quite sophisticated experiments researchers, measured the
Ref. 520 oscillation frequency of light directly. The value, between 375 and 750 THz, is so high
that its detection was impossible for a long time. But with these modern experiments the
Ref. 521 dispersion relation (434) of light has finally been confirmed in all its details.
We are left with one additional question about light. If light oscillates, in which direc-
tion does this occur? The answer is hidden in the parameter A0 in expression (433). Elec-
tromagnetic waves oscillate in directions perpendicular to their motion. Therefore, even
for identical frequency and phase, waves can still differ: they can have different polariza-
tion directions. For example, the polarization of radio transmitters determines whether
radio antennas of receivers have to be kept horizontal or vertical. Also for light, polar-
ization is easily achieved, e.g. by shining it through a stretched plastic film. When the
polarization of light was discovered in 1808 by the French physicist Louis Malus (1775–
1812), it definitively established the wave nature of light. Malus discovered it when he
looked at the strange double images produced by feldspar, a transparent crystal found in
many minerals. Feldspar (KAlSi3 O8 ) splits light beams into two – it is birefringent – and
polarizes them differently. That is the reason that feldspar is part of every crystal collec-
tion. Calcite (CaCO 3 ) shows the same effect. If you ever get hold of a piece of feldspar or
Copyright © Christoph Schiller November 1997–May 2006

transparent calcite, do look through it at some written text.

By the way, the human eye is unable to detect polarization, in contrast to the eyes
of many insects, spiders and certain birds. Honey bees use polarization to deduce the
Ref. 522 position of the Sun, even when it is hidden behind clouds. Some beetles of the genus
Scarabeus use the polarization of the Moon light for navigation, and many insects use
Challenge 996 n polarization to distinguish water surfaces from mirages. Can you find out how? Despite
the human inability to detect polarization, both the cornea and the lens of the human eye
Ref. 523 are birefringent.
Note that all possible polarizations of light form a continuous set. However, a general
plane wave can be seen as the superposition of two orthogonal, linearly polarized waves
with different amplitudes and different phases. Most books show pictures of plane, linear-

Dvipsbugw
 what is light? 563

ized electrodynamic waves. Essentially, electric fields look like water waves generalized
to three dimensions, the same for magnetic fields, and the two are perpendicular to each
Challenge 997 ny other. Can you confirm this?
Interestingly, a generally polarized plane wave can also be seen as the superposition of
right and left circularly polarized waves. However, no illustrations of circularly polarized
Challenge 998 n waves are found in any textbook. Can you explain why?
So far it is clear that light is a wave. To confirm that light waves are indeed electromag-
netic is more difficult. The first argument was given by Bernhard Riemann in 1858;* he
deduced that any electromagnetic wave must propagate with a speed c given by Dvipsbugw

c=º
1
. (435)
ε0 µ0

Already ten years before him, in 1848, Kirchoff had noted that the measured values on
both sides agreed within measurement errors. A few years later, Maxwell gave a beautiful

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

confirmation by deducing the expression from equation (434). You should be able to re-
Challenge 999 ny peat the feat. Note that the right-hand side contains electric and magnetic quantities, and
the left-hand side is an optical entity. Riemann’s expression thus unifies electromagnetism
and optics.
Of course, people were not yet completely convinced. They looked for more ways to
show that light is electromagnetic in nature. Now, since the evolution equations of the elec-
trodynamic field are linear, additional electric or magnetic fields alone do not influence
Challenge 1000 e the motion of light. On the other hand, we know that electromagnetic waves are emit-
ted only by accelerated charges, and that all light is emitted from matter. It thus follows
that matter is full of electromagnetic fields and accelerated electric charges. This in turn
implies that the influence of matter on light can be understood from its internal electro-
magnetic fields and, in particular, that subjecting matter to an external electromagnetic
field should change the light it emits, the way matter interacts with light, or generally, the
material properties as a whole.
Searching for effects of electricity and magnetism on matter has been a main effort of
physicists for over a hundred years. For example, electric fields influence the light trans-
mission of oil, an effect discovered by John Kerr in 1875.** The discovery that certain
gases change colour when subject to a field yielded several Nobel Prizes for physics. With
time, many more influences on light-related properties by matter subjected to fields were
found. An extensive list is given in the table on page 605. It turns out that apart from a
Copyright © Christoph Schiller November 1997–May 2006

few exceptions the effects can all be described by the electromagnetic Lagrangian (425),
Page 553 or equivalently, by Maxwell’s equations (428). In summary, classical electrodynamics in-
deed unifies the description of electricity, magnetism and optics; all phenomena in these
fields, from the rainbow to radio and from lightning to electric motors, are found to be
different aspects of the evolution of the electromagnetic field.

* Bernhard Riemann (b. 1826 Breselenz, d. 1866 Selasca), important German mathematician. He studied
curved space, providing several of the mathematical and conceptual foundations of general relativity, but
then died at an early age.
** John Kerr (1824–1907), Scottish physicist, friend and collaborator of William Thomson.

Dvipsbugw
564 iv cl assical electrodynamics • 14. what is light?

TA B L E 48 The electromagnetic spectrum

Fre- Wa v e - Name Main Appearance U se

quency length properties
3ë 10−18 Hz 1026 m lower frequency limit see the section on cosmology
< 10 Hz A 30 Mm quasistatic fields intergalactic, power transmission,
galactic, stellar and accelerating and
planetary fields, deflecting cosmic
brain, electrical fish radiation Dvipsbugw

radio waves electronic devices

10 Hz– 30 Mm– ELW go round the nerve cells, power transmission,
50 kHz 6 km globe, penetrate electromechanical communication
into water, devices through metal walls,
penetrate metal communication with
submarines http://

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

www.vlf.it
50 – 6 km– LW follow Earth’s emitted by radio
500 kHz 0.6 km curvature, felt by thunderstorms communications,
nerves (‘bad telegraphy, inductive
weather nerves’) heating
500 – 600 m– MW reflected by night radio
1500 kHz 200 m sky
1.5 – 200 m–10 m SW circle world if emitted by stars radio transmissions,
30 MHz reflected by the radio amateurs,
ionosphere, spying
destroy hot air
balloons
15 – 20 m–2 m VHF allow battery emitted by Jupiter remote controls,
150 MHz operated closed networks, tv,
transmitters radio amateurs, radio
navigation, military,
police, taxi
150 – 2 m–0.2 m UHF idem, line of sight radio, walkie-talkies,
1500 MHz propagation tv, mobile phones,
internet via cable,
satellite
Copyright © Christoph Schiller November 1997–May 2006

communication,
bicycle speedometers
microwaves
1.5 – 20 cm–2 cm SHF idem, absorbed night sky, emitted radio astronomy,
15 GHz by water by hydrogen atoms used for cooking
(2.45 GHz),
telecommunications,
radar
15 – 20 mm– EHF idem, absorbed
150 GHz 2 mm by water

Dvipsbugw
what is light? 565

Fre- Wa v e - Name Main Appearance U se

quency length properties

infrared go through emitted by every satellite

clouds warm object photography of Earth,
astronomy
0.3 – 1000 –3 µm IRC or far sunlight, living seeing through
100 THz infrared beings clothes, envelopes
and teeth Dvipsbugw
100 – 3 µm– IRB or sunlight used for optical fibre
210 THz 1.4 µm medium communications for
infrared telephone and cable
TV
210 – 1400– IRA or penetrates for sunlight, radiation healing of wounds,
385 THz 780 nm near several cm into from hot bodies rheumatism, sport

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

infrared human skin physiotherapy,
hidden illumination
375 – 800– light not absorbed by heat (‘hot light’), definition of
750 THz 400 nm air, detected by lasers & chemical straightness,
the eye (up to reactions enhancing
850 nm at e.g. phosphor photosynthesis in
sufficient power) oxidation, fireflies agriculture,
(‘cold light’) photodynamic
therapy,
hyperbilirubinaemia
treatment
375 – 780–627 nm red penetrate flesh blood alarm signal, used
478 THz for breast imaging
700 nm pure red rainbow colour reference for
printing, painting,
illumination and
displays
478 – 627–589 nm orange various fruit attracts birds and
509 THz insects
600 nm standard orange
509 – 589–566 nm yellow majority of flowers idem; best
Copyright © Christoph Schiller November 1997–May 2006

530 THz background for

reading black text
580 nm standard yellow
530 – 566– green maximum eye algae and plants highest brightness
606 THz 495 nm sensitivity per light energy for
the human eye
546.1 nm pure green rainbow colour reference
606 – 495–436 nm blue sky, gems, water
688 THz
488 nm standard cyan

Dvipsbugw
566 iv cl assical electrodynamics • 14. what is light?

Fre- Wa v e - Name Main Appearance U se

quency length properties
435.8 nm pure blue rainbow colour reference
688 – 436– indigo, flowers, gems
789 THz 380 nm violet
ultraviolet
789 – 380–315 nm UVA penetrate 1 mm emitted by Sun and seen by certain birds,
952 THz into skin, darken stars integrated circuit Dvipsbugw
it, produce fabrication
vitamin D,
suppress immune
system, cause
skin cancer,
destroy eye lens

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

0.95 – 315–280 nm UVB idem, destroy idem idem
1.07 PHz DNA, cause skin
cancer
1.07 – 280–100 nm UVC form oxygen idem disinfection, water
3.0 PHz radicals from air, purification, waste
kill bacteria, disposal, integrated
penetrate 10 µm circuit fabrication
into skin
3 –24 PHz 100–13 nm EUV sky maps, silicon
lithography
X-rays penetrate emitted by stars, imaging human
materials plasmas and black tissue
holes
24 – 13–1.3 nm soft idem synchrotron idem
240 PHz X-rays radiation
A 240 PHz < 1.2 nm hard idem emitted when fast crystallography,
or A 1 keV X-rays electrons hit matter structure
determination
A 12 EHz < 24 pm γ-rays idem radioactivity, chemical analysis,
or cosmic rays disinfection,
A 50 keV
Copyright © Christoph Schiller November 1997–May 2006

astronomy
2 ë 1043 Hz  10−35 m Planck limit see part three of this text

The slowness of progress in physics

The well-known expression
c=¼
1
(436)
ε0 µ0

Dvipsbugw
 what is light? 567

for the speed of light is so strange that one should be astonished when one sees it. Some-
thing essential is missing.
Indeed, the speed is independent of the proper motion of the observer measuring the
electromagnetic field. In other words, the speed of light is independent of the speed of
the lamp and independent of the speed of the observer. All this is contained in expression
(436). Incredibly, for five decades, nobody explored this strange result. In this way, the
theory of relativity remained undiscovered from 1848 to 1905. As in so many other cases,
the progress of physics was much slower than necessary.
The constancy of the speed of light is the essential point that distinguishes special Dvipsbugw
relativity from Galilean physics. In this sense, any electromagnetic device, making use of
expression (436), is a working proof of special relativity.

How does the world look when riding on a light beam?

At the end of the nineteenth century, the teenager Albert Einstein read a book discussing

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the speed of light. The book asked what would happen if an observer moved at the same
speed as light.* Einstein thought much about the issue, and in particular, asked himself
what kind of electromagnetic field he would observe in that case. Einstein later explained
that this Gedanken experiment convinced him already at that young age that nothing
could travel at the speed of light, since the field observed would have a property not found
Challenge 1001 n in nature. Can you find out which one he meant?
Riding on a light beam situation would have strange consequences.
— You would have no mirror image, like a vampire.
— Light would not be oscillating, but would be a static field.
— Nothing would move, like in the tale of sleeping beauty.
But also at speeds near the velocity of light observations would be interesting. You would:
— see a lot of light coming towards you and almost no light from the sides or from behind;
the sky would be blue/white in the front and red/black behind;
— observe that everything around happens very very slowly;
— experience the smallest dust particle as a deadly bullet.
Challenge 1002 n Can you think of more strange consequences? It is rather reassuring that our planet moves
rather slowly through its environment. Copyright © Christoph Schiller November 1997–May 2006

Does light travel in a straight line?

Usually light moves in straight lines. Indeed, we even use light to define ‘straightness.’
Ref. 524 However, there are a number of exceptions that every expert on motion should know.
In sugar syrup, light beams curve, as shown in Figure 248. In fact, light beams bend
at any material interface. This effect, called refraction, also changes the appearance of the
shape of our feet when we are in the bath tub and makes aquaria seem less deep than
they actually are. Refraction is a consequence of the change of light speed from material

* This was the book series in twenty volumes by Aaron Bernstein, Naturwissenschaftliche Volksbücher,
Duncker, 1873-1874. The young Einstein read them, between 1892 and 1894, with ‘breathless attention’, as
he wrote later on.

Dvipsbugw
 568 iv cl assical electrodynamics • 14. what is light?

air

light
beam
sugar and water

F I G U R E 248 Sugar water bends light

Dvipsbugw
f

focus

object and real

light departing image

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

from it with
optional
screen

do di

object and focus virtual

light departing image
from it
di

do

F I G U R E 249 The real image produced by a converging lens

and the virtual image produced by a diverging lens

Challenge 1003 n to material. Are you able to explain refraction, and thus explain the syrup effect?
Refraction is chiefly used in the design of lenses. Using glass instead of water, one can
Copyright © Christoph Schiller November 1997–May 2006

produce curved surfaces, so that light can be focused. Focusing devices can be used to
produce images. The two main types of lenses, with their focal points and the images
they produce, are shown in Figure 249. When an object is put between a converging lens
and its focus, works as a magnifying glass. It also produces a real image, i.e., an image that
can be projected onto a screen. In all other cases lenses produce so-called virtual images:
such images can be seen with the eye but not be projected onto a screen. Figure 249 also
allows one to deduce the thin lens formula that connects the lengths d o , d o and f . What
Challenge 1004 n is it?
Even though glasses and lenses have been known since antiquity, the Middle Ages had
to pass by before two lenses were combined to make more elaborate optical instruments.
The telescope was invented in, or just before, 1608 by the German–Dutch lens grinder

Dvipsbugw
 what is light? 569

final, common
enlarged focus
virtual of both
image lenses
object
intermediate,
real
image

Dvipsbugw
objective ocular
lens lens to human eye

F I G U R E 250 Refraction as the basis of the telescope – shown here in the original Dutch design

Johannes Lipperhey (c. 1570–1619), who made a fortune by selling it to the Dutch mil-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

itary. When Galileo heard about the discovery, he quickly took it over and improved it.
In 1609, Galileo performed the first astronomical observations; they made him world-
famous. The Dutch telescope design has a short tube yielding a bright and upright image.
It is still used today in opera glasses. Many other ways of building telescopes have been
developed over the years. *
Another way to combine two lenses leads to the microscope. Can you explain to a non-
Challenge 1005 n physicist how a microscope works?** Werner Heisenberg almost failed his Ph.D. exam
because he could not. The problem is not difficult, though. Indeed, the inventor of the mi-
croscope was an autodidact of the seventeenth century: the Dutch technician Antoni van
Leeuwenhoek (1632–1723) made a living by selling over five hundred of his microscopes
to his contemporaries.
No ray tracing diagram, be it that of a simple lens, of a telescope or a microscope, is
really complete if the eye, with its lens and retina, is missing. Can you add it and convince
Challenge 1006 ny yourself that these devices really work?
Refraction is often colour-dependent. For that reason, microscopes or photographic
cameras have several lenses, made of different materials. They compensate the colour ef-
fects, which otherwise yield coloured image borders. The colour dependence of refraction
Page 574 in water droplets is also the basis of the rainbow, as shown below, and refraction in ice
crystals in the atmosphere is at the basis of the halos and the many other light patterns
Ref. 525 often seen around the Sun and the Moon.
Copyright © Christoph Schiller November 1997–May 2006

Also the lens in the human eye has colour-dependent diffraction. This is effect is not
corrected in the eye, but in the brain. The dispersion of the eye lens can be noticed if this
correction is made impossible, for example when red or blue letters are printed on a black

* A fascinating overview about what people have achieved in this domain up to now is given by Peter
Manly, Unusual Telescopes, Cambridge University Press, 1991. Images can also be made with mirrors.
Since mirrors are cheaper and more easy to fabricate with high precision, most large telescopes have a mirror
instead of the first lens.
By the way, telescopes also exist in nature. Many spiders have two types of eyes. The large ones, made to
see far away, have two lenses arranged in the same way as in the telescope.
** If not, read the beautiful text by Elizabeth M. Slater & Henry S. Slater, Light and Electron
Microscopy, Cambridge University Press, 1993.

Dvipsbugw
 570 iv cl assical electrodynamics • 14. what is light?

Eye lens dispersion

F I G U R E 251 Watching this graphic at higher

magnification shows the dispersion of the lens in
the human eye: the letters float at different depths Dvipsbugw

α
b

M
preliminary drawing
F I G U R E 252 In certain materials, light F I G U R E 253 Masses bend light

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

beams can spiral around each other

background, as shown in Figure 251. One gets the impression that the red letters float in
Challenge 1007 n front of the blue letters. Can you show how dispersion leads to the floating effect?
A second important observation is that light goes around corners, and the more so
the sharper they are. This effect is called diffraction. In fact, light goes around corners in
the same way that sound does. Diffraction is due to the wave nature of light (and sound).
You probably remember the topic from secondary school.
Because of diffraction, it is impossible to produce strictly parallel light beams. For
example, every laser beam diverges by a certain minimum amount, called the diffraction
limit. Maybe you know that the world’s most expensive Cat’s-eyes are on the Moon, where
Ref. 526 they have been deposited by the Lunakhod and the Apollo missions. Can you determine
how wide a laser beam with minimum divergence has become when it arrives at the Moon
and returns back to Earth, assuming that it was 1 m wide when it left Earth? How wide
Challenge 1008 n would it be when it came back if it had been 1 mm wide at the start?
Diffraction implies that there are no perfectly sharp images: there exists a limit on
resolution. This is true for every optical instrument, including the eye. The resolution of
the eye is between one and two minutes of arc, i.e. between 0.3 and 0.6 mrad. The limit
is due to the finite size of the pupil. Therefore, for example, there is a maximum distance
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1009 n at which humans can distinguish the two headlights of a car. Can you estimate it?
Resolution limits also make it impossible to see the Great Wall in northern China from
the Moon, contrary to what is often claimed. In the few parts that are not yet in ruins, the
wall is about 6 metres wide, and even if it casts a wide shadow during the morning or
the evening, the angle it subtends is way below a second of arc, so that it is completely
invisible to the human eye. In fact, three different cosmonauts who travelled to the Moon
Ref. 527 performed careful searches and confirmed that the claim is absurd. The story is one of the
Challenge 1010 ny most tenacious urban legends. (Is it possible to see the Wall from the space shuttle?) The
largest human-made objects are the polders of reclaimed land in the Netherlands; they
are visible from outer space. So are most large cities as well as the highways in Belgium
at night; their bright illumination makes them stand out clearly from the dark side of the

Dvipsbugw
 what is light? 571

cold air

hot air

F I G U R E 254 Reflection at air interfaces is the basis of the Fata Morgana

Dvipsbugw
Earth.
Diffraction also means that behind a small disc illuminated along its axis, the centre
of the shadow shows, against all expectations, a bright spot. This ‘hole’ in the shadow
was predicted in 1819 by Denis Poisson (1781–1840) in order to show to what absurd con-
sequences the wave theory of light would lead. He had just read the mathematical descrip-
tion of diffraction developed by Augustin Fresnel* on the basis of the wave description

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of light. But shortly afterwards, François Arago (1786–1853) actually observed Poisson’s
point, converting Poisson, making Fresnel famous and starting the general acceptance of
the wave properties of light.
Additional electromagnetic fields usually do not influence light directly, since light has
no charge and since Maxwell’s equations are linear. But in some materials the equations
are non-linear, and the story changes. For example, in certain photorefractive materials,
two nearby light beams can even twist around each other, as was shown by Segev and
Ref. 528 coworkers in 1997.
A final way to bend light is gravity, as discussed already in the chapters on universal
gravity and on general relativity. The effect of gravity between two light beams was also
Page 475 discussed there.
In summary, light travels straight only if it travels far from other matter. In everyday
life, ‘far’ simply means more than a few millimetres, because all gravitational and electro-
magnetic effects are negligible at these distances, mainly due to lights’ truly supersonic
speed.

The concentration of light

If one builds a large lens or a curved mirror, one can collect the light of the Sun and focus
it on a single spot. Everybody has used a converging lens as a child to burn black spots on
Copyright © Christoph Schiller November 1997–May 2006

newspapers in this way. In Spain, wealthier researchers have even built a curved mirror
as large as a house, in order to study solar energy use and material behaviour at high
temperature. Essentially, the mirror provides a cheap way to fire an oven. Indeed, ‘focus’
is the Latin word for ‘oven’.
Kids find out quite rapidly that large lenses allow them to burn things more easily than
small ones. It is obvious that the Spanish site is the record holder in this game. However,
building a larger mirror does not make sense. Whatever its size may be, such a set-up

* Augustin Jean Fresnel (1788–1827), engineer and part time physicist; he published in 1818 his great paper
on wave theory for which he got the prize of the French Academy of Sciences in 1819. To improve his finances,
he worked in the commission responsible for lighthouses, for which he developed the well-known Fresnel
lens. He died prematurely, partly of exhaustion due to overwork.

Dvipsbugw
 572 iv cl assical electrodynamics • 14. what is light?

cannot reach a higher temperature than that of the original light source. The surface tem-
perature of the Sun is about 5800 K; indeed, the highest temperature reached so far is
about 4000 K. Are you able to show that this limitation follows from the second law of
Challenge 1011 ny thermodynamics?
In short, nature provides a limit to
the concentration of light energy. In fact,
we will encounter additional limits in the
course of our exploration.
Dvipsbugw
Can one touch light?
If a little glass bead is put on top of a
powerful laser, the bead remains suspended
in mid-air, as shown in Figure 256.* This
means that light has momentum. There-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fore, contrary to what we said in the begin- F I G U R E 255 The last mirror of the solar furnace
ning of our mountain ascent, images can be at Odeillo, in the French Pyrenees (© Gerhard
touched! In fact, the ease with which objects Weinrebe)
can be pushed even has a special name. For
stars, it is called the albedo, and for general
objects it is called the reflectivity, abbreviated as r.
Like each type of electromagnetic field, and like every kind of
wave, light carries energy; the energy flow T per surface and time preliminary figure
Challenge 1012 e is

T= giving an average `Te =

1 1
EB E max B max . (437)
µ0 2µ 0
light
Obviously, light also has a momentum P. It is related to the energy
E by
P= .
E
(438) F I G U R E 256 Levitating
c a small glass bead with
a laser
Challenge 1013 e As a result, the pressure p exerted by light on a body is given by

p= (1 + r)
T
(439)
Copyright © Christoph Schiller November 1997–May 2006

c
where for black bodies we have that a reflectivity r = 0 and for mirrors r = 1; other bodies
have values in between. What is your guess for the amount of pressure due to sunlight on
Challenge 1014 n a black surface of one square metre? Is this the reason that we feel more pressure during
the day than during the night?
In fact, rather delicate equipment is needed to detect the momentum of light, in other
words, its radiation pressure. Already around 1610, Johannes Kepler had suggested in De
cometis that the tails of comets exist only because the light of the Sun hits the small dust
* The heaviest object that has been levitated with a laser had a mass of 20 g; the laser used was enormous,
and the method also made use of a few additional effects, such as shock waves, to keep the object in the air.

Dvipsbugw
 what is light? 573

particles that detach from it. For this reason, the tail always points away from the Sun, as
Challenge 1015 e you might want to check at the next opportunity. Today, we know that Kepler was right;
but proving the hypothesis is not easy.
In 1873, William Crookes * invented the light
mill radiometer. He had the intention of demon-
strating the radiation pressure of light. The light
mill consists of four thin plates, black on one
side and shiny on the other, that are mounted
on a vertical axis, as shown in Figure 257. How- Dvipsbugw
ever, when Crookes finished building it – it was light
similar to those sold in shops today – he found,
like everybody else, that it turned in the wrong
direction, namely with the shiny side towards
Challenge 1016 n the light! (Why is it wrong?) You can check it by
yourself by shining a laser pointer on to it. The

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

behaviour has been a puzzle for quite some time.
Explaining it involves the tiny amount of gas left
over in the glass bulb and takes us too far from
Ref. 529 the topic of our mountain ascent. It was only in
1901, with the advent of much better pumps, that F I G U R E 257 A commercial light mill turns
the Russian physicist Peter/Pyotr Lebedev man- against the light
aged to create a sufficiently good vacuum to al-
Ref. 530 low him to measure the light pressure with such an improved, true radiometer. Lebedev
also confirmed the predicted value of the light pressure and proved the correctness of
Kepler’s hypothesis. Today it is even possible to build tiny propellers that start to turn
Ref. 531 when light shines on to them, in exactly the same way that the wind turns windmills.
But light cannot only touch and be touched, it can also grab. In the 1980s, Arthur
Ashkin and his research group developed actual optical tweezers that allow one to grab,
Ref. 532 suspend and move small transparent spheres of 1 to 20 µm diameter using laser beams. It
is possible to do this through a microscope, so that one can also observe at the same time
what is happening. This technique is now routinely used in biological research around
the world, and has been used, for example, to measure the force of single muscle fibres,
by chemically attaching their ends to glass or Teflon spheres and then pulling them apart
with such optical tweezers.
But that is not all. In the last decade of the twentieth century, several groups even
Copyright © Christoph Schiller November 1997–May 2006

Ref. 532 managed to rotate objects, thus realizing actual optical spanners. They are able to rotate
particles at will in one direction or the other, by changing the optical properties of the
laser beam used to trap the particle.
In fact, it does not take much to deduce that if light has linear momentum, circularly
polarized light also has angular momentum. In fact, for such a wave the angular mo-
mentum L is given by
L=
E nergy
. (440)
ω

* William Crookes (b. 1832 London, d. 1919 London), English chemist and physicist, president of the Royal
Society, discoverer of thallium.

Dvipsbugw
 574 iv cl assical electrodynamics • 14. what is light?

Challenge 1017 e Equivalently, the angular momentum of a wave is λ~2π times its linear momentum. For
Ref. 533 light, this result was already confirmed in the early twentieth century: a light beam can
Challenge 1018 ny put certain materials (which ones?) into rotation, as shown in Figure 258. Of course, the
whole thing works even better with a laser beam. In the 1960s, a beautiful demonstration
was performed with microwaves. A circularly polarized microwave beam from a maser
– the microwave equivalent of a laser – can put a metal piece absorbing it into rotation.
Indeed, for a beam with cylindrical symmetry, depending on the sense of rotation, the
angular momentum is either parallel or antiparallel to the direction of propagation. All
these experiments confirm that light also carries angular momentum, an effect which will Dvipsbugw
play an important role in the second part of our mountain ascent.
We note that not for all waves angular momentum is
energy per angular frequency. This is only the case for
waves made of what in quantum theory will be called spin suspension
wire
1 particles. For example, for gravity waves the angular mo-
mentum is twice this value, and they are therefore expec-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ted to be made of spin 2 particles.
In summary, light can touch and be touched. Obviously,
if light can rotate bodies, it can also be rotated. Could you
Challenge 1019 n imagine how this can be achieved?

War, light and lies circularly

polarized
From the tiny effects of equation (439) for light pressure light beam
we deduce that light is not an efficient tool for hitting ob-
jects. On the other hand, light is able to heat up objects, F I G U R E 258 Light can rotate
as we can feel when the skin is touched by a laser beam of objects
about 100 mW or more. For the same reason even cheap
laser pointers are dangerous to the eye.
In the 1980s, and again in 2001, a group of people who had read too many science
fiction novels managed to persuade the military – who also indulge in this habit – that
lasers could be used to shoot down missiles, and that a lot of tax money should be spent
on developing such lasers. Using the definition of the Poynting vector and a hitting time
of about 0.1 s, are you able to estimate the weight and size of the battery necessary for
Challenge 1020 ny such a device to work? What would happen in cloudy or rainy weather?
Other people tried to persuade NASA to study the possibility of propelling a rocket
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1021 ny using emitted light instead of ejected gas. Are you able to estimate whether this is feasible?

What is colour?
We saw that radio waves of certain frequencies are visible. Within that range, different
frequencies correspond to different colours. (Are you able to convince a friend about
Challenge 1022 n this?) But the story does not finish here. Numerous colours can be produced either by a
single wavelength, i.e. by monochromatic light, or by a mixture of several different colours.
For example, standard yellow can be, if it is pure, an electromagnetic beam of 600 nm
wavelength or it can be a mixture of standard green of 546.1 nm and standard red of
700 nm. The eye cannot distinguish between the two cases; only spectrometers can. In
everyday life, all colours turn out to be mixed, with the exceptions of those of yellow

Dvipsbugw
 what is light? 575

colour-dependent refraction in glass

white red
glass green
violet

colour-dependent refraction in the eye internal reflection and colour-dependent

(watch pattern at 1 cm distance) refraction in the primary rainbow
white (Sun) Dvipsbugw
water droplet
40.5°
42.4°

violet
green
red

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

internal reflection and colour-dependent
refraction in the secondary rainbow
white (Sun)
water droplet

50.3° 53.6°

red
green
violet

F I G U R E 260 Proving that white light is a mixture of colours

street lamps, of laser beams and of the rainbow. You can check this for yourself, using an
Challenge 1023 e umbrella or a compact disc: they decompose light mixtures, but not pure colours.
In particular, white light is a mixture of a continuous range
of colours with a given intensity per wavelength. To check that
white light is a mixture of colours, simply hold the left-hand
side of Figure 260 so near to your eye that you cannot focus
the stripes any more. The unsharp borders of the white stripes
Copyright © Christoph Schiller November 1997–May 2006

have a pink or a green shade. These colours are due to the im-
perfections of the lens in the human eye, its so-called chromatic
aberrations. Aberrations have the consequence that not all light
frequencies follow the same path through the lens of the eye,
and therefore they hit the retina at different spots. This is the F I G U R E 259 Umbrellas
same effect that occurs in prisms or in water drops showing a decompose white light
rainbow. By the way, the shape of the rainbow tells something
Challenge 1024 n about the shape of the water droplets. Can you deduce the connection?
The right side of Figure 260 explains how rainbows form. The main idea is that internal
reflection inside the water droplets in the sky is responsible for throwing back the light
coming from the Sun, whereas the wavelength-dependent refraction at the air–water sur-

Dvipsbugw
 576 iv cl assical electrodynamics • 14. what is light?

face is responsible for the different paths of each colour. The first two persons to verify this
explanation were Theodoricus Teutonicus de Vriberg (c. 1240 to c. 1318), in the years from
Ref. 534 1304 to 1310 and, at the same time, the Persian mathematician Kamal al-Din al-Farisi. To
check the explanation, they did something smart and simple; anybody can repeat this at
Challenge 1025 e home. They built an enlarged water droplet by filling a thin spherical (or cylindrical) glass
container with water; then they shone a beam of white light through it. Theodoricus and
al-Farisi found exactly what is shown in Figure 260. With this experiment, each of them
was able to reproduce the angle of the main or primary rainbow, its colour sequence, as
well as the existence of a secondary rainbow, its observed angle and its inverted colour Dvipsbugw
Page 561 sequence.* All these bows are visible in Figure 246. Theodoricus’s beautiful experiment is
sometimes called the most important contribution of natural science in the Middle Ages.
Even pure air splits white light. This is the reason that
the sky and far away mountains look blue or that the Sun
looks red at sunset and at sunrise. (The sky looks black even
during the day from the Moon.) You can repeat this effect

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1027 e by looking through water at a black surface or at a lamp.
Adding a few drops of milk to the water makes the lamp yel-
low and then red, and makes the black surface blue (like the
sky seen from the Earth as compared to the sky seen from
the Moon). More milk increases the effect. For the same
reason, sunsets are especially red after volcanic eruptions.
Incidentally, at sunset the atmosphere itself also acts as a
prism; that means that the Sun is split into different images,
one for each colour, which are slightly shifted with respect
to each other, a bit like a giant rainbow in which not only the
rim, but the whole disc is coloured. The total shift is about
1/60th of the diameter. If the weather is favourable and if the
air is clear and quiet up to and beyond the horizon, for a few F I G U R E 261 Milk and water
simulate the evening sky (©
seconds it is possible to see, after the red, orange and yellow Antonio Martos)
images of the setting Sun, the rim of the green–blue image.
Ref. 536 This is the famous ‘rayon vert’ described by Jules Verne in his novel of the same title. It is
often seen on islands, for example in Hawaii.**
To clarify the difference between colours in physics and colour in human perception
and language, a famous linguistic discovery deserves to be mentioned: colours in human
language have a natural order. Colours are ordered by all peoples of the world, whether
Copyright © Christoph Schiller November 1997–May 2006

they come from the sea, the desert or the mountains, in the following order: 1st black and
white, 2nd red, 3rd green and yellow, 4th blue, 5th brown; 6th come mauve, pink, orange,
grey and sometimes a twelfth term that differs from language to language. (Colours that
refer to objects, such as aubergine or sepia, or colours that are not generally applicable,

Challenge 1026 ny * Can you guess where the tertiary and quaternary rainbows are to be seen? There are rare reported sightings
of them. The hunt to observe the fifth-order rainbow is still open. (In the laboratory, bows around droplets up
Ref. 535 to the 13th order have been observed.) For more details, see the beautiful website at http://www.sundog.clara.
co.uk/atoptics/phenom.htm. There are several formulae for the angles of the various orders of rainbows; they
follow from straightforward geometric considerations, but are too involved to be given here.
** For this and many other topics on colours in nature, such as, for example, the halos around the Moon and
the Sun or the colour of shadows,, see the beautiful book by Marcel Minnaert mentioned on page 74.

Dvipsbugw
 what is light? 577

v ph

v gr

v So

v fr Dvipsbugw

F I G U R E 262 The definition of important velocities in wave

phenomena

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

such as blond, are excluded in this discussion.) The precise discovery is the following: if a
particular language has a word for any of these colours, then it also has a word for all the
preceding ones. The result also implies that people use these basic colour classes even if
their language does not have a word for each of them. These strong statements have been
Ref. 537 confirmed for over 100 languages.

What is the speed of light? – Again

Physics talks about motion. Talking is the exchange of sound; and sound is an example of
a signal. A (physical) signal is the transport of information using the transport of energy.
There are no signals without a motion of energy. Indeed, there is no way to store inform-
ation without storing energy. To any signal we can thus ascribe a propagation speed. The
highest possible signal speed is also the maximal velocity of the general influences, or, to
use sloppy language, the maximal velocity with which effects spread causes.
If the signal is carried by matter, such as by the written text in a letter, the signal velocity
is then the velocity of the material carrier, and experiments show that it is limited by the
speed of light.
For a wave carrier, such as water waves, sound, light or radio waves, the situation is less
evident. What is the speed of a wave? The first answer that comes to mind is the speed
with which wave crests of a sine wave move. This already introduced phase velocity is
given by the ratio between the frequency and the wavelength of a monochromatic wave,
Copyright © Christoph Schiller November 1997–May 2006

i.e. by
v ph = .
ω
(441)
k
For example, the phase velocity determines interference phenomena. Light in a vacuum
has the same phase velocity v ph = c for all frequencies. Are you able to imagine an exper-
Challenge 1028 n iment to test this to high precision?
On the other hand, there are cases where the phase velocity is greater than c, most
notably when light travels through an absorbing substance, and when at the same time
the frequency is near to an absorption maximum. In these cases, experiments show that
Ref. 538 the phase velocity is not the signal velocity. For such situations, a better approximation

Dvipsbugw
 578 iv cl assical electrodynamics • 14. what is light?

to the signal speed is the group velocity, i.e. the velocity at which a group maximum will
travel. This velocity is given by
v gr = V ,
dω
(442)
dk k 0

where k 0 is the central wavelength of the wave packet. We observe that ω = c(k)k =
2πv ph ~λ implies the relation

dv ph Dvipsbugw
v gr = V = v ph − λ
dω
. (443)
dk k 0 dλ

This means that the sign of the last term determines whether the group velocity is larger
or smaller than the phase velocity. For a travelling group, as shown by the dashed line in
Figure 262, this means that new maxima appear either at the end or at the front of the
group. Experiments show that this is only the case for light passing through matter; for
light in vacuum, the group velocity has the same value v gr = c for all values of the wave

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

vector k.
You should be warned that many publications are still propagating the incorrect state-
ment that the group velocity in a material is never greater than c, the speed of light in
Challenge 1029 ny vacuum. Actually, the group velocity in a material can be zero, infinite or even negative;
this happens when the light pulse is very narrow, i.e. when it includes a wide range of
frequencies, or again when the frequency is near an absorption transition. In many (but
not all) cases the group is found to widen substantially or even to split, making it difficult
to define precisely the group maximum and thus its velocity. Many experiments have con-
firmed these predictions. For example, the group velocity in certain materials has been
Ref. 539 measured to be ten times that of light. The refractive index then is smaller than 1. However,
in all these cases the group velocity is not the same as the signal speed.*
What then is the best velocity describing signal propagation? The German physicist
Arnold Sommerfeld** almost solved the main problem in the beginning of the twentieth
century. He defined the signal velocity as the velocity v So of the front slope of the pulse, as
Ref. 538 shown in Figure 262. The definition cannot be summarized in a formula, but it does have
the property that it describes signal propagation for almost all experiments, in particular
those in which the group and phase velocity are larger than the speed of light. When
studying its properties, it was found that for no material is Sommerfeld’s signal velocity
greater than the speed of light in vacuum.
Copyright © Christoph Schiller November 1997–May 2006

Sometimes it is conceptually easier to describe signal propagation with the help of

the energy velocity. As previously mentioned, every signal transports energy. The energy
velocity v en is defined as the ratio between the power flow density P, i.e. the Poynting

* In quantum mechanics, Schrödinger proved that the velocity of an electron is given by the group velocity
of its wave function. Therefore the same discussion reappeared in quantum theory, as we will find out in the
second part of the mountain ascent.
** Arnold Sommerfeld (b. 1868 Königsberg, d. 1951 München) was a central figure in the spread of special
and general relativity, of quantum theory, and of their applications. A professor in Munich, an excellent
teacher and text book writer, he worked on atomic theory, on the theory of metals and on electrodynamics,
and was the first to understand the importance and the mystery around ‘Sommerfeld’s famous fine structure
constant.’

Dvipsbugw
 what is light? 579

vector, and the energy density W, both taken in the direction of propagation. For electro-
magnetic fields – the only ones fast enough to be interesting for eventual superluminal
signals – this ratio is
Re(P) 2c 2 E  B
ven = = 2 . (444)
W E + c 2 B2
However, as in the case of the front velocity, in the case of the energy velocity we have
to specify if we mean the energy transported by the main pulse or by the front of it. In
vacuum, neither is ever greater than the speed of light.* (In general, the velocity of energy Dvipsbugw
Ref. 538 in matter has a value slightly different from Sommerfeld’s signal velocity.)
In recent years, the progress in light detector technology, allowing one to detect even
the tiniest energies, has forced scientists to take the fastest of all these energy velocities to
describe signal velocity. Using detectors with the highest possible sensitivity we can use as
signal the first point of the wave train whose amplitude is different from zero, i.e. the first
tiny amount of energy arriving. This point’s velocity, conceptually similar to Sommerfeld’s

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

signal velocity, is commonly called the front velocity or, to distinguish it even more clearly
Challenge 1030 n from Sommerfeld’s case, the forerunner velocity. It is simply given by

v fr = lim
ω
. (445)
ω ª k
The forerunner velocity is never greater than the speed of light in a vacuum, even in
materials. In fact it is precisely c because, for extremely high frequencies, the ratio ω~k is
independent of the material, and vacuum properties take over. The forerunner velocity is
the true signal velocity or the true velocity of light. Using it, all discussions on light speed
become clear and unambiguous.
Challenge 1031 n To end this section, here are two challenges for you. Which of all the velocities of light
is measured in experiments determining the velocity of light, e.g. when light is sent to the
Moon and reflected back? And now a more difficult one: why is the signal speed of light
Challenge 1032 n less inside matter, as all experiments show?

200 years too late – negative refraction indices

In 1968 the Soviet physicist Victor Veselago made a strange prediction: the index of re-
fraction could have negative values without invalidating any known law of physics. A
negative index means that a beam is refracted to the same side of the vertical, as shown
Copyright © Christoph Schiller November 1997–May 2006

in Figure 263.
In 1996, John Pendry and his group proposed ways of realizing such materials. In
2000, a first experimental confirmation for microwave refraction was published, but it
Ref. 542 met with strong disbelief. In 2002 the debate was in full swing. It was argued that neg-
ative refraction indices imply speeds greater than that of light and are only possible for
either phase velocity or group velocity, but not for the energy or true signal velocity. The

* Signals not only carry energy, they also carry negative entropy (‘information’). The entropy of a transmitter
increases during transmission. The receiver decreases in entropy (but less than the increase at the transmitter,
Ref. 540 of course).
Note that the negative group velocity implies energy transport against the propagation velocity of light.
Ref. 541 This is possible only in energy loaded materials.

Dvipsbugw
 580 iv cl assical electrodynamics • 14. what is light?

α air α air

water left-handed
(n>0) material
(n<0)
β β

Dvipsbugw
F I G U R E 263 Positive and negative indices of refraction

conceptual problems would arise only because in some physical systems the refraction
angle for phase motion and for energy motion differ.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 543 Today, the consensus is the following: a positive index of refraction less than one is im-
possible, as it implies an energy speed of greater than one. A negative index of refraction,
however, is possible if it is smaller than −1. Negative values have indeed been frequently
observed; the corresponding systems are being extensively explored all over the world.
The materials showing this property are called left-handed. The reason is that the vectors
of the electric field, the magnetic field and the wave vector form a left-handed triplet,
in contrast to vacuum and most usual materials, where the triplet is right-handed. Such
materials consistently have negative magnetic permeability and negative dielectric coeffi-
Ref. 544 cient (permittivity).
Left-handed materials have negative phase velocities, i.e., a phase velocity opposed to
the energy velocity, they show reversed Doppler effects and yield obtuse angles in the
Çerenkov effect (emitting Çerenkov radiation in the backward instead of the forward
direction).
But, most intriguing, negative refraction materials are predicted to allow the construc-
Ref. 545 tion of lenses that are completely flat. In addition, in the year 2000, John Pendry gained
the attention of the whole physics community world-wide by predicting that lenses made
with such materials, in particular for n = −1, would be perfect, thus beating the usual dif-
fraction limit. This would happen because such a lens also images the evanescent parts
of the waves, by amplifying them accordingly. First experiments seem to confirm the
Ref. 544 prediction. Discussion on the topic is still in full swing.
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1033 ny Can you explain how negative refraction differs from diffraction?

Signals and predictions

When one person reads a text over the phone to a neighbour who listens to it and maybe
repeats it, we speak of communication. For any third person, the speed of communica-
tion is always less than the speed of light. But if the neighbour already knows the text, he
can recite it without having heard the readers’ voice. To the third observer such a situation
appears to imply motion that is faster than light. Prediction can thus mimic communic-
ation and, in particular, it can mimic faster-than-light (superluminal) communication.
Ref. 546 Such a situation was demonstrated most spectacularly in 1994 by Günter Nimtz, who
seemingly transported music – all music is predictable for short time scales – through a

Dvipsbugw
 what is light? 581

TA B L E 49 Experimental properties of (flat) vacuum and of the

‘aether’

Physical propert y E x p e r i m e n t a l va l u e

permeability µ 0 =1.3 µH~m

permittivity ε 0 =8.9 pF~m
wave impedance/resistance Z 0 = 376.7 Ω
conformal invariance applies
spatial dimensionality 3 Dvipsbugw
topology R3
mass and energy content not detectable
friction on moving bodies not detectable
motion relative to space-time not detectable

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

‘faster-than-light’ system. To distinguish between the two situations, we note that in the
case of prediction, no transport of energy takes place, in contrast to the case of commu-
nication. In other words, the definition of a signal as a transporter of information is not
as useful and clear-cut as the definition of a signal as a transporter of energy. In the above-
mentioned experiment, no energy was transported faster than light. The same distinction
between prediction on the one hand and signal or energy propagation on the other will
be used later to clarify some famous experiments in quantum mechanics.
If the rate at which physics papers are being

“ published continues to increase, physics

journals will soon be filling library shelves faster
than the speed of light. This does not violate
relativity since no useful information is being
transmitted.
David Mermin

Does the aether exist?

Gamma rays, light and radio waves are moving electromagnetic waves. All exist in empty
”
space. What is oscillating when light travels? Maxwell himself called the ‘medium’ in
which this happens the aether. The properties of the aether measured in experiments are
listed in Table 49.
Copyright © Christoph Schiller November 1997–May 2006

Of course, the values of the permeability and the permittivity of the vacuum are related to
Page 1155 the definition of the units henry and farad. The last item of the table is the most important:
Ref. 547 despite intensive efforts, nobody has been able to detect any motion of the aether. In other
words, even though the aether supposedly oscillates, it does not move. Together with
the other data, all these results can be summed up in one sentence: there is no way to
Challenge 1034 n distinguish the aether from the vacuum: they are one and the same.
Sometimes one hears that certain experiments or even the theory of relativity show
that the aether does not exist. This is not strictly correct. In fact, experiments show some-
thing more important. All the data show that the aether is indistinguishable from the
vacuum. Of course, if we use the change of curvature as the definition for motion of the
vacuum, the vacuum can move, as we found out in the section on general relativity; but

Dvipsbugw
 582 iv cl assical electrodynamics • 14. what is light?

grass dew head Sun

(not to
scale)

F I G U R E 264 The path of light for the dew on grass that is Dvipsbugw
responsible for the aureole

the aether still remains indistinguishable from it.*

Later we will even find out that the ability of the vacuum to allow the propagation of
light and its ability to allow the propagation of particles are equivalent: both require the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

same properties. Therefore the aether remains indistinguishable from a vacuum in the rest
Ref. 548 of our walk. In other words, the aether is a superfluous concept; we will drop it from our
walk from now on. Despite this result, we have not yet finished the study of the vacuum;
vacuum will keep us busy for a long time, starting with the intermezzo following this
chapter. Moreover, quite a few of the aspects in Table 49 will require some amendments
later.

Curiosities and fun challenges about light

How to prove you’re holy
Light reflection and refraction are responsible for many effects. The originally Indian
symbol of holiness, now used throughout most of the world, is the aureole, also called halo
or Heiligenschein, a ring of light surrounding the head. You can easily observe it around
your own head. You need only to get up early in the morning and look into the wet grass
while turning your back to the Sun. You will see an aureole around your shadow. The effect
is due to the morning dew on the grass, which reflects the light back predominantly in
the direction of the light source, as shown in Figure 264. The fun part is that if you do
Ref. 549 this in a group, you will see the aureole around only your own head.
Retroreflective paint works in the same way: it contains tiny glass spheres that play the
role of the dew. A large surface of retroreflective paint, a traffic sign for example, can also
Copyright © Christoph Schiller November 1997–May 2006

Ref. 550 show your halo if the light source is sufficiently far away. Also the so-called ‘glow’ of the
eyes of a cat at night is due to the same effect; it is visible only if you look at the cat with
Challenge 1035 n a light source behind you. By the way, do Cat’s-eyes work like a cat’s eyes?

Ref. 548 * In fact, the term ‘aether’ has been used as an expression for several different ideas, depending on the
author. First of all it was used for the idea that a vacuum is not empty, but full; secondly, that this fullness
can be described by mechanical models, such as gears, little spheres, vortices, etc.; thirdly, it was imagined
that a vacuum is similar to matter, being made of the same substrate. Interestingly, some of these issues will
reappear in the third part of our mountain ascent.

Dvipsbugw
 curiosities and fun challenges abou t light 583

Dvipsbugw

F I G U R E 265 A limitation of the eye

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Do we see what exists?
Sometimes we see less than there is. Close your left eye, look at the white spot in Figure 265,
bring the page slowly towards your eye, and pay attention to the middle lines. At a distance
Challenge 1036 n of about 15 to 20 cm the middle line will seem uninterrupted. Why?
On the other hand, sometimes we see more than there is, as Figures 266 and 267 show.
The first shows that parallel lines can look skewed, and the second show a so-called Her-
mann lattice, named after its discoverer.* The Hermann lattice of Figure 267, discovered
by Elke Lingelbach in 1995, is especially striking. Variations of these lattices are now used
Ref. 551 to understand the mechanisms at the basis of human vision. For example, they can be
used to determine how many light sensitive cells in the retina are united to one signal
pathway towards the brain. The illusions are angle dependent because this number is
also angle dependent.
Our eyes also ‘see’ things differently: the retina sees an inverted image of the world.
There is a simple method to show this, due to Helmholtz.** You need only a needle and
a piece of paper, e.g. this page of text. Use the needle to make two holes inside the two
letters ‘oo’. Then keep the page as close to your eye as possible, look through the two holes
towards the wall, keeping the needle vertical, a few centimetres behind the paper. You
will see two images of the needle. If you now cover the left hole with your finger, the right
Copyright © Christoph Schiller November 1997–May 2006

needle will disappear, and vice versa. This shows that the image inside the eye, on the
Challenge 1037 ny retina, is inverted. Are you able to complete the proof?

* Ludimar Herrmann (1838–1914), Swiss physiologist. The lattices are often falsely called ‘Hering lattices’
after the man who made Hermann’s discovery famous.
** See Hermann von Helmholtz, Handbuch der physiologischen Optik, 1867. This famous classic is
available in English as Handbook of Physiological Optics, Dover, 1962. The Prussian physician, physicist and
science politician born as Hermann Helmholtz (b. 1821 Potsdam, d. 1894 Charlottenburg) was famous for
his works on optics, acoustics, electrodynamics, thermodynamics, epistemology and geometry. He founded
several physics institutions across Germany. He was one of the first to propagate the idea of conservation
of energy. His other important book, Die Lehre von den Tonempfindungen, published in 1863, describes the
basis of acoustics and, like the handbook, is still worth reading.

Dvipsbugw
584 iv cl assical electrodynamics • 14. what is light?

Dvipsbugw

F I G U R E 266 What is the angle between adjacent horizontal lines?

F I G U R E 267 The Lingelbach lattice: do you see white, grey, or black dots? Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net
Copyright © Christoph Schiller November 1997–May 2006

Two other experiments can show the same result. If you push very lightly on the inside
of your eye (careful!), you will see a dark spot appear on the outside of your vision field.
And if you stand in a dark room and ask a friend to look at a burning candle, explore his
eye: you will see three reflections: two upright ones, reflected from the cornea and from
the lens, and a dim third one, upside-down, reflected form the retina.
Another reason that we do not see the complete image of nature is that our eyes have
a limited sensitivity. This sensitivity peaks around 560 nm; outside the red and the violet,
the eye does not detect radiation. We thus see only part of nature. For example, infrared
photographs of nature, such as the one shown in Figure 268, are interesting because they
show us something different from what we see usually.

Dvipsbugw
 curiosities and fun challenges abou t light 585

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 268 An example of an infrared photograph, slightly mixed with a colour
image (© Serge Augustin)

Every expert of motion should also know that the sensitivity of the eye does not cor-
Ref. 552 respond to the brightest part of sunlight. This myth has been spread around the world
by the numerous textbooks that have copied from each other. Depending on whether
frequency or wavelength or wavelength logarithm is used, the solar spectrum peaks at
500 nm, 880 nm or 720 nm. They human eye’s spectral sensitivity, like the completely dif-
ferent sensitivity of birds or frogs, is due to the chemicals used for detection. In short,
the human eye can only be understood by a careful analysis of its particular evolutionary
history.
An urban legend says that newborn babies see everything upside down. Can you ex-
Challenge 1038 n plain why this idea is wrong?
In summary, we have to be careful when maintaining that seeing means observing.
Examples such as these lead to ask whether there are other limitations of our senses which
are less evident. And our walk will indeed uncover several of them.
Copyright © Christoph Schiller November 1997–May 2006

How does one make pictures of the inside of the eye?

The most beautiful pictures so far of a living human retina, such as that of Figure 269, were
made by the group of David Williams and Austin Roorda at the University at Rochester in
Ref. 553 New York. They used adaptive optics, a technique that changes the shape of the imaging
lens in order to compensate for the shape variations of the lens in the human eye.*

* Nature uses another trick to get maximum resolution: the eye continuously performs small movements,
called micronystagmus. The eye continuously oscillates around the direction of vision with around 40 to
50 Hz. In addition, this motion is also used to allow the cells in the retina to recharge.

Dvipsbugw
 586 iv cl assical electrodynamics • 14. what is light?

Dvipsbugw

F I G U R E 269 A high quality photograph of a live human retina, including a measured (false colour)
indication of the sensitivity of each cone cell (© Austin Roorda)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The eyes see colour by averaging the intensity arriving at the red, blue and green sensit-
Page 574 ive cones. This explains the possibility, mentioned above, of getting the same impression
of colour, e.g. yellow, either by a pure yellow laser beam, or by a suitable mixture of red
and green light.
But if the light is focused on to one cone only, the eye makes mistakes. If, using this
adaptive optics, a red laser beam is focused such that it hits a green cone only, a strange
thing happens: even though the light is red, the eye sees a green colour!
Incidentally, Figure 269 is quite puzzling. In the human eye, the blood vessels are loc-
ated in front of the cones. Why don’t they appear in the picture? And why don’t they
Challenge 1039 n disturb us in everyday life? (The picture does not show the other type of sensitive light
cells, the rods, because the subject was in ambient light; rods come to the front of the
retina only in the dark, and then produce black and white pictures.
Of all the mammals, only primates can see colours. Bulls for example, don’t; they can-
not distinguish red from blue. On the other hand, the best colour seers overall are the
birds. They have cone receptors for red, blue, green, UV and, depending on the bird, for
up to three more sets of colours. A number of birds (but not many) also have a better eye
resolution than humans. Several birds also have a faster temporal resolution: humans see
continuous motion when the images follow with 30 to 70 Hz (depending on the image
Copyright © Christoph Schiller November 1997–May 2006

content); some insects can distinguish images up to 300 Hz.

How does one make holograms and other three-dimensional

images?
Our sense of sight gives us the impression of depth mainly due to three effects. First,
the two eyes see different images. Second, the images formed in the eyes are position
dependent. Third, our eye needs to focus differently for different distances.
A simple photograph does not capture any of the three effects. A photograph corres-
ponds to the picture taken by one eye, from one particular spot and at one particular
Challenge 1040 e focus. In fact, all photographic cameras are essentially copies of a single static eye.
Any system wanting to produce the perception of depth must include at least one of

Dvipsbugw
 curiosities and fun challenges abou t light 587

holographic plate developed

virtual film
object observer
object image

reference Dvipsbugw
reconstruction
beam beam

laser laser or point source

F I G U R E 270 The recording and the observation of a hologram

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the three effects just mentioned. In all systems so far, the third and weakest effect, varying
focus with distance, is never used, as it is too weak. Stereo photography and virtual reality
systems extensively use the first effect by sending two different images to the eyes. Also
certain post cards and computer screens are covered by thin cylindrical lenses that allow
one to send two different images to the two eyes, thus generating the same impression of
depth.
But obviously the most spectacular effect is obtained whenever position dependent
images can be created. Some virtual reality systems mimic this effect using a sensor at-
tached to the head, and creating computer–generated images that depend on this pos-
ition. However, such systems are not able to reproduce actual situations and thus pale
when compared with the impression produced by holograms.
Holograms reproduce all that is seen from any point of a region of space. A hologram
is thus a stored set of position dependent pictures of an object. It is produced by storing
amplitude and phase of the light emitted by an object. To achieve this, the object is illu-
minated by a coherent light source, such as a laser, and the interference pattern is stored.
Illuminating the developed photographic film by a coherent light source then allows one
to see a full three-dimensional image. In particular, due to the reproduction of the situ-
Copyright © Christoph Schiller November 1997–May 2006

ation, the image appears to float in free space. Holograms were developed in 1947 by the
Hungarian physicist Dennis Gabor (1900–1979), who received the 1971 Nobel Prize for
physics for this work.
Holograms can be made to work in reflection or transmission. The simplest holograms
use only one wavelength. Most coloured holograms are rainbow holograms, showing false
colours that are unrelated to the original objects. Real colour holograms, made with three
different lasers, are rare but possible.
Is it possible to make moving holograms? Yes; however, the technical set-ups are still
extremely expensive. So far, they exist only in a few laboratories and cost millions of euro.
By the way, can you describe how you would distinguish a moving hologram from a real
Challenge 1041 n body, if you ever came across one, without touching it?

Dvipsbugw
588 iv cl assical electrodynamics • 14. what is light?

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 271 Sub-wavelength optical microscopy using
stimulated emission depletion (© MPI für biophysikalische
Chemie/Stefan Hell)

Imaging
Producing images is an important part of modern society. The quality images depends
on the smart use of optics, electronics, computers and materials science. Despite long
experience in this domain, there are still new results in the field. Images, i.e. two or three-
dimensional reproductions, can be taken by at least four methods:
— Photography uses a light source, lenses and film or another large area detector. Pho-
tography can be used in reflection, in transmission, with phase-dependence and in
many other ways.
— Holography uses lasers and large area detectors, as explained above. Holography allows
to take three-dimensional images of objects. It is usually used in reflection, but can also
be used in transmission.
— Scanning techniques construct images point by point through the motion of the de-
Copyright © Christoph Schiller November 1997–May 2006

tector, the light source or both.

— Tomography, usually in transmission, uses a source and a line detector that are both
rotated around an object. This allows to image cross sections of materials.
In all methods, the race is for images with the highest resolution possible. The techniques
of producing images with resolutions less than the wavelength of light have made great
progress in recent years.
A recent technique, called stimulated emission depletion microscopy, allows spot sizes
of molecular size. The conventional diffraction limit for microscopes is

dE
λ
, (446)
2n sin α

Dvipsbugw
 charges are discrete – the limits of cl assical electrodynamics 589

where λ is the wavelength, n the index of refraction and α is the angle of observation.
The new technique, a special type of fluorescence microscopy developed by Stefan Hell,
modifies this expression to
dE »
λ
, (447)
2n sin α; I~I sat

so that a properly chosen saturation intensity allows one to reduce the diffraction limit
to arbitrary low values. So far, light microscopy with a resolution of 16 nm has been
Dvipsbugw
Ref. 554 performed, as shown in Figure 271. This and similar techniques should become common-
place in the near future.

Light as weapon?
In many countries, there is more money to study assault weapons than to increase the edu-
cation and wealth of their citizen. Several types of assault weapons using electromagnetic

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

radiation are being researched. Two are particularly advanced.
The first weapon is a truck with a movable parabolic antenna on its roof, about 1 m in
size, that emits a high power (a few kW) microwave (95 GHz) beam. The beam, like all
microwave beams, is invisible; depending on power and beam shape, it is painful or lethal,
up to a distance of a 100 m. This terrible device, with which the operator can make many
many victims without even noticing, was ready in 2006. (Who expects that a parabolic
antenna is dangerous?) Efforts to ban it across the world are slowly gathering momentum.
The second weapon under development is the so-called pulsed impulse kill laser. The
idea is to take a laser that emits radiation that is not absorbed by air, steam or similar
obstacles. An example is a pulsed deuterium fluoride laser that emits at 3.5 µm. This laser
burns every material it hits; in addition, the evaporation of the plasma produced by the
burn produces a strong hit, so that people hit by such a laser are hurt and hit at the same
time. Fortunately, it is still difficult to make such a device rugged enough for practical use.
But experts expect battle lasers to appear soon.
In short, it is probable that radiation weapons will appear in the coming years.*

15. charges are discrete – the limits of cl assical

electrodynamics Copyright © Christoph Schiller November 1997–May 2006

One of the most important results of physics: electric charge is discrete has already been
mentioned a number of times. Charge does not vary continuously, but changes in fixed
steps. Not only does nature show a smallest value of entropy and smallest amounts of
matter; nature also shows a smallest charge. Electric charge is quantized.
In metals, the quantization of charge is noticeable in the flow of electrons. In electro-
lytes, i.e. electrically conducting liquids, the quantization of charge appears in the flow
of charged atoms, usually called ions. All batteries have electrolytes inside; also water is
an electrolyte, though a poorly conducting one. In plasmas, like fire or fluorescent lamps,

* What a man working on such developments tells his children when he comes home in the evening is not
clear.

Dvipsbugw
 590 iv cl assical electrodynamics • 15. charges are discrete

both ions and electrons move and show the discreteness of charge. Also in radiation –
from the electron beams inside TVs, channel rays formed in special glass tubes, and cos-
mic radiation, up to radioactivity – charges are quantized.
From all known experiments, the same smallest value for charge change has been
found. The result is
∆q E e = 1.6  10−19 C . (448)

In short, like all flows in nature, the flow of electricity is due to a flow of discrete particles.
A smallest charge change has a simple implication: classical electrodynamics is wrong. Dvipsbugw
A smallest charge implies that no infinitely small test charges exist. But such infinitely
small test charges are necessary to define electric and magnetic fields. The limit on charge
size also implies that there is no correct way of defining an instantaneous electric current
and, as a consequence, that the values of electric and magnetic field are always somewhat
fuzzy. Maxwell’s evolution equations are thus only approximate.
We will study the main effects of the discreteness of charge in the part on quantum

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 753 theory. Only a few effects of the quantization of charge can be treated in classical physics.
An instructive example follows.

How fast do charges move?

In vacuum, such as inside a colour television, charged particles accelerated by a tension
Challenge 1042 n of 30 kV move with a third of the speed of light. In modern particle accelerators charges
move so rapidly that their speed is indistinguishable from that of light for all practical
purposes.
Inside a metal, electric signals move with speeds of the order of the speed of light. The
precise value depends on the capacity and impedance of the cable and is usually in the
range 0.3c to 0.5c. This high speed is due to the ability of metals to easily take in arriving
charges and to let others depart. The ability for rapid reaction is due to the high mobility
of the charges inside metals, which in turn is due to the small mass and size of these
charges, the electrons.
The high signal speed in metals appears to contradict another determination. The drift
speed of the electrons in a metal wire obviously obeys

v=
I
, (449)
Ane
Copyright © Christoph Schiller November 1997–May 2006

where I is the current, A the cross-section of the wire, e the charge of a single electron and
n the number density of electrons. The electron density in copper is 8.5 ë 1028 m−3 . Using
a typical current of 0.5 A and a typical cross-section of a square millimetre, we get a drift
speed of 0.37 µm~s. In other words, electrons move a thousand times slower than ketchup
inside its bottle. Worse, if a room lamp used direct current instead of alternate current,
the electrons would take several days to get from the switch to the bulb! Nevertheless,
the lamp goes on or off almost immediately after the switch is activated. Similarly, the
electrons from an email transported with direct current would arrive much later than a
paper letter sent at the same time; nevertheless, the email arrives quickly. Are you able to
Challenge 1043 n explain the apparent contradiction between drift velocity and signal velocity?

Dvipsbugw
 charges are discrete – the limits of cl assical electrodynamics 591

Inside liquids, charges move with a different speed from that inside metals, and their
charge to mass ratio is also different. We all know this from direct experience. Our nerves
work by using electric signals and take (only) a few milliseconds to respond to a stimulus,
even though they are metres long. A similar speed is observed inside semiconductors
and inside batteries. In all these systems, moving charge is transported by ions; they are
charged atoms. Ions, like atoms, are large and composed entities, in contrast to the tiny
electrons.
In other systems, charges move both as electrons and as ions. Examples are neon lamps,
fire, plasmas and the Sun. Inside atoms, electrons behave even more strangely. One tends Dvipsbugw
to think that they orbit the nucleus (as we will see later) at a rather high speed, as the
orbital radius is so small. However, it turns out that in most atoms many electrons do not
orbit the nucleus at all. The strange story behind atoms and their structure will be told in
Page 754 the second part of our mountain ascent.

Challenges and curiosities about charge discreteness

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Charge discreteness is one of the central results of physics.
**
Challenge 1044 n How would you show experimentally that electrical charge comes in smallest chunks?
**
The discreteness of charge implies that one can estimate the size of atoms by observing
Challenge 1045 ny galvanic deposition. How?
**
Cosmic radiation consists of charged particles hitting the Earth. (We will discuss this
Page 897 in more detail later.) Astrophysicists explain that these particles are accelerated by the
Ref. 555 magnetic fields around the Galaxy. However, the expression of the Lorentz acceleration
shows that magnetic fields can only change the direction of the velocity of a charge, not
Challenge 1046 ny its magnitude. How can nature get acceleration nevertheless?
**
What would be the potential of the Earth in volt if we could take away all the electrons of
Challenge 1047 n a drop of water?
Copyright © Christoph Schiller November 1997–May 2006

**
When a voltage is applied to a resistor, how long does it take until the end value of the cur-
rent, given by Ohm’s ‘law’, is reached? The first to answer this question was Paul Drude.*
in the years around 1900. He reasoned that when the current is switched on, the speed v
of an electron increases as v = (eE~m)t, where E is the electrical field, e the charge and m
the mass of the electron. Drude’s model assumes that the increase of electron speed stops

* Paul Karl Ludwig Drude (1863–1906), German physicist. A result of his electron gas model of metals was
the prediction, roughly correct, that the ratio between the thermal conductivity and the electronic conduct-
ivity at a given temperature should be the same for all metals. Drude also introduced c as the symbol for the
speed of light.

Dvipsbugw
 592 iv cl assical electrodynamics • 16. electromagnetic effects

when the electron hits an atom, loses its energy and begins to be accelerated again. Drude
Challenge 1048 ny deduced that the average time τ up to the collision is related to the specific resistance by

ρ=
2m
, (450)
τe 2 n
with n being the electron number density. Inserting numbers for copper (n =
10.3 ë 1028 ~m−3 and ρ = 0.16 ë 10−7 Ωm), one gets a time τ = 42 ps. This time is so
short that the switch-on process can usually be neglected. Dvipsbugw

16. electromagnetic effects and challenges

Classical electromagnetism and light are almost endless topics. Some aspects are too beau-
tiful to be missed.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Since light is a wave, something must happen if it is directed to a hole less than its
Challenge 1049 n wavelength in diameter. What exactly happens?
**
Electrodynamics shows that light beams always push; they never pull. Can you confirm
Challenge 1050 e that ‘tractor beams’ are impossible in nature?

**
It is well known that the glowing material in light bulbs is tungsten wire in an inert gas.
This was the result of a series of experiments that began with the grandmother of all
lamps, namely the cucumber. The older generation knows that a pickled cucumber, when
attached to the 230 V of the mains, glows with a bright green light. (Be careful; the exper-
iment is dirty and somewhat dangerous.)

**
If you calculate the Poynting vector for a charged magnet – or simpler, a point charge near
a magnet – you get a surprising result: the electromagnetic energy flows in circles around
Challenge 1051 n the magnet. How is this possible? Where does this angular momentum come from?
Copyright © Christoph Schiller November 1997–May 2006

Worse, any atom is an example of such a system – actually of two such systems. Why
Ref. 556 is this effect not taken into account in calculations in quantum theory?

**
Ohm’s law, the observation that for almost all materials the current is proportional to
the voltage, is due to a school teacher. Georg Simon Ohm* explored the question in great
depth; in those days, such measurements were difficult to perform. This has changed now.

* Georg Simon Ohm (b. 1789 Erlangen, d. 1854 München), Bavarian school teacher and physicist. His ef-
forts were recognized only late in his life, and he eventually was promoted to professor at the University in
München. Later the unit of electrical resistance, the proportionality factor between voltage and current, was
named after him.

Dvipsbugw
 electromagnetic effects and challenges 593

insulators
C1 high tension line

C2 wires
neon lamp
F I G U R E 272 F I G U R E 273 Small neon lamps on a high
Capacitors in voltage cable
series Dvipsbugw

Recently, even the electrical resistance of single atoms has been measured: in the case of
Ref. 557 xenon it turned out to be about 105 Ω. It was also found that lead atoms are ten times
Challenge 1052 ny more conductive than gold atoms. Can you imagine why?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Ref. 558 The charges on two capacitors in series are not generally equal, as naive theory states.
For perfect, leak-free capacitors the voltage ratio is given by the inverse capacity ratio
V1 ~V2 = C 2 ~C 1 , due to the equality of the electric charges stored. This is easily deduced
from Figure 272. However, in practice this is only correct for times between a few and a
Challenge 1053 n few dozen minutes. Why?
**
Does it make sense to write Maxwell’s equations in vacuum? Both electrical and magnetic
fields require charges in order to be measured. But in vacuum there are no charges at all.
In fact, only quantum theory solves this apparent contradiction. Are you able to imagine
Challenge 1054 d how?
**
Ref. 559 Grass is usually greener on the other side of the fence. Can you give an explanation based
Challenge 1055 n on observations for this statement?
**
The maximum force in nature limits the maximum charge that a black hole can carry.
Challenge 1056 ny Can you find the relation?
Copyright © Christoph Schiller November 1997–May 2006

**
On certain high voltage cables leading across the landscape, small neon lamps shine when
the current flows, as shown in Figure 273. (You can see them from the train when riding
Challenge 1057 ny from Paris to the Roissy airport.) How is this possible?
**
‘Inside a conductor there is no electric field.’ This statement is often found. In fact the
truth is not that simple. First, a static field or a static charge on the metal surface of a
body does not influence fields and charges inside it. A closed metal surface thus forms a
Challenge 1058 n shield against an electric field. Can you give an explanation? In fact, a tight metal layer is

Dvipsbugw
 594 iv cl assical electrodynamics • 16. electromagnetic effects

not required to get the effect; a cage is sufficient. One speaks of a Faraday cage.
The detailed mechanism allows you to answer the following question: do Faraday cages
for gravity exist? Why?
For moving external fields or charges, the issue is more complex. Fields due to accel-
erated charges – radiation fields – decay exponentially through a shield. Fields due to
charges moving at constant speed are strongly reduced, but do not disappear. The reduc-
tion depends on the thickness and the resistivity of the metal enclosure used. For sheet
metal, the field suppression is very high; it is not necessarily high for metal sprayed plastic.
Ref. 560 Such a device will not necessarily survive a close lightning stroke. Dvipsbugw
In practice, there is no danger if lightning hits an aeroplane or a car, as long they are
made of metal. (There is one film on the internet of a car hit by lightning; the driver does
not even notice.) However, if your car is hit by lightning in dry weather, you should wait
a few minutes before getting out of it. Can you imagine why?
Faraday cages also work the other way round. (Slowly) changing electric fields chan-
ging that are inside a Faraday cage are not felt outside. For this reason, radios, mobile

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

phones and computers are surrounded by boxes made of metal or metal-sprayed plastics.
The metal keeps the so-called electromagnetic smog to a minimum.
There are thus three reasons to surround electric appliances by a grounded shield:
to protect the appliance from outside fields, to protect people and other machines from
electromagnetic smog, and to protect people against the mains voltage accidentally being
fed into the box (for example, when the insulation fails). In high precision experiments,
these three functions can be realized by three separate cages.
For purely magnetic fields, the situation is more complex. It is quite difficult to shield
the inside of a machine from outside magnetic fields. How would you do it? In practice
Challenge 1059 ny one often uses layers of so-called mu-metal; can you guess what this material does?
**
Electric polarizability is the property of matter responsible for the deviation of water flow-
Page 518 ing from a tap caused by a charged comb. It is defined as the strength of electric dipole
induced by an applied electric field. The definition simply translates the observation that
many objects acquire a charge when an electric field is applied. Incidentally, how precisely
combs get charged when rubbed, a phenomenon called electrification, is still one of the
mysteries of modern science.
**
Copyright © Christoph Schiller November 1997–May 2006

A pure magnetic field cannot be transformed into a pure electric field by change of ob-
servation frame. The best that can be achieved is a state similar to an equal mixture of
Challenge 1060 ny magnetic and electric fields. Can you provide an argument elucidating this relation?
**
Ref. 561 Researchers are trying to detect tooth decay with the help of electric currents, using the
observation that healthy teeth are bad conductors, in contrast to teeth with decay. How
Challenge 1061 ny would you make use of this effect in this case? (By the way, it might be that the totally
unrelated technique of imaging with terahertz waves could yield similar results.)

**

Dvipsbugw
 electromagnetic effects and challenges 595

A team of camera men in the middle of the Sahara were using battery-driven electrical
equipment to make sound recordings. Whenever the microphone cable was a few tens
of metres long, they also heard a 50 Hz power supply noise, even though the next power
supply was thousands of kilometres away. An investigation revealed that the high voltage
lines in Europe lose a considerable amount of power by irradiation; these 50 Hz waves are
reflected by the ionosphere around the Earth and thus can disturb recording in the middle
of the desert. Can you estimate whether this observation implies that living directly near
Challenge 1062 ny a high voltage line is dangerous?
Dvipsbugw
**
Ref. 562 When two laser beams cross at a small angle, they can form light pulses that seem to move
Challenge 1063 n faster than light. Does this contradict special relativity?
**
It is said that astronomers have telescopes so powerful that they can see whether some-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1064 ny body is lighting a match on the Moon. Can this be true?

**
When solar plasma storms are seen on the Sun, astronomers first phone the electricity
company. They know that about 24 to 48 hours later, the charged particles ejected by
the storms will arrive on Earth, making the magnetic field on the surface fluctuate. Since
power grids often have closed loops of several thousands of kilometres, additional electric
currents are induced, which can make transformers in the grid overheat and then switch
off. Other transformers then have to take over the additional power, which can lead to
their overheating, etc. On several occasions in the past, millions of people have been left
without electrical power due to solar storms. Today, the electricity companies avoid the
problems by disconnecting the various grid sections, by avoiding large loops, by reducing
the supply voltage to avoid saturation of the transformers and by disallowing load transfer
from failed circuits to others.

**
Is it really possible to see stars from the bottom of a deep pit or of a well, even during the
Challenge 1065 n day, as is often stated?
**
Copyright © Christoph Schiller November 1997–May 2006

If the electric field is described as a sum of components of different frequencies, its so-
Ref. 563 called Fourier components, the amplitudes are given by

Ê(k, t) =
1
(2π)3 ~2 ∫ E(x, t)e −ikx
d3 x (451)

and similarly for the magnetic field. It then turns out that a Lorentz invariant quantity N,
describing the energy per circular frequency ω, can be defined:

SE(k, t)S2 + SB(k, t)S2 3

N=
1
8π ∫ cSkS
d k. (452)

Dvipsbugw
 596 iv cl assical electrodynamics • 16. electromagnetic effects

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 274 How natural colours (top) change for three types of colour blind: deutan, protan and
tritan (© Michael Douma)

Challenge 1066 n Can you guess what N is physically? (Hint: think about quantum theory.)
**
Faraday discovered how to change magnetism into electricity, knowing that electricity
could be transformed into magnetism. (The issue is subtle. Faraday’s law is not the dual of
Ampère’s, as that would imply the use of magnetic monopoles; neither is it the reciprocal,
as that would imply the displacement current. But he was looking for a link and he found
a way to relate the two observations – in a novel way, as it turned out.) Faraday also
discovered how to transform electricity into light and into chemistry. He then tried to
Challenge 1067 ny change gravitation into electricity. But he was not successful. Why not?

**
At high altitudes above the Earth, gases are completely ionized; no atom is neutral. One
speaks of the ionosphere, as space is full of positive ions and free electrons. Even though
both charges appear in exactly the same number, a satellite moving through the iono-
Challenge 1068 n sphere acquires a negative charge. Why? How does the charging stop?
Copyright © Christoph Schiller November 1997–May 2006

**
A capacitor of capacity C is charged with a voltage U. The stored electrostatic energy is
E = CU 2 ~2. The capacitor is then detached from the power supply and branched on to an
empty capacitor of the same capacity. After a while, the voltage obviously drops to U~2.
However, the stored energy now is C(U~2)2 , which is half the original value. Where did
Challenge 1069 n the energy go?
**
Colour blindness was discovered by the great English scientist John Dalton (1766–1844)
Challenge 1070 ny – on himself. Can you imagine how he found out? It affects, in all its forms, one in 20 men.

Dvipsbugw
 electromagnetic effects and challenges 597

Dvipsbugw

F I G U R E 275 Cumulonimbus clouds from ground and from space (NASA)

In many languages, a man who is colour blind is called daltonic. Women are almost never

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 564 daltonic, as the property is linked to defects on the X chromosome. If you are colour blind,
you can check to which type you belong with the help of Figure 274.
**
Perfectly spherical electromagnetic waves are impossible in nature. Can you show this
Challenge 1071 n using Maxwell’s equation of electromagnetism, or even without them?
**
Light beams, such as those emitted from lasers, are usually thought of as lines. However,
light beams can also be tubes. Tubular laser beams, or Bessel beams of high order, are
used in modern research to guide plasma channels.

Is lightning a discharge? – Electricity in the atmosphere

Looking carefully, the atmosphere is full of electrical effects. The most impressive elec-
trical phenomenon we observe, lightning, is now reasonably well understood. Inside a
thunderstorm cloud, especially inside tall cumulonimbus clouds,* charges are separated
by collision between the large ‘graupel’ ice crystals falling due to their weight and the small
Ref. 566 ‘hail’ ice crystallites rising due to thermal upwinds. Since the collision takes part in an elec-
tric field, charges are separated in a way similar to the mechanism in the Kelvin generator.
Copyright © Christoph Schiller November 1997–May 2006

Page 521
Discharge takes place when the electric field becomes too high, taking a strange path influ-
enced by ions created in the air by cosmic rays. It seems that cosmic rays are at least partly

* Clouds have Latin names. They were introduced in 1802 by the English explorer Luke Howard (1772–
1864), who found that all clouds could be seen as variations of three types, which he called cirrus, cumulus
Ref. 565 and stratus. He called the combination of all three, the rain cloud, nimbus (from the Latin ‘big cloud’). Today’s
internationally agreed system has been slightly adjusted and distinguishes clouds by the height of their lower
edge. The clouds starting above a height of 6 km are the cirrus, the cirrocumulus and the cirrostratus; those
starting at heights of between 2 and 4 km are the altocumulus, the altostratus and the nimbostratus; clouds
starting below a height of 2 km are the stratocumulus, the stratus and the cumulus. The rain or thunder
cloud, which crosses all heights, is today called cumulonimbus.

Dvipsbugw
 598 iv cl assical electrodynamics • 16. electromagnetic effects

responsible for the zigzag shape of lightning.* Lightning flashes have strange properties.
First, they appear at fields around 200 kV~m (at low altitude) instead of the 2 MV~m of
normal sparks. Second, lightning emits radio pulses. Third, they emit gamma rays. Rus-
Ref. 568 sian researchers, from 1992 onwards explained all three effects by a newly discovered
discharge mechanism. At length scales of 50 m and more, cosmic rays can trigger the ap-
pearance of lightning; the relativistic energy of these rays allows for a discharge mechan-
ism that does not exist for low energy electrons. At relativistic energy, so-called runaway
breakdown leads to discharges at much lower fields than usual laboratory sparks. The
multiplication of these relativistic electrons also leads to the observed radio and gamma Dvipsbugw
ray emissions.
Incidentally, you have a 75 % chance of survival after being hit by lightning, especially
if you are completely wet, as in that case the current will flow outside the skin. Usually, wet
people who are hit loose all their clothes, as the evaporating water tears them off. Rapid
resuscitation is essential to help somebody to recover after a hit.**
As a note, you might know how to measure the distance of a lightning by counting

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the seconds between the lightning and the thunder and multiplying this by the speed of
sound, 330 m~s; it is less well known that one can estimate the length of the lightning bolt
by measuring the duration of the thunder, and multiplying it by the same factor.
In the 1990s more electrical details about thunderstorms became known. Airline pilots
and passengers sometime see weak and coloured light emissions spreading from the top
of thunderclouds. There are various types of such emissions: blue jets and mostly red
sprites and elves, which are somehow due to electric fields between the cloud top and
the ionosphere. The details are still under investigation, and the mechanisms are not yet
clear.***
All these details are part of the electrical circuit around the Earth. This fascinating part
of geophysics would lead us too far from the aim of our mountain ascent. But every phys-
icist should know that there is a vertical electric field of between 100 and 300 V~m on a
clear day, as discovered already in 1752. (Can you guess why it is not noticeable in everyday
Challenge 1072 n life? And why despite its value it cannot be used to extract large amounts of energy?) The
field is directed from the ionosphere down towards the ground; in fact the Earth is per-
manently negatively charged, and in clear weather current flows downwards through the
clear atmosphere, trying to discharge our planet. The current of about 1 kA is spread over
the whole planet; it is possible due to the ions formed by cosmic radiation. (The resistance
between the ground and the ionosphere is about 200 Ω, so the total voltage drop is about
200 kV.) At the same time, the Earth is constantly being charged by several effects: there
Copyright © Christoph Schiller November 1997–May 2006

is a dynamo effect due to the tides of the atmosphere and there are currents induced by
the magnetosphere. But the most important effect is lightning. In other words, contrary

Ref. 567 * There is no ball lightning even though there is a Physics Report about it. Ball lightning is one of the favourite
myths of modern pseudo-science. Actually, they would exist if we lived in a giant microwave oven. To show
this, just stick a toothpick into a candle, light the toothpick, and put it into (somebody else’s) microwave at
maximum power.
** If you are ever hit by lightning and survive, go to the hospital! Many people died three days later having
failed to do so. A lightning strike often leads to coagulation effects in the blood. These substances block the
kidneys, and one can die three days later because of kidney failure. The remedy is to have dialysis treatment.
*** For images, have a look at the interesting http://sprite.gi.alaska.edu/html/sprites.htm, http://www.
fma-research.com/spriteres.htm and http://paesko.ee.psu.edu/Nature websites.

Dvipsbugw
 electromagnetic effects and challenges 599

Ref. 569 to what one may think, lightning does not discharge the ground, it actually charges it
up!* Of course, lightning does discharge the cloud to ground potential difference; but by
doing so, it actually sends a negative charge down to the Earth as a whole. Thunderclouds
are batteries; the energy from the batteries comes from the the thermal uplifts mentioned
above, which transport charge against the global ambient electrical field.
Using a few electrical measurement stations that measure the variations of the elec-
trical field of the Earth it is possible to locate the position of all the lightning that comes
Ref. 570 down towards the Earth at a given moment. Present research also aims at measuring the
activity of the related electrical sprites and elves in this way. Dvipsbugw
The ions in air play a role in the charging of thunderclouds via the charging of ice
crystals and rain drops. In general, all small particles in the air are electrically charged.
When aeroplanes and helicopters fly, they usually hit more particles of one charge than
of the other. As a result, aeroplanes and helicopters are charged up during flight. When
a helicopter is used to rescue people from a raft in high seas, the rope pulling the people
upwards must first be earthed by hanging it in the water; if this is not done, the people

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

on the raft could die from an electrical shock when they touch the rope, as has happened
a few times in the past.
The charges in the atmosphere have many other effects. Recent experiments have con-
Ref. 571 firmed what was predicted back in the early twentieth century: lightning emits X-rays.
The confirmation is not easy though; it is necessary to put a detector near the lightning
flash. To achieve this, the lightning has to be directed into a given region. This is possible
using a missile pulling a metal wire, the other end of which is attached to the ground.
These experimental results are now being collated into a new description of lightning
which also explains the red-blue sprites above thunderclouds. In particular, the processes
Ref. 572 also imply that inside clouds, electrons can be accelerated up to energies of a few MeV.
Why are sparks and lightning blue? This turns out to be a material property: the colour
comes from the material that happens to be excited by the energy of the discharge, usu-
ally air. This excitation is due to the temperature of 30 kK inside the channel of a typical
lightning flash. For everyday sparks, the temperature is much lower. Depending on the
situation, the colour may arise from the gas between the two electrodes, such as oxygen
or nitrogen, or it may due to the material evaporated from the electrodes by the discharge.
For an explanation of such colours, as for the explanation of all colours due to materials,
we need to wait for the next part of our walk.
But not only electric fields are dangerous. Also time-varying electromagnetic fields can
be. In 1997, in beautiful calm weather, a Dutch hot air balloon approached the powerful
Copyright © Christoph Schiller November 1997–May 2006

radio transmitter in Hilversum. After travelling for a few minutes near to the antenna,
the gondola suddenly detached from the balloon, killing all the passengers inside.
An investigation team reconstructed the facts a few weeks later. In modern gas bal-
loons the gondola is suspended by high quality nylon ropes. To avoid damage by light-
ning and in order to avoid electrostatic charging problems all these nylon ropes contain
thin metal wires which form a large equipotential surface around the whole balloon. Un-
fortunately, in the face of the radio transmitter, these thin metal wires absorbed radio
energy from the transmitter, became red hot, and melted the nylon wires. It was the first
time that this had ever been observed.

Challenge 1073 ny * The Earth is thus charged to about −1 MC. Can you confirm this?

Dvipsbugw
 600 iv cl assical electrodynamics • 16. electromagnetic effects

Does gravity make charges radiate?

We learned in the section on general relativity that gravitation has the same effects as
acceleration. This means that a charge kept fixed at a certain height is equivalent to a
charge accelerated by 9.8 m~s2 , which would imply that it radiates electromagnetically,
since all accelerated charges radiate. However, the world around us is full of charges at
fixed heights, and there is no such radiation. How is this possible?
Ref. 573 The question has been a pet topic for many years. Generally speaking, the concept of
radiation is not observer invariant: If one observer detects radiation, a second one does Dvipsbugw
not necessarily do so as well. The exact way a radiation field changes from one observer
to the other depends on the type of relative motion and on the field itself.
A precise solution of the problem shows that for a uniformly accelerated charge, an ob-
server undergoing the same acceleration only detects an electrostatic field. In contrast, an
Ref. 574 inertial observer detects a radiation field. Since gravity is (to a high precision) equivalent
to uniform acceleration, we get a simple result: gravity does not make electrical charges

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

radiate for an observer at rest with respect to the charge, as is observed. The results holds
true also in the quantum theoretical description.

Research questions
The classical description of electrodynamics is coherent and complete; nevertheless there
are still many subjects of research. Here are a few of them.
The origin of magnetic field of the Earth, the other
planets, the Sun and even of the galaxy is a fascinating ocean
topic. The way that the convection of fluids inside the crust
planets generates magnetic fields, an intrinsically three-
dimensional problem, the influence of turbulence, of mantle
nonlinearities and of chaos makes it a surprisingly com-
plex question. liquid core
The details of the generation of the magnetic field solid core
of the Earth, usually called the geodynamo, began to
appear only in the second half of the twentieth century,
when the knowledge of the Earth’s interior reached a
Ref. 505 sufficient level. The Earth’s interior starts below the
Earth’s crust. The crust is typically 30 to 40 km thick
(under the continents), though it is thicker under high F I G U R E 276 The structure of our
Copyright © Christoph Schiller November 1997–May 2006

mountains and thinner near volcanoes or under the planet

oceans. As already mentioned, the crust consists of
large segments, the plates, that move with respect to one other. The Earth’s interior is
divided into the mantle – the first 2900 km from the surface – and the core. The core is
made up of a liquid outer core, 2210 km thick, and a solid inner core of 1280 km radius.
(The temperature of the core is not well known; it is believed to be 6 to 7 kK. Can you
Challenge 1074 ny find a way to determine it? The temperature might have decreased a few hundred kelvin
during the last 3000 million years.)
The Earth’s core consists mainly of iron that has been collected from the asteroids that
collided with the Earth during its youth. It seems that the liquid and electrically con-
ducting outer core acts as a dynamo that keeps the magnetic field going. The magnetic

Dvipsbugw
 electromagnetic effects and challenges 601

energy comes from the kinetic energy of the outer core, which rotates with respect to the
Earth’s surface; the fluid can act as a dynamo because, apart from rotating, it also convects
from deep inside the Earth to more shallow depths, driven by the temperature gradients
between the hot inner core and the cooler mantle. Huge electric currents flow in com-
plex ways through these liquid layers, maintained by friction, and create the magnetic
field. Why this field switches orientation at irregular intervals of between a few tens of
thousands and a few million years, is one of the central questions. The answers are dif-
ficult; experiments are not yet possible, 150 years of measurements is a short time when
compared with the last transition – about 730 000 years ago – and computer simulations Dvipsbugw
are extremely involved. Since the field measurements started, the dipole moment of the
magnetic field has steadily diminished, presently by 5% a year, and the quadrupole mo-
ment has steadily increased. Maybe we are heading towards a surprise.* (By the way, the
study of galactic magnetic fields is even more complex, and still in its infancy.)
Another important puzzle about electricity results from the equivalence of mass and
energy. It is known from experiments that the size d of electrons is surely smaller than

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 576 10−22 m. This means that the electric field surrounding it has an energy content E given
Challenge 1075 ny by at least

E nergy = ε 0 E lectric field dV = ε 0 ( ) 4πr 2 dr

1 1 ª 1 q 2
2 ∫ 2
2 ∫ d 4πε o r 2

q2 1
= A 1.2 µJ . (453)
8πε o d

On the other hand, the mass of an electron, usually given as 511 keV~c2 , corresponds to
an energy of only 82 fJ, ten million times less than the value just calculated. In other
words, classical electrodynamics has considerable difficulty describing electrons. In fact,
a consistent description of charged point particles within classical electrodynamics is
Ref. 577 impossible. This pretty topic receives only a rare – but then often passionate – interest
nowadays, because the puzzle is solved in a different way in the upcoming, second part
of our mountain ascent.
Even though the golden days of materials science are over, the various electromagnetic
properties of matter and their applications in devices do not seem to be completely ex-
plored yet. About once a year a new effect is discovered that merits inclusion in the list
Page 605 of electromagnetic matter properties of Table 50. Among others, some newer semicon-
ductor technologies will still have an impact on electronics, such as the recent introduc-
Copyright © Christoph Schiller November 1997–May 2006

tion of low cost light detecting integrated circuits built in CMOS (complementary metal
oxide silicon) technology.
The building of light sources of high quality has been a challenge for many centuries
and remains one for the future. Light sources that are intense, tunable and with large co-
herence length or sources that emit extreme wavelengths are central to many research
pursuits. As one example of many, the first X-ray lasers have recently been built; how-
ever, they are several hundred metres in size and use modified particle accelerators. The
construction of compact X-ray lasers is still many years off – if it is possible at all.

* In 2005, it has been reported that the inner core of the Earth seems to rotate faster than the Earth’s crust
Ref. 575 by up to half a degree per year.

Dvipsbugw
 602 iv cl assical electrodynamics • 16. electromagnetic effects

Electrodynamics and general relativity interact in many ways. Only a few cases have
been studied up to now. They are important for black holes and for empty space. For
Ref. 578 example, it seems that magnetic fields increase the stiffness of empty space. Many such
topics will appear in the future.
But maybe the biggest challenge imaginable in classical electrodynamics is to decode
the currents inside the brain. Will it be possible to read our thoughts with an apparatus
Challenge 1076 r placed outside the head? One could start with a simpler challenge: Would it be possible
to distinguish the thought ‘yes’ from the thought ‘no’ by measuring electrical or magnetic
fields around the head? In other words, is mind-reading possible? Maybe the twenty-first Dvipsbugw
century will come up with a positive answer. If so, the team performing the feat will be
instantly famous.

Levitation
We have seen that it is possible to move certain objects without touching them, using a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

magnetic or electric field or, of course, using gravity. Is it also possible, without touching
an object, to keep it fixed, floating in mid-air? Does this type of rest exist?
It turns out that there are several methods of levitating objects. These are commonly
Ref. 579 divided into two groups: those that consume energy and those who do not. Among the
methods that consume energy is the floating of objects on a jet of air or of water, the float-
ing of objects through sound waves, e.g. on top of a siren, or through a laser beam com-
ing from below, and the floating of conducting material, even of liquids, in strong radio-
frequency fields. Levitation of liquids or solids by strong ultrasound waves is presently
Ref. 580 becoming popular in laboratories. All these methods give stationary levitation. Another
group of energy consuming methods sense the way a body is falling and kick it up again in
the right way via a feedback loop; these methods are non-stationary and usually use mag-
netic fields to keep the objects from falling. The magnetic train being built in Shanghai
Ref. 581 by a German consortium is levitated this way. The whole train, including the passengers,
is levitated and then moved forward using electromagnets. It is thus possible, using mag-
nets, to levitate many tens of tonnes of material.
For levitation methods that do not consume energy – all such methods are necessar-
ily stationary – a well-known limitation can be found by studying Coulomb’s ‘law’ of
electrostatics: no static arrangement of electric fields can levitate a charged object in free
space or in air. The same result is valid for gravitational fields and massive objects;* in
other words, we cannot produce a local minimum of potential energy in the middle of a
Copyright © Christoph Schiller November 1997–May 2006

Ref. 582 box using electric or gravitational fields. This impossibility is called Earnshaw’s theorem.
Speaking mathematically, the solutions of the Laplace equation ∆φ = 0, the so-called har-
monic functions, have minima or maxima only at the border, and never inside the domain
of definition. (You proved this yourself on page 121.) The theorem can also be proved by
noting that given a potential minimum in free space, Gauss’ theorem for a sphere around
that minimum requires that a source of the field be present inside, which is in contradic-
tion with the original assumption.

Page 80 * To the disappointment of many science-fiction addicts, this would even be true if a negative mass existed.
And even though gravity is not really due to a field, but to space-time curvature, the result still holds in
general relativity.

Dvipsbugw
 electromagnetic effects and challenges 603

We can deduce that it is also impossible to use electric fields to levitate an electrically
neutral body in air: the potential energy U of such a body, with volume V and dielectric
constant ε, in an environment of dielectric constant ε 0 , is given by

= − (ε − ε 0 )E 2 .
U 1
(454)
V 2
Challenge 1077 ny Since the electric field E never has a maximum in the absence of space charge, and since
for all materials ε A ε 0 , there cannot be a minimum of potential energy in free space for Dvipsbugw
a neutral body.*
To sum up, using static electric or static gravitational fields it is impossible to keep an
object from falling; neither quantum mechanics, which incorporates phenomena such as
antimatter, nor general relativity, including phenomena such as black holes, change this
basic result.
For static magnetic fields, the argument is analogous to electrical fields: the potential

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

energy U of a magnetizable body of volume V and permeability µ in a medium with
Challenge 1079 ny permeability µ 0 containing no current is given by

= − ( − )B 2
U 1 1 1
(455)
V 2 µ µ0

and due to the inequality ∆B 2 E 0, isolated maxima of a static magnetic field are not
possible, only isolated minima. Therefore, it is impossible to levitate paramagnetic (µ A
µ o ) or ferromagnetic (µ Q µ 0 ) materials such as steel, including bar magnets, which are
Challenge 1080 ny all attracted, and not repelled to magnetic field maxima.
There are thus two ways to get magnetic levitation: levitating a diamagnet or using a
Ref. 584 time dependent field. Diamagnetic materials (µ < µ o ) can be levitated by static magnetic
fields because they are attracted to magnetic field minima; the best-known example is the
levitation of superconductors, which are, at least those of type I, perfects diamagnets (µ =
0). Strong forces can be generated, and this method is also being tested for the levitation
Ref. 581 of passenger trains in Japan. In some cases, superconductors can even be suspended in
mid-air, below a magnet. Single atoms with a magnetic moment are also diamagnets; they
Ref. 585 are routinely levitated this way and have also been photographed in this state.
Also single neutrons, which have a magnetic dipole moment, have been kept in mag-
netic bottles in this way, until they decay. Recently, scientists have levitated pieces of wood,
Copyright © Christoph Schiller November 1997–May 2006

plastic, strawberries, water droplets, liquid helium droplets as large as 2 cm, grasshoppers,
fish and frogs (all alive and without any harm) in this way. They are, like humans, all made
Ref. 586 of diamagnetic material. Humans themselves have not yet been levitated, but the feat is
being planned and worked on.
Diamagnets levitate if ∇B 2 A 2µ 0 ρд~χ, where ρ is the mass density of the object and
Challenge 1081 ny χ = 1 − µ~µ 0 its magnetic susceptibility. Since χ is typically about 10−5 and ρ of order
1000 kg~m3 , field gradients of about 1000 T2 ~m are needed. In other words, levitation re-

Ref. 583 * It is possible, however, to ‘levitate’ gas bubbles in liquids – ‘trap’ them to prevent them from rising would
be a better expression – because in such a case the dielectric constant of the environment is higher than that
Challenge 1078 ny of the gas. Can you find a liquid–gas combination where bubbles fall instead of rise?

Dvipsbugw
 604 iv cl assical electrodynamics • 16. electromagnetic effects

Dvipsbugw

F I G U R E 277 Trapping a metal F I G U R E 278 Floating ‘magic’

sphere using a variable speed drill nowadays available in toy shops
and a plastic saddle

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

quires fields changes of 10 T over 10 cm, which is nowadays common for high field labor-
atory magnets.
Finally, time dependent electrical or magnetic fields, e.g. periodic fields, can lead to
Ref. 579 levitation in many different ways without any consumption of energy. This is one of the
methods used in the magnetic bearings of turbomolecular vacuum pumps. Also single
charged particles, such as ions and electrons, are now regularly levitated with Paul traps
Ref. 579 and Penning traps. The mechanical analogy is shown in Figure 277.
Ref. 587 Figure 278 shows a toy that allows you to personally levitate a spinning top in mid-air
above a ring magnet, a quite impressive demonstration of levitation for anybody looking
Ref. 588 at it. It is not hard to build such a device yourself.
Even free electrons can be levitated, letting them float above the surface of fluid helium.
Ref. 589 In the most recent twist of the science of levitation, in 1995 Stephen Haley predicted that
the suspension height of small magnetic particles above a superconducting ring should
be quantized. However, the prediction has not been verified by experiment yet.
For the sake of completeness we mention that nuclear forces cannot be used for levita-
tion in everyday life, as their range is limited to a few femtometres. However, we will see
Page 922 later that the surface matter of the Sun is prevented from falling into the centre by these
interactions; we could thus say that it is indeed levitated by nuclear interactions.

Matter, levitation and electromagnetic effects

Copyright © Christoph Schiller November 1997–May 2006

The levitation used by magicians mostly falls into another class. When David Copper-
field, a magician performing for the MTV generation at the end of the twentieth century,
‘flies’ during his performances, he does so by being suspended on thin fishing lines that
Ref. 590 are rendered invisible by clever lighting arrangements. In fact, if we want to be precise,
we should count fishing lines, plastic bags, as well as every table and chair as levitation
devices. (Journalists would even call them ‘anti-gravity’ devices.) Contrary to our impres-
sion, a hanging or lying object is not really in contact with the suspension, if we look at
the critical points with a microscope.* More about this in the second part of our walk.

Challenge 1082 ny * The issue is far from simple: which one of the levitation methods described above is used by tables or
chairs?

Dvipsbugw
 electromagnetic effects and challenges 605

But if this is the case, why don’t we fall through a table or through the floor? We star-
ted the study of mechanics by stating that a key property of matter its solidity, i.e. the
impossibility of having more than one body at the same place at the same time. But what
is the origin of solidity? Again, we will be able to answer the question only in the second
Page 840 part of our adventure, but we can already collect the first clues at this point.
Solidity is due to electricity. Many experiments show that matter is constituted of
charged particles; indeed, matter can be moved and influenced by electromagnetic fields
Ref. 591 in many ways. Over the years, material scientists have produced a long list of such effects,
all of which are based on the existence of charged constituents. Can you find or imagine Dvipsbugw
Challenge 1083 r a new one? For example, can electric charge change the colour of objects?

TA B L E 50 Selected matter properties related to electromagnetism, showing among other things the
role it plays in the constitution of matter; at the same time a short overview of atomic, solid state, fluid
and business physics
Propert y Example Definition

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

thermal radiation or heat every object temperature-dependent radiation emitted
radiation or incandescence by any macroscopic amount of matter
Interactions with charges and currents
electrification separating metals from spontaneous charging
insulators
triboelectricity glass rubbed on cat fur charging through rubbing
barometer light mercury slipping along gas discharge due to triboelectricity Ref. 592
glass
insulation air no current flow below critical voltage drop
semiconductivity diamond, silicon or current flows only when material is impure
gallium arsenide (‘doped’)
conductivity copper, metals current flows easily
superconductivity niobium current flows indefinitely
ionization fire flames current flows easily
localization (weak, disordered solids resistance of disordered solids
Anderson)
resistivity, Joule effect graphite heating due to current flow
thermoelectric effects: ZnSb, PbTe, PbSe, cooling due to current flow, current flow
Copyright © Christoph Schiller November 1997–May 2006

Peltier effect, Seebeck effect, BiSeTe, Bi2 Te3 , etc. due to temperature difference, or due to
Thomson effect temperature gradients
acoustoelectric effect CdS sound generation by currents, and vice
versa
magnetoresistance iron, metal multilayers resistance changes with applied magnetic
field Ref. 593
recombination fire alarms charge carriers combine to make neutral
atoms or molecules
annihilation positron tomography particle and antiparticle, e.g. electron and
positron, disappear into photons

Dvipsbugw
606 iv cl assical electrodynamics • 16. electromagnetic effects

Propert y Example Definition

Penning effect Ne, Ar ionization through collision with

metastable atoms
Richardson effect, thermal BaO2 , W, Mo, used in emission of electrons from hot metals
emission tv and electron
microscopes
skin effect Cu high current density on exterior of wire
pinch effect InSb, plasmas high current density on interior of wire Dvipsbugw
Josephson effect Nb-Oxide-Nb tunnel current flows through insulator
between two superconductors
Sasaki–Shibuya effect n-Ge, n-Si anisotropy of conductivity due to applied
electric field
switchable magnetism InAs:Mn voltage switchable magnetization Ref. 594
Interactions with magnetic fields

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Hall effect silicon; used for voltage perpendicular to current flow in
magnetic field applied magnetic field
measurements
Zeeman effect Cd change of emission frequency with
magnetic field
Paschen–Back effect atomic gases change of emission frequency in strong
magnetic fields
ferromagnetism Fe, Ni, Co, Gd spontaneous magnetization; material
strongly attracted by magnetic fields
paramagnetism Fe, Al, Mg, Mn, Cr induced magnetization parallel to applied
field; attracted by magnetic fields
diamagnetism water, Au, graphite, induced magnetization opposed to applied
NaCl field; repelled by magnetic fields
magnetostriction CeB6 , CePd2 Al3 change of shape or volume by applied
magnetic field
magnetoelastic effect Fe, Ni change of magnetization by tension or
pressure
acoustomagnetic effect metal alloys, anti-theft excitation of mechanical oscillations
stickers through magnetic field
Copyright © Christoph Schiller November 1997–May 2006

spin valve effect metal multilayers electrical resistance depends on spin

direction of electrons with respect to
applied magnetic field
magneto-optical activity or flint glass polarization angle is rotated with magnetic
Faraday effect or Faraday field; different refraction index for right
rotation and left circularly polarized light, as in
magneto-optic (MO) recording
magnetic circular gases different absorption for right- and
dichroism left-circularly polarized light; essentially
the same as the previous one
Majorana effect colloids specific magneto-optic effect

Dvipsbugw
electromagnetic effects and challenges 607

Propert y Example Definition

photoelectromagnetic InSb current flow due to light irradiation of

effect semiconductor in a magnetic field
Voigt effect vapours birefringence induced by applied magnetic
field
Cotton–Mouton effect liquids birefringence induced by applied magnetic
field
Hanle effect Hg change of polarization of fluorescence with Dvipsbugw
magnetic field
Shubnikov–de Haas effect Bi periodic change of resistance with applied
magnetic field
thermomagnetic effects: BiSb alloys relation between temperature, applied
Ettinghausen effect, fields and electric current
Righi–Leduc effect, Nernst

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

effect, magneto–Seebeck
effect
Ettinghausen–Nernst effect Bi appearance of electric field in materials
with temperature gradients in magnetic
fields
photonic Hall effect CeF3 transverse light intensity depends on the
applied magnetic field Ref. 595
magnetocaloric effect gadolinium, GdSiGe material cools when magnetic field is
alloys switched off Ref. 596
cyclotron resonance semiconductors, selective absorption of radio waves in
metals magnetic fields
magnetoacoustic effect semiconductors, selective absorption of sound waves in
metals magnetic fields
magnetic resonance most materials, used selective absorption of radio waves in
for imaging in magnetic fields
medicine for structure
determination of
molecules
magnetorheologic effect liquids, used in change of viscosity with applied magnetic
advanced car fields
suspensions
Copyright © Christoph Schiller November 1997–May 2006

Meissner effect type 1 superconductors, expulsion of magnetic field from

used for levitation superconductors
Interactions with electric fields
polarizability all matter polarization changes with applied electric
field
ionization, field emission, all matter, tv charges are extracted at high fields
Schottky effect
paraelectricity BaTiO3 applied field leads to polarization in same
direction
dielectricity water in opposite direction

Dvipsbugw
608 iv cl assical electrodynamics • 16. electromagnetic effects

Propert y Example Definition

ferroelectricity BaTiO3 spontaneous polarization below critical

temperature
piezoelectricity the quartz lighter used polarization appears with tension, stress, or
in the kitchen pressure
electrostriction platinum sponges in shape change with applied voltage Ref. 597
acids
pyroelectricity CsNO3 , tourmaline, change of temperature produces charge Dvipsbugw
crystals with polar separation
axes; used for infrared
detection
electro-osmosis or many ionic liquids liquid moves under applied electric field
electrokinetic effect Ref. 598
electrowetting salt solutions on gold wetting of surface depends on applied

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

voltage
electrolytic activity sulphuric acid charge transport through liquid
liquid crystal effect watch displays molecules turn with applied electric field
electro-optical activity: liquids (e.g. oil), material in electric field rotates light
Kerr effect, Pockels effect crystalline solids polarization, i.e. produces birefringence
Freederichsz effect, nematic liquid crystals electrically induced birefringence
Schadt–Helfrichs effect
Stark effect hydrogen, mercury colour change of emitted light in electric
field
field ionization helium near tungsten ionization of gas atoms in strong electric
tips in field ion fields
microscope
Zener effect Si energy-free transfer of electrons into
conduction band at high fields
field evaporation W evaporation under strong applied electric
fields
Interactions with light
absorption coal, graphite transformation of light into heat or other
energy forms (which ones?)Challenge 1084 n
blackness coal, graphite complete absorption in visible range
Copyright © Christoph Schiller November 1997–May 2006

colour, metallic shine ruby absorption depending on light frequency

photostriction PbLaZrTi light induced piezoelectricity
photography AgBr, AgI light precipitates metallic silver
photoelectricity, Cs current flows into vacuum due to light
photoeffect irradiation
internal photoelectric effect Si p–n junctions, solar voltage generation and current flow due to
cells light irradiation
photon drag effect p-Ge current induced by photon momentum
emissivity all bodies ability to emit light

Dvipsbugw
electromagnetic effects and challenges 609

Propert y Example Definition

transparency glass, quartz, diamond low reflection, low absorption, low

scattering
reflectivity metals light bounces on surface
polarization pulled polymer sheets light transmission depending on
polarization angle
optical activity sugar dissolved in rotation of polarization
water, quartz Dvipsbugw
birefringence feldspar,cornea refraction index depends on polarization
direction, light beams are split into two
beams
dichroism feldspar, andalusite absorption depends on polarization
optically induced AgCl optically induced birefringence and
anisotropy, Weigert effect dichroism

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

second harmonic LiNbO3 , KPO4 light partially transformed to double
generation frequency
luminescence: general term GaAs, tv cold light emission
for opposite of
incandescence
fluorescence CaF2 , X-ray light emission during and after light
production, light tubes, absorption or other energy input
cathode ray tubes
phosphorescence TbCl3 light emission due to light, electrical or
chemical energy input, continuing long
after stimulation
electroluminescence ZnS emission of light due to alternating
electrical field
photoluminescence ZnS  Cu, light emission triggered by UV light, used
SrAlO4  Eu, Dy, in safety signs
hyamine
chemoluminescence H2 O2 , phenyl oxalate cold light emission used in light sticks for
ester, dye divers and fun
bioluminescence glow-worm, deep sea cold light emission in animals
fish
Copyright © Christoph Schiller November 1997–May 2006

triboluminescence sugar light emission during friction or crushing

thermoluminescence quartz, feldspar light emission during heating, used e.g. for
archaeological dating of pottery Ref. 599
bremsstrahlung X-ray generation radiation emission through fast
deceleration of electrons
Compton effect momentum change of wavelength of light, esp. X-rays
measurements and gamma radiation, colliding with
matter

Dvipsbugw
610 iv cl assical electrodynamics • 16. electromagnetic effects

Propert y Example Definition

Čerenkov effect water, polymer particle light emission in a medium due to

detectors particles, e.g. emitted by radioactive
processes, moving faster than the speed of
light in that medium
transition radiation any material light emission due to fast particles moving
from one medium to a second with
different refractive index Dvipsbugw
electrochromicity wolframates colour change with applied electric field
scattering gases, liquids light changes direction
Mie scattering dust in gases light changes direction
Raleigh scattering sky light changes direction, sky is blue
Raman effect or molecular gases scattered light changes frequency
Smekal–Raman effect

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

laser activity, beer, ruby, He–Ne emission of stimulated radiation
superradiation
sonoluminescence air in water light emission during cavitation
gravitoluminescence does not exist; Challenge
1085 n why?
switchable mirror LaH voltage controlled change from reflection
to transparency Ref. 600
radiometer effect bi-coloured windmills irradiation turns mill (see page 573)
luminous pressure idem irradiation turns mill directly
solar sail effect future satellites motion due to solar wind
acoustooptic effect LiNbO3 diffraction of light by sound in transparent
materials
photorefractive materials LiNbO3 , GaAs, InP light irradiation changes refractive index
Auger effect Auger electron electron emission due to atomic
spectroscopy reorganization after ionization by X-rays
Bragg reflection crystal structure X-ray diffraction by atomic planes
determination
Mößbauer effect Fe, used for recoil-free resonant absorption of gamma
spectroscopy radiation
pair creation Pb transformation of a photon in a charged
Copyright © Christoph Schiller November 1997–May 2006

particle–antiparticle pair
photoconductivity Se, CdS change of resistivity with light irradiation
optoacoustic affect, gases, solids creation of sound due to absorption of
photoacoustic effect pulsed light
optogalvanic effect plasmas change of discharge current due to light
irradiation
optical nonlinear effects: parametric amplification, frequency mixing, saturable absorption, n-th
harmonic generation, optical Kerr effect, etc.
phase conjugated mirror gases reflection of light with opposite phase
activity

Dvipsbugw
 electromagnetic effects and challenges 611

Propert y Example Definition

Material properties
solidity, impenetrability floors, columns, ropes, at most one object per place at a given time
buckets
Interactions with vacuum
Casimir effect metals attraction of uncharged, conducting bodies
Dvipsbugw

All matter properties in the list can be influenced by electric or magnetic fields or
directly depend on them. This shows that the nature of all these material properties is
electromagnetic. In other words, charges and their interactions are an essential and fun-
damental part of the structure of objects. The table shows so many different electromag-
netic properties that the motion of charges inside each material must be complex indeed.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Most effects are the topic of solid state physics,* fluid and plasma physics.
Solid state physics is by far the most important part of physics, when measured by the
impact it has on society. Almost all effects have applications in technical products, and
give employment to many people. Can you name a product or business application for
Challenge 1086 e any randomly chosen effect from the table?
In our mountain ascent however, we look at only one example from the above list:
thermal radiation, the emission of light by hot bodies.
Earnshaw’s theorem about the impossibility of a stable equilibrium for charged
particles at rest implies that the charges inside matter must be moving. For any charged
particle in motion, Maxwell’s equations for the electromagnetic field show that it radiates
energy by emitting electromagnetic waves. In short, classical mechanics thus predicts that
matter must radiate electromagnetic energy.
Interestingly, everybody knows from experience that this is indeed the case. Hot bod-
ies light up depending on their temperature; the working of light bulbs thus proves that
metals are made of charged particles. Incandescence, as it is called, requires charges. Ac-
tually, every body emits radiation, even at room temperature. This radiation is called
thermal radiation; at room temperature it lies in the infrared. Its intensity is rather weak
Ref. 601 in everyday life; it is given by the general expression

2π 5 k 4
I(T) = f T 4 or I(T) = f σ T 4 with σ = 56.7 nW~K4 m2 ,
Copyright © Christoph Schiller November 1997–May 2006

(456)
15c 2 h 3

where f is a material-, shape- and temperature-dependent factor, with a value between

zero and one, and is called the emissivity. The constant σ is called the Stefan–Boltzmann
black body radiation constant or black body radiation constant. A body whose emissivity
is given by the ideal case f = 1 is called a black body, because at room temperature such a
Challenge 1087 n body also has an ideal absorption coefficient and thus appears black. (Can you see why?)
The heat radiation such a body emits is called black body radiation.
Ref. 602 By the way, which object radiates more energy: a human body or an average piece of
Challenge 1088 n the Sun of the same mass? Guess first!
* Probably the best and surely the most entertaining introductory English language book on the topic is the

Dvipsbugw
 612 iv cl assical electrodynamics • 16. electromagnetic effects

Why can we see each other?

Physicists have a strange use of the term ‘black’. Most bodies at temperatures at which
they are red hot or even hotter are excellent approximations of black bodies. For example,
the tungsten in incandescent light bulbs, at around 2000 K, emits almost pure black body
radiation; however, the glass then absorbs much of the ultraviolet and infrared compon-
ents. Black bodies are also used to define the colour white. What we commonly call pure
white is the colour emitted by a black body of 6500 K, namely the Sun. This definition
Ref. 603 is used throughout the world, e.g. by the Commission Internationale d’Eclairage. Hotter Dvipsbugw
black bodies are bluish, colder ones are yellow, orange or red.* The stars in the sky are
classified in this way, as summarized on page 186.
Let us make a quick summary of black body radiation. Black body radiation has two
important properties: first, the emitted light power increases with the fourth power of
the temperature. With this relation alone you can check the temperature of the Sun, men-
tioned above, simply by comparing the size of the Sun with the width of your thumb when

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1089 d your arm is stretched out in front of you. Are you able to do this? (Hint: use the excellent
Ref. 604 approximation that the Earth’s average temperature of about 14.0°C is due to the Sun’s
irradiation.)
The precise expression for the emitted energy density u per frequency ν can be de-
duced from the radiation law for black bodies discovered by Max Planck**

ν3
u(ν, T) =
8πh
. (457)
c 3 e hν~k T − 1
He made this important discovery, which we will discuss in more detail in the second
part of our mountain ascent, simply by comparing this curve with experiment. The new
constant h, quantum of action or Planck’s constant, turns out to have the value 6.6 ë 10−34 Js,
Page 705 and is central to all quantum theory, as we will see. The other constant Planck introduced,
the Boltzmann constant k, appears as a prefactor of temperature all over thermodynamics,
as it acts as a conversion unit from temperature to energy.
Challenge 1090 ny The radiation law gives for the total emitted energy density the expression

8π 5 k 4
u(T) = T 4 (458)
15c 3 h 3

from which equation (456) is deduced using I = uc~4. (Why?)

Copyright © Christoph Schiller November 1997–May 2006

Challenge 1091 ny

one by Neil Ashcroft & David Mermin, Solid State Physics, Holt Rinehart & Winston, 1976.
* Most bodies are not black, because colour is not only determined by emission, but also by absorption of
light.
** Max Planck (1858–1947), professor of physics in Berlin, was a central figure in thermostatics. He dis-
covered and named Boltzmann’s constant k and the quantum of action h, often called Planck’s constant. His
introduction of the quantum hypothesis gave birth to quantum theory. He also made the works of Einstein
known in the physical community, and later organized a job for him in Berlin. He received the Nobel Prize
for physics in 1918. He was an important figure in the German scientific establishment; he also was one of
the very few who had the courage to tell Adolf Hitler face to face that it was a bad idea to fire Jewish profess-
ors. (He got an outburst of anger as answer.) Famously modest, with many tragedies in his personal life, he
was esteemed by everybody who knew him.

Dvipsbugw
 electromagnetic effects and challenges 613

Figure to be inserted

F I G U R E 279 Bodies inside a oven at room temperature (left) and red hot (right)
Dvipsbugw

The second property of black body radiation is the value of the peak wavelength, i.e. the
wavelength emitted with the highest intensity. This wavelength determines their colour;
Challenge 1092 ny it is deduced from equation (457) to be

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

λ max = = ħν max = 2.82 kT = (3.9 ë 10−23 J~K) ë T . (459)
hc 1 2.9 mm K
but
4.956 k T T
Either of these expressions is called Wien’s colour displacement after its discoverer.* The
colour change with temperature is used in optical thermometers; this is also the way the
temperatures of stars are measured. For 37°C, human body temperature, it gives a peak
wavelength of 9.3 µm or 115 THz, which is therefore the colour of the bulk of the radiation
emitted by every human being. (The peak wavelength does not correspond to the peak
Challenge 1093 ny frequency. Why?) On the other hand, following the telecommunication laws of many
countries, any radiation emitter needs a licence to operate; it follows that strictly in Ger-
many only dead people are legal, and only if their bodies are at absolute zero temperature.
Note that a black body or a star can be blue, white, yellow, orange or red. It is never
Challenge 1094 ny green. Can you explain why?
Above, we predicted that any material made of charges emits radiation. Are you able
to find a simple argument showing whether heat radiation is or is not this classically
Challenge 1095 ny predicted radiation?
But let us come back to the question in the section title. The existence of thermal radi-
ation implies that any hot body will cool, even if it is left in the most insulating medium
there is, namely in vacuum. More precisely, if the vacuum is surrounded by a wall, the
temperature of a body in the vacuum will gradually approach that of the wall.
Interestingly, when the temperature of the wall and of the body inside have become the
Copyright © Christoph Schiller November 1997–May 2006

same, something strange happens. The effect is difficult to check at home, but impressive
Ref. 605 photographs exist in the literature.
One arrangement in which walls and the objects inside them are at the same temper-
ature is an oven. It turns out that it is impossible to see objects in an oven using the light
coming from thermal radiation. For example, if an oven and all its contents are red hot,
taking a picture of the inside of the oven (without a flash!) does not reveal anything; no
contrast nor brightness changes exist that allow one to distinguish the objects from the
Challenge 1096 n walls or their surroundings. Can you explain the finding?

* Wilhelm Wien (b. 1864 Gaffken, d. 1824 München), East-Prussian physicist; he received the Nobel Prize
for physics in 1911 for the discovery of this relation.

Dvipsbugw
 614 iv cl assical electrodynamics • 17. cl assical physics in a nu tshell

In short, we are able to see each other only because the light sources we use are at
a different temperature from us. We can see each other only because we do not live in
thermal equilibrium with our environment.

A summary of classical electrodynamics and of its limits

In general, classical electrodynamics can be summarized in a few main ideas.
— The electromagnetic field is a physical observable, as shown e.g. by compass needles.
Dvipsbugw
— The field sources are the (moving) charges and the field evolution is described by Max-
well’s evolution equations, as shown, for example, by the properties of amber, lode-
stone, batteries and remote controls.
— The electromagnetic field changes the motion of electrically charged objects via the
Lorentz expression as, for example, shown by electric motors.
— The field behaves like a continuous quantity, a distribution of little arrows, and propag-
ates as a wave, as shown, for example, by radios and mobile phones.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

— The field can exist and move in empty space, as shown, for example, by the stars.
As usual, the motion of the sources and the field is reversible, continuous, conserved and
deterministic. However, there is quite some fun in the offing; even though this description
is correct in everyday life, during the rest of our mountain ascent we will find that each
of the bullet points is in fact wrong. A simple example shows this.
At a temperature of zero kelvin, when matter does not radiate thermally, we have the
paradoxical situation that the charges inside matter cannot be moving, since no emitted
radiation is observed, but they cannot be at rest either, due to Earnshaw’s theorem. In
short, the simple existence of matter – with its discrete charge values – shows that classical
electrodynamics is wrong.
Page 605 In fact, the overview of material properties of Table 50 makes the same point even
more strongly; classical electrodynamics can describe many of the effects listed, but it
cannot explain the origin of any of them. Even though few of the effects will be studied
in our walk – they are not essential for our adventure – the general concepts necessary
for their description will be the topic of the second part of this mountain ascent, that on
quantum theory.

17. cl assical physics in a nu tshell – one and a half

Copyright © Christoph Schiller November 1997–May 2006

steps ou t of three
The description of general relativity and classical electrodynamics concludes our walk
hrough classical physics. In order to see its limitations, we summarize what we have found
out. In nature, we learned to distinguish and to characterize objects, radiation and space-
time. All these three can move. In all motion we distinguish the fixed, intrinsic properties
from the varying state. All motion happens in such a way as to minimize change.
Looking for all the fixed, intrinsic aspects of objects, we find that all sufficiently small
objects or particles are described completely by their mass and their electric charge. There
is no magnetic charge. Mass and electric charge are thus the only localized intrinsic prop-
erties of classical, everyday objects. Both mass and electric charge are defined by the ac-

Dvipsbugw
cl assical physics in a nu tshell – one and a half steps ou t of three 615

celerations they produce around them. Both quantities are conserved; thus they can be
added. Mass, in contrast to charge, is always positive. Mass describes the interaction of
objects with their environment, charge the interaction with radiation.
All varying aspects of objects, i.e. their state, can be described using momentum and
position, as well as angular momentum and orientation. All can vary continuously in
amount and direction. Therefore the set of all possible states forms a space, the so-called
phase space. The state of extended objects is given by the states of all its constituent
particles. These particles make up all objects and somehow interact electromagnetically.
The state of a particle depends on the observer. The state is useful to calculate the Dvipsbugw
change that occurs in motion. For a given particle, the change is independent of the ob-
server, but the states are not. The states found by different observers are related: the re-
lations are called the ‘laws’ of motion. For example, for different times they are called
evolution equations, for different places and orientations they are called transformation
relations, and for different gauges they are called gauge transformations. All can be con-
densed in the principle of least action.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We also observe the motion of a massless entity: radiation. Everyday types of radiation,
such as light, radio waves and their related forms, are travelling electromagnetic waves.
They are described by same equations that describe the interaction of charged or magnetic
objects. The speed of massless entities is the maximum possible speed in nature and is
the same for all observers. The intrinsic properties of radiation are its dispersion relation
and its energy–angular momentum relation. The state of radiation is described by its
electromagnetic field strength, its phase, its polarization and its coupling to matter. The
motion of radiation describes the motion of images.
The space-time environment is described by space and time coordinates. Space-time
is also able to move, by changing its curvature. The intrinsic properties of space-time are
the number of dimensions, its signature and its topology. The state is given by the metric,
which describes distances and thus the local warpedness. The warpedness can oscillate
and propagate, so that empty space can move like a wave.
Our environment is finite in age. It has a long history, and on large scales, all matter
in the universe moves away from all other matter. The large scale topology of our envir-
onment is unclear, as is unclear what happens at its spatial and temporal limits.
Motion follows a simple rule: change is always as small as possible. This applies to
matter, radiation and space-time. All energy moves in the way space-time dictates it, and
space moves the way energy dictates it. This relation describes the motion of the stars,
of thrown stones, of light beams and of the tides. Rest and free fall are the same, and
Copyright © Christoph Schiller November 1997–May 2006

gravity is curved space-time. Mass breaks conformal symmetry and thus distinguishes
space from time.
Energy and mass speed is bound from above by a universal constant c, and energy
change per time is bound from above by a universal constant c 5 ~4G. The speed value c is
realized for the motion of massless particles. It also relates space to time. The power value
c 5 ~4G is realized by horizons. They are found around black holes and at the border of the
universe. The value also relates space-time curvature to energy flow and thus describes
the elasticity of space-time.
No two objects can be at the same spot at the same time. This is the first statement that
humans encounter about electromagnetism. More detailed investigation shows that elec-
tric charge accelerates other charges, that charge is necessary to define length and time

Dvipsbugw
 616 iv cl assical electrodynamics • 17. cl assical physics in a nu tshell

intervals, and that charges are the source of electromagnetic fields. Also light is such a
field. Light travels at the maximum possible velocity. In contrast to objects, light can in-
terpenetrate. In summary, we learned that of the two naive types of object motion, namely
motion due to gravity – or space-time curvature – and motion due to the electromagnetic
field, only the latter is genuine.
Above all, classical physics showed us that motion, be it linear or rotational, be it that
of matter, radiation or space-time, is conserved. Motion is continuous. More than that,
motion is similar to a continuous substance: it is never destroyed, never created, but al-
ways redistributed. Owing to conservation, all motion, that of objects, images and empty Dvipsbugw
space, is predictable and reversible. Owing to conservation of motion, time and space
can be defined. In addition, we found that classical motion is also right–left symmetric.
Classical physics showed us that motion is predictable: there are no surprises in nature.

The future of planet Earth

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Maybe nature shows no surprises, but it still provides many adventures. On the 8th of
March 2002, a 100 m sized body almost hit the Earth. It passed at a distance of only
450 000 km from our planet. On impact, it would have destroyed a region the size of
London. A few months earlier, a 300 m sized body missed the Earth by 800 000 km; the
record for closeness so far was in 1994, when the distance was only 100 000 km.* Several
other adventures can be predicted by classical physics, as shown in Table 51. Many are
problems facing humanity in the distant future, but some, such as volcanic eruptions or
Ref. 606 asteroid impacts, could happen at any time. All are research topics.

TA B L E 51 Examples of disastrous motion of possible future importance

C r i t i c a l si t uat i o n Ye a r s f r o m n o w

End of fundamental physics c. 30 (around year 2030)

Giant tsunami from volcanic eruption at Canary islands c. 10-200
Major nuclear material accident or weapon use unknown
Ozone shield reduction c. 100
Rising ocean levels due to greenhouse warming c. 100-1 000
End of applied physics A 200
Explosion of volcano in Greenland, leading to long darkening of unknown
sky
Copyright © Christoph Schiller November 1997–May 2006

Several magnetic north and south poles appear, allowing solar c. 800
storms to disturb radio and telecommunications, to interrupt
electricity supplies, to increase animal mutations and to disori-
ent migrating animals such as wales, birds and tortoises
Our interstellar gas cloud detaches from the solar systems, chan- c. 3 000
ging the size of the heliosphere, and thus expose us more to au-
rorae and solar magnetic fields

* The web pages around http://cfa-www.harvard.edu/iau/lists/Closest.html provide more information on

such events.

Dvipsbugw
cl assical physics in a nu tshell – one and a half steps ou t of three 617

C r i t i c a l si t uat i o n Ye a r s f r o m n o w

Reversal of Earth’s magnetic field, implying a time with almost no unknown

magnetic field, with increased cosmic radiation levels and thus
more skin cancers and miscarriages
Atmospheric oxygen depletion due to forest reduction and exag- A 1000
gerated fuel consumption
Upcoming ice age c. 15 000
Possible collision with interstellar gas cloud assumed to be c. 50 000 Dvipsbugw
crossed by the Earth every 60 million years, maybe causing mass
extinctions
Explosion of Yellowstone or other giant volcano leading to year- 0 to 100 000
long volcanic winter
Possible genetic degeneration of homo sapiens due to Y chromo- c. 200 000
some reduction

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Africa collides with Europe, transforming the Mediterranean around 3 ë 106
into a lake that starts evaporating
Gamma ray burst from within our own galaxy, causing radiation between 0 and 5 ë 106
damage to many living beings
Asteroid hitting the Earth, generating tsunamis, storms, darken- between 0 and 50 ë 106
ing sunlight, etc.
Neighbouring star approaching, starting comet shower through A 106
destabilization of Oort cloud and thus risk for life on Earth
American continent collides with Asia A 100 ë 106
Instability of solar system A 100 ë 106
Low atmospheric CO2 content stops photosynthesis A 100 ë 106
Collision of Milky Way with star cluster or other galaxy A 150 ë 106
Sun ages and gets hotter, evaporating seas A 250 ë 106
Ocean level increase due to Earth rotation slowing/stopping (if A 109
not evaporated before)
Temperature rise/fall (depending on location) due to Earth rota- A 109
tion stop
Sun runs out of fuel, becomes red giant, engulfs Earth 5.0 ë 109
Sun stops burning, becomes white dwarf 5.2 ë 109
Earth core solidifies, removing magnetic field and thus Earth’s 10.0 ë 109
Copyright © Christoph Schiller November 1997–May 2006

cosmic radiation shield

Nearby nova (e.g. Betelgeuse) bathes Earth in annihilation radi- unknown
ation
Nearby supernova (e.g. Eta Carinae) blasts over solar system unknown
Galaxy centre destabilizes rest of galaxy unknown
Universe recollapses – if ever (see page 377) A 20 ë 109
Matter decays into radiation – if ever (see Appendix C) A 1033
Problems with naked singularities unknown, controversial
Vacuum becomes unstable unknown, controversial

Dvipsbugw
618 iv cl assical electrodynamics • 17. cl assical physics in a nu tshell

Despite the fascination of the predictions, we leave aside these literally tremendous issues
and continue on our adventure.

The essence of classical physics – the infinitely small implies the

lack of surprises
We can summarize classical physics with a simple statement: nature lacks surprises be-
cause classical physics is the description of motion using the concept of the infinitely small.
All concepts used so far, be they for motion, space, time or observables, assume that the
Dvipsbugw
infinitely small exists. Special relativity, despite the speed limit, still allows infinitely small
velocities; general relativity, despite its black hole limit, still allows infinitely small force
and power values. Similarly, in the description of electrodynamics and gravitation, both
integrals and derivatives are abbreviations of mathematical processes that use infinitely
small intermediate steps.
In other words, the classical description of nature introduces the infinitely small in the
description of motion. The classical description then discovers that there are no surprises

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

in motion. The detailed study of this question lead us to a simple conclusion: the infinitely
small implies determinism.* Surprises contradict the existence of the infinitely small.
On the other hand, both special and general relativity have eliminated the existence
of the infinitely large. There is no infinitely large force, power, size, age or speed.

Why have we not yet reached the top of the mountain?

The more important fundamental laws and facts

“ of physical science have all been discovered, and

these are now so firmly established that the
possibility of their ever being supplanted in
consequence of new discoveries is exceedingly
remote... Our future discoveries must be looked
for in the sixth place of decimals.
Albert Michelson.**

We might think that we know nature now, as did Albert Michelson at the end of the
nineteenth century. He claimed that electrodynamics and Galilean physics implied that
the major laws of physics were well known. The statement is often quoted as an example
”
of flawed predictions, since it reflects an incredible mental closure to the world around
him. General relativity was still unknown, and so was quantum theory.
At the end of the nineteenth century, the progress in technology due to the use of
Copyright © Christoph Schiller November 1997–May 2006

electricity, chemistry and vacuum technology had allowed better and better machines
and apparatuses to be built. All were built with classical physics in mind. In the years
between 1890 and 1920, these classical machines completely destroyed the foundations
of classical physics. Experiments with these apparatuses showed that matter is made of
atoms, that electrical charge comes in the smallest amounts and that nature behaves ran-
domly. Nature does produce surprises – through in a restricted sense, as we will see. Like
* No surprises also imply no miracles. Classical physics is thus in opposition to many religions. Indeed, many
religions argue that infinity is the necessary ingredient to perform miracles. Classical physics shows that this
is not the case.
** From his 1894 address at the dedication ceremony for the Ryerson Physical Laboratory at the University
of Chicago.

Dvipsbugw
cl assical physics in a nu tshell – one and a half steps ou t of three 619

the British Empire, the reign of classical physics collapsed. Speaking simply, classical phys-
ics does not describe nature at small scales.
But even without machines, the Victorian physicist could have predicted the situ-
ation. (In fact, many more progressive minds did so.) He had overlooked a contradiction
between electrodynamics and nature, for which he had no excuse. In our walk so far we
found that clocks and metre bars are necessarily made of matter and based on electro-
magnetism. But as we just saw, classical electrodynamics does not explain the stability of
matter. Matter is made of small particles, but the relation between these particles, electri-
city and the smallest charges is not clear. If we do not understand matter, we do not yet Dvipsbugw
understand space and time, since they are defined using measurement devices made of
matter.
Worse, the Victorian physicist overlooked a simple fact: the classical description of
nature does not allow one to understand life. The abilities of living beings – growing,
seeing, hearing, feeling, thinking, being healthy or sick, reproducing and dying – are all
unexplained by classical physics. In fact, all these abilities contradict classical physics.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Understanding matter and its interactions, including life itself, is therefore the aim of the
second part of our ascent of Motion Mountain. The understanding will take place at small
scales; to understand nature, we need to study particles. Indeed, the atomic structure of
matter, the existence of a smallest charge and the existence of a smallest entropy makes
us question the existence of the infinitely small. There is something to explore. Doing so
will lead us from surprise to surprise. To be well prepared, we first take a break.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
620 iv cl assical electrodynamics

Biblio graphy

484 Julian Schwinger, L.L. DeR aad, K.A. Milton & W.Y. Tsai, Classical Electrody-
namics, Perseus, 1998. An excellent text on the topic by one of its greatest masters.
See also the beautiful problem book by André Bu toli & Jean-Marc Lév y-
Leblond, La physique en questions – électricité et magnétisme, Vuibert, 1999. Cited on
pages 517 and 548.
485 K. Hagiwara & al., Physical Review D 66, p. 010001, 2002, or H.V. Kl apd or-
Dvipsbugw
Kleingrothaus & K. Zuber, Particle Astrophysics, The Institute of Physics, UK, 1999.
See also H. Jeon & M. Longo, Search for magnetic monopoles trapped in matter, Phys-
ical Review Letters 75, pp. 1443–1447, 1995. See also A.S. Goldhaber & W.P. Trower,
Resource letter MM-1: magnetic monopoles, American Journal of Physics 58, pp. 429–439,
1990. Cited on page 519.
486 R. Edwards, Filling station fires spark cars’ recall, New Scientist, pp. 4–5, 4 March 1995.
Cited on page 521.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

487 S. Desmet, F. Orban & F. Grandjean, On the Kelvin electrostatic generator, European
Journal of Physics 10, pp. 118–122, 1989. You can also find construction plans for it in various
places on the internet. Cited on page 521.
488 For an etching of Franklin’s original ringing rod, see E.P. Krider, Benjamin Franklin and
lightning rods, Physics Today 59, pp. 42–48, 2006. Cited on page 522.
489 For more details on various electromagnetic units, see the standard text by J.D. Jackson,
Classical Electrodynamics, 3rd edition, Wiley, 1998. Cited on page 525.
490 See the old but beautiful papers by R ichard C. Tolman & T. Dale Stewart, The
electromotive force produced by the acceleration of metals, Physical Review 8, pp. 97–116,
1916, R ichard C. Tolman & T. Dale Stewart, The mass of the electric carrier in
copper, silver and aluminium, Physical Review 9, pp. 164–167, 1917, and the later but much
more precise experiment by C.F. Kettering & G.G. Scott, Inertia of the carrier of
electricity in copper and aluminum, Physical Review 66, pp. 257–267, 1944. (Obviously the
American language dropped the ‘i’ from aluminium during that period.) The first of these
papers is also a review of the preceding attempts, and explains the experiment in detail. The
last paper shows what had to be taken into consideration to achieve sufficient precision.
Cited on page 526.
491 This effect has first been measured by S.J. Barnett, A new electron-inertia effect and the
determination of m/e for the free electron in copper, Philosophical Magazine 12, p. 349, 1931.
Cited on page 526.
492 See for example C. Schiller, A.A. Koomans, T.L. van Rooy, C. Schönenberger
Copyright © Christoph Schiller November 1997–May 2006

& H.B. Elswijk, Decapitation of tungsten field emitter tips during sputter sharpening,
Surface Science Letters 339, pp. L925–L930, 1996. Cited on page 526.
493 L.I. Schiff & M.V. Barnhill, Gravitational-induced electric field near a metal, Physical
Review 151, pp. 1067–1071, 1966. F.C. Witteborn & W.M. Fairbank, Experimental
comparison of the gravitational force on freely falling electrons and metallic electrons, Phys-
ical Review Letters 19, pp. 1049–1052, 1967. Cited on page 527.
494 J. Lepak & M. Crescimanno, Speed of light measurement using ping, electronic pre-
print available at http://www.arxiv.org/abs/physics/0201053. Cited on page 528.
495 Pierre de Maricourt, Tractatus de magnete, 1269. Cited on page 529.
496 R. Wiltschko & W. Wiltschko, Magnetic Orientation in Animals, Springer, Berlin,
1995. Cited on page 530.

Dvipsbugw
 bibliography 621

497 The ratio of angular L to magnetic M moment is

=
L 2m 1
ë , (460)
M e д
where e is the electron charge and m its mass. Both L and M are measurable. The first meas-
urements were published with a д-value of 1, most probably because the authors expected
the value. In later experiments, de Haas found other values. Measurements by other research-
ers gave values nearer to 2 than to 1, a fact that was only understood with the discovery of
spin. The original publications are A. Einstein & W.J. de Haas, Proefondervinderlijk Dvipsbugw
bewijs voor het bestaan der moleculaire stroomen van Ampère, Konninklijke Akademie der
Wetenschappen te Amsterdam, Verslagen 23, p. 1449, 1915, and A. Einstein & W.J. de
Haas, Experimental proof of the existence of Ampère’s molecular currents, Konninklijke
Akademie der Wetenschappen te Amsterdam, Proceedings 18, p. 696, 1916. Cited on page 532.
498 S.J. Barnett, Magnetization by rotation, Physical Review 6, pp. 171–172, 1915, and S.J.
Barnett, Magnetization by rotation, Physical Review 6, pp. 239–270, 1915. Cited on page
532.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

499 See J.D. Jackson, Classical Electrodynamics, 3rd edition, Wiley, 1998, or also R.F. Har-
rington, Time Harmonic Electromagnetic Fields, McGraw–Hill, 1961. Cited on pages 535
and 548.
500 The http://suhep.phy.syr.edu/courses/modules/MM/Biology/biology2.html website gives
an introduction into brain physiology. Cited on page 535.
501 N. Salingaros, Invariants of the electromagnetic field and electromagnetic waves, Amer-
ican Journal of Physics 53, pp. 361–363, 1985. Cited on page 535.
502 A.C. de l a Torre, VD c in 1820?, European Journal of Physics 20, pp. L23–L24, March
1999. Cited on page 536.
503 R.H. Tyler, S. Maus & H. Lühr, Magnetic signal due to ocean tidal flow identified in
satellite observations, Science 299, pp. 239–241, 2003. The films derived from the data can
be found on the http://www.tu-bs.de/institute/geophysik/spp/publikationen.html website.
Cited on page 541.
504 H. Montgomery, Unipolar induction: a neglected topic in the teaching of electromag-
netism, European Journal of Physics 20, pp. 271–280, 1999. Cited on page 542.
505 On the geodynamo status, see the article by P.H. Roberts & G.A. Gl atzmaier, Geody-
namo theory and simulations, Reviews of Modern Physics 72, pp. 1081–1123, 2000. An older
article is R. Jeanloz & B. Romanowicz, Geophysical dynamics at the center of the
Earth, Physics Today pp. 22–27, August 1997. Cited on pages 543 and 600.
506 J. Yang, F. Lu, L.W. Kostiuk & D.Y. Kwok, Electrokinetic microchannel battery by
Copyright © Christoph Schiller November 1997–May 2006

means of electrokinetic and microfluidic phenomena, Journal of Micromechanics and Mi-

croengineering 13, pp. 963–970, 2003. Cited on page 545.
507 Oleg D. Jefimenko, A relativistic paradox seemingly violating conservation of mo-
mentum law in electromagnetic systems, European Journal of Physics 20, pp. 39–44, 1999.
Of course, the missing momentum goes into the electromagnetic field. Given that the elec-
tromagnetic momentum is given by the vector potential, are you able to check whether
Challenge 1097 ny everything comes out right? Cited on page 548.
508 H. Van Dam & E.P. Wigner, Classical relativistic mechanics of interacting point
particles, Physical Review 136B, pp. 1576–1582, 1965. Cited on page 548.
509 Mark D. Semon & John R. Taylor, Thoughts on the magnetic vector potential, Amer-
ican Journal of Physics 64, pp. 1361–1369, 1996. Cited on pages 549, 550, and 551.

Dvipsbugw
622 iv cl assical electrodynamics

510 Jean Sivardière, Simple derivation of magnetic vector potentials, European Journal of
Physics 14, pp. 251–254, 1993. Cited on page 549.
511 T.T. Wu & C.N. Yang, 1975, Concept of nonintegrable phase factors and global formula-
tion of gauge fields, Physical Review D 12, pp. 3845–3857, Cited on page 552.
512 An electrodynamics text completely written with (mathematical) forms is Kurt Meetz
& Walter L. Engl, Elektromagnetische Felder – mathematische und physikalische
Grundlagen, Springer, 1980. Cited on page 553.
513 J. Travis, Twirl those organs into place – getting to the heart of how a heart knows left Dvipsbugw
from right, Science News 156, 21 August, 1999. A good book on asymmetries in nature is H.
Brunner, Rechts oder links, Wiley–Vch, 1999. Cited on page 555.
514 See for example the discussion by M.C. Corballis & I.L. Beale, On telling left from
right, Scientific American 224, pp. 96–104, March 1971. Cited on page 555.
515 Wolf gang R indler, Essential Relativity – Special, General, and Cosmological, revised
2nd edition, Springer Verlag, 1977, page 247. There is also the beautiful paper by M. Le
Bell ac & J.-M. Lév y-Leblond, Galilean electrodynamics, Nuovo Cimento B 14, p. 217,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1973, that explains the possibilities but also the problems appearing when trying to define
the theory non-relativistically. Cited on page 557.
516 L.-C. Tu, J. Luo & G.T. Gilles, The mass of the photon, Reports on Progress of Physics
68, pp. 77–130, 2005. Cited on page 557.
517 For a captivating account on the history of the ideas on light, see David Park, The Fire
Within the Eye: a Historical Essay on the Nature and Meaning of Light, Princeton University
Press, 1997. Cited on page 558.
518 See the text by R aymond L. Lee & Alistair B. Fraser, The Rainbow Bridge: Rainbows
in Art, Myth, and Science, Pennsylvania State University Press, 2000. A chapter can be found
at the http://www.nadn.navy.mil/Oceanography/RainbowBridge/Chapter_8.html website.
Cited on page 561.
519 The beautiful slit experiment was published by E.A. Montie, E.C. Cosman, G.W. ’t
Hooft, M.B. van der Mark & C.W.J. Beenakker, Observation of the optical
analogue of quantized conductance of a point contact, Nature 350, pp. 594–595, 18
April 1991, and in the longer version E.A. Montie, E.C. Cosman, G.W. ’t Hooft,
M.B. van der Mark & C.W.J. Beenakker, Observation of the optical analogue of
the quantised conductance of a point contact, Physica B 175, pp. 149–152, 1991. The result
was also publicized in numerous other scientific magazines. Cited on pages 561 and 892.
520 A recent measurement of the frequency of light is presented in Th. Udem, A. Huber, B.
Gross, J. R eichert, M. Prevedelli, M. Weitz & T.W. Hausch, Phase-coherent
Copyright © Christoph Schiller November 1997–May 2006

measurement of the hydrogen 1S–2S transition frequency with an optical frequency inter-
val divider chain, Physical Review Letters 79, pp. 2646–2649, 6 October 1997. Another is
C. Schwob, L. Jozefowski, B. de Beauvoir, L. Hilico, F. Nez, L. Julien, F.
Biraben, O. Acef & A. Cl airon, Optical frequency measurement of the 2S-12D trans-
itions in hydrogen and deuterium: Rydberg constant and Lamb shift determinations, Phys-
ical Review Letters 82, pp. 4960–4963, 21 June 1999. Cited on page 562.
521 The discoverors of two such methods were awarded the 2005 Nobel Prize for physics. Cited
on page 562.
522 See for example G. Horváth, J. Gál & R. Wehner, Why are water-seeking insects not
attracted by mirages? The polarization pattern of mirages, Naturwissenschaften 83, pp. 300–
303, 1997. Cited on page 562.

Dvipsbugw
bibliography 623

523 On the birefringence of the eye, see L. B our, Een eigenaardige speling der natuur, Neder-
lands tijdschrift voor natuurkunde 67, pp. 362–364, December 2001. In particular, a photo-
graph of the eye using linear polarized illumination and taken through an analyser shows a
black cross inside the pupil. Cited on page 562.
524 The standard reference on the propagation of light is Max B orn & Emil Wolf, Principles
of Optics – Electromagnetic Theory of Propagation, Interference and Diffraction of Light, Per-
gamon Press, 6th edition, 1998. Cited on page 567.
525 An introduction to the topic of the 22° halo, the 46° halo, Sun dogs, and the many other
arcs and bows that can be seen around the Sun, see the beautifully illustrated paper by R. Dvipsbugw
Greenler, Lichterscheinungen, Eiskristalle und Himmelsarchäologie, Physikalische Blät-
ter 54, pp. 133–139, 1998, or his book Rainbows, Halos, and Glories, Cambridge University
Press 1980. Cited on page 569.
526 James E. Faller & E. Joseph Wampler, The lunar laser reflector, Scientific American
pp. 38–49, March 1970. Cited on page 570.
527 Neil Armstrong of Apollo 11, Jim Lovell of Apollo 8 and Apollo 13, and Jim Irwin of Apollo

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

15 extensively searched for it and then made negative statements, as told in Science News
p. 423, 24 & 31 December 1994. From the space shuttle however, which circles only a few
hundred kilometres above the Earth, the wall can be seen when the Sun is low enough such
that the wall appears wider through its own shadow, as explained in Science News 149, p. 301,
1996. Cited on page 570.
528 M. Shih, M. Segev & G. Sal amo, Physical Review Letters 78, pp. 2551–2554, 1997.
See also the more readable paper by Mordechai Segev & George Stegeman, Self-
trapping of optical beams: spatial solitons, Physics Today 51, pp. 42–48, August 1998. Cited
on page 571.
529 The first correct explanation of the light mill was given by Osborne R eynolds, On cer-
tain dimensional properties of matter in the gaseous state, Royal Society Philosophical Trans-
actions Part 2, 1879. The best discussion is the one given on the web by Phil Gibbs, in the
frequently asked question list of the usenet news group sci.physics; it is available at the http://
www.desy.de/user/projects/Physics/light-mill.html website. Cited on page 573.
530 P. Lebedev, Untersuchungen über die Druckkräfte des Lichtes, Annalen der Physik 6,
pp. 307–458, 1901. He was also the first who understood that this effect is the basis for the
change of direction of the tails of comets when they circle around the Sun. Cited on page
573.
531 P. Gal ajda & P. Ormos, Applied Physics Letters 78, p. 249, 2001. Cited on page 573.
532 A short overview is given by Miles Pad gett & Les Allen, Optical tweezers and span-
ners, Physics World pp. 35–38, September 1997. The original papers by Ashkin’s group are
Copyright © Christoph Schiller November 1997–May 2006

A. Ashkin, J.M. Dziedzic, J.E. Bjorkholm & S. Chu, Observation of a gradient

force optical trap for dielectric particles, Optics Letters 11, p. 288, 1986, and A. Askin,
J.M. Dziedzic & T. Yamane, Optical trapping and manipulation of single cells using in-
frared laser beams, Nature 330, p. 769, 1987. A pedagogical explanation on optical spanners,
together with a way to build one, can be found in D.N. Moothoo, J. Arlt, R.S. Con-
roy, F. Akerboom, A. Voit & K. Dhol akia, Beth’s experiment using optical tweezers,
American Journal of Physics 69, pp. 271–276, 2001, and in S.P. Smith, S.R. Bhalotra,
A.L. Brody, B.L. Brown, E.K. B oyda & M. Prentiss, Inexpensive optical tweezers
for undergraduate laboratories, American Journal of Physics 67, pp. 26–35, 1999. Cited on
page 573.
533 R. Beth, Physical Review 50, p. 115, 1936. For modern measurements, see N.B. Simpson,
K. Dhol akia, L. Allen & M.J. Pad gett, Optics Letters 22, p. 52, 1997, and M.E.J.

Dvipsbugw
624 iv cl assical electrodynamics

Frieze, T.A. Nieminen, N.R. Heckenberg & H. Rubinsztein-Dunlop, Optics

Letters 23, p. 1, 1998. Cited on page 574.
534 See the Latin text by Dietrich von Freiberg De iride et radialibus impressionibus,
c. 1315. Cited on page 576.
535 J. Walker, Multiple rainbows from single drops of water and other liquids, American
Journal of Physics 44, pp. 421–433, 1976, and his How to create and observe a dozen rainbows
in a single drop of water, Scientific American 237, pp. 138–144, 1977. See also K. Sassen, An-
gular scattering and rainbow formation in pendant drops, Journal of the Optical Society of
America 69, pp. 1083–1089, 1979. Cited on page 576. Dvipsbugw
536 There are also other ways to see the green ray, for longer times, namely when a fata morgana
appears at sunset. An explanation with colour photograph is contained in M. Vollmer,
Gespiegelt in besonderen Düften ...– Oasen, Seeungeheuer und weitere Spielereien der Fata
Morgana, Physikalische Blätter 54, pp. 903–909, 1998. Cited on page 576.
537 This famous discovery is by Brent Berlin & Paul Kay, Basic Color Terms: Their Uni-
versality and Evolution, University of California Press, 1969. Their ongoing world colour sur-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

vey is eagerly awaited. Of course there are also ongoing studies to find possible exceptions;
the basic structure is solid, as shown in the conference proceedings C.L. Hardin & Lu-
isa Maffi, Colour Categories in Thought and Language, Cambridge University Press, 1997.
Cited on page 577.
538 For a thorough discussion of the various velocities connected to wave trains, see the classic
text by Louis Brillouin, Wave Propagation and Group Velocity, Academic Press, New
York, 1960. It expands in detail the theme discussed by Arnold Sommerfeld, Über die
Fortpflanzung des Lichtes in dispergierenden Medien, Annalen der Physik, 4th series, 44,
pp. 177–202, 1914. See also Arnold Sommerfeld, Optik, Dietrichssche Verlagsbuchand-
lung, Wiesbaden 1950, section 22. An English translation Arnold Sommerfeld, Lec-
tures on Theoretical Physics: Optics, 1954, is also available. Cited on pages 577, 578, and 579.
539 Changing the group velocity in fibers is now even possible on demand, as shown by M.
González-Herráez, K.-Y. Song & L. Thévenaz, Optically controlled slow and
fast light in optical fibers using stimulated Brillouin scattering, Applied Physics Letters 87,
p. 081113, 2005. They demonstrate group velocities from 0.24c to plus infinity and beyond,
to negative values. Cited on page 578.
Another experiment was carried out by S. Chu & S. Wong, Linear pulse propagation
in an absorbing medium, Physical Review Letters 48, pp. 738–741, 1982. See also S. Chu
& D. Styer, Answer to question #52. Group velocity and energy propagation, American
Journal of Physics 66, pp. 659–661, 1998. Another example was described in 1993 by the group
of Raymond Chiao for the case of certain nonlinear materials in R. Chiao, P.G. Kwait
& A.M. Steinberg, Faster than light?, Scientific American 269, p. 52, August 1993, and
Copyright © Christoph Schiller November 1997–May 2006

R.Y. Chiao, A.E. Kozhekin & G. Kurizki, Tachyonlike excitations in inverted two-
level media, Physical Review Letters 77, pp. 1254–1257, 1996. On still another experimental
set-up using anomalous dispersion in caesium gas, see L.J. Wang, A. Kuzmich & A.
Dogarin, Gain-assisted superluminal light propagation, Nature 406, pp. 277–279, 20 July
2000.
540 Y.P. Terletskii, Paradoxes in the Theory of Relativity, Plenum Press, 1968. Cited on page
579.
541 See the excellent explanation by Kirk T. McDonald, Negative group velocity, American
Journal of Physics 69, pp. 607–614, 2001. Cited on page 579.
542 The prediction of negative refraction is due to V.G. Vesel ago, The electrodynamics of
substances with simultaneously negative values of ε and µ, Soviet Physics Uspekhi 10, p. 509,

Dvipsbugw
bibliography 625

1968. The explanation with different refraction directions was published by P.M. Val anju,
R.M. Walser & A.P. Val anju, Wave refraction in negative-index media: always positive
and very inhomogeneous, Physical Review Letters 88, p. 187401, 8 May 2002. Also Fermat’s
principle is corrected, as explained in V.G. Vesel ago, About the wording of Fermat’s
principle for light propagation in media with negative refraction index, http://www.arxiv.
org/abs/cond-mat/0203451. Cited on page 579.
543 The first example of material system with a negative refraction index were presented by
David Smith and his team. R.A. Schelby, D.R. Smith & S. Schultz, Experimental
verification of a negative index of refraction, Science 292, p. 77-79, 2001. More recent ex- Dvipsbugw
amples are A.A. Houck, J.B. Brock & I.L. Chuang, Experimental observations of a
left-handed material that obeys Snell’s law, Physical Review Letters 90, p. 137401, 2003, C.G.
Parazzoli, R.B. Greegor, K. Li, B.E.C. Koltenbah & M. Tanielian, Experi-
mental verification and simulation of negative index of refraction using Snell’s law, Phys-
ical Review Letters 90, p. 107401, 2003. S. Foteinopoulou, E.N. Economou & C.M.
Soukoulis, Refraction in media with a negative refractive index, Physical Review Letters
90, p. 107402, 2003. Cited on page 580.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

544 S.A. R amakrishna, Physics of negative refractive index materials, Reorts on Progress of
Physics 68, pp. 449–521, 2005. Cited on page 580.
545 J. Pendry, Negegative refraction makes a perfect lens, Physical Review Letters 85, p. 3966,
2000. Cited on page 580.
546 G. Nimtz, A. Enders & H. Spieker, Journal de Physique I (Paris) 4, p. 565, 1994. Unfor-
tunately, Nimtz himself seems to believe that he transported energy or signals faster than
light; he is aided by the often badly prepared critics of his quite sophisticated experiments.
See A. Enders & G. Nimtz, Physikalische Blätter 49, p. 1119, Dezember 1993, and the weak
replies in Physikalische Blätter 50, p. 313, April 1994. See also A.M. Steinberg, Journal de
Physique I (Paris) 4, p. 1813, 1994, A.M. Steinberg, P.G. Kwiat & R.Y. Chiao, Phys-
ical Review Letters 71, pp. 708–711, 1993, and A. R anfagni, P. Fabeni, G.P. Pazzi & D.
Mugnai, Physical Review E 48, p. 1453, 1993. Cited on page 580.
547 A summary of all evidence about the motion of the aether is given by R.S. Shankl and,
S.W. McCuskey, F.C. Leone & G. Kuerti, New analysis of the interferometer obser-
vations of Dayton C. Miller, Review of Modern Physics 27, pp. 167–178, 1955. An older text
is H. Witte, Annalen der Physik 26, p. 235, 1908. Cited on page 581.
548 The history of the concept of vacuum can be found in the book by E. Grant, Much Ado
About Nothing, Cambridge University Press, 1981, and in the extensive reference text by Ed-
mund T. Whittaker, A History of the Theories of Aether and Electricity, Volume 1: The
Classical Theories, Volume 2: The Modern Theories, Tomash Publishers, American Institute
Copyright © Christoph Schiller November 1997–May 2006

of Physics 1951, 1987. Cited on page 582.

The various aether models – gears, tubes, vortices – proposed in the nineteenth century
were dropped for various reasons. Since many models used to explain electric and magnetic
fields as motion of some entities, it was concluded that the speed of light would depend
on electric or magnetic fields. One type of field was usually described by linear motion of
the entities, the other by rotatory or twisting motion; both assignments are possible. As a
consequence, aether must be a somewhat strange fluid that flows perfectly, but that resists
rotation of volume elements, as McCullogh deduced in 1839. However, experiments show
that the speed of light in vacuum does not depend on electromagnetic field intensity. Vor-
tices were dropped because real world vortices were found out to be unstable. All models
received their final blow when they failed to meet the requirements of special relativity.
549 This happened to Giovanni Bellini (c. 1430–1516) the great Venetian Renaissance painter,

Dvipsbugw
626 iv cl assical electrodynamics

who even put this experience into writing, thus producing one of the greatest ‘gaffes’ ever. If
you take a photograph of the effect with a remotely controlled camera, you can prove that
your camera is holy as well. Cited on page 582.
550 S.R. Wilk, How retroreflectors really work, Optics & Photonics News, pp. 6–7, December
1993. Cited on page 582.
551 W.H. Ehrenstein & B. Lingelbach, Das Hermann–Gitter, Physik in unserer Zeit 6,
pp. 263–268, 2002. The journal also shows a colour variation of these lattices. Cited on page
583.
552 The eye sensitivity myth is debunked in detail by B.H. Soffer & D.K. Lynch, Some Dvipsbugw
paradoxes, errors, and resolutions concerning the spectral optimization of human vision,
American Journal of Physics 67, pp. 946–953, 1999. Cited on page 585.
553 David R. Williams, Supernormal Vision, Science News 152, pp. 312–313, 15 November
1997. See also http://www.cvs.rochester.edu/people/d_williams/d_williams.html as well as
the photographs at http://www.cvs.rochester.edu/people/~aroorda/ao_research.html of the
interior of living human eyes. Their last publication is A. Roorda, A. Metha, P. Lennie

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

& D.R. Williams, Packing arrangement of the three cone classes in the primate retina,
Vision Research 41, pp. 1291–1306, 2001. Cited on page 585.
554 S.W. Hell, Strategy for far-field optical imaging and writing without diffraction limit, Phys-
ics Letters A 326, pp. 140–145, 2004, see also V. Westphal & S.W. Hell, Nanoscale resol-
ution in the focal plane of an optical microscope, Physical Review Letters 94, p. 143903, 2005,
and V. Westphal, J. Seeger, T. Salditt & S.W. Hell, Stimulated emission depletion
microscopy on lithographic microstructures, Journal of Physics B 38, pp. S695–S705, 2005.
Cited on page 589.
555 A.D. Erlykin & A.W. Wolfendale, The origin of cosmic rays, European Journal of
Physics 20, pp. 409–418, 1999, Cited on page 591.
556 D. Singleton, Electromagnetic angular momentum and quantum mechanics, American
Journal of Physics 66, pp. 697–701, 1998, Cited on page 592.
557 A. Yazdani, D.M. Eigler & N.D. L ang, Off-resonance conduction through atomic
wires, Science 272, pp. 1921–1924, 28 June 1996. For aluminium, gold, lead, niobium, as
well as the influence of chemical properties, see Elke Scheer, The signature of chem-
ical valence in the electric conduction through a single-atom contact, Nature 394, pp. 154–
157, 9 July 1998. Cited on page 593.
558 See L. Kowalski, A myth about capacitors in series, The Physics Teacher 26, pp. 286–287,
1988, and A.P. French, Are the textbook writers wrong about capacitors?, The Physics
Teacher 31, pp. 156–159, 1993. Cited on page 593.
559 This problem was suggested by Vladimir Surdin. Cited on page 593.
Copyright © Christoph Schiller November 1997–May 2006

560 See for example, J.M. Aguirregabiria, A. Hernandez & M. R ivas, Velocity fields
inside a conducting sphere near a slowly moving charge, American Journal of Physics 62,
pp. 462–466, 1994. Cited on page 594.
561 Philip Cohen, Open wide, this won’t hurt a bit, New Scientist p. 5, 3 February 1996. Cited
on page 594.
562 Such a claim was implicitly made by D. Mugnai, A. R anfagni & R. Ruggieri,
Observation of superluminal behaviors in wave propagation, Physical Review Letters 84,
p. 4830, 2000. An excellent explanation and rebuttal was given by W.A. Rodrigues, D.S.
Thober & A.L. X avier, Causal explanation for observed superluminal behavior of mi-
crowave propagation in free space, http://www.arxiv.org/abs/physics/0012032. Cited on
page 595.

Dvipsbugw
bibliography 627

563 J.E. Avron, E. Berg, D. Goldsmith & A. Gord on, Is the number of photons a clas-
sical invariant?, European Journal of Physics 20, pp. 153–159, 1999. Cited on page 595.
564 If you want to see more on how the world looks for the different types of colour blind, have
a look at the http://webexhibits.org/causesofcolor/2.html or the http://www.vischeck.com/
examples web pages. Cited on page 597.
565 About the life-long passion that drove Luke Howard, see the book by R ichard Hamblyn,
The Invention of Clouds, Macmillan 2001. Cited on page 597.
566 J. L atham, The electrification of thunderstorms, Quartely Journal of the Royal Meteorolo- Dvipsbugw
gical Society 107, pp. 277–289, 1981. For a more recent and wider review, see Earle R.
Williams, The tripole structure of thunderstorms, Journal of Geophysical Research 94,
pp. 13151–13167, 1989. See also the book by the National R esearch Council Staff,
The Earth’s Electrical Environment, Studies in Geophysics, National Academy Press, 1986.
Cited on page 597.
567 B.M. Smirnov, Physics of ball lightning, Physics Reports 224, pp. 151–236, 1993. See also D.
Finkelstein & J. Rubinstein, Ball lightning, Physical Review 135, pp. 390–396, 1964.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

For more folklore on the topic, just search the world wide web. Cited on page 598.
568 A.V. Gurevich & K.P. Zybin, Runaway breakdown and the mysteries of lightning, Phys-
ics Today 58, pp. 37–43, May 2005. Cited on page 598.
569 To learn more about atmospheric currents, you may want to have a look at the popularizing
review of the US work by E.A. Bering, A.A. Few & J.R. Benbrook, The global electric
circuit, Physics Today 51, pp. 24–30, October 1998, or the more technical overview by E.
Bering, Reviews of Geophysics (supplement) 33, p. 845, 1995. Cited on page 599.
570 The use of Schumann resonances in the Earth–ionosphere capacitor for this research field
is explained in K. Schlegel & M. Füllerkrug, Weltweite Ortung von Blitzen, Physik
in unserer Zeit 33, pp. 257–261, 2002. Cited on page 599.
571 J.R. Dw yer, M.A. Uman, H.K. R assoul, M. Al-Dayeh, E.L. Caraway, J.
Jerauld, V.A. R akov, D.M. Jordan, K.J. R ambo, V. Corbin &B. Wright,
Energetic radiation produced by rocket-triggered lightning, Science 299, pp. 694–697, 2003.
Cited on page 599.
572 J.R. Dw yer, A fundamental limit on electric fields in air, Geophysical Research Letters 30,
p. 2055, 2003. Cited on page 599.
573 For a recent summary, see S. Parrott, http://www.arxiv.org/abs/gr-qc/9711027. See also
T.A. Abbott & D.J. Griffiths, Acceleration without radiation, American Journal of
Physics 53, pp. 1203–1211, 1985. See also A. Kovetz & G.E. Tauber, Radiation from an
accelerated charge and the principle of equivalence, American Journal of Physics 37, pp. 382–
Copyright © Christoph Schiller November 1997–May 2006

385, 1969. Cited on page 600.

574 C. de Almeida & A. Saa, The radiation of a uniformly accelerated charge is beyond the
horizon: a simple derivation, American Journal of Physics 74, pp. 154–158, 2006. Cited on
page 600.
575 J. Zhang, X.D. Song, Y.C. Li, P.G. R ichards, X.L. Sun & F. Waldhauser, Inner
core differential motion confirmed by earthquake doublet waveform doublets, Science 309,
pp. 1357–1360, 2005. Cited on page 601.
576 This is deduced from the д − 2 measurements, as explained in his Nobel Prize talk by Hans
Dehmelt, Experiments with an isolated subatomic particle at rest, Reviews of Modern Phys-
ics 62, pp. 525–530, 1990, and in Hans Dehmelt, Is the electron a composite particle?,
Hyperfine Interactions 81, pp. 1–3, 1993. Cited on page 601.

Dvipsbugw
628 iv cl assical electrodynamics

577 A good and short introduction is the paper F. Rohrlich, The self-force and radiation
reaction, American Journal of Physics 68, pp. 1109–1112, 2000. Cited on page 601.
578 C.G. Tsagas, Magnetic tension and the geometry of the universe, Physical Review Letters
86, pp. 5421–5424, 2001. An overview of the topic is C.G. Tsagas, Geometrical aspects of
cosmic magnetic fields, http://www.arxiv.org/abs/gr-qc/0112077. Cited on page 602.
579 An excellent review is E.H. Brandt, Levitation in Physics, Science 243, pp. 349–355, 1989.
Cited on pages 602 and 604.
580 See the article by R. Tuckermann, S. Bauerecker & B. Neidhart, Levitation in
Ultraschallfeldern – Schwebende Tröpfchen, Physik in unserer Zeit 32, pp. 69–75, February Dvipsbugw
2001. Liquid drops up to 1 g have been levitated in this way. Cited on page 602.
581 F.C. Moon & P.Z. Chang, Superconducting Levitation – Applications to Bearings and
Magnetic Transportation, Wiley & Sons, 1994. Cited on pages 602 and 603.
582 W.T. Scott, Who was Earnshaw?, American Journal of Physics 27, pp. 418–419, 1959. Cited
on page 602.
583 The trick is to show that div E = 0, curl E = 0, thus E∇2 E = 0 and, from this, ∇2 E 2 E 0;

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

there are thus no local electric field maxima in the absence of free charges. The same proof
works for the magnetic field. However, bodies with dielectric constants lower than their
environment can be levitated in static electric fields. An example is gas bubbles in liquids, as
shown by T.B. Jones & G.W. Bliss, Bubble dielectrophoresis, Journal of Applied Physics
48, pp. 1412–1417, 1977. Cited on page 603.
584 However, it is possible to levitate magnets if one uses a combination containing diamagnets.
See A.K. Geim, M.D. Simon, M.I. B oamfa & L.O. Heflinger, Magnet levitation at
your fingertips, Nature 400, pp. 323–324, 1999. Cited on page 603.
585 The first photographs of a single ion were in W. Neuhauser, M. Hohenstatt, P.E.
Toschek & H. Dehmelt, Localized visible Ba + mono-ion oscillator, Physical Review A
22, pp. 1137–1140, 1980. See also D.J. Winel and & W.M. Itano, Physics Letters A 82,
p. 75, 1981, as well as F. Dietrich & H. Walter, Physical Review Letters 58, p. 203, 1987.
For single atoms, see photographs in Z. Hu & H.J. Kimble, Optics Letters 1, p. 1888,
1994, F. Ruschewitz, D. Bettermann, J.L. Peng & W. Ertmer, Europhysics Letters
34, p. 651, 1996, D. Haubrich, H. Schadwinkel, F. Strauch, B. Ueberholz, R.
Wynands & D. Meschede, Europhysics Letters 34, p. 663, 1996. Cited on page 603.
586 See for example Mark Buchanan, And God said...let there be levitating strawberries,
flying frogs and humans that hover over Seattle, New Scientist pp. 42–43, 26 July 1997, or C.
Wu, Floating frogs, Science News 152, pp. 632–363, 6 December 1997, and C. Wu, Molecular
magnetism takes off, Physics World April 1997, page 28. The experiments by Andre Geim,
Jan Kees Maan, Humberto Carmona and Peter Main were made public by P. Rod gers,
Copyright © Christoph Schiller November 1997–May 2006

Physics World 10, p. 28, 1997. Some of the results can be found in M.V. Berry & A.K.
Geim, Of flying frogs and levitrons, European Journal of Physics 18, pp. 307–313, 1997. See
also their http://www-hfml.sci.kun.nl/hfml/levitate.html website. Cited on page 603.
587 The well-known toy allows levitation without the use of any energy source and is called the
‘Levitron’. It was not invented by Bill Hones of Fascination Toys & Gifts in Seattle, as the
http://www.levitron.com website explains. The toy is discussed by Ron Ed ge, Levitation
using only permanent magnets, Physics Teacher 33, p. 252, April 1995. It is also discussed
in M.V. Berry, The LevitronT M : an adiabatic trap for spins, Proceedings of the Royal So-
ciety A 452, pp. 1207–1220, 1996, (of Berry’s phase fame) as well as by M.D. Simon, L.O.
Heflinger & S.L. R id geway, Spin stabilized magnetic levitation, American Journal of
Physics 65, pp. 286–92, 1997, and by T.B. Jones, M. Washizu & R. Gans, Simple theory
for the Levitron, Journal of Applied Physics 82, pp. 883–889, 1997. Cited on page 604.

Dvipsbugw
bibliography 629

588 The drill trick and the building of a Levitron are described in the beautiful lecture script
by Josef Zweck, Physik im Alltag, Skript zur Vorlesung im WS 1999/2000 der Universität
Regensburg. Cited on page 604.
589 The prediction about quantized levitation is by Stephen B. Haley, Length quantization
in levitation of magnetic microparticles by a mesoscopic superconducting ring, Physical Re-
view Letters 74, pp. 3261–3264, 1995. The topic is discussed in more detail in Stephen B.
Haley, Magnetic levitation, suspension, and superconductivity: macroscopic and meso-
scopic, Physical Review B 53, p. 3506, 1996, reversed in order with Stephen B. Haley,
Quantized levitation of superconducting multiple-ring systems, Physical Review B 53, Dvipsbugw
p. 3497, 1996, as well as Stephen B. Haley, Quantized levitation by multiply-connected
superconductors, LT-21 Proceedings, in Czechoslovak Journal of Physics 46, p. 2331, 1996. In
1998, there was not yet an experimental confirmation (Stephen Haley, private communica-
tion). Cited on page 604.
590 All the illusions of the flying act look as if the magician is hanging on lines, as observed by
many, including the author. (Photographic flashes are forbidden, a shimmery background is
set up to render the observation of the lines difficult, no ring is ever actually pulled over the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

magician, the aquarium in which he floats is kept open to let the fishing lines pass through,
always the same partner is ‘randomly’ chosen from the public, etc.) Information from eye-
witnesses who have actually seen the fishing lines used by David Copperfield explains the
reasons for these set-ups. The usenet news group alt.magic.secrets, in particular Tilman
Hausherr, was central in clearing up this issue in all its details, including the name of the
company that made the suspension mechanism. Cited on page 604.
591 Detailed descriptions of many of these effects can be found in the excellent overview edited
by Manfred von Ardenne, Gerhard Musiol & Siegfried R eball, Effekte der
Physik und ihre Anwendungen, Harri Deutsch, 2004. Cited on page 605.
592 R. Buddakian, K. Weninger, R.A. Hiller & Seth J. Pu tterman, Picosecond
discharges and stick–slip friction at a moving meniscus of mercury in glass, Nature 391,
pp. 266–268, 15 January 1998. See also Science News 153, p. 53, 24 January 1998. Cited on
page 605.
593 Henk Swagten & R einder Coehoorn, Magnetische tunneljuncties, Nederlands tijd-
schrift voor natuurkunde 64, pp. 279–283, November 1998. Cited on page 605.
594 H. Ohno, D. Chiba, F. Matsukura, T. Omiya, E. Abe, T. Dietl, Y. Ohno & K.
Ohtani, Electric-field control of ferromagnetism, Nature 408, pp. 944–946, 21-28 Decem-
ber 2000. Cited on page 606.
595 This effect was discovered by Geert R ikken, Bart van Tiggelen & Anja Sparen-
berg, Lichtverstrooiing in een magneetveld, Nederlands tijdschrift voor natuurkunde 63,
Copyright © Christoph Schiller November 1997–May 2006

pp. 67–70, maart 1998. Cited on page 607.

596 Vitalij Pecharsky & Karl A. Gschneidner, Giant magnetocaloric effect in
Gd5(Si2Ge2), Physical Review Letters 78, pp. 4494–4497, 1995, and, from the same authors,
Tunable magnetic regenerator alloys with a giant magnetocaloric effect for magnetic refri-
geration from 20 to 2990 K, Applied Physics Letters 70, p. 3299, 1997. Cited on page 607.
597 J. Weissmüller, R.N. Viswanath, D. Kramer, P. Zimmer, R. Würschum & H.
Gleiter, Charge-induced reversible strain in a metal, Science 300, pp. 312–315, 11 April
2003. Cited on page 608.
598 A. Ajdari, Electro-osmosis on inhomogeneously charged surfaces, Physical Review Letters
75, pp. 755–758, 1995. Cited on page 608.
599 M.J. Aitken, Thermoluminescence Dating, Academic Press, 1985. The precision of the

Dvipsbugw
630 iv cl assical electrodynamics

method is far worse that C14 dating, however, as shown by H. Huppertz, Thermolu-
mineszenzdatierung: eine methodologische Analyse aufgrund gesicherter Befunde, Peter Lang
Verlag, 2000. Cited on page 609.
600 This effect was discovered by J.N. Huiberts, R. Griessen, J.H. R ector, R.J.
Wijngarden, J.P. Dekker, D.G. de Groot & N.J. Koeman, Yttrium and
lanthanum hydride films with switchable optical properties, Nature 380, pp. 231–234, 1996.
A good introduction is R. Griessen, Schaltbare Spiegel aus Metallhydriden, Physikalische
Blätter 53, pp. 1207–1209, 1997. Cited on page 610.
601 See any book on thermostatics, such as Linda R eichl, A Modern Course in Statistical Dvipsbugw
Physics, Wiley, 2nd edition, 1998. Cited on page 611.
602 The Sun emits about 4 ë 1026 W from its mass of 2 ë 1030 kg, about 0.2 mW~kg; a person with
an average mass of 75 kg emits about 100 W (you can check this in bed at night), i.e. about
500 times more. Cited on page 611.
603 See its http://www.cie.co.at/cie website. Cited on page 612.
604 P.D. Jones, M. New, D.E. Parker, S. Martin & I.G. R igor, Surface air temperature

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

and its changes over the past 150 years, Reviews of Geophysics 37, pp. 173–199, May 1999.
Cited on page 612.
605 Pictures of objects in a red hot oven and at room temperature are shown in C.H. Bennett,
Demons, engines and the second law, Scientific American 255, pp. 108–117, November 1987.
Cited on page 613.
606 Bill McGuire, A Guide to the End of the World: Everything you Never Wanted to Know,
Oxford University Press, 2002. On past disasters, see introduction by Tony Hall am, Cata-
strophes and Lesser Calamities – the Causes of Mass Extinctions, Oxford University Press,
2004. Cited on page 616.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 Intermezzo

THE BR AI N, L ANGUAGE AND THE

HUMAN C ONDI TION
Dvipsbugw

Alles was überhaupt gedacht werden kann,

“ kann klar gedacht werden.*

Ludwig Wittgenstein, Tractatus, 4.116

”
I
n our quest for increased precision in the description of all motion around us, it
s time to take a break, sit down and look back. In our walk so far, which has led us to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

nvestigate mechanics, general relativity and electrodynamics, we used several con-
cepts without defining them. Examples are ‘information’, ‘memory’, ‘measurement’, ‘set’,
‘number’, ‘infinity’, ‘existence’, ‘universe’ and ‘explanation’. Each of these is a common and
important term. In this intermezzo, we take a look at these concepts and try to give some
simple, but sufficiently precise definitions, keeping them as provocative and entertaining
Challenge 1098 e as possible. For example, can you explain to your parents what a concept is?
The reason for studying definitions is simple. We need the clarific-
ations in order to get to the top of Motion Mountain. Many have lost
their way because of lack of clear concepts. In this situation, physics
has a special guiding role. All sciences share one result: every type
of change observed in nature is a form of motion. In this sense, but
in this sense only, physics, focusing on motion itself, forms the basis
for all the other sciences. In other words, the search for the famed
‘theory of everything’ is an arrogant expression for the search for a
theory of motion. Even though the knowledge of motion is basic, its Ludwig Wittgenstein
precise description does not imply a description of ‘everything’: just
try to solve a marriage problem using the Schrödinger equation to note the difference.
Given the basic importance of motion, it is necessary that in physics all statements on
observations be as precise as possible. For this reason, many thinkers have investigated
physical statements with particular care, using all criteria imaginable. Physics is detailed
Copyright © Christoph Schiller November 1997–May 2006

prattle by curious people about moving things. The criteria for precision appear once we
Challenge 1099 e ask: which abilities does this prattle require? You might want to fill in the list yourself.
The abilities necessary for talking are a topic of research even today. The way that
the human species acquired the ability to chat about motion is studied by evolutionary
biologists. Child psychologists study how the ability develops in a single human being.
Physiologists, neurologists and computer scientists are concerned with the way the brain
and the senses make this possible; linguists focus on the properties of the language we use,

* ‘Everything that can be thought at all can be thought clearly.’ This and other quotes of Ludwig Wittgenstein
are from the equally short and famous Tractatus logico-philosophicus, written in 1918, first published in 1921;
it has now been translated into many other languages.

Dvipsbugw
 632 intermezzo: the brain, l anguage and the human condition

while logicians, mathematicians and philosophers of science study the general properties
of statements about nature. All these fields investigate tools that are essential for the devel-
opment of physics, the understanding of motion and the specification of the undefined
concepts listed above. The fields structure this intermezzo.

Evolution
A hen is only an egg’s way of making another

“ egg.
Samuel Butler, Life and Habit.

The evolution of the human species is the result of a long story that has been told in ”
Dvipsbugw

Ref. 607 many excellent books. A summarizing table on the history of the universe is given in the
Page 450 chapter on general relativity. The almost incredible chain of events that has lead to one’s
own existence includes the formation of atoms, of the galaxies, the stars, the planets, the
Moon, the atmosphere, the first cells, the water animals, the land animals, the mammals,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 608 the hominids, the humans, the ancestors, the family and finally oneself.
The way the particles we are made of moved during this sequence, being blown
through space, being collected on Earth, becoming organized to form people, is one of the
most awe-inspiring examples of motion. Remembering this fantastic sequence of motion
Challenge 1100 e every now and then can be an enriching experience.
In particular, without biological evolution, we would not be able to talk about motion;
only moving bodies can study moving bodies. Evolution was also the fount of childhood
and curiosity. In this intermezzo we will discover that most concepts of classical phys-
ics have already been introduced by little children, in the experiences they have while
growing up.

Children and physics

Physicists also have a shared reality. Other than

“ that, there isn’t really a lot of difference between

being a physicist and being a schizophrenic.
Richard Bandler

Ref. 609
During childhood, everybody is a physicist. When we follow our own memories back-
wards in time as far as we can, we reach a certain stage, situated before birth, which forms
the starting point of human experience. In that magic moment, we sensed somehow that
”
apart from ourselves, there is something else. The first observation we make about the
Copyright © Christoph Schiller November 1997–May 2006

world, during the time in the womb, is thus the recognition that we can distinguish two
parts: ourselves and the rest of the world. This distinction is an example – perhaps the
first – of a large number of ‘laws of nature’ that we stumble upon in our lifetime. By dis-
covering more and more distinctions we bring structure in the chaos of experience. We
quickly find out that the world is made of related parts, such as mama, papa, milk, Earth,
toys, etc.
Later, when we learn to speak, we enjoy using more difficult words and we call the
surroundings the environment. Depending on the context, we call the whole formed by
oneself and the environment together the (physical) world, the (physical) universe, nature,

Dvipsbugw
 children and physics 633

or the cosmos. These concepts are not distinguished from each other in this walk;* they
are all taken to designate the sum of all parts and their relations. They are simply taken
here to designate the whole.
The discovery of the first distinction starts a chain of similar discoveries. We extract
the numerous distinctions that are possible in the environment, in our own body and in
the various types of interactions between them. The ability to distinguish is the central
ability that allows us to change our view from that of the world as chaos, i.e. as a big
mess, to that of the world as a system, i.e. a structured set, in which parts are related in
Challenge 1101 n specific ways. (If you like precision, you may ponder whether the two choices of ‘chaos’ Dvipsbugw
and ‘system’ are the only possible ones. We will return to this issue in the third part of our
mountain ascent.)
In particular, the observation of the differences between oneself and the environment
goes hand in hand with the recognition that not only are we not independent of the en-
vironment, but we are firmly tied to it in various inescapable ways: we can fall, get hurt,
feel warm, cold, etc. Such relations are called interactions. Interactions express the obser-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

vation that even though the parts of nature can be distinguished, they cannot be isolated.
Page 676 In other words, interactions describe the difference between the whole and the sum of its
Challenge 1102 e parts. No part can be defined without its relation to its environment. (Do you agree?)
Interactions are not arbitrary; just take touch, smell or sight as examples. They differ
in reach, strength and consequences. We call the characteristic aspects of interactions
patterns of nature, or properties of nature, or rules of nature or, equivalently, with their
historical but unfortunate name, ‘laws’ of nature. The term ‘law’ stresses their general
validity; unfortunately, it also implies design, aim, coercion and punishment for infringe-
ment. However, no design, aim or coercion is implied in the properties of nature, nor is
infringement possible. The ambiguous term ‘law of nature’ was made popular by René
Descartes (1596–1650) and has been adopted enthusiastically because it gave weight to
the laws of the state – which were far from perfect at that time – and to those of other
organizations – which rarely are. The expression is an anthropomorphism coined by an
authoritarian world view, suggesting that nature is ‘governed’. We will therefore use the
term as rarely as possible in our walk and it will, if we do, be always between ‘ironical’
parentheses. Nature cannot be forced in any way. The ‘laws’ of nature are not obligations
for nature or its parts, they are obligations only for physicists and all other people: the
patterns of nature oblige us to use certain descriptions and to discard others. Whenever
one says that ‘laws govern nature’ one is talking nonsense; the correct expression is rules
describe nature.
Copyright © Christoph Schiller November 1997–May 2006

During childhood we learn to distinguish between interactions with the environment

(or perceptions): some are shared with others and called observations, others are uniquely
personal and are called sensations.** A still stricter criterion of ‘sharedness’ is used to
divide the world into ‘reality’ and ‘imagination’ (or ‘dreams’). Our walk will show that this

* The differences in usage can be deduced from their linguistic origins. ‘World’ is derived from old Germanic
‘wer’ – person – and ‘ald’ – old – and originally means ‘lifetime’. ‘Universe’ is from the Latin, and designates
the one – ‘unum’ – which one sees turning – ‘vertere’, and refers to the starred sky at night which turns
around the polar star. ‘Nature’ comes also from the Latin, and means ‘what is born’. ‘Cosmos’ is from Greek
κόσµος and originally means ‘order’.
** A child that is unable to make this distinction among perceptions – and who is thus unable to lie – almost
Ref. 610 surely develops or already suffers from autism, as recent psychological research has shown.

Dvipsbugw
 634 intermezzo: the brain, l anguage and the human condition

distinction is not essential, provided that we stay faithful to the quest for ever increasing
precision: we will find that the description of motion that we are looking for does not
depend on whether the world is ‘real’ or ‘imagined’, ‘personal’ or ‘public’.
Humans enjoy their ability to distinguish parts, which in other contexts they also call
details, aspects or entities, and enjoy their ability to associate them or to observe the re-
Ref. 611 lations between them. Humans call this activity classification. Colours, shapes, objects,
mother, places, people and ideas are some of the entities that humans discover first.
Our anatomy provides a handy tool to make efficient use of these discoveries: memory.
It stores a large amount of input that is called experience afterwards. Memory is a tool used Dvipsbugw
by both young and old children to organize their world and to achieve a certain security
in the chaos of life.
Memorized classifications are called concepts. Jean Piaget was the first researcher to
describe the influence of the environment on the concepts that a child forms. Step by
step, children learn that objects are localized in space, that space has three dimensions,
that objects fall, that collisions produce noise, etc. In particular, Piaget showed that space

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

and time are not a priori concepts, but result from the interactions of every child with its
environment.*
Around the time that a child goes to school, it starts to understand the idea of per-
manence of substances, e.g. liquids, and the concept of contrary. Only at that stage does
Ref. 613 its subjective experience becomes objective, with abstract comprehension. Still later, the
child’s description of the world stops to be animistic: before this step, the Sun, a brook or
a cloud are alive. In short, only after puberty does a human become ready for physics.

* An overview of the origin of developmental psychology is given by J.H. Flavell, The Developmental
Psychology of Jean Piaget, 1963. This work summarizes the observations by the French speaking Swiss Jean
Piaget (1896–1980), the central figure in the field. He was one of the first researchers to look at child devel-
opment in the same way that a physicist looks at nature: carefully observing, taking notes, making exper-
iments, extracting hypotheses, testing them, deducing theories. His astonishingly numerous publications,
based on his extensive observations, cover almost all stages of child development. His central contribution
is the detailed description of the stages of development of the cognitive abilities of humans. He showed that
all cognitive abilities of children, the formation of basic concepts, their way of thinking, their ability to talk,
etc., result from the continuous interaction between the child and the environment.
In particular, Piaget described the way in which children first learn that they are different from the ex-
ternal environment, and how they then learn about the physical properties of the world. Of his many books
related to physical concepts, two especially related to the topic of this walk are J. Piaget, Les notions de
mouvement et de vitesse chez l’enfant, Presses Universitaires de France, 1972 and Le developpement de la
Copyright © Christoph Schiller November 1997–May 2006

notion de temps chez l’enfant, Presses Universitaires de France, 1981, this last book being born from a sugges-
tion by Albert Einstein. These texts should be part of the reading of every physicist and science philosopher
interested in these questions.
Piaget also describes how in children the mathematical and verbal intelligence derives from sensomo-
torial, practical intelligence, which itself stems from habits and acquired associations to construct new con-
cepts. Practical intelligence requires the system of reflexes provided by the anatomical and morphological
structure of our organism. Thus his work shows in detail that our faculty for mathematical description of
the world is based, albeit indirectly, on the physical interaction of our organism with the world.
Some of his opinions on the importance of language in development are now being revised, notably
Ref. 612 through the rediscovery of the work of Lev Vigotsky, who argues that all higher mental abilities, emotions,
recollective memory, rational thought, voluntary attention and self-awareness, are not innate, but learned.
This learning takes place through language and culture, and in particular through the process of talking to
oneself.

Dvipsbugw
 children and physics 635

Even though everyone has been a physicist in their youth, most people remain classical
physicists. In this adventure we continue, using all the possibilities of a toy with which
nature provides us: the brain.
Experience is the name everyone gives to their

“ mistakes.
Oscar Wilde, Lady Windermere’s Fan.

”
Why a brain?
Dvipsbugw
Denken ist bereits Plastik.*
Ref. 614

“ Joseph Beuys, sculptor.

Numerous observations show that sense input is processed, i.e. classified, stored and re- ”
trieved in the brain. Notably, lesions of the brain can lead to the loss of part or all of these
Ref. 615 functions. Among the important consequences of these basic abilities of the brain are

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

thought and language. All such abilities result from the construction, from the ‘hardware’
Ref. 616 of the brain.
Systems with the ability to deduce classifications from the input they receive are called
Ref. 617 classifiers, and are said to be able to learn. Our brain shares this property with many com-
plex systems; the brain of many animals, but also certain computer algorithms, such as
the so-called ‘neural networks’, are examples of such classifiers. Such systems are studied
Ref. 619 in several fields, from biology to neurology, mathematics and computer science. Classi-
fiers have the double ability to discriminate and to associate; both are fundamental to
thinking.
Machine classifiers have a lot in common with the brain. As an example, following an
Ref. 618 important recent hypothesis in evolutionary biology, the necessity to cool the brain in
an effective way is responsible for the upright, bipedal walk of humans. The brain needs
a powerful cooling system to work well. In this it resembles modern computers, which
usually have powerful fans or even water cooling systems built into them. It turns out that
the human species has the most powerful cooling system of all mammals. An upright
posture allowed the air to cool the body most effectively in the tropical environment
where humans evolved. For even better cooling, humans have also no body hair, except
on their head, where it protects the brain from direct heating by the Sun.**
All classifiers are built from smallest classifying entities, sometimes large numbers of
them. Usually, the smallest units can classify input into only two different groups. The lar-
Copyright © Christoph Schiller November 1997–May 2006

ger the number of these entities, often called ‘neurons’ by analogy to the brain, the more
sophisticated classifications can be produced by the classifier.*** Classifiers thus work by
applying more or less sophisticated combinations of ‘same’ and ‘different’. The distinction
by a child of red and blue objects is such a classification; the distinction of compact and
non-compact gauge symmetry groups in quantum theory is a more elaborate classifica-
tion, but relies on the same fundamental ability.

* Thinking is already sculpture.

** The upright posture in turn allowed humans to take breath independently of their steps, a feat that many
animals cannot perform. This is turn allowed humans to develop speech. Speech in turn developed the brain.
*** A good introduction to neural nets is J. Hertz, A. Krogh & R. Palmer, Introduction to the Theory
of Neural Computation, Addison Wesley, 1991.

Dvipsbugw
 636 intermezzo: the brain, l anguage and the human condition

In all classifiers, the smallest classifying units interact with each other. Often these in-
teractions are channelled via connections, and the set is then called a network. In these
connections, signals are exchanged, via moving objects, such as electrons or photons.
Thus we arrive at the conclusion that the ability of the brain to classify the physical world,
for example to distinguish moving objects interacting with each other, is a consequence
of the fact that it itself consists of moving objects interacting with each other. Without
a powerful classifier, humans would not have become such a successful animal species.
And only the motion inside our brain allows us to talk about motion in general.
Numerous researchers are identifying the parts of the brain used when different intel- Dvipsbugw
lectual tasks are performed. The experiments become possible using magnetic resonance
Page 897 imaging and other methods. Other researchers are studying how thought processes can
be modelled from the brain structure. Neurology is still making regular progress. In par-
ticular, it is steadily destroying the belief that thinking is more than a physical process.
This belief results from personal fears, as you might want to test by introspection. It will
Challenge 1103 n disappear as time goes by. How would you argue that thought is just a physical process?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What is information?
These thoughts did not come in any verbal

“ formulation. I rarely think in words at all. A

thought comes, and I may try to express it in
words afterward.
Albert Einstein

We started by stating that studying physics means to talk about motion. To talk is to
transmit information. Can information be measured? Can we measure the progress of
physics in this way? Is the universe made of information?
”
Information is the result of classification. A classification is the answer to one or to
several yes–no questions. Such yes–no questions are the simplest classifications possible;
they provide the basic units of classification, from which all others can be built. The
simplest way to measure information is therefore to count the implied yes–no questions,
the bits, leading to it. Are you able to say how many bits are necessary to define the place
where you live? Obviously, the number of bits depends on the set of questions with which
we start; that could be the names of all streets in a city, the set of all coordinates on the
surface of the Earth, the names of all galaxies in the universe, the set of all letter combin-
Challenge 1104 n ations in the address. What is the most efficient method you can think of? A variation
of the combination method is used in computers. For example, the story of this walk re-
Copyright © Christoph Schiller November 1997–May 2006

quired about a thousand million bits. But since the amount of information in a normal
letter depends on the set of questions with which we start, it is impossible to define a
precise measure for information in this way.
The only way to measure information precisely is to take the largest possible set of
questions that can be asked about a system, and to compare it with what is known about
the system. In this case, the amount of unknown information is called entropy, a concept
Page 237 that we have already encountered. With this approach you should able to deduce yourself
Challenge 1105 n whether it is really possible to measure the advance of physics.
Since categorization is an activity of the brain and other, similar classifiers, informa-
tion as defined here is a concept that applies to the result of activities by people and by

Dvipsbugw
 children and physics 637

other classifiers. In short, information is produced when talking about the universe – the
universe itself is not the same as information. There is a growing number of publications
based on the opposite of this view; however, this is a conceptual short circuit. Any trans-
mission of information implies an interaction; physically speaking, this means that any
information needs energy for transmission and matter for storage. Without either of these,
there is no information. In other words, the universe, with its matter and energy, has to
exist before transmission of information is possible. Saying that the universe is made of
information is as meaningful as saying that it is made of toothpaste.
The aim of physics is to give a complete classification of all types and examples of mo- Dvipsbugw
tion, in other words, to know everything about motion. Is this possible? Or are you able
Challenge 1106 n to find an argument against this endeavour?

What is memory?
The brain is my second favorite organ.

“ Woody Allen

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Memory is the collection of records of perceptions. The production of such records is
the essential aspect of observation. Records can be stored in human memory, i.e. in the
brain, or in machine memory, as in computers, or in object memory, such as notes on
paper. Without memory, there is no science, no life – since life is based on the records
inside the DNA – and especially, no fun, as proven by the sad life of those who lose their
Ref. 615 memory.
Obviously every record is an object. But under which conditions does an object qualify
as a record? A signature can be the record of the agreement on a commercial transaction.
A single small dot of ink is not a record, because it could have appeared by mistake, for
example by an accidental blot. In contrast, it is improbable that ink should fall on paper
exactly in the shape of a signature. (The simple signatures of physicians are obviously
exceptions.) Simply speaking, a record is any object, which, in order to be copied, has
to be forged. More precisely, a record is an object or a situation that cannot arise nor
disappear by mistake or by chance. Our personal memories, be they images or voices,
have the same property; we can usually trust them, because they are so detailed that they
cannot have arisen by chance or by uncontrolled processes in our brain.
Can we estimate the probability for a record to appear or disappear by chance? Yes, we
can. Every record is made of a characteristic number N of small entities, for example the
number of the possible ink dots on paper, the number of iron crystals in a cassette tape,
Copyright © Christoph Schiller November 1997–May 2006

the electrons in a bit of computer memory, the silver iodide grains in a photographic
negative, etc. The chance disturbances in any memory are due to internal fluctuations,
also called noise. Noise makes the record unreadable; it can be dirt on a signature, thermal
magnetization changes in iron crystals, electromagnetic noise inside a solid state memory,
etc. Noise is found in all classifiers, since it is inherent in all interactions and thus in all
information processing.
It is a general property that internal fluctuations due to noise decrease when the size,
i.e. the number of components of the record is increased. In fact, the probability p mis for
Challenge 1107 ny a misreading or miswriting of a record changes as

Dvipsbugw
 638 intermezzo: the brain, l anguage and the human condition

p mis  1~N , (461)

where N is the number of particles or subsystems used for storing it. This relation appears
because, for large numbers, the so-called normal distribution is a good approximation of
almost any process; the width of the normal distribution, which determines the probab-
ility of record errors, grows less rapidly than its integral when the number of entities is
Challenge 1108 ny increased. (Are you able to confirm this?)
We conclude that any good record must be made from a large number of entities. The
larger the number, the less sensitive the memory is to fluctuations. Now, a system of large Dvipsbugw
size with small fluctuations is called a (physical) bath. Only baths make memories possible.
In other words, every record contains a bath. We conclude that any observation of a system
is the interaction of that system with a bath. This connection will be used several times in
the following, in particular in quantum theory. When a record is produced by a machine,
the ‘observation’ is usually called a (generalized) measurement. Are you able to specify the
Challenge 1109 n bath in the case of a person looking at a landscape?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

From the preceding discussion we can deduce a powerful conclusion: since we have
such a good memory at our disposition, we can deduce that we are made of many small
parts. And since records exist, the world must also be made of a large number of small
parts. No microscope of any kind is needed to confirm the existence of molecules or
similar small entities; such a tool is only needed to determine the sizes of these particles.
Their existence can be deduced simply from the observation that we have memory. (Of
course, another argument proving that matter is made of small parts is the ubiquity of
Page 222 noise.)
A second conclusion was popularized in the late 1920s by Leo Szilard. Writing a mem-
ory does not produce entropy; it is possible to store information into a memory without
increasing entropy. However, entropy is produced in every case that the memory is erased.
Challenge 1110 ny It turns out that the (minimum) entropy created by erasing one bit is given by

Sper erased bit = k ln 2 , (462)

and the number ln 2  0.69 is the natural logarithm of 2. Erasing thus on one hand re-
duces the disorder of the data – the local entropy–, but on the other hand increases the
total entropy. As is well known, energy is needed to reduce the entropy of a local system.
In short, any system that erases memory requires energy. For example, a logical AND gate
effectively erases one bit per operation. Logical thinking thus requires energy. It is also
Copyright © Christoph Schiller November 1997–May 2006

known that dreaming is connected with the erasing and reorganization of information.
Could that be the reason that, when we are very tired, without any energy left, we do not
Challenge 1111 n dream as much as usual?
Entropy is thus necessarily created when we forget. This is evident when we remind
Ref. 620 ourselves that forgetting is similar to the deterioration of an ancient manuscript. Entropy
increases when the manuscript is not readable any more, since the process is irreversible
and dissipative.* Another way to see this is to recognize that to clear a memory, e.g. a

Ref. 621 * As Wojciech Zurek clearly explains, the entropy created inside the memory is the main reason that even
Maxwell’s demon cannot reduce the entropy of two volumes of gases by opening a door between them in
such a way that fast molecules accumulate on one side and slow molecules accumulate on the other. (Maxwell

Dvipsbugw
 children and physics 639

magnetic tape, we have to put energy into it, and thus increase its entropy. Conversely,
writing into a memory can often reduce entropy; we remember that signals, the entities
that write memories, carry negative entropy. For example, the writing of magnetic tapes
usually reduces their entropy.

The capacity of the brain

Computers are boring. They can give only

“ answers.
(Wrongly) attributed to Pablo Picasso

The human brain is built in such a way that its fluctuations cannot destroy its contents. ”
Dvipsbugw

The brain is well protected by the skull for exactly this reason. In addition, the brain
literally grows connections, called synapses, between its various neurons, which are the
cells doing the signal processing. The neuron is the basic processing element of the brain,
performing the basic classification. It can only do two things: to fire and not to fire. (It

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is possible that the time at which a neuron fires also carries information; this question is
Ref. 623, Ref. 624 not yet settled.) The neuron fires depending on its input, which comes via the synapses
from hundreds of other neurons. A neuron is thus an element that can distinguish the
inputs it receives into two cases: those leading to firing and those that do not. Neurons
are thus classifiers of the simplest type, able only to distinguish between two situations.
Every time we store something in our long term memory, such as a phone number,
new synapses are grown or the connection strength of existing synapses is changed. The
connections between the neurons are much stronger than the fluctuations in the brain.
Only strong disturbances, such as a blocked blood vessel or a brain lesion, can destroy
neurons and lead to loss of memory.
As a whole, the brain provides an extremely efficient memory. Despite intense efforts,
engineers have not yet been able to build a memory with the capacity of the brain in
the same volume. Let us estimated this memory capacity. By multiplying the number of
neurons, about 1011 ,* by the average number of synapses per neuron, about 100, and also
by the estimated number of bits stored in every synapse, about 10, we arrive at a storage
capacity for the brain of about

M rewritable  1014 bit  104 GB . (463)

(One byte, abbreviated B, is the usual name for eight bits of information.) Note that evolu-
Copyright © Christoph Schiller November 1997–May 2006

tion has managed to put as many neurons in the brain as there are stars in the galaxy, and
that if we add all the synapse lengths, we get a total length of about 1011 m, which cor-
responds to the distance to from the Earth to the Sun. Our brain truly is astronomically
complex.
In practice, the capacity of the brain seems almost without limit, since the brain frees

had introduced the ‘demon’ in 1871, to clarify the limits posed by nature to the gods.) This is just another
way to rephrase the old result of Leo Szilard, who showed that the measurements by the demon create more
Ref. 622 entropy than they can save. And every measurement apparatus contains a memory.
To play being Maxwell’s demon, click on the http://www.wolfenet.com/~zeppelin/maxwell.htm website.
* The number of neurons seems to be constant, and fixed at birth. The growth of interconnections is highest
between age one and three, when it is said to reach up to 107 new connections per second.

Dvipsbugw
 640 intermezzo: the brain, l anguage and the human condition

memory every time it needs some new space, by forgetting older data, e.g. during sleep.
Note that this standard estimate of 1014 bits is not really correct! It assumes that the only
component storing information in the brain is the synapse strength. Therefore it only
measures the erasable storage capacity of the brain. In fact, information is also stored in
the structure of the brain, i.e. in the exact configuration in which every cell is connected
to other cells. Most of this structure is fixed at the age of about two years, but it continues
to develop at a lower level for the rest of human life. Assuming that for each of the N cells
»
with n connections there are f n connection possibilities, this write once capacity of the
brain can be estimated as roughly N f n f n log f n bits. For N = 1011 , n = 102 , f = 6, this
Dvipsbugw
Challenge 1112 ny
gives
M writeonce  1016 bit  106 GB . (464)

Ref. 625 Recent measurements confirmed that bilingual persons, especially early bilinguals, have
a higher density of grey mass in the small parietal cortex on the left hemisphere of the
brain. This is a region mainly concerned with language processing. The brain thus makes

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

also use of structural changes for optimized storage and processing.
Incidentally, even though the brains of sperm whales and of elephants can be five to
six times as heavy as those of humans, the number of neurons and connections, and thus
the capacity, is lower than for humans.
Sometimes it is claimed that people use only between 5% or 10% of their brain capacity.
This myth, which goes back to the nineteenth century, would imply that it is possible
to measure the actual data stored in the brain and compare it with its capacity to an
impossible accuracy. Alternatively, the myth implies that the processing capacity can be
measured. It also implies that nature would develop and maintain an organ with 90 %
overcapacity, wasting all the energy and material to build, repair and maintain it. The
myth is wrong.
The large storage capacity of the brain also shows that human memory is filled by the
environment and is not inborn: one human ovule plus one sperm have a mass of about
1 mg, which corresponds to about 3ë1016 atoms. Obviously, fluctuations make it impossible
to store 1016 bits in it. In fact, nature stores only about 3 ë 109 bits in the genes of an ovule,
using 107 atoms per bit. In contrast, a typical brain has a mass of 1.5 to 2 kg, containing
about 5 to 7 ë 1025 atoms, which makes it as efficient as the ovule. The difference between
the number of bits in human DNA and those in the brain nicely shows that almost all
information stored in the brain is taken from the environment; it cannot be of genetic
origin, even allowing for smart decompression of stored information.
Copyright © Christoph Schiller November 1997–May 2006

In total, all these tricks used by nature result in the most powerful classifier yet
known.* Are there any limits to the brain’s capacity to memorize and to classify? With
the tools that humans have developed to expand the possibilities of the brain, such as
paper, writing and printing to help memory, and the numerous tools available to sim-
plify and to abbreviate classifications explored by mathematicians, brain classification is
Ref. 626 only limited by the time spent practising it.Without tools, there are strict limits, of course.
The two-millimetre thick cerebral cortex of humans has a surface of about four sheets of
A4 paper, a chimpanzee’s yields one sheet and a monkey’s is the size of a postcard. It is

* Also the power consumption of the brain is important: even though it contains only about 2% of the body’s
mass, is uses 25% of the energy taken in by food.

Dvipsbugw
what is l anguage? 641

estimated that the total intellectually accessible memory is of the order of

M intellectual  1 GB , (465)

though with a large experimental error.

The brain is also unparalleled in its processing capacity. This is most clearly demon-
strated by the most important consequence deriving from memory and classification:
thought and language. Indeed, the many types of thinking or language we use, such
as comparing, distinguishing, remembering, recognizing, connecting, describing, dedu- Dvipsbugw
cing, explaining, imagining, etc., all describe different ways to classify memories or per-
ceptions. In the end all thinking and talking directly or indirectly classify observations.
But how far are computers from achieving this! To talk to a computer program, such as
to the famous program Eliza or its successors that mimic a psychoanalyst, is still a disap-
pointing experience. To understand the reasons for this slow development, we ask:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What is l anguage?
Reserve your right to think, for even to think

“ wrongly is better than not to think at all.

Hypatia of Alexandria
Ein Satz kann nur sagen, wie ein Ding ist, nicht
”
“ was es ist.*
Ludwig Wittgenstein, Tractatus, 3.221

Language possibly is the most wonderful gift of human nature. Using the ability to pro- ”
duce sounds and to put ink on paper, people attach certain symbols,** also called words or
terms in this context, to the many partitions they specify with the help of their thinking.
Such a categorization is then said to define a concept or notion, and is set in italic typeface
in this text. A standard set of concepts forms a language.*** In other words, a (human)
language is a standard way of symbolic interaction between people.**** There are human
languages based on facial expressions, on gestures, on spoken words, on whistles, on writ-
ten words, and more. The use of spoken language is considerably younger than the human
species; it seems that it appeared only about two hundred thousand years ago. Written lan-
guage is even younger, namely only about six thousand years old. But the set of concepts
* Propositions can only say how things are, not what they are.
** A symbol is a type of sign, i.e. an entity associated by some convention to the object it refers. Following
Copyright © Christoph Schiller November 1997–May 2006

Charles Peirce (1839–1914) – see http://www.peirce.org – the most original philosopher born in the United
States, a symbol differs from an icon (or image) and from an index, which are also attached to objects by
convention, in that it does not resemble the object, as does an icon, and in that it has no contact with the
object, as is the case for an index.
*** The recognition that language is based on a partition of ideas, using the various differences between
them to distinguish them from each other, goes back to the Swiss thinker Ferdinand de Saussure (1857–
1913), who is regarded as the founder of linguistics. His textbook Cours de linguistique générale, Editions
Payot, 1985, has been the reference work of the field for over half a century. Note that Saussure, in contrast
to Peirce, prefers the term ‘sign’ to ‘symbol’, and that his definition of the term ‘sign’ includes also the object
to which it refers.
**** For slightly different definitions and a wealth of other interesting information about language, see the
beautiful book by David Crystal, The Cambridge Encyclopedia of Language, Cambridge University Press,
1987.

Dvipsbugw
 642 intermezzo: the brain, l anguage and the human condition

used, the vocabulary, is still expanding. For humans, the understanding of language be-
gins soon after birth (perhaps even before), the active use begins at around a year of age,
the ability to read can start as early as two, and personal vocabulary continues to grow as
long as curiosity is alive.
Physics being a lazy way to chat about motion, it needs language as an essential tool.
Of the many aspects of language, from literature to poetry, from jokes to military orders,
from expressions of encouragement, dreams, love and emotions, physics uses only a small
and rather special segment. This segment is defined by the inherent restriction to talk
about motion. Since motion is an observation, i.e. an interaction with the environment Dvipsbugw
that several people experience in the same way, this choice puts a number of restrictions
on the contents – the vocabulary – and on the form – the grammar – of such discussions.
For example, from the definition that observations are shared by others, we get the
requirement that the statements describing them must be translatable into all languages.
But when can a statement be translated? On this question two extreme points of view are
possible: the first maintains that all statements can be translated, since it follows from the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

properties of human languages that each of them can express every possible statement.
In this view, only sign systems that allow one to express the complete spectrum of human
messages form a human language. This property distinguishes spoken and sign language
from animal languages, such as the signs used by apes, birds or honey bees, and also from
computer languages, such as Pascal or C. With this meaning of language, all statements
can be translated by definition.
It is more challenging for a discussion to follow the opposing view, namely that precise
translation is possible only for those statements which use terms, word types and gram-
matical structures found in all languages. Linguistic research has invested considerable
effort in the distillation of phonological, grammatical and semantic universals, as they are
called, from the 7000 or so languages thought to exist today.*
The investigations into the phonological aspect, which showed for example that every
language has at least two consonants and two vowels, does not provide any material for
the discussion of translation.** Studying the grammatical (or syntactic) aspect, one finds
that all languages use smallest elements, called ‘words’, which they group into sentences.
They all have pronouns for the first and second person, ‘I’ and ‘you’, and always contain
nouns and verbs. All languages use subjects and predicates or, as one usually says, the three
entities subject, verb and object, though not always in this order. Just check the languages
Challenge 1113 e you know.
On the semantic aspect, the long list of lexical universals, i.e. words that appear in all
Copyright © Christoph Schiller November 1997–May 2006

languages, such as ‘mother’ or ‘Sun’, has recently been given a structure. The linguist Anna
Wierzbicka performed a search for the building blocks from which all concepts can be

* A comprehensive list with 6 800 languages (and with 41 000 language and dialect names) can be found on
the world wide website by Barbara Grimes, Ethnologue – Languages of the World, to be found at the address
http://www.ethnologue.com or in the printed book of the same name.
It is estimated that 15 000  5 000 languages have existed in the past.
Nevertheless, in today’s world, and surely in the sciences, it is often sufficient to know one’s own language
plus English. Since English is the language with the largest number of words, learning it well is a greater
Ref. 627 challenge than learning most other languages.
** Studies explore topics such as the observation that in many languages the word for ‘little’ contains an ‘i’
(or high pitched ‘e’) sound: petit, piccolo, klein, tiny, pequeño, chiisai; exceptions are: small, parvus.

Dvipsbugw
 what is l anguage? 643

TA B L E 52 The semantic primitives, following Anna Wierzbicka

I, you, someone, something, people [substantives]

this, the same, one, two, all, much/many [determiners and quantifiers]
know, want, think, feel, say [mental predicates]
do, happen [agent, patient]
good, bad [evaluative]
big, small [descriptors]
very [intensifier] Dvipsbugw
can, if (would) [modality, irrealis]
because [causation]
no (not) [negation]
when, where, after (before), under (above) [time and place]
kind of, part of [taxonomy, partonomy]
like [hedge/prototype]

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

built. She looked for the definition of every concept with the help of simpler ones, and
continued doing so until a fundamental level was reached that cannot be further reduced.
The set of concepts that are left over are the primitives. By repeating this exercise in many
languages, Wierzbicka found that the list is the same in all cases. She thus had discovered
universal semantic primitives. In November 1992, the list contained the terms given in
Table 52.
Following the life-long research of Anna Wierzbicka and her research school, all these
concepts exist in all languages of the world studied so far.* They have defined the mean-
ing of each primitive in detail, performed consistency checks and eliminated alternative
approaches. They have checked this list in languages from all language groups, in lan-
guages from all continents, thus showing that the result is valid everywhere. In every
Ref. 628 language all other concepts can be defined with the help of the semantic primitives.
Simply stated, learning to speak means learning these basic terms, learning how to
combine them and learning the names of these composites. The definition of language
given above, namely as a means of communication that allows one to express everything
one wants to say, can thus be refined: a human language is any set of concepts that includes
the universal semantic primitives.
For physicists – who aim to talk in as few words as possible – the list of semantic
Copyright © Christoph Schiller November 1997–May 2006

primitives has three facets. First, the approach is appealing, as it is similar to physics’
own aim: the idea of primitives gives a structured summary of everything that can be
said, just as the atomic elements structure all objects that can be observed. Second, the
list of primitives can be structured. In fact, the list of primitives can be divided into two
groups: one group contains all terms describing motion (do, happen, when, where, feel,

* It is easy to imagine that this research steps on the toes of many people. A list that maintains that ‘true’,
‘good’, ‘creation’, ‘life’, ‘mother’ or ‘god’ are composite will elicit violent reactions, despite the correctness
of the statements. Indeed, some of these terms were added in the 1996 list, which is somewhat longer. In
addition, a list that maintains that we only have about thirty basic concepts in our heads is taken by many
to be offensive.

Dvipsbugw
 644 intermezzo: the brain, l anguage and the human condition

small, etc. – probably a term from the semantic field around light or colour should be
added) and the other group contains all terms necessary to talk about abstract sets and
relations (this, all, kind of, no, if, etc.). Even for linguistics, aspects of motion and logical
concepts are the basic entities of human experience and human thinking. To bring the
issue to a point, the semantic primitives contain the basic elements of physics and the
basic elements of mathematics. All humans are thus both physicists and mathematicians.
The third point is that the list of primitives is too long. The division of the list into two
groups directly suggests shorter lists; we just have to ask physicists and mathematicians
for concise summaries of their respective fields. To appreciate this aim, try to define what Dvipsbugw
Challenge 1114 d ‘if ’ means, or what an ‘opposite’ is – and explore your own ways of reducing the list.
Reducing the list of primitives is also one of our aims in this adventure. We will explore
the mathematical group of primitives in this intermezzo; the physical group will occupy
us in the rest of our adventure. However, a shorter list of primitives is not sufficient. Our
goal is to arrive at a list consisting of only one basic concept. Reaching this goal is not
simple, though. First, we need to check whether the set of classical physical concepts

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that we have discovered so far is complete. Can classical physical concepts describe all
observations? The second part of our adventure is devoted to this question. The second
task is to reduce the list. This task is not straightforward; we have already discovered that
physics is based on a circular definition: in Galilean physics, space and time are defined
Page 157 using matter, and matter is defined using space and time. We will need quite some effort
to overcome this obstacle. The third part of this text tells the precise story. After numerous
adventures we will indeed discover a basic concept on which all other concepts are based.
We can summarize all the above-mentioned results of linguistics in the following way.
By constructing a statement made only of subject, verb and object, consisting only of
nouns and verbs, using only concepts built from the semantic primitives, we are sure that
it can be translated into all languages. This explains why physics textbooks are often so
boring: the authors are often too afraid to depart from this basic scheme. On the other
hand, research has shown that such straightforward statements are not restrictive: with
them one can say everything that can be said.
Every word was once a poem.

What is a concept?
“ Ralph Waldo Emerson*

”
Concepts are merely the results, rendered

“
Copyright © Christoph Schiller November 1997–May 2006

permanent by language, of a previous process of

comparison.
William Hamilton

There is a group of people that has taken the strict view on translation and on precision
to the extreme. They build all concepts from an even smaller set of primitives, namely
only two: ‘set’ and ‘relation’, and explore the various possible combinations of these two
”
concepts, studying their classifications. Step by step, this radical group, commonly called
mathematicians, came to define with full precision concepts such as numbers, points,

* Ralph Waldo Emerson (1803–1882), US-American essayist and philosopher.

Dvipsbugw
 what is l anguage? 645

curves, equations, symmetry groups and more. The construction of these concepts is sum-
marized partly in the following and partly in Appendix D.
However, despite their precision, in fact precisely because of it, no mathematical
concept talks about nature or about observations.* Therefore the study of motion needs
other, more useful concepts. What properties must a useful concept have? For example,
what is ‘freedom’ or what is a ‘parachute’? Obviously, a useful concept implies a list of its
parts, its aspects and their internal relations, as well as their relation to the exterior world.
Thinkers in various fields, from philosophy to politics, agree that the definition of any
concept requires: Dvipsbugw

— explicit and fixed content,

— explicit and fixed limits,
— explicit and fixed domain of application.
The inability to state these properties or keep them fixed is often the easiest way to distin-
guish crackpots from more reliable thinkers. Unclearly defined terms, which thus do not

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

qualify as concepts, regularly appear in myths, e.g. ‘dragon’ or ‘sphinx’, or in ideologies,
e.g. ‘worker’ or ‘soul’. Even physics is not immune. For example, we will discover later that
Challenge 1115 n neither ‘universe’ nor ‘creation’ are concepts. Are you able to argue the case?
But the three defining properties of any concepts are interesting in their own right.
Explicit content means that concepts are built one onto another. In particular, the most
fundamental concepts appear to be those that have no parts and no external relations, but
Challenge 1116 n only internal ones. Can you think of one? Only the last part of this walk will uncover the
final word on the topic.
The requirements of explicit limits and explicit contents also imply that all concepts
describing nature are sets, since sets obey the same requirements. In addition, explicit
domains of application imply that all concepts also are relations.** Since mathematics is
based on the concepts of ‘set’ and of ‘relation’, one follows directly that mathematics can
provide the form for any concept, especially whenever high precision is required, as in the
study of motion. Obviously, the content of the description is only provided by the study
of nature itself; only then do concepts become useful.
In the case of physics, the search for sufficiently precise concepts can be seen as the
single theme structuring the long history of the field. Regularly, new concepts have been
proposed, explored in all their properties, and tested. Finally, concepts are rejected or
adopted, in the same way that children reject or adopt a new toy. Children do this un-
consciously, scientists do it consciously, using language.*** For this reason, concepts are
Copyright © Christoph Schiller November 1997–May 2006

universally intelligible.
* Insofar as one can say that mathematics is based on the concepts of ‘set’ and ‘relation’, which are based on
experience, one can say that mathematics explores a section of reality, and that its concepts are derived from
experience. This and similar views of mathematics are called platonism. More concretely, platonism is the
view that the concepts of mathematics exist independently of people, and that they are discovered, and not
created, by mathematicians.
In short, since mathematics makes use of the brain, which is a physical system, actually mathematics is
applied physics.
** We see that every physical concept is an example of a (mathematical) category, i.e. a combination of objects
and mappings. For more details about categories, with a precise definition of the term, see page 650.
*** Concepts formed unconsciously in our early youth are the most difficult to define precisely, i.e. with
language. Some who were unable to define them, such as the Prussian philosopher Immanuel Kant (1724–

Dvipsbugw
646 intermezzo: the brain, l anguage and the human condition

Note that the concept ‘concept’ itself is not definable independently of experience; a
concept is something that helps us to act and react to the world in which we live. Moreover,
concepts do not live in a world separate from the physical one: every concept requires
memory from its user, since the user has to remember the way in which it was formed;
therefore every concept needs a material support for its use and application. Thus all
thinking and all science is fundamentally based on experience.
In conclusion, all concepts are based on the idea that nature is made of related parts.
This idea leads to complementing couples such as ‘noun–verb’ in linguistics, ‘set–relation’
or ‘definition–theorem’ in mathematics, and ‘aspect of nature–pattern of nature’ in phys- Dvipsbugw
ics. These couples constantly guide human thinking, from childhood onwards, as devel-
opmental psychology can testify.

What are sets? What are relations?

Alles, was wir sehen, könnte auch anders sein.

“ Alles, was wir überhaupt beschreiben können,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

könnte auch anders sein. Es gibt keine Ordnung
der Dinge a priori.*
Ludwig Wittgenstein, Tractatus, 5.634

Defining sets and defining relations are the two fundamental acts of our thinking. This
can be seen most clearly in any book about mathematics; such a book is usually divided
into paragraphs labelled ‘definition’, ‘theorem’, ‘lemma’ and ‘corollary’. The first type of
”
paragraph defines concepts, i.e. defines sets, and the other three types of paragraphs ex-
press relations, i.e. connections between these sets. Mathematics is thus the exploration
of the possible symbolic concepts and their relations. Mathematics is the science of sym-
bolic necessities.
Sets and relations are tools of classification; that is why they are also the tools of any
bureaucrat. (See Figure 280.) This class of humans is characterized by heavy use of pa-
per clips, files, metal closets, archives – which all define various types of sets – and by
the extensive use of numbers, such as reference numbers, customer numbers, passport
numbers, account numbers, law article numbers – which define various types of relations
between the items, i.e. between the elements of the sets.
Both the concepts of set and of relation express, in different ways, the fact that nature
can be described, i.e. that it can be classified into parts that form a whole. The act of group-
ing together aspects of experience, i.e. the act of classifying them, is expressed in formal
Copyright © Christoph Schiller November 1997–May 2006

language by saying that a set is defined. In other words, a set is a collection of elements
of our thinking. Every set distinguishes the elements from each other and from the set
itself. This definition of ‘set’ is called the naive definition. For physics, the definition is
sufficient, but you won’t find many who will admit this. In fact, mathematicians have re-
fined the definition of the concept ‘set’ several times, because the naive definition does

1804) used to call them ‘a priori’ concepts (such as ‘space’ and ‘time’) to contrast them with the more clearly
defined ‘a posteriori’ concepts. Today, this distinction has been shown to be unfounded both by the study
of child psychology (see the footnote on page 634) and by physics itself, so that these qualifiers are therefore
not used in our walk.
* Whatever we see could be other than it is. Whatever we can describe at all could be other than it is. There
is no a priori order of things.

Dvipsbugw
 what is l anguage? 647

TA B L E 53 The defining properties of a set – the ZFC axioms

Th e ax i om s of Z e r m e l o –Fr a e n k e l –C set t h e ory

– Two sets are equal if and only if they have the same elements. (Axiom of extensionality)
– The empty set is a set. (Axiom of the null set)
– If x and y are sets, then the unordered pair ˜x, y is a set. (Axiom of unordered pairs)
– If x is a set of sets, the union of all its members is a set. (Union or sum set axiom)
– The entity ˜g, ˜g, ˜˜g, ˜˜˜g, ... is a set a – in other words, infinite collections, such
as the natural numbers, are sets. (Axiom of infinity) Dvipsbugw
– An entity defined by all elements having a given property is a set, provided this property is reas-
onable; some important technicalities defining ‘reasonable’ are necessary. (Axiom of separation)
– If the domain of a function is a set, so is its range. (Axiom of replacement)
– The entity y of all subsets of x is also a set, called the power set. (Axiom of the power set)
– A set is not an element of itself – plus some technicalities. (Axiom of regularity)
– The product of a family of non-empty sets is non-empty. Equivalently, picking elements from a
list of sets allows one to construct a new set – plus technicalities. (Axiom of choice)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

a. The more common formulation (though equivalent to the above) is: The entity
˜g, ˜g, ˜g, ˜g, ˜g, ˜g, ˜g, ˜g, ... is a set.

not work well for infinite sets. A famous example is the story about sets which do not con-
tain themselves. Obviously, any set is of two sorts: either it contains itself or it does not.
Challenge 1117 n If we take the set of all sets that do not contain themselves, to which sort does it belong?
To avoid problems with the concept of ‘set’,
mathematics requires a precise definition. The
first such definition was given by the German
mathematician Ernst Zermelo (b. 1871 Berlin,
d. 1951 Freiburg i.B.) and the German–Israeli
mathematician Adolf/Abraham Fraenkel (b. 1891
München, d. 1965 Jerusalem). Later, the so-called F I G U R E 280 Devices for the definition
axiom of choice was added, in order to make it of sets (left) and of relations (right)
possible to manipulate a wider class of infinite sets.
The result of these efforts is called the ZFC definition.* From this basic definition we
can construct all mathematical concepts used in physics. From a practical point of view,
it is sufficient to keep in mind that for the whole of physics, the naive definition of a
Copyright © Christoph Schiller November 1997–May 2006

set is equivalent to the precise ZFC definition, actually even to the simper ZF definition.
Subtleties appear only for some special types of infinite sets, but these are not used in
physics. In short, from the basic, naive set definition we can construct all concepts used

* A global overview of axiomatic set theory is given by Paul J. Cohen & Reuben Hersch, Non-
Cantorian set theory, Scientific American 217, pp. 104–116, 1967. Those were the times when Scientific Amer-
ican was a quality magazine.
Ref. 629 Other types of entities, more general than standard sets, obeying other properties, can also be defined,
and are also subject of (comparatively little) mathematical research. To find an example, see the section
Page 649 on cardinals later on. Such more general entities are called classes whenever they contain at least one set.
Challenge 1118 n Can you give an example? In the third part of our mountain ascent we will meet physical concepts that are
described neither by sets nor by classes, containing no set at all. That is were the real fun starts.

Dvipsbugw
 648 intermezzo: the brain, l anguage and the human condition

in physics.
Ref. 630 The naive set definition is far from boring. To satisfy two people when dividing a cake,
we follow the rule: I cut, you choose. The method has two properties: it is just, as every-
body thinks that they have the share that they deserve, and it is fully satisfying, as every-
body has the feeling that they have at least as much as the other. What rule is needed for
Challenge 1119 d three people? And for four?
Apart from defining sets, every child and every brain creates links between the dif-
ferent aspects of experience. For example, when it hears a voice, it automatically makes
the connection that a human is present. In formal language, connections of this type are Dvipsbugw
called relations. Relations connect and differentiate elements along other lines than sets:
the two form a complementing couple. Defining a set unifies many objects and at the
same time divides them into two: those belonging to the set and those that do not; defin-
ing a (binary) relation unifies elements two by two and divides them into many, namely
into the many couples it defines.
Sets and relations are closely interrelated concepts. Indeed, one can define (mathem-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

atical) relations with the help of sets. A (binary) relation between two sets X and Y is a
subset of the product set, where the product set or Cartesian product X  Y is the set of all
ordered pairs (x, y) with x > X and y > Y. An ordered pair (x, y) can easily be defined
Challenge 1120 n with the help of sets. Can you find out how? For example, in the case of the relation ‘is
wife of ’, the set X is the set of all women and the set Y that of all men; the relation is given
by the list all the appropriate ordered pairs, which is much smaller than the product set,
i.e. the set of all possible woman–man combinations.
It should be noted that the definition of relation just given is not really complete, since
every construction of the concept ‘set’ already contains certain relations, such as the rela-
tion ‘is element of.’ It does not seem to be possible to reduce either one of the concepts ‘set’
or ‘relation’ completely to the other one. This situation is reflected in the physical cases
of sets and relations, such as space (as a set of points) and distance, which also seem
impossible to separate completely from each other. In other words, even though math-
ematics does not pertain to nature, its two basic concepts, sets and relations, are taken
from nature. In addition, the two concepts, like those of space-time and particles, are
each defined with the other.

Infinity
Mathematicians soon discovered that the concept of ‘set’ is only useful if one can also
call collections such as ˜0, 1, 2, 3..., i.e. of the number 0 and all its successors, a ‘set’. To
Copyright © Christoph Schiller November 1997–May 2006

achieve this, one property in the Zermelo–Fraenkel list defining the term ‘set’ explicitly
specifies that this collection can be called a set. (In fact, also the axiom of replacement
states that sets may be infinite.) Infinity is thus put into mathematics and into the tools
of our thought right at the very beginning, in the definition of the term ‘set’. When de-
scribing nature, with or without mathematics, we should never forget this fact. A few
additional points about infinity should be of general knowledge to any expert on motion.
Only sets can be infinite. And sets have parts, namely their elements. When a thing or
a concept is called ‘infinite’ one can always ask and specify what its parts are: for space
the parts are the points, for time the instants, for the set of integers the integers, etc. An

Dvipsbugw
 what is l anguage? 649

indivisible or a finitely divisible entity cannot be called infinite.*

A set is infinite if there is a function from it into itself that is injective (i.e. different
elements map to different results) but not onto (i.e. some elements do not appear as images
of the map); e.g. the map n ( 2n shows that the set of integers is infinite. Infinity also
can be checked in another way: a set is infinite if it remains so also after removing one
element, even repeatedly. We just need to remember that the empty set is finite.
There are many types of infinities, all of different sizes.** This important result was dis-
covered by the Danish-Russian-German mathematician Georg Cantor (1845–1918). He
showed that from the countable set of natural numbers one can construct other infinite Dvipsbugw
sets which are not countable. He did this by showing that the power set P(ω), namely the
set of all subsets, of a countably infinite set is infinite, but not countably infinite. Sloppily
speaking, the power set is ‘more infinite’ than the original set. The real numbers R, to be
defined shortly, are an example of an uncountably infinite set; there are many more of
Challenge 1121 n them than there are natural numbers. (Can you show this?) However, any type of infinite
set contains at least one subset which is countably infinite.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Even for an infinite set one can define size as the number of its elements. Cantor called
this the cardinality of a set. The cardinality of a finite set is simply given by the number
of its elements. The cardinality of a power set is 2 exponentiated by the cardinality of the
set. The cardinality of the set of integers is called ℵ0 , pronounced ‘aleph zero’, after the
Page 1146 first letter of the Hebrew alphabet. The smallest uncountable cardinal is called ℵ1 . The
next cardinal is called ℵ2 etc. A whole branch of mathematics is concerned with the ma-
nipulation of these infinite ‘numbers’; addition, multiplication, exponentiation are easily
defined. For some of them, even logarithms and other functions make sense.***
The cardinals defined in this way, including ℵn , ℵω , ℵℵℵ are called accessible, be-
cause since Cantor, people have defined even larger types of infinities, called inaccess-
ible. These numbers (inaccessible cardinals, measurable cardinals, supercompact cardin-
als, etc.) need additional set axioms, extending the ZFC system. Like the ordinals and the
cardinals, they form examples of what are called transfinite numbers.
The real numbers have the cardinality of the power set of the integers, namely 2ℵ0 .
Challenge 1122 n Can you show this? The result leads to the famous question: Is ℵ1 = 2ℵ0 or not? The
statement that this be so is called the continuum hypothesis and was unproven for several
generations. The surprising answer came in 1963: the usual definition of the concept of
Ref. 631 set is not specific enough to fix the answer. By specifying the concept of set in more detail,
with additional axioms – remember that axioms are defining properties – you can make
the continuum hypothesis come out either right or wrong, as you prefer.
Copyright © Christoph Schiller November 1997–May 2006

Another result of research into transfinites is important: for every definition of a type
of infinite cardinal, it seems to be possible to find a larger one. In everyday life, the idea
of infinity is often used to stop discussions about size: ‘My big brother is stronger than
yours.’ ‘But mine is infinitely stronger than yours!’ Mathematics has shown that questions
on size do continue afterwards: ‘The strength of my brother is the power set of that of

* Therefore, most gods, being concepts and thus sets, are either finite or, in the case where they are infinite,
they are divisible. It seems that only polytheistic world views are not disturbed by this conclusion.
** In fact, there is such a huge number of types of infinities that none of these infinities itself actually de-
scribes this number. Technically speaking, there are as many infinities as there are ordinals.
*** Many results are summarized in the excellent and delightful paperback by Rudy Rucker, Infinity and
the Mind – the Science and Philosophy of the Infinite, Bantam, Toronto, 1983.

Dvipsbugw
 650 intermezzo: the brain, l anguage and the human condition

Ref. 632 yours!’ Rucker reports that mathematicians conjecture that there is no possible nor any
conceivable end to these discussions.
For physicists, a simple question appears directly. Do infinite quantities exist in nature?
Or better, is it necessary to use infinite quantities to describe nature? You might want to
Challenge 1123 e clarify your own opinion on the issue. It will be settled during the rest of our adventure.

Functions and structures

Which relations are useful to describe patterns in nature? A typical example is ‘larger Dvipsbugw
stones are heavier’. Such a relation is of a specific type: it relates one specific value of an
observable ‘volume’ to one specific value of the observable ‘weight’. Such a one-to-one rela-
tion is called a (mathematical) function or mapping. Functions are the most specific types
of relations; thus they convey a maximum of information. In the same way as numbers
are used for observables, functions allow easy and precise communication of relations
between observations. All physical rules and ‘laws’ are therefore expressed with the help

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of functions and, since physical ‘laws’ are about measurements, functions of numbers are
their main building blocks.
A function f , or mapping, is a thus binary relation, i.e. a set f = ˜(x, y) of ordered
pairs, where for every value of the first element x, called the argument, there is only one
pair (x, y). The second element y is called the value of the function at the argument x.
The set X of all arguments x is called the domain of definition and the set Y of all second
arguments y is called the range of the function. Instead of f = ˜(x, y) one writes

f X Y and f x(y or y = f (x) , (466)

where the type of arrow – with initial bar or not – shows whether we are speaking about
sets or about elements.
We note that it is also possible to use the couple ‘set’ and ‘mapping’ to define all math-
ematical concepts; in this case a relation is defined with the help of mappings. A modern
school of mathematical thought formalized this approach by the use of (mathematical)
categories, a concept that includes both sets and mappings on an equal footing in its defin-
ition.*
To think and talk more clearly about nature, we need to define more specialized con-
cepts than sets, relations and functions, because these basic terms are too general. The
most important concepts derived from them are operations, algebraic structures and
Copyright © Christoph Schiller November 1997–May 2006

numbers.
A (binary) operation is a function that maps the Cartesian product of two copies of a
set X into itself. In other words, an operation w takes an ordered couple of arguments

* A category is defined as a collection of objects and a collection of ‘morphisms’, or mappings. Morphisms

can be composed; the composition is associative and there is an identity morphism. The strange world of cat-
egory theory, sometimes called the abstraction of all abstractions, is presented in F. William L aw vere
& Stephen H. S chanuel, Conceptual Mathematics: a First Introduction to Categories, Cambridge Uni-
versity Press, 1997.
Note that every category contains a set; since it is unclear whether nature contains sets, as we will discuss
on page 681, it is questionable whether categories will be useful in the unification of physics, despite their
intense and abstract charm.

Dvipsbugw
 what is l anguage? 651

x > X and assigns to it a value y > X:

wXX X and w  (x, x) ( y . (467)

Challenge 1124 n Is division of numbers an operation in the sense just defined?

Now we are ready to define the first of three basic concepts of mathematics. An al-
gebraic structure, also called an algebraic system, is (in the most restricted sense) a set
together with certain operations. The most important algebraic structures appearing in
Page 1204 physics are groups, vector spaces, and algebras. Dvipsbugw
In addition to algebraic structures, mathematics is based on order structures and on
topological structures. Order structures are building blocks of numbers and necessary
to define comparisons of any sort. Topological structures are built, via subsets, on the
concept of neighbourhood. They are necessary to define continuity, limits, dimensional-
Page 1213 ity, topological spaces and manifolds.
Obviously, most mathematical structures are combinations of various examples of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

these three basic structure types. For example, the system of real numbers is given by the
set of real numbers with the operations of addition and multiplication, the order relation
‘is larger than’ and a continuity property. They are thus built by combining an algebraic
Ref. 633 structure, an order structure and a topological structure. Let us delve a bit into the details.

Numbers
Which numbers are multiplied by six when

Challenge 1125 n
“ ”their last digit is taken away and transferred to
the front?

Numbers are the oldest mathematical concept and are found in all cultures. The notion
of number, in Greek ἀριθµός, has been changed several times. Each time the aim was to
include wider classes of objects, but always retaining the general idea that numbers are
entities that can be added, subtracted, multiplied and divided.
The modern way to write numbers, as e.g. in 12 345 679 ë 45 = 666 666 666, is essential
for science.* It can be argued that the lack of a good system for writing down and for
calculating with numbers delayed the progress of science by several centuries. (By the
way, the same can be said for the affordable mass reproduction of written texts.)
The simplest numbers, 0, 1, 2, 3, 4, ..., are usually seen as being taken directly from ex-
perience. However, they can also be constructed from the notions of ‘relation’ and ‘set’.
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1126 n One of the many possible ways to do this (can you find another?) is by identifying a nat-
ural number with the set of its predecessors. With the relation ‘successor of ’, abbreviated
S, this definition can be written as

0 = g , 1 = S 0 = ˜0 = ˜g ,
2 = S 1 = ˜0, 1 = ˜g, ˜g and n + 1 = S n = ˜0, ..., n . (468)

This set, together with the binary operations ‘addition’ and ‘multiplication,’ constitutes

* However, there is no need for written numbers for doing mathematics, as shown by Marcia Ascher,
Ethnomathematics – A Multicultural View of Mathematical Ideas, Brooks/Cole, 1991.

Dvipsbugw
 652 intermezzo: the brain, l anguage and the human condition

the algebraic system N = (N, +, ë, 1) of the natural numbers. For all number systems
the algebraic system and the set are often sloppily designated by the same symbol. The
Page 1194 algebraic system N is a so-called semi-ring, as explained in Appendix D. (Some authors
prefer not to count the number zero as a natural number.) Natural numbers are fairly
useful.
TA B L E 54 Some large numbers

Number E x a m p l e i n n at u r e
Dvipsbugw
Around us
1 number of angels that can be in one place at the same time, following
Thomas Aquinas Ref. 634
8 number of times a newspaper can be folded in alternate perpendicular dir-
ections
12 largest number of times a paper strip has been folded in the same direction

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 635
20 number of digits in precision measurements that will probably never be
achieved
34, 55, 89 petals of common types of daisy and sunflower Ref. 636
57 faces of a diamond with brilliant cut
2000 stars visible in the night sky
105 leaves of a tree (10 m beech)
6 to 7 ë109 humans in the year 2000
1017 ants in the world
c. 1020 number of snowflakes falling on the Earth per year
c. 1024 grains of sand in the Sahara desert
1022 stars in the universe
1025 cells on Earth
1.1 ë 1050 atoms making up the Earth (63703 km3 ë 4 ë 3.14~3 ë 5500 kg~m3 ë 30 mol~kg ë
6 ë 1023 ~mol)
1081 atoms in the visible universe
1090 photons in the visible universe
10169 number of atoms fitting in the visible universe
10244 number of space-time points inside the visible universe
Copyright © Christoph Schiller November 1997–May 2006

Information
51 record number of languages spoken by one person
c. 5000 words spoken on an average day by a man
c. 7000 words spoken on an average day by a woman
c. 7000 number of languages on Earth
c. 350 000 words of the English language (more than any other language, with the pos-
sible exception of German)
c. 2 000 000 number of scientists on Earth around the year 2000
3 ë 108 words spoken during a lifetime (2/3 time awake, 30 words per minute)
4 ë 109 pulses exchanged between both brain halves every second

Dvipsbugw
 what is l anguage? 653

Number E x a m p l e i n n at u r e
109 words heard and read during a lifetime
1017 image pixels seen in a lifetime (3 ë 109 s ë (1~15 ms) ë 2~3 (awake) ë106 (nerves
to the brain) Ref. 637
1019 bits of information processed in a lifetime (the above times 32)
c. 5 ë 1012 printed words available in (different) books around the world (c. 100 ë 106
books consisting of 50 000 words)
210 ë 37 ë 8! ë 12! Dvipsbugw
= 4.3 ë 1019 possible positions of the 3  3  3 Rubik’s Cube Ref. 638
5.8 ë 1078 possible positions of the 4  4  4 Rubik-like cube
5.6 ë 10117 possible positions of the 5  5  5 Rubik-like cube
c. 10200 possible games of chess
c. 10800 possible games of go
7
c. 1010 possible states in a personal computer

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Parts of us
600 numbers of muscles in the human body, of which about half are in the face
150 000  50 000 hairs on a healthy head
900 000 neurons in the brain of a grasshopper
126 ë 106 light sensitive cells per retina (120 million rods and 6 million cones)
1010 to 1011 neurons in the human brain
A 1016 memory bits in the human brain
500 ë 106 blinks of the eye during a lifetime (about once every four seconds when
awake)
300 ë 106 breaths taken during human life
3 ë 109 heart beats during a human life
3 ë 109 letters (base pairs) in haploid human DNA
6.1 ë 109 bits in a compact disc
1 ë 1011 humans who have ever lived
10151 cells in the human body
10161 bacteria carried in the human body

The system of integers Z = (..., −2, −1, 0, 1, 2, ..., +, ë, 0, 1) is the minimal ring that is an
extension of the natural numbers. The system of rational numbers Q = (Q, +, ë, 0, 1) is the
Copyright © Christoph Schiller November 1997–May 2006

minimal field that is an extension of the ring of the integers. (The terms ‘ring’ and ‘field’ are
Page 1194 explained in Appendix D.) The system of real numbers R = (R, +, ë, 0, 1, A) is the minimal
extension of the rationals that is continuous and totally ordered. (For the definition of
continuity, see page 1214 and 1195.) Equivalently, the reals are the minimal extension of
the rationals forming a complete, totally strictly-Archimedean ordered field. This is the
historical construction – or definition – of the integer, rational and real numbers from
the natural numbers. However, it is not the only one construction possible. The most
beautiful definition of all these types of numbers is the one discovered in 1969 by John
Ref. 639 Conway, and popularized by him, Donald Knuth and Martin Kruskal.
— A number is a sequence of bits. The two bits are usually called ‘up’ and ‘down’. Examples

Dvipsbugw
654 intermezzo: the brain, l anguage and the human condition

.
.
.

ω = smallest infinite 1=1

1
11111 11 2 ω
1=1111... ω e
ω+4 2ω
1111 ω 1
1111 11
111 4 π 1 1 1 1
1 ω/2 ω/4 1
3 111 ω-4 sqrt(ω)
11 1 8/3
11
2 1 1 Dvipsbugw
1 3/2
1 1 1
1
1 11 1 2/3+2ι/3
1 3/4 2/3
11
1/2 1 ι = 1/ω = simplest infinitesimal
1 1
1/4 1 1 1111 1 1 sqrt(ι)
1
0 ι 4ι 11 2
1 ι
-1/4
1 1
11

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

-1/2
-1 1 -3/4 -1/3
1 1 11
1 1 1
-3/2
-2 -4/5
11 11 1 1 1
11 11 1 1 11111 11...
-3
111 -sqrt(2) 1
-4 1
-ω -ω/2
smaller 1111
earlier 1=1111 -2ω 2 ω
... -ω -e
11
1
1

F I G U R E 281 The surreal numbers in conventional and in bit notation

of numbers and the way to write them are given in Figure 281.
— The empty sequence is zero.
— A finite sequence of n ups is the integer number n, and a finite sequence of n downs
is the integer −n. Finite sequences of mixed ups and downs give the dyadic rational
numbers. Examples are 1, 2, 3, −7, 19/4, 37~256, etc. They all have denominators with
a power of 2. The other rational numbers are those that end in an infinitely repeating
string of ups and downs, such as the reals, the infinitesimals and simple infinite num-
Copyright © Christoph Schiller November 1997–May 2006

bers. Longer countably infinite series give even more crazy numbers. The complete
class is called the class of surreal numbers.*
There is a second way to write surreal numbers. The first is the just mentioned se-
quence of bits. But in order to define addition and multiplication, another notation is
usually used, deduced from Figure 281. A surreal α is defined as the earliest number of

* The surreal numbers do not form a set since they contain all ordinal numbers, which themselves do not
form a set, even though they of course contain sets. In short, ordinals and surreals are classes which are
larger than sets.

Dvipsbugw
 what is l anguage? 655

all those between two series of earlier surreals, the left and the right series:

α = ˜a, b, c, ...SA, B, C, ... with a, b, c, < α < A, B, C . (469)

For example, we have

˜0S = 1 , ˜0, 1 S = 2 , ˜S0 = −1 , ˜S − 1, 0 = −2 , ˜0 S1 = 1~2 ,

˜0 S1~2, 1~4 = 1 , ˜0, 1, 3~2, 25~16 S 41~16, 13~8, 7~4, 2 = 1 + 37~64 , (470) Dvipsbugw
showing that the finite surreals are the dyadic numbers m~2n (n and m being integers).
Given two surreals α = ˜..., a, ...S..., A, ... with a < α < A and β = ˜..., b, ...S..., B, ...
with b < β < B, addition is defined recursively, using earlier, already defined numbers, as

α + β = ˜..., a + β, ..., α + b, ...S..., A + β, ..., α + B, ... . (471)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

This definition is used simply because it gives the same results as usual addition for in-
tegers and reals. Can you confirm this? By the way, addition is not always commutative.
Challenge 1127 n Are you able to find the exceptions, and to find the definition for subtraction? Multiplic-
ation is also defined recursively, namely by the expression

αβ =˜..., aβ + αb − ab, ..., Aβ + αB − AB, ...S

..., aβ + αB − aB, ..., Aβ + αb − Ab, ... . (472)
º
These definitions allow one to write ι = 1~ω, and to talk about numbers such as ω , the
square root of infinity, about ω + 4, ω − 1, 2ω, eω and about other strange numbers shown
Ref. 639 in Figure 281. However, the surreal numbers are not commonly used. More common is
one of their subsets.
The real numbers are those surreals whose length is not larger than infinity and that
do not have periodic endings with a period of length 1. In other words, the surreals dis-
tinguish the number 0.9999999 from the number 1, whereas the reals do not. In fact,
Challenge 1128 n between the two, there are infinitely many surreal numbers. Can you name a few?
Reals are more useful for describing nature than surreals, first because they form a set –
Appendix D which the surreals do not – and secondly because they allow the definition of integration.
Other numbers defined with the help of reals, e.g. the complex numbers C and the qua-
Copyright © Christoph Schiller November 1997–May 2006

ternions H, are presented in Appendix D. A few more elaborate number systems are also
presented there.
To conclude, in physics it is usual to call numbers the elements of any set that is a semi-
ring (e.g. N), a ring (e.g. Z) or a field (Q, R, C or H). All these concepts are defined in
Appendix D. Since numbers allow one to compare magnitudes and thus to measure, they
play a central role in the description of observations.

Dvipsbugw
 656 intermezzo: the brain, l anguage and the human condition

A series of equal balls is packed in such a way

Ref. 640
“ that the area of needed wrapping paper is
minimal. For small numbers of balls the linear
package, with all balls in one row, is the most
efficient. For which number of balls is the linear
package no longer a minimum?

”
Challenge 1129 n

Why use mathematics?

Die Forderung der Möglichkeit der einfachen Dvipsbugw

“ Zeichen ist die Forderung der Bestimmtheit des

Sinnes.*
Ludwig Wittgenstein, Tractatus, 3.23

Ref. 641
Several well-known physicists have repeatedly asked why mathematics is so important.
For example, Niels Bohr is quoted as having said: ‘We do not know why the language of
mathematics has been so effective in formulating those laws in their most succinct form.’
”

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Eugene Wigner wrote an often cited paper entitled The unreasonable effectiveness of math-
Ref. 642 ematics. At the start of science, many centuries earlier, Pythagoras and his contemporaries
were so overwhelmed by the usefulness of numbers in describing nature, that Pythagoras
was able to organize a sect based on this connection. The members of the inner circle of
this sect were called ‘learned people,’ in Greek ‘mathematicians’, from the Greek µάθηµα
‘teaching’. This sect title then became the name of the modern profession.
These men forgot that numbers, as well as a large part of mathematics, are concepts
developed precisely with the aim of describing nature. Numbers and mathematical con-
cepts were developed right from the start to provide as succinct a description as possible.
That is one consequence of mathematics being the science of symbolic necessities.
Perhaps we are being too dismissive. Perhaps these thinkers mainly wanted to express
their feeling of wonder when experiencing that language works, that thinking and our
brain works, and that life and nature are so beautiful. This would put the title question
nearer to the well-known statement by Albert Einstein: ‘The most incomprehensible fact
about the universe is that it is comprehensible.’ Comprehension is another word for de-
scription, i.e. for classification. Obviously, any separable system is comprehensible, and
there is nothing strange about it. But is the universe separable? As long as is it described
as being made of particles and vacuum, this is the case.
We will find in the third part of this adventure that the basic assumption made at our
start is built on sand. The assumption that observations in nature can be counted, and thus
Copyright © Christoph Schiller November 1997–May 2006

that nature is separable, is an approximation. The quoted ‘incomprehensibility’ becomes

amazement at the precision of this approximation. Nevertheless, Pythagoras’ sect, which
was based on the thought that ‘everything in nature is numbers’, was wrong. Like so many
beliefs, observation will show that it was wrong.
Die Physik ist für Physiker viel zu schwer.**

“ David Hilbert

”
* The requirement that simple signs be possible is the requirement that sense be determinate.
** Physics is much too difficult for physicists.

Dvipsbugw
 what is l anguage? 657

Is mathematics a language?
Die Sätze der Mathematik sind Gleichungen,

“ also Scheinsätze. Der Satz der Mathematik

drückt keinen Gedanken aus.*
Ludwig Wittgenstein, Tractatus, 6.2, 6.21

Surely, mathematics is a vocabulary that helps us to talk with precision. Mathematics can
be seen as the exploration of all possible concepts that can be constructed from the two
fundamental bricks ‘set’ and ‘relation’ (or some alternative, but equivalent pair). Mathem-
” Dvipsbugw
atics is the science of symbolic necessities. Rephrased again, mathematics is the explor-
ation of all possible types of classifications. This explains its usefulness in all situations
where complex, yet precise classifications of observations are necessary, such as in phys-
ics.
However, mathematics cannot express everything that humans want to communicate,
such as wishes, ideas or feelings. Just try to express the fun of swimming using mathem-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

atics. Indeed, mathematics is the science of symbolic necessities; thus mathematics is not a
language, nor does it contain one. Mathematical concepts, being based on abstract sets
and relations, do not pertain to nature. Despite its beauty, mathematics does not allow us
to talk about nature or the observation of motion. Mathematics does not tell what to say
about nature; it does tell us how to say it.
In his famous 1900 lecture in Paris, the German mathematician David Hilbert** gave a
list of 23 great challenges facing mathematics. The sixth of Hilbert’s problems was to find
a mathematical treatment of the axioms of physics. Our adventure so far has shown that
physics started with circular definitions that has not yet been eliminated after 2500 years
of investigations: space-time is defined with the help of objects and objects are defined
Page 157 with the help of space and time. Being based on a circular definition, physics is thus
not modelled after mathematics, even if many physicists and mathematicians, including
Ref. 643 Hilbert, would like it to be so. Physicists have to live with logical problems and have to
walk on unsure ground in order to achieve progress. In fact, they have done so for 2500
years. If physics were an axiomatic system, it would not contain contradictions; on the
other hand, it would cease to be a language and would cease to describe nature. We will
Page 1049 return to this issue later.

Curiosities and fun challenges about mathematics

**
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1130 n What is the largest number that can be written with four digits of 2 and no other sign?

* The propositions of mathematics are equations, and therefore pseudo-propositions. A proposition of math-
ematics does not express a thought.
** David Hilbert (1862 Königsberg–1943 Göttingen), professor of mathematics in Göttingen, greatest math-
ematician of his time. He was a central figure to many parts of mathematics, and also played an important
role both in the birth of general relativity and of quantum theory. His textbooks are still in print. His famous
personal credo was: ‘Wir müssen wissen, wir werden wissen.’ (We must know, we will know.) His famous
Paris lecture is published e.g. in Die Hilbertschen Probleme, Akademische Verlagsgesellschaft Geest & Portig,
1983. The lecture galvanized all of mathematics. (Despite efforts and promises of similar fame, nobody in the
world had a similar overview of mathematics that allowed him or her to repeat the feat in the year 2000.) In
his last decade he suffered the persecution of the Nazi regime; the persecution eliminated Göttingen from

Dvipsbugw
 658 intermezzo: the brain, l anguage and the human condition

And with four 4s?

**
Pythagorean triplets are integers that obey a 2 + b 2 = c 2 . Give at least ten examples. Then
Challenge 1131 e show the following three properties: at least one number in a triplet is a multiple of 3; at
least one number in a triplet is a multiple of 4; at least one number in a triplet is a multiple
of 5.
** Dvipsbugw
The number 1~n, when written in decimal notation, has a periodic sequence of digits.
The period is at most n − 1 digits long, as for 1~7 = 0.142857 142857 1428.... Which other
Challenge 1132 d numbers 1~n have periods of length n − 1?
**
Felix Klein was a famous professor of mathematics at Göttingen University. There were

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

two types of mathematicians in his department: those who did research on whatever they
wanted and those for which Klein provided the topic of research. To which type did Klein
Challenge 1133 n belong?
Obviously, this is a variation of another famous puzzle. A barber shaves all those
Challenge 1134 n people who do not shave themselves. Does the barber shave himself?
**
Everybody knows what a magic square is: a square array of numbers, in the simplest case
from 1 to 9, that are distributed in such a way that the sum of all rows, columns (and
possibly all diagonals) give the same sum. Can you write down the simplest 3  3  3
Challenge 1135 n magic cube?
**
The digits 0 to 9 are found on keyboards in two different ways. Calculators and keyboards
have the 7 at the top left, whereas telephones and automatic teller machines have the digit
1 at the top left. The two standards, respectively by the International Standards Organiz-
ation (ISO) and by the International Telecommunication Union (ITU, formerly CCITT),
Ref. 644 evolved separately and have never managed to merge.
**
Copyright © Christoph Schiller November 1997–May 2006

Leonhard Euler in his notebooks sometimes wrote down equations like

1 + 22 + 24 + 26 + 28 + ... = − .
1
(473)
3
Challenge 1136 d Can this make sense?

the list of important science universities, without recovering its place up to this day.

Dvipsbugw
 physical concepts, lies and patterns of nature 659

Physical concepts, lies and patterns of nature

Die Grenzen meiner Sprache bedeuten die

“ Grenzen meiner Welt.*

Ludwig Wittgenstein, Tractatus, 5.6
Der Satz ist ein Bild der Wirklichkeit. Der Satz
”
“ ist ein Modell der Wirklichkeit, so wie wir sie
uns denken.**
Ludwig Wittgenstein, Tractatus, 4.01

Ref. 645 In contrast to mathematics, physics does aim at being a language. Through the descrip-
tion of motion it aims to express everything observed and, in particular, all examples and
possibilities of change.*** Like any language, physics consists of concepts and sentences.
” Dvipsbugw

In order to be able to express everything, it must aim to use few words for a lot of facts.****
Physicists are essentially lazy people: they try to minimize the effort in everything they do.
The concepts in use today have been optimised by the combined effort of many people to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

be as practical, i.e. as powerful as possible. A concept is called powerful when it allows one
to express in a compact way a large amount of information, meaning that it can rapidly
convey a large number of details about observations.
General statements about many examples of motion are called rules or patterns. In the
past, it was often said that ‘laws govern nature’, using an old and inappropriate ideology.
A physical ‘law’ is only a way of saying as much as possible with as few words as possible.
When saying ‘laws govern nature’ we actually mean to say ‘being lazy, we describe obser-
vations with patterns’. Laws are the epitome of laziness. Formulating laws is pure sloth. In
fact, the correct expression is patterns describe nature.
Physicists have defined the laziness necessary for their field in much detail. In order
to become a master of laziness, we need to distinguish lazy patterns from those which are
not, such as lies, beliefs, statements that are not about observations, and statements that
Page 663 are not about motion. We do this below.
The principle of extreme laziness is the origin, among others, of the use of numbers in
physics. Observables are often best described with the help of numbers, because numbers
allow easy and precise communication and classification. Length, velocity, angles, tem-
perature, voltage or field strength are of this type. The notion of ‘number’, used in every
* The limits of my language are the limits of my world.
** A proposition is a picture of reality. A proposition is a model of reality as we imagine it.
*** All observations are about change or variation. The various types of change are studied by the various
Copyright © Christoph Schiller November 1997–May 2006

sciences; they are usually grouped in the three categories of human sciences, formal sciences and natural
sciences. Among the latter, the oldest are astronomy and metallurgy. Then, with the increase of curiosity
in early antiquity, came the natural science concerned with the topic of motion: physics. In the course of
our walk it will become clear that this seemingly restrictive definition indeed covers the whole set of topics
studied in physics. In particular it includes the more common definition of physics as the study of matter,
its properties, its components and their interactions.
**** A particular, specific observation, i.e. a specific example of input shared by others, is called a fact, or
in other contexts, an event. A striking and regularly observed fact is called a phenomenon, and a general
observation made in many different situations is called a (physical) principle. (Often, when a concept is
introduced that is used with other meaning in other fields, in this walk it is preceded by the qualifier ‘physical’
or ‘mathematical’ in parentheses.) Actions performed towards the aim of collecting observations are called
experiments. The concept of experiment became established in the sixteenth century; in the evolution of a
child, it can best be compared to that activity that has the same aim of collecting experiences: play.

Dvipsbugw
 660 intermezzo: the brain, l anguage and the human condition

measurement, is constructed, often unconsciously, from the notions of ‘set’ and ‘relation’,
as shown above. Apart from the notion of number, other concepts are regularly defined to
allow fast and compact communication of the ‘laws’ of nature; all are ‘abbreviation tools.’
In this sense, the statement ‘the level of the Kac–Moody algebra of the Lagrangian of the
heterotic superstring model is equal to one’ contains precise information, explainable to
everybody; however, it would take dozens of pages to express it using only the terms ‘set’
and ‘relation.’ In short, the precision common in physics results from its quest for laziness.

Are physical concepts discovered or created? Dvipsbugw

Das logische Bild der Tatsachen ist der

“ Gedanke.*
Ludwig Wittgenstein, Tractatus, 3

The title question is often rephrased as: are physical concepts free of beliefs, taste or per- ”
sonal choices? The question has been discussed so much that it even appears in Holly-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

wood movies. We give a short summary that can help you to distinguish honest from
dishonest teachers.
Creation of concepts, in contrast to their discovery, would imply free choice between
many alternative possibilities. The chosen alternative would then be due to the beliefs or
tastes used. In physics (in obvious contrast to other, more ideological fields of enquiry),
we know that different physical descriptions of observations are either equivalent or, in
the opposite case, imprecise or even wrong. A description of observations is thus essen-
tially unique: any choices of concepts are only apparent. There is no real freedom in the
definition of physical concepts. In this property, physics is in strong contrast to artistic
activity.
If two different concepts can be used to describe the same aspect of observations, they
must be equivalent, even if the relation that leads to the equivalence is not immediately
clear. In fact, the requirement that people with different standpoints and observing the
same event deduce equivalent descriptions lies at the very basis of physics. It expresses
the requirement that observations are observer independent. In short, the strong require-
ment of viewpoint independence makes the free choice of concepts a logical impossibility.
The conclusion that concepts describing observations are discovered rather than cre-
ated is also reached independently in the field of linguistics by the above-mentioned re-
search on semantic primitives,** in the field of psychology by the observations on the
formation of the concepts in the development of young children, and in the field of eth-
Copyright © Christoph Schiller November 1997–May 2006

ology by the observations of animal development, especially in the case of mammals. In

all three fields detailed observations have been made of how the interactions between an
individual and its environment lead to concepts, of which the most basic ones, such as
space, time, object or interaction, are common across the sexes, cultures, races and across
many animal species populating the world. Curiosity and the way that nature works leads
to the same concepts for all people and even the animals; the world offers only one pos-
sibility, without room for imagination. Imagining that physical concepts can be created
at your leisure is a belief – or a useful exercise, but never successful.
* A logical picture of facts is a thought.
** Anna Wierzbicka concludes that her research clearly indicates that semantic primitives are discovered, in
Ref. 628 particular that they are deduced from the fundamentals of human experience, and not invented.

Dvipsbugw
 physical concepts, lies and patterns of nature 661

Physical concepts are classifications of observations. The activity of classification itself

follows the patterns of nature; it is a mechanical process that machines can also perform.
This means that any distinction, i.e. any statement that A is different from B, is a theory-
free statement. No belief system is necessary to distinguish different entities in nature.
Cats and pigs can also do so. Physicists can be replaced by animals, even by machines.
Our mountain ascent will repeatedly confirm this point.
As already mentioned, the most popular physical concepts allow to describe observa-
tions as succinctly and as accurately as possible. They are formed with the aim of having
the largest possible amount of understanding with the smallest possible amount of effort. Dvipsbugw
Page 1115 Both Occam’s razor – the requirement not to introduce unnecessary concepts – and the
drive for unification automatically reduce the number and the type of concepts used in
physics. In other words, the progress of physical science was and is based on a programme
that reduces the possible choice of concepts as drastically as possible.
In summary, we found that physical concepts are the same for everybody and are free
of beliefs and personal choices: they are first of all boring. Moreover, as they could stem

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

from machines instead of people, they are born of laziness. Despite these human analo-
gies – not meant to be taken too seriously – physical concepts are not created; they are
discovered. If a teacher tells you the opposite, he is lying.
Having handled the case of physical concepts, let us now turn to physical statements.
The situation is somewhat similar: physical statements must be lazy, arrogant and boring.
Let us see why.
Wo der Glaube anfängt, hört die Wissenschaft

“ auf.*
Ernst Haeckel, Natürliche Schöpfungsgeschichte,
1879.

How do we find physical patterns and rules?

Grau, treuer Freund, ist alle Theorie,
”
“ Und grün des Lebens goldner Baum.**
J.W. v. Goethe, Faust.
Physics is usually presented as an objective
”
“ science, but I notice that physics changes and
the world stays the same, so there must be
something subjective about physics.
Richard Bandler

”
Copyright © Christoph Schiller November 1997–May 2006

Progressing through the study of motion reflects a young child’s attitude towards life. The
progress follows the simple programme on the left of Table 55.
Adult scientists do not have much more to add, except the more fashionable terms
on the right, plus several specialized professions to make money from them. The experts
of step 7 are variously called lobbyists or fund raisers; instead of calling this program
‘curiosity’, they call it the ‘scientific method.’ They mostly talk. Physics being the talk about
motion,*** and motion being a vast topic, many people specialize in this step.
* Where belief starts, science ends.
** ‘Grey, dear friend, is all theory, and green the golden tree of life.’ Johann Wolfgang von Goethe (1749–
1832), the most influential German poet.
*** Several sciences have the term ‘talk’ as part of their name, namely all those whose name finishes in ‘-logy’,

Dvipsbugw
662 intermezzo: the brain, l anguage and the human condition

TA B L E 55 The ‘scientific method’

Normal description Lobbyist description

Curiosity Scientific method
1. look around a lot 1. interact with the world
2. don’t believe anything told 2. forget authority
3. choose something interesting and explore it your- 3. observe
self Dvipsbugw
4. make up your own mind and describe precisely 4. use reason, build hypothesis
what you saw
5. check if you can also describe similar situations in 5. analyse hypothesis
the same way
6. increase the precision of observation until the 6. perform experiments until hypo-
checks either fail or are complete thesis is proved false or established

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

7. depending on the case, continue with step 4 or 1 7. ask for more money

The experts of step 6 are called experimental physicists or simply experimentalists, a

term derived from the Latin ‘experiri’, meaning ‘to try out’. Most of them are part of the
category ‘graduate students’. The experts of steps 5 and 4 are called theoretical physicists
or simply theoreticians.* This is a rather modern term; for example, the first professors of
theoretical physics were appointed around the start of the twentieth century. The term is
derived from the Greek θεωρία meaning ‘observation, contemplation’. Finally, there are
the people who focus on steps 1 to 3, and who induce others to work on steps 4 to 6; they
are called geniuses.
Obviously an important point is hidden in step 6: how do all these people know
whether their checks fail? How do they recognize truth?
All professions are conspiracies against laymen.

What is a lie?
“ George Bernard Shaw

”
Get your facts straight, and then you can distort

“ them at your leisure.

Mark Twain

”
Copyright © Christoph Schiller November 1997–May 2006

such as e.g. biology. The ending stems from ancient Greek and is deduced from λήγηιν meaning ‘to say, to
talk’. Physics as the science of motion could thus be called ‘kinesiology’ from κίνησις, meaning ‘motion’; but
for historical reasons this term has a different meaning, namely the study of human muscular activity. The
term ‘physics’ is either derived from the Greek φύσικη (τέχνη is understood) meaning ‘(the art of) nature’,
or from the title of Aristotle’ works τά φυσικά meaning ‘natural things’. Both expressions are derived from
φύσις, meaning ‘nature’.
* If you like theoretical physics, have a look at the refreshingly candid web page by Nobel Prize winner
Gerard ‘t Hooft with the title How to become a good theoretical physicist. It can be found at http://www.phys.
uu.nl/~thooft/theorist.html.

Dvipsbugw
 physical concepts, lies and patterns of nature 663

The pure truth is always a lie.

“
Lies are useful statements, as everybody learns during their youth. One reason that they
Bert Hellinger

”
are useful is because we can draw any imaginable conclusion from them. A well-known
discussion between two Cambridge professors early in the twentieth century makes the
point. McTaggart asked: ‘If 2 + 2 = 5, how can you prove that I am the pope?’ Godfrey
Hardy: ‘If 2+2 = 5, then 4 = 5; subtract 3; then 1 = 2; but McTaggart and the pope are two;
therefore McTaggart and the pope are one.’ As noted long ago, ex falso quodlibet. From Dvipsbugw
what is wrong, anything imaginable can be deduced. It is true that in our mountain ascent
we need to build on previously deduced results and that our trip could not be completed
if we had a false statement somewhere in our chain of arguments. But lying is such an
important activity that one should learn to perform it well.
There are various stages in the art of lying. Many animals have been shown to de-
Ref. 610 ceive their kin. Children start lying just before their third birthday, by hiding experiences.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Adults cheat on taxes. And many intellectuals or politicians even claim that truth does
not exist.
However, in most countries, everybody must know what ‘truth’ is, since in a law court
for example, telling an untruth can lead to a prison sentence. The courts are full of experts
in lie detection. If you lie in court, you better do it well; experience shows that you might
get away with many criminal activities. In court, a lie is a statement that knowingly con-
trasts with observations.* The truth of a statement is thus checked by observation. The
check itself is sometimes called the proof of the statement. For law courts, as for physics,
truth is thus the correspondence with facts, and facts are shared observations. A ‘good’ lie
is thus a lie whose contrast with shared observations is hard to discover.
The first way of lying is to put an emphasis on the sharedness only. Populists and po-
lemics do this regularly. (‘Every foreigner is a danger for the values of our country.’) Since
almost any imaginable opinion, however weird, is held by some group – and thus shared
– one can always claim it as true. Unfortunately, it is no secret that ideas also get shared
because they are fashionable, imposed or opposed to somebody who is generally disliked.
Often a sibling in a family has this role – remember Cassandra.** For a good lie we thus
need more than sharedness, more than intersubjectivity alone.
A good lie should be, like a true statement, really independent of the listener and the
observer and, in particular, independent of their age, their sex, their education, their civil-
ization or the group to which they belong. For example, it is especially hard – but not im-
Copyright © Christoph Schiller November 1997–May 2006

possible – to lie with mathematics. The reason is that the basic concepts of mathematics,
be they ‘set’, ‘relation’ or ‘number’, are taken from observation and are intersubjective, so

* Statements not yet checked are variously called speculations, conjectures, hypotheses, or – wrongly – simply
theses. Statements that are in correspondence with observations are called correct or true; statements that
contrast with observations are called wrong or false.
** The implications of birth order on creativity in science and on acceptance of new ideas has been studied in
the fascinating book by Frank J. Sulloway, Born to Rebel – Birth Order, Family Dynamics and Creative
Lives, Panthon Books, 1996. This exceptional book tells the result of a life-long study correlating the personal
situations in the families of thousands of people and their receptivity to about twenty revolutions in the
recent history. The book also includes a test in which the reader can deduce their own propensity to rebel,
on a scale from 0 to 100 %. Darwin scores 96 % on this scale.

Dvipsbugw
 664 intermezzo: the brain, l anguage and the human condition

that statements about them are easily checked. Usually, lies thus avoid mathematics.*
Secondly, a ‘good’ lie should avoid statements about observations and use interpreta-
tions instead. For example, some people like to talk about other universes, which implies
talking about fantasies, not about observations. A good lie has to avoid, however, to fall
in the opposite extreme, namely to make statements which are meaningless; the most
destructive comment that can be made about a statement is the one used by the great
Page 749 Austrian physicist Wolfgang Pauli: ‘That is not even wrong.’
Thirdly, a good lie doesn’t care about observations, only about imagination. Only truth
needs to be empirical, to distinguish it from speculative statements. If you want to lie Dvipsbugw
‘well’ even with empirical statements, you need to pay attention. There are two types of
empirical statements: specific statements and universal statements. For example, ‘On the
31st of August 1960 I saw a green swan swimming on the northern shore of the lake of
Varese’ is specific, whereas ‘All ravens are black’ is universal, since it contains the term
‘all’. There is a well-known difference between the two, which is important for lying well:
specific statements cannot be falsified, they are only verifiable, and universal statements

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 646 cannot be verified, they are only falsifiable. Why is this so?
Universal statements such as ‘the speed of light is constant’ cannot be tested for all
possible cases. (Note that if they could, they would not be universal statements, but just
a list of specific ones.) However, they can be reversed by a counter-example. Another
example of the universal type is: ‘Apples fall upwards.’ Since it is falsified by an observation
conducted by Newton several centuries ago, or by everyday experience, it qualifies as an
(easily detectable) lie. In general therefore, lying by stating the opposite of a theory is
usually unsuccessful. If somebody insists on doing so, the lie becomes a superstition, a
belief, a prejudice or a doctrine. These are the low points in the art of lying. A famous case
of insistence on a lie is that of the colleagues of Galileo, who are said to have refused to
look through his telescope to be convinced that Jupiter has moons, an observation that
would have shaken their belief that everything turns around the Earth. Obviously these
astronomers were amateurs in the art of lying. A good universal lie is one whose counter-
example is not so easily spotted.
There should be no insistence on lies in physics. Unfortunately, classical physics is full
of lies. We will dispel them during the rest of our walk.
Lying by giving specific instead of universal statements is much easier. (‘I can’t remem-
ber.’) Even a specific statement such as ‘yesterday the Moon was green, cubic and smelled
of cheese’ can never be completely falsified: there is no way to show with absolute cer-
tainty that this is wrong. The only thing that we can do is to check whether the statement
Copyright © Christoph Schiller November 1997–May 2006

is compatible with other observations, such as whether the different shape affected the
tides as expected, whether the smell can be found in air collected that day, etc. A good
specific lie is thus not in contrast with other observations.**

* In mathematics, ‘true’ is usually specified as ‘deducible’ or ‘provable’; this is in fact a special case of the
usual definition of truth, namely ‘correspondence with facts’, if one remembers that mathematics studies
the properties of classifications.
** It is often difficult or tedious to verify statements concerning the past, and the difficulty increases with
the distance in time. That is why people can insist on the occurrence of events which are supposed to be
exceptions to the patterns of nature (‘miracles’). Since the advent of rapid means of communication these
checks are becoming increasingly easy, and no miracles are left over. This can be seen in Lourdes in France,
where even though today the number of visitors is much higher than in the past, no miracles have been seen

Dvipsbugw
 physical concepts, lies and patterns of nature 665

Incidentally, universal and specific statements are connected: the opposite of a univer-
sal statement is always a specific statement, and vice versa. For example, the opposite of
the general statement ‘apples fall upwards’, namely ‘some apples fall downwards’, is specific.
Similarly, the the specific statement ‘the Moon is made of green cheese’ is in opposition
to the universal statement ‘the Moon is solid since millions of years and has almost no
smell or atmosphere.’
In other words, law courts and philosophers disagree. Law courts have no problem
with calling theories true, and specific statements lies. Many philosophers avoid this. For
example, the statement ‘ill-tempered gaseous vertebrates do not exist’ is a statement of Dvipsbugw
the universal type. If a universal statement is in agreement with observations, and if it
is falsifiable, law courts call it true. The opposite, namely the statement: ‘ill-tempered
gaseous vertebrates do exist’, is of the specific type, since it means ‘Person X has observed
an ill-tempered gaseous vertebrate in some place Y at some time Z’. To verify this, we
need a record of the event. If such a record, for example a photographs or testimony does
not exist, and if the statement can be falsified by other observations, law courts call the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

specific statement a lie. Even though these are the rules for everyday life and for the law,
there is no agreement among philosophers and scientists that this is acceptable. Why?
Intellectuals are a careful lot, because many of them have lost their lives as a result of
exposing lies too openly.
In short, specific lies, like all specific statements, can never be falsified with certainty.
This is what makes them so popular. Children learn specific lies first. (‘I haven’t eaten the
jam.’) General lies, like all general statements, can always be corroborated by examples.
This is the reason for the success of ideologies. But the criteria for recognizing lies, even
general lies, have become so commonplace that beliefs and lies try to keep up with them. It
became fashionable to use expressions such as ‘scientific fact’ – there are no non-scientific
facts –, or ‘scientifically proven’ – observations cannot be proven otherwise – and similar
empty phrases. These are not ‘good’ lies; whenever we encounter sentences beginning
with ‘science says ...’ or ‘science and religion do ...’, replacing ‘science’ by ‘knowledge’ or
‘experience’ is an efficient way of checking whether such statements are to be taken seri-
ously or not.*
Lies differ from true statements in their emotional aspect. Specific statements are usu-
ally boring and fragile, whereas specific lies are often sensational and violent. In contrast,
general statements are often daring and fragile whereas general lies are usually boring
and violent. The truth is fragile. True statements require the author to stick his neck out
to criticism. Researchers know that if one doesn’t stick the neck out, it can’t be an observa-
Copyright © Christoph Schiller November 1997–May 2006

tion or a theory. (A theory is another name for one or several connected, not yet falsified
in decades.
In fact, all modern ‘miracles’ are kept alive only by consciously eschewing checks, such as the supposed
yearly liquefaction of blood in Napoli, the milk supposedly drunk by statues, the supposed healers in tele-
Ref. 647 vision evangelism, etc. Most miracles only remain because many organizations make money out of the dif-
ficulty of falsifying specific statements. For example, when the British princess Diana died in a car crash
in 1997, even though the events were investigated in extreme detail, the scandal press could go on almost
without end about the ‘mysteries’ of the accident.
* To clarify the vocabulary usage of this text: religion is spirituality plus a varying degree of power abuse. The
mixture depends on each person’s history, background and environment. Spirituality is the open participa-
tion in the whole of nature. Most, maybe all, people with a passion for physics are spiritual. Most are not
religious.

Dvipsbugw
666 intermezzo: the brain, l anguage and the human condition

universal statements about observations.)* Telling the truth does make vulnerable. For
this reason, theories are often daring, arrogant or provoking; at the same time they have
to be fragile and vulnerable. For men, theories thus resemble what they think about wo-
men. Darwin’s The origin of the species, which developed daring theories, illustrates the
stark contrast between the numerous boring and solid facts that Darwin collected and
the daring theory that he deduced. Boredom of facts is a sign of truth.
In contrast, the witch-hunters propagating the so-called ‘intelligent design’ are ex-
amples of liars. The specific lies they propagate, such as ‘the world was created in October
4004 bce’, are sensational, whereas the general lies they propagate, such as ‘there have Dvipsbugw
not been big changes in the past’, are boring. This is in full contrast with common sense.
Moreover, lies, in contrast to true statements, make people violent. The worse the lie, the
more violent the people. This connection can be observed regularly in the news. In other
words, ‘intelligent design’ is not only a lie, it is a bad lie. A ‘good’ general lie, like a good
physical theory, seems crazy and seems vulnerable, such as ‘people have free will’. A ‘good’
specific lie is boring, such as ‘this looks like bread, but for the next ten minutes it is not’.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Good lies do not induce violence. Feelings can thus be a criterion to judge the quality of
lies, if we pay careful attention to the type of statement. A number of common lies are
discussed later in this intermezzo.
An important aspect of any ‘good’ lie is to make as few public statements as possible, so
that critics can check as little as possible. (For anybody sending corrections of mistakes in
this text, the author provides a small reward.) To detect lies, public scrutiny is important,
though not always reliable. Sometimes, even scientists make statements which are not
based on observations. However, a ‘good’ lie is always well prepared and told on purpose;
accidental lies are frowned upon by experts. Examples of good lies in science are ‘aether’,
‘UFOs’, ‘creation science’, or ‘cold fusion’. Sometimes it took many decades to detect the
lies in these domains.
To sum up, the central point of the art of lying without being caught is simple: do
not divulge details. Be vague. All the methods used to verify a statement ask for details,
for precision. For any statement, its degree of precision allows one to gauge the degree
to which the author is sticking his neck out. The more precision that is demanded, the
weaker a statement becomes, and the more likely a fault will be found, if there is one. This
is the main reason that we chose an increase in precision as a guide for our mountain
ascent. By the way, the same method is used in criminal trials. To discover the truth,
investigators typically ask all the witnesses a large number of questions, allowing as many
details as possible come to light. When sufficient details are collected, and the precision is
Copyright © Christoph Schiller November 1997–May 2006

high enough, the situation becomes clear. Telling ‘good’ lies is much more difficult than

* In other words, a set of not yet falsified patterns of observations on the same topic is called a (physical)
theory. The term ‘theory’ will always be used in this sense in this walk, i.e. with the meaning ‘set of correct
general statements’. This use results from its Greek origin: ‘theoria’ means ‘observation’; its original meaning,
‘passionate and emphatic contemplation’, summarizes the whole of physics in a single word. (‘Theory’, like
‘theatre’, is formed from the root θέ, meaning ‘the act of contemplating’.) Sometimes, however, the term
‘theory’ is used – being confused with ‘hypothesis’ – with the meaning of ‘conjecture’, as in ‘your theory
is wrong’, sometimes with the meaning of ‘model’, as in ‘Chern–Simons’ theory and sometimes with the
meaning of ‘standard procedure’, as in ‘perturbation theory’. These incorrect uses are avoided here. To bring
the issue to a point: the theory of evolution is not a conjecture, but a set of correct statements based on
observation.

Dvipsbugw
 physical concepts, lies and patterns of nature 667

telling the truth; it requires an excellent imagination.

Truth is an abyss.

“ Democritus

To teach superstitions as truth is a most terrible

”
“ thing.
Hypatia of Alexandria (c. 355–415)
[Absolute truth:] It is what scientists say it is
”
Ref. 649

Is this statement true?

“ when they come to the end of their labors.
Charles Peirce

”
Dvipsbugw

Truth is a rhetorical concept.

“
Not all statements can be categorized as true or false. Statements can simply make no
Paul Feyerabend

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

sense. There are even such statements in mathematics, where they are called undecid-
able. An example is the continuum hypothesis. This hypothesis is undecidable because
it makes a statement that depends on the precise meaning of the term ‘set’; in standard
mathematical usage the term is not defined sufficiently precisely so that a truth value
can be assigned to the continuum hypothesis. In short, statements can be undecidable
because the concepts contained in them are not sharply defined.
Statements can also be undecidable for other reasons. Phrases such as ‘This statement
is not true’ illustrate the situation. Kurt Gödel* has even devised a general way of con-
structing such statements in the domain of logic and mathematics. The different vari-
ations of these self-referential statements, especially popular both in the field of logic and
computer science, have captured a large public.** Similarly undecidable statements can
Ref. 650 be constructed with terms such as ‘calculable’, ‘provable’ and ‘deducible’.
In fact, self-referential statements are undecidable because they are meaningless. If the
usual definition of ‘true’, namely corresponding to facts, is substituted into the sentence
‘This statement is not true’, we quickly see that it has no meaningful content. The most
famous meaningless sentence of them all was constructed by the linguist Noam Chom-
sky:
Colorless green ideas sleep furiously.
Ref. 616 It is often used as an example for the language processing properties of the brain, but
Copyright © Christoph Schiller November 1997–May 2006

nobody sensible elevates it to the status of a paradox and writes philosophical discussions
about it. To do that with the title of this section is a similar waste of energy.
The main reason for the popular success of self-reference is the difficulty in perceiving
the lack of meaning.*** A good example is the statement:
* Kurt Gödel (1906–1978), famous Austrian logician.
** A general introduction is given in the beautiful books by R aymond Smullyan: Satan, Cantor and
Infinity and Other Mind-boggling Puzzles, Knopf, 1992; What is the Name of This Book? The Riddle of Dracula
and Other Logical Puzzles, Touchstone, 1986, and The Lady or the Tiger? And Other Puzzles, Times Books,
1982. Also definitions can have no content, such as David Hilbert’s ‘smallest number that has not been
mentioned this century’ or ‘the smallest sequence of numbers that is described by more signs than this
sentence’.
*** A well-known victim of this difficulty is Paulus of Tarsus. The paradox of the Cretan poet Epimenedes

Dvipsbugw
 668 intermezzo: the brain, l anguage and the human condition

This statement is false or you are an angel.

Challenge 1138 n We can actually deduce from it that ‘you are an angel.’ Can you see how? If you want, you
can change the second half and get even more interesting statements. Such examples show
that statements referring to themselves have to be treated with great care when under
investigation. In short, whenever you meet somebody who tries to use the self-referential
construction by Kurt Gödel to deduce another statement, take a step back, or better, a few
more. Self-reference, especially the type defined by Gödel, is a hard but common path
– especially amongst wannabe-intellectuals – to think, tell and write nonsense. Nothing Dvipsbugw
useful can be deduced from nonsense. Well, not entirely; it does help to meet psychiatrists
on a regular basis.
In physics, in the other natural sciences, and in legal trials these problems do not
emerge, because self-referential statements are not used.* In fact, the work of logicians
confirms, often rather spectacularly, that there is no way to extend the term ‘truth’ beyond
the definition of ‘correspondence with facts.’

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ein Satz kann unmöglich von sich selbst

“ aussagen, daß er wahr ist.**

Ludwig Wittgenstein, Tractatus, 4.442

”
Curiosities and fun challenges about lies
Some lies are entertaining, others are made with criminal intent; some are good, others
are bad.

**
Challenge 1140 e ‘Yesterday I drowned.’ Is this a good or a bad lie?
**
In the 1990s, so-called crop circles were formed by people walking with stilts, a piece of
wood and some rope in fields of crops. Nevertheless, many pretended and others believed
that these circles were made by extraterrestrial beings. Is this a good or a bad lie? Can you
Challenge 1141 ny give a reason why this is impossible?
**
Sometimes it is heard that a person whose skin is completely covered with finest metal
powder will die, due to the impossibility of the skin to breathe. Can you show that this is
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1142 ny wrong?

**

(6th century bce) who said ‘All Cretans lie’ is too difficult for the notoriously humour-impaired Paulus, who
in his letter to Titus (chapter 1, verses 12 and 13, in the christian bible) calls Epimenedes a ‘prophet’, adds
Ref. 651 some racist comments, and states that this ‘testimony’ is true. But wait; there is a final twist to this story.
The statement ‘All Cretans lie’ is not a paradox at all; a truth value can actually be ascribed to it, because
Challenge 1137 n the statement is not really self-referential. Can you confirm this? The only genuine paradox is ‘I am lying’, to
which it is indeed impossible to ascribe a truth value.
Challenge 1139 ny * Why are circular statements, like those of Galilean physics, not self-referential?
** It is quite impossible for a proposition to state that it itself is true.

Dvipsbugw
 physical concepts, lies and patterns of nature 669

A famous mixture of hoax and belief premises that the Earth was created about six thou-
sand years ago. (Some believers even use this false statement as justification for violence
Challenge 1143 ny against non-believers.) Can you explain why the age is wrong?
**
A famous provocation: the world has been created last Saturday. Can you decide whether
Challenge 1144 ny this is wrong?
** Dvipsbugw
Hundreds of hoaxes are found on the http://www.museumofhoaxes.com website. It gives
an excellent introduction into the art of lying; of course it exposes only those who have
been caught. Enjoy the science stories, especially those about archaeology. (Several other
sites with similar content can be found on the internet.)
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In the 1990s, many so-called ‘healers’ in the Philippines made millions by suggesting
patients that they were able to extract objects from their bodies without operating. Why
Challenge 1145 e is this not possible? (For more information on health lies, see the http://www.quackwatch.
com website.)
**
Since the 1980s, people have claimed that it is possible to acquire knowledge simply from
somebody 1000 km away, without any communication between the two people. However,
Challenge 1146 e the assumed ‘morphogenetic fields’ cannot exist. Why not?
**
It is claimed that a Fire Brigade building in a city in the US hosts a light bulb that has
been burning without interruption since 1901 (at least it was so in 2005). Can this be
Challenge 1147 n true? Hundreds of such stories, often called ‘urban legends,’ can be found on the http://
www.snopes.com website. However, some of the stories are not urban legends, but true,
as the site shows.
**
‘This statement has been translated from French into English’. Is the statement true, false
or neither?
Copyright © Christoph Schiller November 1997–May 2006

**
Aeroplanes have no row 13. Many tall hotels have no floor 13. What is the lie behind this
Challenge 1148 ny habit? What is the truth behind it?
**
‘In the middle age and in antiquity, people believed in the flat earth.’ This is a famous lie
that is rarely questioned. The historian Reinhard Krüger has shown that the lie is most
of all due to the writers Thomas Paine (1794) and Washington Irving (1928). Fact is that
since Aristotle, everybody believed in a spherical Earth.

Dvipsbugw
 670 intermezzo: the brain, l anguage and the human condition

Observations
Knowledge is a sophisticated statement of

“ ignorance.
Attributed to Karl Popper

The collection of a large number of true statements about a type of observations, i.e. of ”
a large number of facts, is called knowledge. Where the domain of observations is suffi-
Ref. 652 ciently extended, one speaks of a science. A scientist is thus somebody who collects know-
ledge.* We found above that an observation is classified input into the memory of several Dvipsbugw
people. Since there is motion all around, to describe all these observations is a mammoth
task. As for every large task, to a large extent the use of appropriate tools determines the
degree of success that can be achieved. These tools, in physics and in all other sciences,
fall in three groups: tools for the collection of observations, tools to communicate obser-
vations and tools to communicate relations between observations. The latter group has
been already discussed in the section on language and on mathematics. We just touch on

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the other two.

Have enough observations been recorded?

Every generation is inclined to define ‘the end of

“ physics’ as coincident with the end of their

scientific contributions.
Julian Schwinger**

Physics is an experimental science; it rests on the collection of observations. To realize

this task effectively, all sorts of instruments, i.e. tools that facilitate observations, have
been developed and built. Microscopes, telescopes, oscilloscopes, as well as thermomet-
”
ers, hygrometers, manometers, pyrometers, spectrometers amongst others are familiar
examples. The precision of many of these tools is being continuously improved even
today; their production is a sizeable part of modern industrial activity, examples being
electrical measuring apparatus and diagnostic tools for medicine, chemistry and biology.
Instruments can be as small as a tip of a few tungsten atoms to produce an electron beam
of a few volts, and as large as 27 km in circumference, producing an electron beam with
more than 100 GV effective accelerating voltage. Instruments have been built that contain
and measure the coldest known matter in the universe. Other instruments can measure
length variations of much less than a proton diameter over kilometre long distances. In-
Copyright © Christoph Schiller November 1997–May 2006

* The term ‘scientist’ is a misnomer peculiar to the English language. Properly speaking, a ‘scientist’ is a
follower of scientism, an extremist philosophical school that tried to resolve all problems through science.
For this reason, some religious sects have the term in their name. Since the English language did not have a
shorter term to designate ‘scientific persons’, as they used to be called, the term ‘scientist’ started to appear in
the United States, from the eighteenth century onwards. Nowadays the term is used in all English-speaking
countries – but not outside them, fortunately.
** Julian Seymour Schwinger (1918–1994), US-American infant prodigy. He was famous for his clear think-
ing and his excellent lectures. He worked on waveguides and synchroton radiation, made contributions to
nuclear physics and developed quantum electrodynamics. For the latter he received the 1965 Nobel Prize
in physics together with Tomonaga and Feynman. He was a thesis advisor to many famous physicists and
Ref. 653 wrote several excellent and influential textbooks. Nevertheless, at the end of his life, he became strangely
interested in a hoax turned sour: cold fusion.

Dvipsbugw
 observations 671

struments have been put deep inside the Earth, on the Moon, on several planets, and have
been sent outside the Solar system.
In this walk, instruments are not described; many good textbooks on this topic are
Ref. 654, Ref. 655 available. Most observations collected by instruments are not mentioned here. The most
important results in physics are recorded in standard publications, such as the Landolt–
Ref. 656 Börnstein series and the physics journals (Appendix E gives a general overview of inform-
ation sources).
Will there be significant new future observations in the domain of the fundament-
als of motion? At present, in this specific domain, even though the number of physicists Dvipsbugw
and publications is at an all-time high, the number of new experimental discoveries has
been steadily diminishing for many years and is now fairly small. The sophistication and
investment necessary to obtain new results has become extremely high. In many cases,
measuring instruments have reached the limits of technology, of budgets or even those
of nature. The number of new experiments that produce results showing no deviation
from theoretical predictions is increasing steadily. The number of historical papers that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

try to enliven dull or stalled fields of enquiry are increasing. Claims of new effects which
turn out to be false, due to measurement errors, self-deceit or even fraud have become
so frequent that scepticism has become a common response. Although in many domains
of science, including physics, discoveries are still expected, on the fundamentals of mo-
tion the arguments just presented seem to show that new observations are only a remote
possibility. The task of collecting observations on the foundations of motion (though not
on other topics of physics) seems to be complete. Indeed, most observations described
here were obtained before the end of the twentieth century. We are not too early with our
walk.
Measure what is measurable; make measurable

Ref. 657
“ what is not.
Wrongly attributed to Galileo.

”
Are all physical observables known?
Scientists have odious manners, except when

“ you prop up their theory; then you can borrow

money from them.
Mark Twain

The most practical way to communicate observations was developed a long time ago:
”
Copyright © Christoph Schiller November 1997–May 2006

by measurements. A measurement allows effective communication of an observation to

other times and places. This is not always as trivial as it sounds; for example, in the Middle
Ages people were unable to compare precisely the ‘coldness’ of the winters of two different
years! The invention of the thermometer provided a reliable solution to this requirement.
A measurement is thus the classification of an observation into a standard set of observa-
tions; to put it simply, a measurement is a comparison with a standard. This definition of
a measurement is precise and practical, and has therefore been universally adopted. For
example, when the length of a house is measured, this aspect of the house is classified
into a certain set of standard lengths, namely the set of lengths defined by multiples of a
unit. A unit is the abstract name of the standard for a certain observable. Numbers and
units allow the most precise and most effective communication of measurement results.

Dvipsbugw
672 intermezzo: the brain, l anguage and the human condition

For all measurable quantities, practical standard units and measurement methods
have been defined; the main ones are listed and defined in Appendix B. All units are
derived from a few fundamental ones; this is ultimately due to our limited number of
senses: length, time and mass are related to sight, hearing and touch. Our limited num-
ber of senses is, in turn, due to the small number of observables of nature.
We call observables the different measurable aspects of a system. Most observables,
such as size, speed, position, etc. can be described by numbers, and in this case they
are quantities, i.e. multiples of some standard unit. Observables are usually abbreviated
by (mathematical) symbols, usually letters from some alphabet. For example, the sym- Dvipsbugw
bol c commonly specifies the velocity of light. For most observables, standard symbols
have been defined by international bodies.* The symbols for the observables that describe
the state of an object are also called variables. Variables on which other observables de-
pend are often called parameters. (Remember: a parameter is a variable constant.) For
example, the speed of light is a constant, the position a variable, the temperature is often
a parameter, on which the length of an object, for example, can depend. Note that not all

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

observables are quantities; in particular, parities are not multiples of any unit.
Today the task of defining tools for the communication of observations can be con-
sidered complete. (For quantities, this is surely correct; for parity-type observables there
could be a few examples to be discovered.) This is a simple and strong statement. Even
the BIPM, the Bureau International des Poids et Mesures, has stopped adding new units.**
As a note, the greatness of a physicist can be ranked by the number of observables
he has introduced. Even a great scientist such as Einstein, who discovered many ‘laws’ of
nature, only introduced one new observable, namely the metric tensor for the description
of gravity. Following this criterion – as well as several others – Maxwell is the most import-
ant physicist, having introduced electric and magnetic fields, the vector potential, and
several other material dependent observables. For Heisenberg, Dirac and Schrödinger,
the wave function describing electron motion could be counted as half an observable (as
it is a quantity necessary to calculate measurement results, but not itself an observable).
Incidentally, even the introduction of any term that is taken up by others is a rare event;
‘gas’, ‘entropy’ and only a few others are such examples. It has always been much more
difficult to discover an observable than to discover a ‘law’; usually, observables are de-
veloped by many people cooperating together. Indeed, many ‘laws’ bear people’s names,
but almost no observables.
If the list of observables necessary to describe nature is complete, does this mean that
all the patterns or rules of nature are known? No; in the history of physics, observables
Copyright © Christoph Schiller November 1997–May 2006

were usually defined and measured long before the precise rules connecting them were
found. For example, all observables used in the description of motion itself, such as time,
position and its derivatives, momentum, energy and all the thermodynamic quantities,
were defined before or during the nineteenth century, whereas the most precise versions

* All mathematical symbols used in this walk, together with the alphabets from which they are taken, are
listed in Appendix A on notation. They follow international standards whenever they are defined. The stand-
ard symbols of the physical quantities, as defined by the International Standards Organization (ISO), the
International Union of Pure and Applied Physics (IUPAP) and the International Union of Pure and Applied
Chemistry (IUPAC), can be found for example in the bible, i.e. the CRC Handbook of Chemistry and Physics,
CRC Press, Boca Raton, 1992.
** The last, the katal or mol~s, was introduced in 1999. Physical units are presented in Appendix B.

Dvipsbugw
observations 673

of the patterns or ‘laws’ of nature connecting them, special relativity and non-equilibrium
thermodynamics, have been found only in the twentieth century. The same is true for
all observables connected to electromagnetic interaction. The corresponding patterns
of nature, quantum electrodynamics, was discovered long after the corresponding ob-
servables. The observables that were discovered last were the fields of the strong and the
weak nuclear interactions. Also, in this case, the patterns of nature were formulated much
later.*

Do observations take time? Dvipsbugw

An observation is an interaction with some part of nature leading to the production of a

record, such as a memory in the brain, data on a tape, ink on paper, or any other fixed
process applied to a support. The necessary irreversible interaction process is often called
writing the record. Obviously, writing takes a certain amount of time; zero interaction
time would give no record at all. Therefore any recording device, including our brain,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

always records some time average of the observation, however short it may be.
What we call a fixed image, be it a mental image or a photograph, is always the time
average of a moving situation. Without time averaging, we would have no fixed memories.
On the other hand, any time averaging introduces a blur that hides certain details; and in
our quest for precision, at a certain moment, these details are bound to become important.
The discovery of these details will begin in the second part of the walk, the one centred
on quantum theory. In the third part of our mountain ascent we will discover that there
is a shortest possible averaging time. Observations of that short duration show so many
details that even the distinction between particles and empty space is lost. In contrast,
our concepts of everyday life appear only after relatively long time averages. The search
for an average-free description of nature is one of the big challenges of our adventure.

Is induction a problem in physics?

Nur gesetzmäßige Zusammenhänge sind

“ denkbar.**
Ludwig Wittgenstein, Tractatus, 6.361
There is a tradition of opposition between
”
“ adherents of induction and of deduction. In my
view it would be just as sensible for the two ends
of a worm to quarrel.
Alfred North Whitehead

”
Copyright © Christoph Schiller November 1997–May 2006

Induction is the usual term used for the act of making, from a small and finite number
of experiments, general conclusions about the outcome of all possible experiments per-
formed in other places, or at other times. In a sense, it is the technical term for sticking
out one’s neck, which is necessary in every scientific statement. Induction has been a ma-
jor topic of discussion for science commentators. Frequently one finds the remark that

* Is it possible to talk about observations at all? It is many a philosopher’s hobby to discuss whether there
actually is an example for an ‘Elementarsatz’ mentioned by Wittgenstein in his Tractatus. There seems to
be at least one that fits: Differences exist. It is a simple sentence; in the third part of our walk, it will play a
central role.
** Only connexions that are subject to law are thinkable.

Dvipsbugw
 674 intermezzo: the brain, l anguage and the human condition

knowledge in general, and physics in particular, relies on induction for its statements. Ac-
cording to some, induction is a type of hidden belief that underlies all sciences but at the
same time contrasts with them.
To avoid wasting energy, we make only a few remarks. The first can be deduced from
a simple experiment. Try to convince a critic of induction to put their hand into a fire.
Nobody who honestly calls induction a belief should conclude from a few unfortunate
experiences in the past that such an act would also be dangerous in the future... In short,
somehow induction works.
A second point is that physical universal statements are always openly stated; they are Dvipsbugw
never hidden. The refusal to put one’s hand into a fire is a consequence of the invariance
of observations under time and space translations. Indeed, general statements of this type
form the very basis of physics. However, no physical statement is a belief only because
it is universal; it always remains open to experimental checks. Physical induction is not
a hidden method of argumentation, it is an explicit part of experimental statements. In
fact, the complete list of ‘inductive’ statements used in physics is given in the table on

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

page 200. These statements are so important that they have been given a special name:
they are called symmetries. The table lists all known symmetries of nature; in other words,
it lists all inductive statements used in physics.
Perhaps the best argument for the use of induction is that there is no way to avoid it
when one is thinking. There is no way to think, to talk or to remember without using con-
cepts, i.e. without assuming that most objects or entities have the same properties over
time. There is also no way to communicate with others without assuming that the obser-
vations made from the other’s viewpoint are similar to one’s own. There is no way to think
without symmetry and induction. Indeed, the concepts related to symmetry and induc-
Page 643 tion, such as space and time, belong to the fundamental concepts of language. The only
Ref. 652 sentences which do not use induction, the sentences of logic, do not have any content
Challenge 1149 n (Tractatus, 6.11). Indeed, without induction, we cannot classify observations at all! Evolu-
tion has given us memory and a brain because induction works. To criticize induction is
not to criticize natural sciences, it is to criticize the use of thought in general. We should
never take too seriously people who themselves do what they criticize in others; sporadic-
ally pointing out the ridicule of this endeavour is just the right amount of attention they
deserve.
The topic could be concluded here, were it not for some interesting developments in
modern physics that put two additional nails in the coffin of arguments against induction.
First, in physics whenever we make statements about all experiments, all times or all
Copyright © Christoph Schiller November 1997–May 2006

velocities, such statements are actually about a finite number of cases. We know today
that infinities, both in size and in number, do not occur in nature. The infinite number of
cases appearing in statements in classical physics and in quantum mechanics are apparent,
not real, and due to human simplifications and approximations. Statements that a certain
experiment gives the same result ‘everywhere’ or that a given equation is correct for ‘all
times’, always encompass only a finite number of examples. A great deal of otherwise often
instinctive repulsion to such statements is avoided in this way. In the sciences, as well as
in this book, ‘all’ never means an infinite number of cases.
Finally, it is well known that extrapolating from a few cases to many is false when the
few cases are independent of each other. However, this conclusion is correct if the cases
are interdependent. From the fact that somebody found a penny on the street on two

Dvipsbugw
 the quest for precision and its implications 675

subsequent months, cannot follow that he will find one the coming month. Induction
is only correct if we know that all cases have similar behaviour, e.g. because they follow
from the same origin. For example, if a neighbour with a hole in his pocket carries his
salary across that street once a month, and the hole always opens at that point because of
the beginning of stairs, then the conclusion would be correct. It turns out that the results
of modern physics encountered in the third part of our walk show that all situations in
nature are indeed interdependent, and thus we prove in detail that what is called ‘induc-
tion’ is in fact a logically correct conclusion.
In the progress of physics, the exception usually
Dvipsbugw

The quest for precision and its implications

“ turned out to be the general case.

”
Der Zweck der Philosophie ist die logische

“ Klärung der Gedanken.*

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ludwig Wittgenstein, Tractatus, 4.112

To talk well about motion means to talk precisely. Precision requires avoiding
hree common mistakes in the description of nature.
First, concepts should never have a contradiction built into their definition. For ex-
”
ample, any phenomenon occurring in nature evidently is a ‘natural’ phenomenon; there-
fore, to talk about either ‘supernatural’ phenomena or ‘unnatural’ phenomena is a mis-
take that nobody interested in motion should let go unchallenged; such terms contain
a logical contradiction. Naturally, all observations are natural. Incidentally, there is a re-
Ref. 658 ward of more than a million dollars for anybody proving the opposite. In over twenty
years, nobody has yet been able to collect it.
Second, concepts should not have unclear or constantly changing definitions. Their
content and their limits must be kept constant and explicit. The opposite of this is often
Ref. 659 encountered in crackpots or populist politicians; it distinguishes them from more reliable
thinkers. Physicists can also fall into the trap; for example, there is, of course, only one
single (physical) universe, as even the name says. To talk about more than one universe
is an increasingly frequent error.
Third, concepts should not be used outside their domain of application. It is easy to suc-
cumb to the temptation to transfer results from physics to philosophy without checking
the content. An example is the question: ‘Why do particles follow the laws of nature?’ The
flaw in the question is due to a misunderstanding of the term ‘laws of nature’ and to a con-
Copyright © Christoph Schiller November 1997–May 2006

fusion with the laws of the state. If nature were governed by ‘laws’, they could be changed
by parliament. Remembering that ‘laws of nature’ simply means ‘pattern’, ‘property’ or ‘de-
scription of behaviour’, and rephrasing the question correctly as ‘Why do particles behave
in the way we describe their behaviour?’ one can recognize its senselessness.
In the course of our walk, we will often be tempted by these three mistakes. A few such
situations follow, with the ways of avoiding them.
Consistency is the last refuge of the

“ unimaginative.

* Philosophy aims at the logical clarification of thoughts.

Oscar Wilde

”
Dvipsbugw
 676 intermezzo: the brain, l anguage and the human condition

What are interactions? – No emergence

The whole is always more than the sum of its

“ parts.
Aristotle, Metaphysica, 10f–1045a.

In the physical description of nature, the whole is always more than the sum of its parts. ”
Actually, the difference between the whole and the sum of its parts is so important that
it has a special name: the interaction between the parts. For example, the energy of the
whole minus the sum of the energies of its parts is called the energy of interaction. In fact, Dvipsbugw
the study of interactions is the main topic of physics. In other words, physics is concerned
primarily with the difference between the parts and the whole, contrary to what is often
suggested by bad journalists or other sloppy thinkers.
Note that the term ‘interaction’ is based on the general observation that anything that
affects anything else is, in turn, affected by it; interactions are reciprocal. For example, if
one body changes the momentum of another, then the second changes the momentum

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of the first by the same (negative) amount. The reciprocity of interactions is a result of
conservation ‘laws’. The reciprocity is also the reason that somebody who uses the term
‘interaction’ is considered a heretic by monotheistic religions, as theologians regularly
point out. They repeatedly stress that such a reciprocity implicitly denies the immutability
Challenge 1150 ny of the deity. (Are they correct?)
The application of the definition of interaction also settles the frequently heard ques-
tion of whether in nature there are ‘emergent’ properties, i.e. properties of systems that
cannot be deduced from the properties of their parts and interactions. By definition, there
are no emergent properties. ‘Emergent’ properties can only appear if interactions are ap-
Page 260 proximated or neglected. The idea of ‘emergent’ properties is a product of minds with
restricted horizons, unable to see or admit the richness of consequences that general prin-
Ref. 660 ciples can produce. In defending the idea of emergence, one belittles the importance of
interactions, working, in a seemingly innocuous, maybe unconscious, but in fact sneaky
way, against the use of reason in the study of nature. ‘Emergence’ is a belief.
The simple definition of interaction given above sounds elementary, but it leads to
surprising conclusions. Take the atomic idea of Democritus in its modern form: nature
is made of vacuum and of particles. The first consequence is the paradox of incomplete
description: experiments show that there are interactions between vacuum and particles.
However, interactions are differences between parts and the whole, in this case between
vacuum and particles on the one hand, and the whole on the other. We thus have deduced
Copyright © Christoph Schiller November 1997–May 2006

that nature is not made of vacuum and particles alone.

The second consequence is the paradox of overcomplete description: experiments also
show that interactions happen through exchange of particles. However, we have counted
particles already as basic building blocks. Does this mean that the description of nature
Challenge 1151 n by vacuum and particles is an overdescription, counting things twice?
Page 1018 We will resolve both paradoxes in the third part of the mountain ascent.

Dvipsbugw
 the quest for precision and its implications 677

What is existence?
You know what I like most?

Ref. 661
“ Rhetorical questions.

”
Assume a friend tells you ‘I have seen a grampus today!’ You would naturally ask what it
looks like. What answer do we expect? We expect something like ‘It’s an animal with a
certain number of heads similar to a X, attached to a body like a Y, with wings like a Z, it
make noises like a U and it felt like a V ’ – the letters denoting some other animal or object. Dvipsbugw
Generally speaking, in the case of an object, this scene from Darwin’s voyage to South
America shows that in order to talk to each other, we first need certain basic, common
concepts (‘animal’, ‘head’, ‘wing’, etc.). In addition, for the definition of a new entity we
need a characterization of its parts (‘size’, ‘colour’), of the way these parts relate to each
other, and of the way that the whole interacts with the outside world (‘feel’, ‘sound’). In
other words, for an object to exist, we must be able to give a list of relations with the
outside world. An object exists if we can interact with it. (Is observation sufficient to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1152 n determine existence?)
For an abstract concept, such as ‘time’ or ‘superstring’, the definition of existence has to
be refined only marginally: (physical) existence is the effectiveness to describe interactions
accurately. This definition applies to trees, time, virtual particles, imaginary numbers,
entropy and so on. It is thus pointless to discuss whether a physical concept ‘exists’ or
whether it is ‘only’ an abstraction used as a tool for descriptions of observations. The two
possibilities coincide. The point of dispute can only be whether the description provided
by a concept is or is not precise.
For mathematical concepts, existence has a somewhat different meaning: a mathem-
atical concept is said to exist if it has no built-in contradictions. This is a much weaker
requirement than physical existence. It is thus incorrect to deduce physical existence from
mathematical existence. This is a frequent error; from Pythagoras’ times onwards it was of-
ten stated that since mathematical concepts exist, they must therefore also exist in nature.
Historically, this error occurred in the statements that planet orbits ‘must’ be circles, that
planet shapes ‘must’ be spheres or that physical space ‘must’ be Euclidean. Today this is
still happening with the statements that space and time ‘must’ be continuous and that
nature ‘must’ be described by sets. In all these cases, the reasoning is wrong. In fact, the
continuous attempts to deduce physical existence from mathematical existence hide that
the opposite is correct: a short reflection shows that mathematical existence is a special
case of physical existence.
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1153 n
We note that there is also a different type of existence, namely psychological existence.
A concept can be said to exist psychologically if it describes human internal experience.
Thus a concept can exist psychologically even if it does not exist physically. It is easy to find
Challenge 1154 n examples from the religions or from systems that describe inner experiences. Also myths,
legends and comic strips define concepts that only exist psychologically, not physically.
In our walk, whenever we talk about existence, we mean physical existence only.

Dvipsbugw
 678 intermezzo: the brain, l anguage and the human condition

Do things exist?
Wer Wissenschaft und Kunst besitzt,

“ Hat auch Religion;

Wer jene beiden nicht besitzt,
Der habe Religion.*
Johann Wolfgang von Goethe, Zahme Xenien,
IX

Using the above definition of existence, the question becomes either trivial or imprecise.
It is trivial in the sense that things necessarily exist if they describe observations, since
they were defined that way. But perhaps the questioner meant to ask: Does reality exist
” Dvipsbugw

independently of the observer?

Using the above, this question can be rephrased: ‘Do the things we observe exist in-
dependently of observation?’ After thousands of years of extensive discussion by profes-
sional philosophers, logicians, sophists and amateurs the answer is the same: it is ‘Yes’,
because the world did not change after great-grandmother died. The disappearance of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

observers does not seem to change the universe. These experimental findings can be cor-
roborated by inserting the definition of ‘existence’ into the question, which then becomes:
‘Do the things we observe interact with other aspects of nature when they do not interact
with people?’ The answer is evident. Recent popular books on quantum mechanics fan-
tasize about the importance of the ‘mind’ of observers – whatever this term may mean;
they provide pretty examples of authors who see themselves as irreplaceable, seemingly
having lost the ability to see themselves as part of a larger entity.
Of course there are other opinions about the existence of things. The most famous is
that of the Irishman George Berkeley (1685–1753) who rightly understood that thoughts
based on observation alone, if spread, would undermine the basis of the religious organ-
ization of which he was one of the top managers. To counteract this tendency, in 1710
he published A Treatise Concerning the Principles of Human Knowledge, a book denying
the existence of the material world. This reactionary book became widely known in like-
minded circles (it was a time when few books were written) even though it is based on
a fundamentally flawed idea: it assumes that the concept of ‘existence’ and that of ‘world’
Challenge 1155 e can be defined independently. (You may be curious to try the feat.)
Berkeley had two aims when he wrote his book. First, he tried to deny the capacity of
people to arrive at judgements on nature or on any other matter from their own experience.
Second, he also tried to deny the ontological reach of science, i.e. the conclusions one can
draw from experience on the questions about human existence. Even though Berkeley is
Copyright © Christoph Schiller November 1997–May 2006

generally despised nowadays, he actually achieved his main aim: he was the originator of
the statement that science and religion do not contradict, but complement each other. By
religion, Berkeley did not mean either morality or spirituality; every scientists is a friend
of both of these. By religion, Berkeley meant that the standard set of beliefs that he stood
for is above the deductions of reason. This widely cited statement, itself a belief, is still
held dearly by many even to this day. However, when searching for the origin of motion,
all beliefs stand in the way, including this one. Carrying beliefs is like carrying oversized
baggage: it prevents one from reaching the top of Motion Mountain.

* He who possesses science and art, also has religion; he who does not possess the two, better have religion.

Dvipsbugw
 the quest for precision and its implications 679

Does the void exist?

Teacher: ‘What is found between the nucleus

“ and the electrons?’

Student: ‘Nothing, only air.’
Natura abhorret vacuum.
”
“
In philosophical discussions ‘void’ is usually defined as ‘non-existence’. It then becomes
Antiquity

” Dvipsbugw
a game of words to ask for a yes or no answer to the question ‘Does the void exist?’ The
expression ‘the existence of non-existence’ is either a contradiction of terms or is at least
unclearly defined; the topic would not seem to be of great interest. However, similar ques-
tions do appear in physics, and a physicist should be prepared to notice the difference
of this from the previous one. Does a vacuum exist? Does empty space exist? Or is the
world ‘full’ everywhere, as the more conservative biologist Aristotle maintained? In the
past, people have been killed for giving an answer that was unacceptable to authorities.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

It is not obvious, but it is nevertheless important, that the modern physical concepts
of ‘vacuum’ and ‘empty space’ are not the same as the philosophical concept of ‘void’.
‘Vacuum’ is not defined as ‘non-existence’; on the contrary, it is defined as the absence
of matter and radiation. Vacuum is an entity with specific observable properties, such
as its number of dimensions, its electromagnetic constants, its curvature, its vanishing
mass, its interaction with matter through curvature and through its influence on decay,
etc. (A table of the properties of a physical vacuum is given on page 581.) Historically, it
took a long time to clarify the distinction between a physical vacuum and a philosophical
void. People confused the two concepts and debated the existence of the vacuum for more
than two thousand years. The first to state that it existed, with the courage to try to look
through the logical contradiction at the underlying physical reality, were Leucippus and
Democritus, the most daring thinkers of antiquity. Their speculations in turn elicited the
reactionary response of Aristotle, who rejected the concept of vacuum. Aristotle and his
disciples propagated the belief about nature’s horror of the vacuum.
The discussion changed completely in the seventeenth century, when the first experi-
mental method to realize a vacuum was devised by Torricelli.* Using mercury in a glass
Challenge 1156 n tube, he produced the first laboratory vacuum. Can you guess how? Arguments against
the existence of the vacuum again appeared around 1900, when it was argued that light
needed ‘aether’ for its propagation, using almost the same arguments that had been used
two hundred years earlier, but in different words. However, experiments failed to detect
Copyright © Christoph Schiller November 1997–May 2006

any of the supposed properties of this unclearly defined concept. Experiments in the field
of general relativity showed that a vacuum can move – though in a completely different
way from the way in which the aether was expected to move – that the vacuum can be
bent, but it then tends to return to its shape. Then, in the late twentieth century, quantum
field theory again argued against the existence of a true vacuum and in favour of a space
full of virtual particle–antiparticle pairs, culminating in the discussions around the cos-
Page 996 mological constant.

* Evangelista Torricelli (b. 1608 Faenza, d. 1647 Florence), Italian physicist, pupil and successor to Galileo.
The (non-SI) pressure unit ‘torr’ is named after him.

Dvipsbugw
 680 intermezzo: the brain, l anguage and the human condition

The question ‘Does the void exist?’ is settled conclusively only in the third part of this
Page 1020 walk, in a rather surprising way.

Is nature infinite?
It is certain and evident to our senses, that in

“ the world some things are in motion. Now

whatever is moved is moved by another... If that
by which it is moved be itself moved, then this
also needs to be to be moved by another, and Dvipsbugw
that by another again. But this cannot go on to
infinity, because then there would be no first
mover and consequently, no other mover, seeing
that subsequent movers move only inasmuch as
they are moved by the first mover, as the staff
moves only because it is moved by the hand.
Therefore it is necessary to arrive at a first

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

mover, moved by no other; and this everyone
understands to be god.
Thomas Aquinas (c. 1225–1274) Summa
Theologiae, I, q. 2.

Most of the modern discussions about set theory centre on ways to defining the term ‘set’
for various types of infinite collections. For the description of motion this leads to two
questions: Is the universe infinite? Is it a set? We begin with the first one. Illuminating
”
the question from various viewpoints, we will quickly discover that it is both simple and
imprecise.
Do we need infinite quantities to describe nature? Certainly, in classical and quantum
physics we do, e.g. in the case of space-time. Is this necessary? We can say already a few
things.
Any set can be finite in one aspect and infinite in another. For example, it is possible
to proceed along a finite mathematical distance in an infinite amount of time. It is also
possible to travel along any distance whatsoever in a given amount of mathematical time,
making infinite speed an option, even if relativity is taken into account, as was explained
Page 297 earlier.
Despite the use of infinities, scientists are still limited. We saw above that many types of
Page 648 infinities exist. However, no infinity larger than the cardinality of the real numbers plays
a role in physics. No space of functions or phase space in classical physics and no Hilbert
Copyright © Christoph Schiller November 1997–May 2006

Ref. 648 space in quantum theory has higher cardinality. Despite the ability of mathematicians to
define much larger kinds of infinities, the description of nature does not need them. Even
the most elaborate descriptions of motion use only the infinity of the real numbers.
But is it possible at all to say of nature or of one of its aspects that it is indeed infin-
Challenge 1157 n ite? Can such a statement be compatible with observations? No. It is evident that every
statement that claims that something in nature is infinite is a belief, and is not backed by
observations. We shall patiently eliminate this belief in the following.
The possibility of introducing false infinities make any discussion on whether human-
Ref. 649 ity is near the ‘end of science’ rather difficult. The amount of knowledge and the time
required to discover it are unrelated. Depending on the speed with which one advances
through it, the end of science can be near or unreachable. In practice, scientists have

Dvipsbugw
 the quest for precision and its implications 681

thus the power to make science infinite or not, e.g. by reducing the speed of progress. As
scientists need funding for their work, one can guess the stand that they usually take.
In short, the universe cannot be proven to be infinite. But can it be finite? At first sight,
this would be the only possibility left. (It is not, as we shall see.) But even though many
have tried to describe the universe as finite in all its aspects, no one has yet been successful.
In order to understand the problems that they encountered, we continue with the other
question mentioned above:

Is the universe a set? Dvipsbugw

A simple observation leads us to question whether the universe is a set. For 2500 years
Ref. 662 it has been said that the universe is made of vacuum and particles. This implies that the
universe is made of a certain number of particles. Perhaps the only person to have taken
this conclusion to the limit was the English astrophysicist Arthur Eddington (1882–1944),
who wrote:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

I believe there are 15,747,724,136,275,002,577,605,653,961,181,555,468,044,717,914-
,527,116,709,366,231,425,076,185,631,031,296 protons in the universe and the same
Ref. 663 number of electrons.

Eddington was ridiculed over and over again for this statement and for his beliefs that
lead up to it. His arguments were indeed based on his personal preferences for certain
pet numbers. However, we should not laugh too loudly. In fact, for 2500 years almost all
scientists have thought along the same line, the only difference being that they have left
the precise number unspecified! In fact, any other number put into the above sentence
would be equally ridiculous. Avoiding specifying it is just a cowards’ way of avoiding
looking at this foggy aspect of the particle description of nature.
Is there a particle number at all in nature? If you smiled at Eddington’s statement, or
if you shook your head over it, it may mean that you instinctively believe that nature
is not a set. Is this so? Whenever we define the universe as the totality of events, or as
the totality of all space-time points and objects, we imply that space-time points can be
distinguished, that objects can be distinguished and that both can be distinguished from
each other. We thus assume that nature is separable and a set. But is this correct? The
question is important. The ability to distinguish space-time points and particles from
each other is often called locality. Thus the universe is separable or a set if and only if our
Copyright © Christoph Schiller November 1997–May 2006

description of it is local.* And in everyday life, locality is observed without exception.

In daily life we also observe that nature is separable and a whole at the same time.
It is a ‘many that can be thought as one’: in daily life nature is a set. Indeed, the basic
characteristic of nature is its diversity. In the world around us we observe changes and
differences; we observe that nature is separable. Furthermore, all aspects of nature belong
together: there are relations between these aspects, often called ‘laws,’ stating that the
different aspects of nature form a whole, usually called the universe.

* In quantum mechanics also other, less clear definitions of locality are used. We will mention them in the
second part of this text. The issue mentioned here is a different, more fundamental one, and not connected
with that of quantum theory.

Dvipsbugw
 682 intermezzo: the brain, l anguage and the human condition

In other words, the possibility of describing observations with the help of ‘laws’ follows
from our experience of the separability of nature. The more precisely the separability
is specified, the more precisely the ‘laws’ can be formulated. Indeed, if nature were not
separable or were not a unity, we could not explain why stones fall downwards. Thus we
are led to speculate that we should be able to deduce all ‘laws’ from the fact that nature is
separable.
In addition, only the separability allows us to describe nature at all. A description is a
classification, that is, a mapping between certain aspects of nature and certain concepts.
All concepts are sets and relations. Since the universe is separable, it can be described Dvipsbugw
with the help of sets and relations. Both are separable entities with distinguishable parts.
A precise description is commonly called an understanding. In short, the universe is com-
prehensible only because it is separable.
Moreover, only the separability of the universe makes our brain such a good instru-
ment. The brain is built from a large number of connected components, and only the
brain’s separability allows it to function. In other words, thinking is only possible because

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

nature is separable.
Finally, only the separability of the universe allows us to distinguish reference frames,
and thus to define all symmetries at the basis of physical descriptions. And in the same
way that separability is thus necessary for covariant descriptions, the unity of nature is
necessary for invariant descriptions. In other words, the so-called ‘laws’ of nature are
based on the experience that nature is both separable and unifiable – that it is a set.
These arguments seem overwhelmingly to prove that the universe is a set. However,
these arguments apply only to everyday experience, everyday dimensions and everyday
energies. Is nature a set also outside the domains of daily life? Are objects different at all
energies, even when they are looked at with the highest precision possible? We have three
open issues left: the issue of the number of particles in the universe; the circular defini-
tion of space, time and matter; and the issue as to whether describing nature as made of
particles and void is an overdescription, an underdescription, or neither. These three is-
sues make us doubt whether objects are countable at all energies. We will discover in the
Page 1046 third part of our mountain ascent that this is not the case in nature. The consequences will
be extensive and fascinating. As an example, try to answer the following: if the universe
Challenge 1158 n is not a set, what does that mean for space and time?

Does the universe exist?

Each progressive spirit is opposed by a
Copyright © Christoph Schiller November 1997–May 2006

“ thousand men appointed to guard the past.

Maurice Maeterlink

Following the definition above, existence of a concept means its usefulness to describe
”
interactions. There are two common definitions of the concept of ‘universe’. The first is
the totality of all matter, energy and space-time. But this usage results in a strange con-
sequence: since nothing can interact with this totality, we cannot claim that the universe
exists.
So let us take the more restricted view, namely that the universe is only the totality of
all matter and energy. But also in this case it is impossible to interact with the universe.
Challenge 1159 n Can you give a few arguments to support this?

Dvipsbugw
 the quest for precision and its implications 683

In short, we arrive at the conclusion that the universe does not exist. We will indeed
Page 1049 confirm this result in more detail later on in our walk. In particular, since the universe
does not exist, it does not make sense to even try to answer why it exists. The best answer
Ref. 616 might be: because of furiously sleeping, colourless green ideas.

What is creation?
(Gigni) De nihilo nihilum, in nihilum nil posse

Ref. 664
“ reverti.*
Persius, Satira, III, v. 83-84.

Anaxagoras, discovering the ancient theory that

”
Dvipsbugw

“ nothing comes from nothing, decided to

abolish the concept of creation and introduced
in its place that of discrimination; he did not
hesitate to state, in effect, that all things are
mixed to the others and that discrimination

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

produces their growth.
Anonymous fragment, Middle Ages.

”
Ref. 665

The term ‘creation’ is often heard when talking about nature. It is used in various contexts
with different meanings.
One speaks of creation as the characterization of human actions, such as observed
in an artist painting or a secretary typing. Obviously, this is one type of change. In the
classification of change introduced at the beginning of our walk, the changes cited are
movements of objects, such as the electrons in the brain, the molecules in the muscles,
the material of the paint, or the electrons inside the computer. This type of creation is
thus a special case of motion.
One also speaks of creation in the biological or social sense, such as in ‘the creation
of life’, or ‘creation of a business’, or ‘the creation of civilization’. These events are forms of
growth or of self-organization; again, they are special cases of motion.
Physicists one often say that a lamp ‘creates’ light or that a stone falling into a pond
‘creates’ water ripples. Similarly, they talk of ‘pair creation’ of matter and antimatter. It
Page 759 was one of the important discoveries of physics that all these processes are special types
Page 850 of motion, namely excitation of fields.
In popular writing on cosmology, ‘creation’ is also a term commonly applied, or better
misapplied, to the big bang. However, the expansion of the universe is a pure example
of motion, and contrary to a frequent misunderstanding, the description of the big bang
Copyright © Christoph Schiller November 1997–May 2006

contains no process that does not fall into one of the previous three categories, as shown in
Page 464 the chapter on general relativity. Quantum cosmology provides more reasons to support
Page 463 the fact that the term ‘creation’ is not applicable to the big bang. First, it turns out that the
big bang was not an event. Second, it was not a beginning. Third, it did not provide a choice
from a large set of possibilities. The big bang does not have any properties attributed to
the term ‘creation’.
In summary, we conclude that in all cases, creation is a type of motion. (The same
applies to the notions of ‘disappearance’ and ‘annihilation’.) No other type of creation is
observed in nature. In particular, the naive sense of ‘creation’, namely ‘appearance from

* Nothing (can appear) from nothing, nothing can disappear into nothing.

Dvipsbugw
 684 intermezzo: the brain, l anguage and the human condition

nothing’ – ex nihilo in Latin – is never observed in nature. All observed types of ‘creation’
require space, time, forces, energy and matter for their realization. Creation requires
something to exist already, in order to take place. In addition, precise exploration shows
that no physical process and no example of motion has a beginning. Our walk will show
us that nature does not allow us to pinpoint beginnings. This property alone is sufficient
to show that ‘creation’ is not a concept applicable to what happens in nature. Worse still,
creation is applied only to physical systems; we will discover that nature is not a system
and that systems do not exist.
The opposite of creation is conservation. The central statements of physics are conser- Dvipsbugw
vation theorems: for energy, mass, linear momentum, angular momentum, charge, etc.
In fact, every conservation ‘law’ is a detailed and accurate rejection of the concept of cre-
ation. The ancient Greek idea of atoms already contains this rejection. Atomists stated
that there is no creation and no disappearance, but only motion of atoms. Every trans-
formation of matter is a motion of atoms. In other words, the idea of the atom was a
direct consequence of the negation of creation. It took humanity over 2000 years before

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 220 it stopped locking people in jail for talking about atoms, as had happened to Galileo.
However, there is one exception in which the naive concept of creation does apply:
it describes what magicians do on stage. When a magician makes a rabbit appear from
nowhere, we indeed experience ‘creation’ from nothing. At its best such magic is a form of
entertainment, at its worst, a misuse of gullibility. The idea that the universe results from
either of these two does not seem appealing; on second thought though, maybe looking
at the universe as the ultimate entertainment could open up a fresh and more productive
approach to life.
Voltaire (1694–1778) popularized an argument against creation often used in the past:
we do not know whether creation has taken place or not. Today the situation is different:
we do know that it has not taken place, because creation is a type of motion and, as we
will see in the third part of our mountain ascent, motion did not exist near the big bang.
Have you ever heard the expression ‘creation of the laws of nature’? It is one of the
most common examples of disinformation. First of all, this expression confuses the ‘laws’
with nature itself. A description is not the same as the thing itself; everybody knows that
giving their beloved a description of a rose is different from giving an actual rose. Second,
the expression implies that nature is the way it is because it is somehow ‘forced’ to follow
the ‘laws’ – a rather childish and, what is more, incorrect view. And third, the expression
assumes that it is possible to ‘create’ descriptions of nature. But a ‘law’ is a description,
and a description by definition cannot be created: so the expression makes no sense at all.
Copyright © Christoph Schiller November 1997–May 2006

The expression ‘creation of the laws of nature’ is the epitome of confused thinking.
It may well be that calling a great artist ‘creative’ or ‘divine’, as was common during
the Renaissance, is not blasphemy, but simply an encouragement to the gods to try to
do as well. In fact, whenever one uses the term ‘creation’ to mean anything other than
some form of motion, one is discarding both observations and human reason. It is one
of the last pseudo-concepts of our modern time; no expert on motion should forget this.
It is impossible to escalate Motion Mountain without getting rid of ‘creation’. This is not
Page 962 easy. We will encounter the next attempt to bring back creation in the study of quantum
theory.

Dvipsbugw
 the quest for precision and its implications 685

Every act of creation is first of all an act of

“ destruction.
Pablo Picasso

”
Is nature designed?
In the beginning the universe was created. This

“ has made a lot of people very angry and has

been widely regarded as a bad move.
Douglas Adams

”
Dvipsbugw
The tendency to infer the creation of an object from its simple existence is widespread.
Some people jump to this conclusion every time they see a beautiful landscape. This habit
stems from the triple prejudice that a beautiful scene implies a complex description, in
turn implying complex building instructions, and therefore pointing to an underlying
design.
This chain of thought contains several mistakes. First, in general, beauty is not a con-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

sequence of complexity. Usually it is the opposite: indeed, the study of chaos and of self-
Ref. 666 organization demonstrated how beautifully complex shapes and patterns can be gener-
ated with extremely simple descriptions. True, for most human artefacts, complex descrip-
tions indeed imply complex building processes; a personal computer is a good example
of a complex object with a complex production process. But in nature, this connection
does not apply. We have seen above that even the amount of information needed to con-
struct a human body is about a million times smaller than the information stored in the
brain alone. Similar results have been found for plant architecture and for many other ex-
amples of patterns in nature. The simple descriptions behind the apparent complexities
of nature have been and are still being uncovered by the study of self-organization, chaos,
turbulence and fractal shapes. In nature, complex structures derive from simple processes.
Beware of anyone who says that nature has ‘infinite complexity’: first of all, complexity
is not a measurable entity, despite many attempts to quantify it. In addition, all known
complex system can be described by (relatively) few parameters and simple equations.
Finally, nothing in nature is infinite.
The second mistake in the argument for design is to link a description with an ‘instruc-
tion’, and maybe even to imagine that some unknown ‘intelligence’ is somehow pulling
the strings of the world’s stage. The study of nature has consistently shown that there is
no hidden intelligence and no instruction behind the processes of nature. An instruction
is a list of orders to an executioner. But there are no orders in nature, and no executioners.
Copyright © Christoph Schiller November 1997–May 2006

There are no ‘laws’ of nature, only descriptions of processes. Nobody is building a tree;
the tree is an outcome of the motion of molecules making it up. The genes in the tree do
contain information; but no molecule is given any instructions. What seem to be instruc-
tions to us are just natural movements of molecules and energy, described by the same
patterns taking place in non-living systems. The whole idea of instruction – like that of
‘law’ of nature – is an ideology, born from an analogy with monarchy or even tyranny,
and a typical anthropomorphism.
The third mistake in the argument for design is the suggestion that a complex descrip-
tion for a system implies an underlying design. This is not correct. A complex descrip-
tion only implies that the system has a long evolution behind it. The correct deduction is:

Dvipsbugw
 686 intermezzo: the brain, l anguage and the human condition

something of large complexity exists; therefore it has grown, i.e. it has been transformed
through input of (moderate) energy over time. This deduction applies to flowers, moun-
tains, stars, life, people, watches, books, personal computers and works of art; in fact it
applies to all objects in the universe. The complexity of our environment thus points out
the considerable age of our environment and reminds us of the shortness of our own life.
The lack of basic complexity and the lack of instructions in nature confirm a simple
result: there is not a single observation in nature that implies or requires design or cre-
ation. On the other hand, the variety and intensity of nature’s phenomena fills us with
deep awe. The wild beauty of nature shows us how small a part of nature we actually are, Dvipsbugw
Page 961 both in space and in time.* We shall explore this experience in detail. We shall find that
remaining open to nature’s phenomena in all their overwhelming intensity is central to
the rest of our adventure.
There is a separation between state and church,

“ but not yet between state and science.

Paul Feyerabend

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What is a description?
In theory, there is no difference between theory

“ and practice. In practice, there is.

Following standard vocabulary usage, a description of an observation is a list of the de-

”
Page 677 tails. The above example of the grampus showed this clearly. In other words, a description
of an observation is the act of categorizing it, i.e. of comparing, by identifying or distin-
guishing, the observation with all the other observations already made. A description is
a classification. In short, to describe means to see as an element of a larger set.
A description can be compared to the ‘you are here’ sign on a city tourist map. Out
of a set of possible positions, the ‘you are here’ sign gives the actual one. Similarly, a
description highlights the given situation in comparison with all other possibilities. For
example, the formula a = GM~r 2 is a description of the observations relating motion
to gravity, because it classifies the observed accelerations a according to distance to the
central body r and to its mass M; indeed such a description sees each specific case as
an example of a general pattern. The habit of generalizing is one reason for the often
disturbing dismissiveness of scientists: when they observe something, their professional
training usually makes them classify it as a special case of a known phenomenon and thus
Copyright © Christoph Schiller November 1997–May 2006

keeps them from being surprised or from being exited about it.
A description is thus the opposite of a metaphor; the latter is an analogy relating an
observation with another special case; a description relates an observation with a general
case, such as a physical theory.

* The search for a ‘sense’ in life or in nature is a complicated (and necessary) way to try to face the smallness
of human existence.

Dvipsbugw
 the quest for precision and its implications 687

Felix qui potuit rerum cognoscere causas,

“ atque metus omnis et inexorabile fatum

subjecit pedibus strepitumque acherontis avari.
Vergilius*

Reason, purpose and explanation

Der ganzen modernen Weltanschauung liegt die
”
“ Täuschung zugrunde, daß die sogenannten
Naturgesetze die Erklärungen der
Naturerscheinungen seien.**
Ludwig Wittgenstein, Tractatus, 6.371
Dvipsbugw

— Why are the leaves of most trees green? Because they absorb red and blue light. Why
do they absorb those colours? Because they contain chlorophyll. Why is chlorophyll
green? Because all chlorophyll types contain magnesium between four pyrrole groups,
”
and this chemical combination gives the green colour, as a result of its quantum mech-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

anical energy levels. Why do plants contain chlorophyll? Because this is what land
plants can synthesize. Why only this? Because all land plants originally evolved from
the green algae, who are only able to synthesize this compound, and not the com-
pounds found in the blue or in the red algae, which are also found in the sea.
— Why do children climb trees, and why do some people climb mountains? Because of
the sensations they experience during their activity: the feelings of achievement, the
symbolic act to go upwards, the wish to get a wider view of the world are part of this
type of adventure.
The ‘why’-questions in the last two paragraphs show the general difference between
reasons and purposes (although the details of these two terms are not defined in the
same way by everybody). A purpose or intention is a classification applied to the actions
of humans or animals; strictly speaking, it specifies the quest for a feeling, namely for
achieving some type of satisfaction after completion of the action. On the other hand, a
reason is a specific relation of a fact with the rest of the universe, usually its past. What
we call a reason always rests outside the observation itself, whereas a purpose is always
internal to it.
Reasons and purposes are the two possibilities of explanations, i.e. the two possible
answers to questions starting with ‘why’. Usually, physics is not concerned with purpose or
with people’s feeling, mainly because its original aim, to talk about motion with precision,
Copyright © Christoph Schiller November 1997–May 2006

does not seem to be achievable in this domain. Therefore, physical explanations of facts
are never purposes, but are always reasons. A physical explanation of an observation is
Ref. 667 always the description of its relation with the rest of nature.***

* ‘Happy he who can know the causes of things and who, free of all fears, can lay the inexorable fate and the
noise of Acheron to his feet.’ (Georg. 2, 490 ss.) Publius Vergilius Maro (70–19 bce ), the great roman poet,
is author of the Aeneis. Acheron was the river crossed by those who had just died and were on their way to
the Hades.
** The whole modern conception of the world is founded on the illusion that the so-called laws of nature
are the explanations of natural phenomena.
*** It is important to note that purposes are not put aside because they pertain to the future, but because
they are inadmissible anthropomorphisms. In fact, for deterministic systems, we can equally say that the

Dvipsbugw
 688 intermezzo: the brain, l anguage and the human condition

This means that – contrary to common opinion – a question starting with ‘why’ is ac-
cessible to physical investigation as long as it asks for a reason and not for a purpose. In
particular, questions such as ‘why do stones fall downwards and not upwards?’ or ‘why do
electrons have that value of mass, and why do they have mass at all?’ or ‘why does space
have three dimensions and not thirty-six?’ can be answered, as these ask for the connec-
tion between specific observations and more general ones. Of course, not all demands
for explanation have been answered yet, and there are still problems to be solved. Our
present trail only leads from a few answers to some of the more fundamental questions
about motion. Dvipsbugw
The most general quest for an explanation derives from the question: why is the uni-
verse the way it is? The topic is covered in our mountain ascent using the two usual ap-
proaches, namely:

Unification and demarcation

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Tout sujet est un; et, quelque vaste qu’il soit, il

“ peut être renfermé dans un seul discours.*

Buffon, Discours sur le style.

Studying the properties of motion, constantly paying attention to increase the accuracy ”
of description, we find that explanations are generally of two types:**
— ‘It is like all such cases; also this one is described by ...’ The situation is recognized as a
special case of a general behaviour.
— ‘If the situation were different, we would have a conclusion in contrast with observa-
tions.’ The situation is recognized as the only possible case.***
In other words, the first approach is to formulate rules or ‘laws’ that describe larger and
larger numbers of observations, and compare the observation with them. This endeavour
is called the unification of physics – by those who like it; those who don’t like it, call it ‘re-
ductionism’. For example, the same rule describes the flight of a tennis ball, the motion of
the tides at the sea shore, the timing of ice ages, and the time at which the planet Venus
ceases to be the evening star and starts to be the morning star. These processes are all con-
sequences of universal gravitation. Similarly, it is not evident that the same rule describes
the origin of the colour of the eyes, the formation of lightning, the digestion of food and
the working of the brain. These processes are described by quantum electrodynamics.
Unification has its most impressive successes when it predicts an observation that has
Copyright © Christoph Schiller November 1997–May 2006

not been made before. A famous example is the existence of antimatter, predicted by Dirac

future is actually a reason for the present and the past, a fact often forgotten.
* Every subject is one and, however vast it is, it can be comprised in a single discourse.
Challenge 1160 n ** Are these the only possible ones?
*** These two cases have not to be confused with similar sentences that seem to be explanations, but that
aren’t:

— ‘It is like the case of ...’ A similarity with another single case is not an explanation.
— ‘If it were different, it would contradict the idea that ...’ A contradiction with an idea or with a theory is
not an explanation.

Dvipsbugw
 the quest for precision and its implications 689

when he investigated the solutions of an equation that describes the precise behaviour of
common matter.
The second procedure in the search for explanations is the elimination of all other
imaginable alternatives in favour of the actually correct one. This endeavour has no com-
monly accepted name: it could be called the demarcation of the ‘laws’ of physics – by those
who like it; others call it ‘anthropocentrism’, or simply ‘arrogance’.
When we discover that light travels in such a way that it takes the shortest possible
time to its destination, when we describe motion by a principle of least action, or when
we discover that trees are branched in such a way that they achieve the largest effect with Dvipsbugw
the smallest effort, we are using a demarcation viewpoint.
In summary, unification, answering ‘why’ questions, and demarcation, answering ‘why
not’ questions, are typical for the progress throughout the history of physics. We can
say that the dual aspects of unification and demarcation form the composing and the
opposing traits of physics. They stand for the desire to know everything.
However, neither demarcation nor unification can explain the universe. Can you see

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1161 n why? In fact, apart from unification and demarcation, there is a third possibility that
Challenge 1162 n merges the two and allows one to say more about the universe. Can you find it? Our
walk will automatically lead to it later.

Pigs, apes and the anthropic principle

Das wichtigste Instrument des Wissenschaftlers

“ ist der Papierkorb.*

”
The wish to achieve demarcation of the patterns of nature is most interesting when we
follow the consequences of different rules of nature until we find them in contradiction
with the most striking observation: our own human existence. In this special case the pro-
gram of demarcation is often called the anthropic principle – from the Greek ἄνθρωπος,
meaning ‘man’.
For example, if the Sun–Earth distance were different from what it is, the resulting tem-
perature change on the Earth would have made impossible the appearance of life, which
needs liquid water. Similarly, our brain would not work if the Moon did not circle the
Earth. It is only because the Moon revolves around our planet that the Earth’s magnetic
field is large enough to protect the Earth by deviating most of the cosmic radiation that
would otherwise make all life on Earth impossible. It is only because the Moon revolves
around our planet that the Earth’s magnetic field is small enough to leave enough radi-
Copyright © Christoph Schiller November 1997–May 2006

ation to induce the mutations necessary for evolution. It is also well-known that if there
were fewer large planets in the solar system, the evolution of humans would have been
impossible. The large planets divert large numbers of comets, preventing them from hit-
ting the Earth. The spectacular collision of comet Shoemaker–Levy-9 with Jupiter, the
astronomical event of July 1994, was an example of this diversion of a comet.**
Also the anthropic principle has its most impressive successes when it predicts un-
known observations. The most famous example stems from the study of stars. Carbon
atoms, like all other atoms except most hydrogen, helium or lithium atoms, are formed

* The most important instrument of a scientist is the waste paper basket.

** For a collection of pictures of this event, see e.g. the http://garbo.uwasa.fi/pc/gifslevy.html website.

Dvipsbugw
 690 intermezzo: the brain, l anguage and the human condition

in stars through fusion. While studying the mechanisms of fusion in 1953, the well-known
British astrophysicist Fred Hoyle* found that carbon nuclei could not be formed from the
alpha particles present inside stars at reasonable temperatures, unless they had an excited
state with an increased cross-section. From the fact of our existence, which is based on
carbon, Hoyle thus predicted the existence of a previously unknown excited state of the
Ref. 668 carbon nucleus. And, indeed, the excited state was found a few months later by Willy
Fowler.**
In its serious form, the anthropic principle is therefore the quest to deduce the descrip-
Ref. 669 tion of nature from the experimental fact of our own existence. In the popular literature, Dvipsbugw
however, the anthropic principle is often changed from a simple experimental method to
deduce the patterns of nature, to its perverted form, a melting pot of absurd metaphysical
ideas in which everybody mixes up their favourite beliefs. Most frequently, the experi-
mental observation of our own existence has been perverted to reintroduce the idea of
‘design’, i.e. that the universe has been constructed with the aim of producing humans;
often it is even suggested that the anthropic principle is an explanation – a gross example

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of disinformation.
How can we distinguish between the serious and the perverted form? We start with
an observation. We would get exactly the same rules and patterns of nature if we used
the existence of pigs or monkeys as a starting point. In other words, if we would reach
different conclusions by using the porcine principle or the simian principle, we are using
the perverted form of the anthropic principle, otherwise we are using the serious form.
(The carbon-12 story is thus an example of the serious form.) This test is effective because
there is no known pattern or ‘law’ of nature that is particular to humans but unnecessary
apes or for pigs.***
Er wunderte sich, daß den Katzen genau an den

“ Stellen Löcher in den Pelz geschnitten wären,

wo sie Augen hätten.
Georg Christoph Lichtenberg****

”
* Fred Hoyle (b. 1915 Bingley, Yorkshire, d. 2001), important British astronomer and astrophysicist. He was
the first and maybe only physicist who ever made a specific prediction – namely the existence of an excited
Copyright © Christoph Schiller November 1997–May 2006

state of the carbon nucleus – from the simple fact that humans exist. A permanent maverick, he coined the
term ‘big bang’ even though he did not accept the evidence for it, and proposed another model, the ‘steady
state’. His most important and well-known research was on the formation of atoms inside stars. He also
propagated the belief that life was brought to Earth from extraterrestrial microbes.
** William A. Fowler (1911–1995) shared the 1983 Nobel Prize in physics with Subramanyan
Chandrasekhar for this and related discoveries.
*** Though apes do not seem to be good physicists, as described in the text by D.J. Povinelli, Folk Physics
for Apes: the Chimpanzee’s Theory of How the World Works, Oxford University Press, 2000.
**** ‘He was amazed that cats had holes cut into their fur precisely in those places where they had eyes.’
Georg Christoph Lichtenberg (1742–1799), German physicist and intellectual, professor in Göttingen, still
famous today for his extremely numerous and witty aphorisms and satires. Among others of his time,
Lichtenberg made fun of all those who maintained that the universe was made exactly to the measure of
man, a frequently encountered idea in the foggy world of the anthropic principle.

Dvipsbugw
 the quest for precision and its implications 691

Does one need cause and effect in explanations?

In nature there are neither rewards nor

“ punishments – there are consequences.

Ivan Illich
The world owes you nothing. It was there first.
”
“ Mark Twain

No matter how cruel and nasty and evil you may

”
“ be, every time you take a breath you make a
flower happy.
Mort Sahl
Dvipsbugw

Ref. 670
Historically, the two terms ‘cause’ and ‘effect’ have played an important role in philosoph-
ical discussions. In particular, during the birth of modern mechanics, it was important
to point out that every effect has a cause, in order to distinguish precise thought from
”
thought based on beliefs, such as ‘miracles’, ‘divine surprises’ or ‘evolution from noth-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ing’. It was equally essential to stress that effects are different from causes; this distinction
avoids pseudo-explanations such as the famous example by Molière where the doctor ex-
plains to his patient in elaborate terms that sleeping pills work because they contain a
dormitive virtue.
But in physics, the concepts of cause and effect are not used at all. That miracles do not
appear is expressed every time we use symmetries and conservation theorems. The obser-
vation that cause and effect differ from each other is inherent in any evolution equation.
Moreover, the concepts of cause and effect are not clearly defined; for example, it is espe-
cially difficult to define what is meant by one cause as opposed to several of them, and the
same for one or several effects. Both terms are impossible to quantify and to measure. In
other words, useful as ‘cause’ and ‘effect’ may be in personal life for distinction between
events that regularly succeed each other, they are not necessary in physics. In physical
explanations, they play no special roles.

“
Ὰγαθον καὶ ξαξόν ë ἔν καὶ ταὐτό.*
Heraclitus

Wenn ein Arzt hinter dem Sarg seines Patienten ”

“ geht, so folgt manchmal tatsächlich die Ursache
der Wirkung.**
Robert Koch

”
Copyright © Christoph Schiller November 1997–May 2006

Is consciousness required?
Variatio delectat.***

Ref. 671
“
A lot of mediocre discussions are going on about this topic, and we will skip them here.
What is consciousness? Most simply and concretely, consciousness means the possession
Cicero

”
* ‘Good and bad – one and the same.’
** ‘When a doctor walks behind the coffin of his patient, indeed the cause sometimes follows the effect.’
*** ‘Change pleases.’ Marcus Tullius Cicero (106–43 bce ), important lawyer, orator and politician at the
end of the Roman republic.

Dvipsbugw
 692 intermezzo: the brain, l anguage and the human condition

of a small part of oneself that is watching what the rest of oneself is perceiving, feeling,
thinking and doing. In short, consciousness is the ability to observe oneself, and in par-
ticular one’s inner mechanisms and motivations. Consciousness is the ability of introspec-
tion. For this reason, consciousness is not a prerequisite for studying motion. Indeed,
animals, plants or machines are also able to observe motion. For the same reason, con-
sciousness is not necessary to observe quantum mechanical motion. On the other hand,
both the study of motion and that of oneself have a lot in common: the need to observe
carefully, to overcome preconceptions, to overcome fear and the fun of doing so.
For the time being, we have put enough emphasis on the precision of concepts. Talking Dvipsbugw
about motion is also something to be deeply enjoyed. Let us see why.
Precision and clarity obey the indeterminacy

“ relation: their product is constant.

”
Curiosity

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Precision is the child of curiosity.

“ ”
Like the history of every person, also the history of mankind charts a long struggle to
avoid the pitfalls of accepting the statements of authorities as truth, without checking the
facts. Indeed, whenever curiosity leads us to formulate a question, there are always two
general ways to proceed. One is to check the facts personally, the other is to ask somebody.
However, the last way is dangerous: it means to give up a part of oneself. Healthy people,
children whose curiosity is still alive, as well as scientists, choose the first way. After all,
science is adult curiosity.
Curiosity, also called the exploratory drive, plays strange games with people. Starting
with the original experience of the world as a big ‘soup’ of interacting parts, curiosity can
drive one to find all the parts and all the interactions. It drives not only people. It has been
observed that when rats show curious behaviour, certain brain cells in the hypothalamus
get active and secrete hormones that produce positive feelings and emotions. If a rat has
the possibility, via some implanted electrodes, to excite these same cells by pressing a
Ref. 672 switch, it does so voluntarily: rats get addicted to the feelings connected with curiosity.
Like rats, humans are curious because they enjoy it. They do so in at least four ways:
because they are artists, because they are fond of pleasure, because they are adventurers
and because they are dreamers. Let us see how.
Copyright © Christoph Schiller November 1997–May 2006

Originally, curiosity stems from the desire to interact in a positive way with the envir-
onment. Young children provide good examples: curiosity is a natural ingredient of their
life, in the same way that it is for other mammals and a few bird species; incidentally, the
same taxonomic distribution is found for play behaviour. In short, all animals that play
are curious, and vice versa. Curiosity provides the basis for learning, for creativity and
thus for every human activity that leaves a legacy, such as art or science. The artist and art
theoretician Joseph Beuys (1920–1986) had as his own guiding principle that every creat-
Ref. 673 ive act is a form of art. Humans, and especially children, enjoy curiosity because they feel
its importance for creativity, and for growth in general.
Curiosity regularly leads one to exclaim: ‘Oh!’, an experience that leads to the second
reason to be curious: relishing feelings of wonder and surprise. Epicurus (Epikuros) (341–

Dvipsbugw
 the quest for precision and its implications 693

271 bce ) maintained that this experience, θαυµάζειν, is the origin of philosophy. These
feelings, which nowadays are variously called religious, spiritual, numinous, etc., are the
same as those to which rats can become addicted. Among these feelings, Rudolf Otto
has introduced the now classical distinction into the fascinating and the frightening.
He named the corresponding experiences ‘mysterium fascinans’ and ‘mysterium tremen-
dum’.* Within these distinctions, physicists, scientists, children and connoisseurs take a
clear stand: they choose the fascinans as the starting point for their actions and for their
approach to the world. Such feelings of fascination induce some children who look at the
night sky to dream about becoming astronomers, some who look through a microscope Dvipsbugw
to become biologists or physicists, and so on. (It could also be that genetics plays a role
Ref. 674 in this pleasure of novelty seeking.)
Perhaps the most beautiful moments in the study of physics are those appearing after
new observations have shaken our previously held thinking habits, have forced us to give
up a previously held conviction, and have engendered the feeling of being lost. When,
in this moment of crisis, we finally discover a more adequate and precise description of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the observations, which provide a better insight into the world, we are struck by a feeling
usually called illumination. Anyone who has kept alive the memory and the taste for
these magic moments knows that in these situations one is pervaded by a feeling of union
between oneself and the world.** The pleasure of these moments, the adventures of the
change of thought structures connected with them, and the joy of insight following them
provides the drive for many scientists. Little talk and lots of pleasure is their common
denominator. In this spirit, the well-known Austrian physicist Victor Weisskopf (1908–
2002) liked to say jokingly: ‘There are two things that make life worth living: Mozart and
quantum mechanics.’
The choice of moving away from the tremendum towards the fascinans stems from
an innate desire, most obvious in children, to reduce uncertainty and fear. This drive is
the father of all adventures. It has a well-known parallel in ancient Greece, where the first
men studying observations, such as Epicurus, stated explicitly that their aim was to free
people from unnecessary fear by deepening knowledge and transforming people from
frightened passive victims into fascinated, active and responsible beings. Those ancient
thinkers started to popularize the idea that, like the common events in our life, the rarer
events also follow rules. For example, Epicurus underlined that lightning is a natural phe-
nomenon caused by interactions between clouds, and stressed that it was a natural pro-
cess, i.e. a process that followed rules, in the same way as the falling of a stone or any other
familiar process of everyday life.
Copyright © Christoph Schiller November 1997–May 2006

Investigating the phenomena around them, philosophers and later scientists suc-
ceeded in freeing people from most of their fears caused by uncertainty and a lack of
knowledge about nature. This liberation played an important role in the history of hu-

* This distinction is the basis of Rudolf Otto, Das Heilige – Über das Irrationale in der Idee des Göttlichen
und sein Verhältnis zum Rationalen, Beck, München, 1991. This is a new edition of the epoch-making work
originally published at the beginning of the twentieth century. Rudolf Otto (1869–1937) was one of the most
important theologians of his time.
** Several researchers have studied the situations leading to these magic moments in more detail, notably
the Prussian physician and physicist Hermann von Helmholtz (1821–1894) and the French mathematician
Henri Poincaré (1854–1912). They distinguish four stages in the conception of an idea at the basis of such
Ref. 675 a magic moment: saturation, incubation, illumination and verification.

Dvipsbugw
 694 intermezzo: the brain, l anguage and the human condition

man culture and still pervades in the personal history of many scientists. The aim to ar-
rive at stable, rock-bottom truths has inspired (but also hindered) many of them; Albert
Einstein is a well-known example for this, discovering relativity, helping to start up but
then denying quantum mechanics.
Interestingly, in the experience and in the development of every human being, curios-
ity, and therefore the sciences, appears before magic and superstition. Magic needs deceit
to be effective, and superstition needs indoctrination; curiosity doesn’t need either. Con-
flicts of curiosity with superstitions, ideologies, authorities or the rest of society are thus
preprogrammed. Dvipsbugw
Curiosity is the exploration of limits. For every limit, there are two possibilities: the
limit can turn out to be real or apparent. If the limit is real, the most productive attitude is
that of acceptance. Approaching the limit then gives strength. If the limit is only apparent
and in fact non-existent, the most productive attitude is to re-evaluate the mistaken view,
extract the positive role it performed, and then cross the limit. Distinguishing between
real and apparent limits is only possible when the limit is investigated with great care,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

openness and unintentionality. Most of all, exploring limits need courage.
Das gelüftete Geheimnis rächt sich.*

Courage
“ Bert Hellinger

”
It is dangerous to be right in matters on which

“ the established authorities are wrong.

Voltaire
Manche suchen Sicherheit, wo Mut gefragt ist,
”
“ und suchen Freiheit, wo das Richtige keine
Wahl läßt.**
Bert Hellinger

Ref. 676
Most of the material in this intermezzo is necessary in the adventure to get to the top of
Motion Mountain. But we need more. Like any enterprise, curiosity also requires courage,
and complete curiosity, as aimed for in our quest, requires complete courage. In fact, it is
”
easy to get discouraged on this trip. The journey is often dismissed by others as useless,
uninteresting, childish, confusing, damaging or, most often, evil. For example, between
the death of Socrates in 399 bce and Paul Thierry, Baron d’Holbach, in the eighteenth
century, no book was published with the statement ‘gods do not exist’, because of the
Copyright © Christoph Schiller November 1997–May 2006

threats to the life of anyone who dared to make the point. Even today, this type of attitude
still abounds, as the newspapers show.
Through the constant elimination of uncertainty, both curiosity and scientific activity
are implicitly opposed to any idea, person or organization that tries to avoid the com-
parison of statements with observations. These ‘avoiders’ demand to live with supersti-
tions and beliefs. But superstitions and beliefs produce unnecessary fear. And fear is the
basis of all unjust authorities. One gets into a vicious circle: avoiding comparison with
* The unveiled secret takes revenge.
** ‘Some look for security where courage is required and look for freedom where the right way doesn’t leave
any choice.’ This is from the beautiful booklet by Bert Hellinger, Verdichtetes, Carl-Auer Systeme Verlag,
1996.

Dvipsbugw
the quest for precision and its implications 695

observation produces fear – fear keeps unjust authority in place – unjust authority avoids
comparison with observation – etc.
As a consequence, curiosity and science are fundamentally opposed to unjust author-
ity, a connection that made life difficult for people such as Anaxagoras (500–428 bce )
in ancient Greece, Hypatia in the Christian Roman empire, Galileo Galilei in the church
state, Antoine Lavoisier in France and Albert Einstein in Germany. In the second half
of the twentieth century, victims were Robert Oppenheimer, Melba Phillips and Chand-
ler Davis in the United States and Andrei Sakharov in the Soviet Union. Each of them
tell a horrible but instructive story, as have, more recently, Fang Lizhi, Xu Liangying, Liu Dvipsbugw
Gang and Wang Juntao in China, Kim Song-Man in South Korea, Otanazar Aripov in
Uzbekistan, Ramadan al-Hadi al-Hush in Libya, Bo Bo Htun in Burma, as well as many
hundreds of others. In many authoritarian societies the antagonism between curiosity
and injustice has hindered or even completely suppressed the development of physics
and other sciences, with extremely negative economic, social and cultural consequences.
When embarking on this ascent, we need to be conscious of what we are doing. In

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fact, external obstacles can be avoided or at least largely reduced by keeping the project
to oneself. Other difficulties still remain, this time of personal nature. Many have tried to
embark on this adventure with some hidden or explicit intention, usually of an ideological
nature, and then have got entangled by it before reaching the end. Some have not been
prepared to accept the humility required for such an endeavour. Others were not prepared
for the openness required, which can shatter deeply held beliefs. Still others were not
ready to turn towards the unclear, the dark and the unknown, confronting them at every
occasion.
On the other hand, the dangers are worth it. By taking curiosity as a maxim, facing
disinformation and fear with all one’s courage, one achieves freedom from all beliefs. In
exchange, you come to savour the fullest pleasures and the deepest satisfaction that life
has to offer.
We thus continue our hike. At this point, the trail towards the top of Motion Mountain
is leading us towards the next adventure: discovering the origin of sizes and shapes in
nature.
And the gods said to man: ‘Take what you want,

“ and pay the price.’

(Popular saying)

It is difficult to make a man miserable while he

”
“ feels he is worthy of himself.
Copyright © Christoph Schiller November 1997–May 2006

Abraham Lincoln

Dvipsbugw
696 intermezzo: the brain, l anguage and the human condition

Biblio graphy

607 See for example the beautiful textbook by Stephen C. Stearns & Rolf F. Hoekstra,
Evolution: An Introduction, Oxford University Press, 2000. For fascinating story of evolu-
tion for non-specialists, see R ichard Fortey, Life – An Unauthorized Biography, Harper
Collins, 1997, or also Menno Schilthuizen, Frogs, Flies & Dandelions – the Making of
Species, Oxford University Press/ 2001. See also Stephen J. Gould, The Panda’s thumb,
W.W. Norton & Co., 1980, one of the several interesting and informative books on evolution-
ary biology by the best writer in the field. An informative overview over the results of evolu- Dvipsbugw
tion, with the many-branched family tree that it produced, is given on the http://phylogeny.
arizona.edu/tree website. About the results of evolution for human beings, see the informat-
ive text by K. Kusch & S. Kusch, Der Mensch in Zahlen, Spektrum Akademischer Verlag,
2nd edn., 2000. The epochal work by Charles Darwin, On the Origin of Species, can be
found on the web, e.g. on the http://entisoft.earthlink.net/origspec.htm website. Cited on
page 632.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

608 A simple description is Malcolm Ross Macd onald, The Origin of Johnny, Jonathan
Cape, 1976. Cited on page 632.
609 There is disagreement among experts about the precise timing of this experience. Some say
that only birth itself is that moment. However, there are several standard methods to re-
call memories of early life, even of the time before birth; one is described and used by J.K.
Stettbacher, Making Sense of Suffering, Penguin, New York, 1991, translated from the
German original. Even more impressive examples are given by N.J. Mayer, Der Kainkom-
plex – neue Wege der systemischen Familientherapie, Integral Verlag, 1998. Cited on page
632.
610 Sanjida O’Connell, Mindreading – How We Learn to Love and Lie, Arrow, 1998. This
interesting book describes the importance of lying in the development of a human being,
and explains the troubles those people have who cannot read other minds and thus cannot
lie, such as autistics. Cited on pages 633 and 663.
611 The approach to describe observations as related parts is called structuralism; the starting
point for this movement was de Saussure’s Cours de linguistique générale (see the footnote
on page 641). A number of thinkers have tried to use the same approach in philosophy,
mythology and literature theory, though with little success. An overview of the (modest)
success of structuralism in linguistics and its failure in other fields is given by L. Jackson,
The Poverty of Structuralism: Literature and Structuralist Theory, Longman, 1991. The author
argues that when one reduces systems to interactions, one neglects the specific content and
properties of the elements of the system, and this approach prevents a full understanding of
Copyright © Christoph Schiller November 1997–May 2006

the system under discussion. Cited on page 634.

612 For a view of the mental abilities different from that of Piaget (described on page 634),
a presently much discussed author is the Soviet experimental psychologist Lev Vigotsky,
whose path-breaking ideas and complicated life are described, e.g., in Lev Vigotsky,
Mind in Society, Harvard University Press, 1978, or in R ené van der Veer & Jaan
Valsiner, Understanding Vigotsky: a Quest for Synthesis, Blackwell Publishers, 1994. More
extensive material can be found in the extensive work by R ené van der Veer & Jaan
Valsinger, The Vigotsky Reader, Blackwell, 1994. Cited on page 634.
613 A somewhat unconventional source for more details is the beautiful text by Bruno
Bettelheim, The Uses of Enchantment: the Meaning and Importance of Fairy Tales,
Knopf, 1976. Cited on page 634.

Dvipsbugw
bibliography 697

614 Quoted in V. Harlen, R. R appmann & P. Schata, Soziale Plastik – Materialien zu

Joseph Beuys, Achberger Verlag, 1984, p. 61. Cited on page 635.
615 The problems appearing when one loses the ability to classify or to memorise are told in
the beautiful book by the neurologist Oliver Sacks, The Man Who Mistook His Wife for
a Hat, Picador, 1985, which collects many case studies he encountered in his work. More
astonishing cases are collected in his equally impressive text An Anthropologist on Mars, Pic-
ador, 1995.
See also the beautiful text Donald D. Hoffman, Visual Intelligence – How We Cre-
ate What We See, W.W. Norton & Co., 1998, and the http://ari.ss.uci.edu/cogsci/personnel/ Dvipsbugw
hoffman.html website associated to it. Cited on pages 635 and 637.
616 For a passionate introduction to the connections between language and the brain from a
Chomskian perspective, see the bestselling book by Steven Pinker, The Language In-
stinct – How the Mind Creates Language, Harper Perennial, 1994. The green idea sentence is
discussed in a chapter of the book. Cited on pages 635, 667, and 683.
617 An introduction to neurology is Joseph Led oux, Synaptic Self: How Our Brains Become

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Who We Are, Viking Press, 2002. Cited on page 635.
618 An overview of the status of the research into the origin of bipedalism is given by B. Wood,
Four legs good, two legs better, Nature 363, pp. 587–588, 17 June 1983. Cited on page 635.
619 A good introduction into the study of classifiers is James A. Anderson, An Introduc-
tion to Neural Networks, MIT Press, 1995. An introduction to computer science is given in
J. Glenn Brookshear, Computer Science, An Overview, 6th edition, Addison Wesley,
2000, or in R ick Decker & Stuart Hirshfield, The Analytical Engine: An Introduc-
tion to Computer Science Using the Internet, Brooks/Cole Publishers, 1998. Cited on page
635.
620 More about the connection between entropy and computers can be found in the classic
paper by R. L andauer, Irreversibility and heat generation in the computing process,
IBM Journal of Research and Development 5, pp. 183–191, 1961, and in C.H. Bennett &
R. L andauer, The fundamental physical limits of computation, Scientific American 253,
pp. 48–56, 1985. Cited on page 638.
621 W.H. Zurek, Thermodynamic cost of computation, algorithmic complexity and the in-
formation metric, Nature 341, pp. 119–124, 14 August 1989. Cited on page 638.
622 L. Szil ard, Über die Entropieverminderung in einem thermodynamischen System bei
Eingriffen intelligenter Wesen, Zeitschrift für Physik 53, p. 840, 1929. This classic paper can
also be found in English translation in the collected works by Leo Szilard. Cited on page
639.
623 J.J. Hopfield/ Nature 376, pp. 33–36, 1995. This paper by one of the fathers of the field
Copyright © Christoph Schiller November 1997–May 2006

presents one possibility by which the timing of nerve signals, instead of the usually assumed
firing frequency, could also carry information. Cited on page 639.
624 The details of the properties of the firing patterns of neurons are nicely described in the
article by M. Mahowald & R. Dougl as, A silicon neuron, Nature 354, pp. 515–518,
19/26 December 1991, in which they show how to simulate a neuron’s electrical behaviour
using a silicon circuit. Cited on page 639.
625 A. Mechelli, J.T. Crinion, U. Noppeney, J. O’Doberty, J. Ashburner, R.S.
Frackowiak & C.J. Price, Neurolinguistics: structural plasticity in the bilingual brain,
Nature 431, p. 757, 2004. Cited on page 640.
626 The discussion whether the brain is or is not superior to a computer is nicely summarised
by G. Vollmer, Algorithmen, Gehirne, Computer – Was sie können und was sie nicht

Dvipsbugw
 698 intermezzo: the brain, l anguage and the human condition

können, Teil I und Teil II, Naturwissenschaften 78, p. 481, 1991, and 78, pp. 533–542, 1991.
Cited on page 640.
627 However, the language with the largest available dictionary is Dutch, with the 40 volumes
of the Wordenboek der Nederlandsche Taal, which appeared between 1864 and 1998. It has
almost 400 000 entries. Cited on page 642.
628 The list and the remark on discovery on concepts is due to a personal communication from
Anna Wierzbicka. A longer list is published in her book Semantics, Primes and Universals,
Oxford University Press, 1996. Cited on pages 643 and 660.
629 W.S. Hatcher, Foundations of Mathematics, W.B. Saunders Co., Philadelphia, 1968. There Dvipsbugw
is also the article by P. J. Cohen & R. Hersch, Non-Cantorian set theory, Scientific
American 217, pp. 104–116, 1967. Cohen was the mathematician who in 1963 proved that
the negation of the continuum hypothesis could be added to the axioms of set theory and
still produce a consistent theory; he calls such sets non-Cantorian. Cited on page 647.
630 See the beautiful article by I. Stewart, Fair shares for all, New Scientist, pp. 42–46, 17 June
1995. Cited on page 648.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

631 The proof of the independence of the continuum hypothesis came in two parts. First, Kurt
Gödel proved in 1940 that an axiom can be consistently added to ZFC set theory so that the
continuum hypothesis is correct. Then, in 1963, Paul Cohen proved that an axiom can be
consistently added to ZFC set theory so that the continuum hypothesis is false. Cited on
page 649.
632 Rudy Rucker, Infinity and the Mind – the Science and Philosophy of the Infinite, Bantam,
Toronto, 1983. Cited on page 650.
633 This general division of mathematics is nicely explained in the text by Pierre Basieux,
Die Architektur der Mathematik – Denken in Strukturen, Rororo, 2000. Cited on page 651.
634 The issue is treated in Thomas Aquinas, Summa Theologica, in question 52 of the first
part. The complete text, several thousand pages, can be found on the http://www.newadvent.
org website. We come back to it in the part on quantum theory, in the section on the Pauli
Page 785 exclusion principle. It seems that the whole question goes back to Peter (the) Lombard,
Liber Sententiarum c. 1150. Cited on page 652.
635 B.C. Gallivan, How to fold paper in half twelve times:an “impossible challenge” solved
and explained, Histrical Society of Pomona Valley, 2002, also found at http://www.osb.net/
Pomona/12times.htm. See also http://www.sciencenews.org/20040124/mathtrek.asp. Cited
on page 652.
636 I. Stewart, Daisy, daisy, give me your answer, do, Scientific American, pp. 76–79, Janu-
ary 1995. This pedagogical article explains how the growth of plants usually leads to flowers
whose number of petals is from the Fibonacci series 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, etc.
Copyright © Christoph Schiller November 1997–May 2006

(The figure on page 175 is an example.) Deviations from this ideal case are also explained.
The original work are two articles by S. Douady & Y. Couder, La physique des spirales
végétales, La Recherche 24, pp. 26–36, 1993, and Phyllotaxis as a self-organised growth pro-
cess, in Growth Patterns in Physical Sciences and Biology, edited by J.M. Garcia-Ruiz &
al., Plenum Press, 1993. Cited on page 652.
637 H. Davson, The Eye, Academic Press, 1962. Cited on page 653.
638 See the http://akbar.marlboro.edu/~mahoney/cube/NxN.txt website. Cited on page 653.
639 An introduction to the surreal numbers is given by the article by Polly Shulman, Infinity
plus one, and other surreal numbers, Discover, pp. 99–105, December 1995. There is also the
text by D. Knu th, Surreal Numbers: How two ex-Students Turned on to Pure Mathematics
and Found Total Happiness, Addison Wesley, 1974, or http://www-cs-faculty.stanford.edu/

Dvipsbugw
bibliography 699

~knuth/sn.html. The usually quoted references on the topic include John H. Conway,
On Numbers and Games, Academic Press, 1976, E.R. Berlekamp, J.H. Conway & R.K.
Guy, Winning Ways for Your Mathematical Plays, Volume I: Games in General, Academic
Press, 1982, and H. Gonshor, An Introduction to Surreal Numbers, Cambridge University
Press, 1986. Cited on pages 653 and 655.
640 This beautiful problem is discussed by Ian Stewart, A bundling fool beats the wrap, Sci-
entific American, pp. 109–111, June 1993. In four dimensions, the answer is known to lie some-
where between 50 000 and 100 000, whereas the five-dimensional answer is conjectured to
be ‘never’. Cited on page 656. Dvipsbugw
641 A. Pais, Niels Bohr’s Times: in Physics, Philosophy and Polity, Oxford University Press, 1991,
page 176. Cited on page 656.
642 Eugene Wigner, Symmetries and Reflections, Indiana University Press, 1962. Cited on
page 656.
643 Alfred Tarski, Introduction to Modern Logic, Dover, 1946. See also the famous chil-
dren’s book by the mathematician and photographer Lewis Carroll, Alice in Wonder-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

land. Cited on page 657.
644 Göran Wikell, The layout of digits on pushbutton telephones – a review of the literature,
Tele 34, pp. 34–42, 1982. Cited on page 658.
645 A clear overview of philosophy of science, often called epistemology, without unnecessary
detail, is given by Robert Bl anché, L’Epistémologie, Presses Universitaires de France,
1972. Cited on page 659.
646 About the different aspects of falsifiability of general statements it is commonplace to cite
the work by the epitemologist Karl Popper (1902–1994), especially his long and boring book
Logik der Forschung, first published in 1934. The reason for this boredom is that Popper’s
work is simply a warming-up of Pierre Duhem’s ideas. Cited on page 664.
647 For a good way of making blood that liquefies, see L. Garl aschelli, F. R amaccini
& S. Dell a Scal a, Working bloody miracles, Nature 353, p. 507, 1991. The Grand dic-
tionnaire universel du XIXe siècle, by Pierre L arousse, also contains a recipe; it was
again shown to the public in the 1980s by Henri Broch. A wonderful and classic text is
Harry Houdini, Miracle Mongers and their Methods, Prometheus Books, Buffalo, 1981.
The original, written in 1920, by the world famous magician named ‘The Great Houdini’, is
also available on the http://etext.lib.virginia.edu/toc/modeng/public/HouMirM.html web-
site. The milk drinking Indian statues were common around the world in 1994 and 1995.
About healers, see James R andi, Flim-flam!, Prometheus Books, Buffalo, New York, 1987,
and the exquisite book by Hans Conrad Z ander, Warum ich Jesus nicht leiden kann,
Copyright © Christoph Schiller November 1997–May 2006

Rowohlt, 1994. Cited on page 665.

648 Roger Penrose, The Road to Reality: A Complete Guide to the Laws of the Universe,
Jonathan Cape, 2004, page 378. Cited on page 680.
649 John Horgan, The End of Science – Facing the Limits of Knowledge in the Twilight of the
Scientific Age, Broadway Books, 1997, pp. 31–32, and chapter 2, note 2. Cited on pages 667
and 680.
650 For an opinion completely contrary to the one described here, see the book by Gregory J.
Chaitin, The Limits of Mathematics, Springer Verlag, 1997, which can also be found on
the author’s website http://www.cs.auckland.ac.nz/CDMTCS/chaitin/lm.html, along with
his other works. Chaitin has devoted most of his life to the questions discussed in the section,
especially on computability. Cited on page 667.

Dvipsbugw
700 intermezzo: the brain, l anguage and the human condition

651 See the book by J. Barwise & J. Etchemendy, The Liar, Oxford University Press, New
York, 1987. Cited on page 668.
652 This definition (statement 4.11) and many other statements about science are in the beau-
tiful and rightly famous text by Ludwig Wittgenstein, Tractatus logico-philosophicus,
Edition Suhrkamp, 1963. It gives a condensed summary of the basis of science, thought and
language in a collection of highly structured and numbered sentences. Cited on pages 670
and 674.
653 See M. Dresden, The Klopsteg memorial lecture, American Journal of Physics 66, pp. 468–
482, 1998. Cited on page 670. Dvipsbugw
654 Well-known books are e.g. Friedrich Kohlrausch, Praktische Physik, Teubner, 24. Au-
flage, 1996. Cited on page 671.
655 Results are described e.g. in L. Bergmann & C. Schäfer, Lehrbuch der Experimental-
physik, Band I, II, III und IV, W. de Gruyter. Cited on page 671.
656 L and olt-B örnstein, edited by K.-H. Hellwege & O. Madelung, Zahlenwerte
und Funktionen aus Naturwissenschaften und Technik, Neue Serie, Springer Verlag, Berlin,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1984. This series of more than one hundred volumes contains all important observations in
the domain of physics. Cited on page 671.
657 The origin of this incorrect attribution is the book by Gerhard Szczesny, Brecht, Leben
des Galilei – Dichtung und Wirklichkeit, Ullstein, Berlin 1966, p. 6. The statement has never
been made by Galilei; this issue has been discussed at length in specialist circles, e.g. by
F. Kleinert, "Messen was meßbar ist" - Über ein angebliches Galilei-Zitat, Berichte zur
Wissenschaftgeschichte 11, p. 221, 1988, or on the internet by Peter Jaencke. Cited on page
671.
658 The strange and sometimes dangerous consequences of beliefs can be found e.g. in Martin
Gardner, Fads and Fallacies, Dover, 1957, and in James R andi, Faith Healers, Prometh-
eus Books, 1989. The million dollar prize for showing any paranormal or supernormal effect
is available from his http://www.randi.org website. Cited on page 675.
659 See the nice collection of cranks on the http://www.crank.net website. Cited on page 675.
660 As an example, the opposite view on the emergence of properties is strongly defended by P.
Jensen, Particle physics and our everyday world, Physics Today pp. 58–59, July 1998. Cited
on page 676.
661 See page 133 of the bibliography by John B owlby, Charles Darwin, Pimlico, 1991. Cited
on page 677.
662 A beautiful introduction to Greek philosophy is Eduard Zeller, Outlines of the History
of Greek Philosophy, Dover, 1980, a reprint of a book published in 1928. Among others, it
Copyright © Christoph Schiller November 1997–May 2006

gives a clear exposition of the philosophy of Democritus and the other presocratics. Cited
on page 681.
663 The famous quote is found at the beginning of chapter XI, ‘The Physical Universe’, in Ar-
thur Eddington, The Philosophy of Physical Science, Cambridge, 1939. Cited on page
681.
664 Giuseppe Fumagalli, Chi l’ha detto?, Hoepli, Milano, 1983. Cited on page 683.
665 See Jean-Paul Dumont, Les écoles présocratiques, Folio Essais, Gallimard, p. 653, 1991.
Cited on page 683.
666 For a beautiful text on fractals, see the footnote on page 53. Cited on page 685.
667 As has been pointed out by René Descartes. Cited on page 687.

Dvipsbugw
bibliography 701

12
668 The famous carbon C resonance was found by Willy Fowler, as described in E. Mar-
garet Burbrid ge, G.R. Burbrid ge, W.A. Fowler & F. Hoyle, Synthesis of the
elements in stars, Reviews of Modern Physics 29, pp. 547–560, 1957. Cited on page 690.
669 An extensive overview of the topic is given in the thick book by John D. Barrow &
Frank J. Tipler, The Anthropic Cosmological Principle, Oxford University Press, 1986.
The term itself is due to Brandon Carter, who coined it in 1973 and presented it in a sym-
posium devoted to the 500th anniversary of Nicolaus Copernicus. For more literature, see
Yuri I. Bal ashov, Resource Letter AP-1: the anthropic principle, American Journal of
Physics 59, pp. 1069–1076, 1991. Cited on page 690. Dvipsbugw
670 Voltaire, Candide ou l’optimisme, 1759. See the footnote on page 182. The book is so good
that it was still being seized by the US customs in 1930, and the US post office refused to
transport it as late as 1944. For more details, search for ‘banned books online’ on the world-
wide web. Cited on page 691.
671 The number of books on consciousness is large and the contents not always interesting, and
often not based on fact, as shown by Karl R. Popper & John Eccles, The Self and its

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Brain – an Argument for Interactionism, Rutledge, 1993. Cited on page 691.
672 See e.g. the Encyclopedia Britannica, Macropaedia, in the entry on animal behaviour. Cited
on page 692.
673 A straight and informative introduction to the work and ideas of Joseph Beuys (in German)
is by R enate Georgi, Joseph Beuys, RAAbits Kunst, Raabe Fachverlag, September 1996.
Cited on page 692.
674 Two studies, one by R.P. Ebstein & al., Dopamine D4 receptor (D4DR) exon III poly-
morphism associated with human personality trait of novelty seeking, Nature Genetics 12,
pp. 78–80, January 1996, and another study by J. Benjamin & al., Population and fa-
milial association between the D4 dopamine receptor gene and measures of novelty seeking,
Nature Genetics 12, pp. 81–84, January 1996, found that people with a special form of the D4
dopamine receptor gene, or D4DR, are more prone to novelty seeking than people with the
usual form. The D4DR gene regulates the formation of dopamine receptors, a chemical mes-
senger in the brain that has been a candidate for some time for a substance involved in the
propensity for novelty seeking. Cited on page 693.
675 More can be found in Jacques Hadamard, The Mathematician’s Mind – The Psychology
of Invention in the Mathematical Field, Princeton Science Library, 1996. Cited on page 693.
676 For example, one needs the courage to face envy. About this topic see the classic text by
Helmu t Schoeck, Der Neid, 1966, published in English as Envy: A Theory of Social Be-
havior, 1969. It is the standard work in the field. Cited on page 694.
Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
Second Part

Quantum Theory: What Is Mat-

ter?
What Are Interactions?

Where the existence of a minimal change is deduced,

implying that motion is fuzzy,
that matter is not permanent,
that boxes are never tight,
that matter is composed of elementary units
and that light and interactions are streams of particles,
thus explaining why antimatter exists,
why the floor does not fall but keeps on carrying us,
why particles are unlike gloves,
why empty space pulls mirrors together
and why the stars shine.
 C hap t e r V

QUANTA OF LIGHT AND MAT TER

Dvipsbugw

18. minimum action – quantum theory for p oets and

l awyers
Natura [in operationibus suis] non facit saltus.*
Ref. 677

“ 15th century

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

E
scalating Motion Mountain up to this point, we completed three legs. We first
ncountered Galileo’s mechanics, the description of motion for kids, then
instein’s relativity, the description of motion for science fiction enthusiasts, and
finally Maxwell’s electrodynamics, the description of motion valuable to craftsmen and
businessmen.
These three classical descriptions of motion are im-
pressive, beautiful and useful. However, they also have
a small problem: they are wrong. The reason is simple:
none of them describes life. Whenever we observe a
flower such as the one of Figure 282, we enjoy its bright
colours, its wild smell, its soft and delicate shape or
the fine details of its symmetry. None of the three clas-
sical descriptions can explain any of these properties;
neither do they explain the impression they make on
our senses. Classical physics can describe them partly,
but it cannot explain their origins. For an explanation,
we need quantum theory. In fact we will discover that in
life, every type of pleasure is an example of quantum mo-
tion. Just try; take any example of a pleasant situation,
Copyright © Christoph Schiller November 1997–May 2006

such as a beautiful evening sky, a waterfall, a caress or F I G U R E 282 An example of a

Challenge 1163 n a happy child. Classical physics is not able to explain it: quantum system (© Ata Masafumi)
the involved colours, shapes and sizes remain mysteri-
ous.
In the beginning of physics this limitation was not seen as a shortcoming: in those
times neither senses nor material properties were imagined to be related to motion. And
of course, in older times the study of pleasure was not deemed a serious topic of invest-
Page 253 igation for a respectable researcher. However, in the meantime we know that the senses
of touch, smell and sight are first of all detectors of motion. Without motion, no senses!

* ‘Nature [in its workings] makes no jumps.’

Dvipsbugw
 minimum action – quantum theory for poets and l aw yers 705

In addition, all detectors are built of matter. In the chapter on electromagnetism we star-
ted to understand that all properties of matter are due to motion of charged constituents.
Density, stiffness, colour and all other material properties result from the electromagnetic
behaviour of the Lego bricks of matter, namely the molecules, the atoms and the electrons.
Thus, also matter properties are consequences of motion. In addition, we saw that these
tiny constituents are not correctly described by classical electrodynamics. We even found
Page 612 that light itself behaves unclassically. Therefore the inability of classical physics to describe
matter and the senses is indeed due to its intrinsic limitations.
In fact, every failure of classical physics can be traced back to a single, fundamental Dvipsbugw
Ref. 678 discovery made in 1899 by Max Planck:
U In nature, actions smaller than the value ħ~2 = 0.53 ë 10−34 Js are not observed.
All experiments trying to do so invariably fail.* In other words, in nature there is always
some action – like in a good movie. This existence of a minimal action, the so-called
Challenge 1164 n quantum principle, is in full contrast with classical physics. (Why?) However, it has passed

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the largest imaginable number of experimental confirmations, many of which we will en-
counter in this second part of our mountain ascent. Planck discovered the principle when
Page 612 studying the properties of incandescent light, i.e. the light emanating from hot bodies. But
the quantum principle also applies to motion of matter, and even, as we will see later, to
motion of space-time. Incidentally, the factor 1~2 results from the historical accidents in
the definition of the constant ħ, which is read as ‘eitch-bar’. Despite the missing factor, the
constant ħ is called the quantum of action or also, after its discoverer, (reduced) Planck’s
constant.
The quantum principle states that no experiment whatsoever can measure an action
value smaller than ħ~2. For a long time, even Einstein tried to devise experiments to
overcome the limit. But he failed: nature does not allow it.
Interestingly, since action in physics, like action in the film industry, is a way to meas-
ure the change occurring in a system, a minimum action implies that there is a minimum
change in nature. The quantum of action thus would be better named the quantum of
change. Comparing two observations, there always is change. (What is called ‘change’ in
everyday life is often called ‘change of state’ by physicists; the content is the same.) Before
we explore experiments confirming this statement, we give an introduction to some of
its more surprising consequences.

The effects of the quantum of action on motion

Copyright © Christoph Schiller November 1997–May 2006

Since action measures change, a minimum observable action means that two subsequent
observations of the same system always differ by at least ħ~2. In every system, there is
always something happening. As a consequence, in nature there is no rest. Everything
Page 704 moves, all the time, at least a little bit. Natura facit saltus. True, it is only a tiny bit, as
the value of ħ~2 is so small. But for example, the quantum of action implies that in a

* In fact, the cited quantum principle is a simplification; the constant originally introduced by Planck was
the (unreduced) constant h = 2πħ. The factors 2π and 1/2 leading to the final quantum principle were found
somewhat later, by other researchers. This somewhat unconventional, but useful didactic approach is due to
Niels Bohr. Nowadays, the approach is almost never found in the literature; it might be used in a teaching
Ref. 679 text for the first time here. About Max Planck and his accomplishments, see the footnote on page 612.

Dvipsbugw
 706 v quanta of light and matter • 18. minimum action

mountain, a system at rest if there is any, all atoms and all electrons are continuously
buzzing around. Rest can be observed only macroscopically, and only as a long time or
many particle average.
Since there is a minimum action for all observers, and since there is no rest, in nature
there is no perfectly straight and no perfectly uniform motion. Forget all you have learned
so far. Every object moves in straight and uniform motion only approximately, and only
when observed over long distances or long times. We will see later that the more massive
Challenge 1165 ny the object is, the better the approximation is. Can you confirm this? As a consequence,
macroscopic observers can still talk about space-time symmetries. Special relativity can Dvipsbugw
thus easily be reconciled with quantum theory.
Obviously, also free fall, i.e. motion along geodesics, exists only as a long time average.
In this sense, general relativity, being based on the existence of freely falling observers,
cannot be correct when actions of the order of ħ are involved. Indeed, the reconciliation
of the quantum principle with general relativity – and thus with curved space – is a big
challenge. The issues are so mind-shattering that the topic forms a separate, third, part of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

this mountain ascent.

The consequences of the quantum of action for objects

Have you ever wondered why leaves are green? Probably you know that they are green
because they absorb blue light, of short wavelengths, and red light, of long wavelengths,
and let green, medium wavelength light be reflected. How can a system filter out the
small and the large, and let the middle pass through? To do so, leaves must somehow
measure the wavelength. But we have seen that classical physics does not allow to measure
length or time intervals, as any measurement requires a measurement unit, and classical
physics does not allow to define units for them. On the other hand, it takes only a few lines
to confirm that with the help of the quantum of action ħ (and the Boltzmann constant
k, which Planck discovered at the same time), fundamental measurement units of all
measurable quantities can be defined, including length and thus wavelength. Can you find
Challenge 1166 n a combination of c, G and ħ giving a length? It only will take a few minutes. When Planck
found the combination, he was happy like a child; he knew straight away that he had made
a fundamental discovery, even though in 1899 quantum theory did not exist yet. He even
told his seven-year-old son Erwin abut it, while walking with him through the forests
around Berlin. Planck explained to his son that he had made a discovery as important as
universal gravity. Indeed, Planck knew that he had found the key to understanding most
Copyright © Christoph Schiller November 1997–May 2006

of the effects which were unexplained so far. In particular, without the quantum of action,
colours would not exist. Every colour is a quantum effect.*
Planck also realized that the quantum of action allows to understand the size of all
things. With the quantum of action, it was finally possible to answer the question on the
Page 221 maximum size of mountains, of trees and of humans. Planck knew that the quantum
of action confirmed the answer Galileo had deduced already long before him: sizes are
due to fundamental, minimal scales in nature. The way the quantum of action allows to
understand the sizes of physical systems will be uncovered step by step in the following.
The size of objects is related to the size of atoms; in turn, the size of atoms is a direct
consequence of the quantum of action. Can you deduce an approximation for the size
Challenge 1167 n * It is also possible to define all units using c, G and e, the electron charge. Why is this not satisfactory?

Dvipsbugw
 minimum action – quantum theory for poets and l aw yers 707

of atoms, knowing that it is given by the motion of electrons of mass m e and charge e,
Challenge 1168 n constrained by the quantum of action? This connection, a simple formula, was discovered
in 1910 by Arthur Erich Haas, 15 years before quantum theory was formulated; at the time,
everybody made fun of him. Nowadays, the expression is found in all textbooks.*
By determining the size of atoms, the quantum of action has an important con-
sequence: Gulliver’s travels are impossible. There are no tiny people and no giant ones.
Classically, nothing speaks against the idea; but the quantum of action does. Can you
Challenge 1169 n provide the detailed argument?
But if rest does not exist, how can shapes exist? Any shape, Dvipsbugw
also that of a flower, is the result of body parts remaining
at rest with respect to each other. Now, all shapes result
from the interactions of matter constituents, as shown most
clearly in the shape of molecules. But how can a molecule,
such as the water molecule H2 O, have a shape? In fact, it does
not have a fixed shape, but its shape fluctuates, as expected

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

from the quantum of action. Despite the fluctuations it does
have an average shape, because different angles and distances
correspond to different energies. And again, these average
length and angle values only result because the quantum of F I G U R E 283 An artistic
impression of a water
action leads to fundamental length scales in nature. Without molecule
the quantum of action, there would be no shapes in nature.
As we will discover shortly, quantum effects surround us from all sides. However, since
the minimum action is so small, its effects on motion appear mostly, but not exclusively,
in microscopic systems. The study of such systems has been called quantum mechanics
by Max Born, one of the main figures of the field.** Later on, the term quantum theory
became more popular. In any case, quantum physics is the description of microscopic
motion. But when is quantum theory necessary? Table 56 shows that all processes on
atomic and molecular scale, including biological and chemical ones, involve action values
near the quantum of action. So do processes of light emission and absorption. All these
phenomena can be described only with quantum theory.

Copyright © Christoph Schiller November 1997–May 2006

* Before the discovery of ħ, the only simple length scale for the electron was the combination
e 2 ~(4πε 0 mc 2 )  3 fm; this value is ten thousand times smaller than an atom.
** Max Born (b. 1882 Breslau, d. 1970 Göttingen) first studied mathematics, then turned to physics. Pro-
fessor in Göttingen, he made the city one of the world centres of physics. He developed quantum mechanics
with his assistants Werner Heisenberg and Pascual Jordan, then applied it to scattering, to solid state phys-
ics, to optics and to liquids. He was the first to understood that the state function describes a probability
Ref. 680 amplitude. He is one of the authors of the famous Born & Wolf textbook on optics; it still remains the main
book of the field. Born attracted to Göttingen the most brilliant talents of the time, receiving as visitors
Hund, Pauli, Nordheim, Oppenheimer, Goeppert–Mayer, Condon, Pauling, Fock, Frenkel, Tamm, Dirac,
Mott, Klein, Heitler, London, von Neumann, Teller, Wigner and dozens of others. Being Jewish, Max Born
lost his job in 1933; he emigrated and became professor in Edinburgh, where he stayed for twenty years.
Physics at Göttingen University never recovered from this loss. For his elucidation of the meaning of the
wave function he received the 1954 Nobel prize in physics.

Dvipsbugw
 708 v quanta of light and matter • 18. minimum action

What does ‘quantum’ mean?

Quantum theory results from the existence of minimum measurable values in nature, pre-
cisely in the way that Galileo already speculated about in the seventeenth century. As
Page 220 mentioned in detail earlier on, it was Galileo’s insistence on these ‘piccolissimi quanti’
that got him accused. Of course, we will discover that only the idea of a smallest change
leads to a precise and accurate description of nature. The term ‘quantum’ theory does not
mean that all measurement values are multiples of a smallest one; this is correct only in
certain special cases. Dvipsbugw
Table 56 also shows that the term ‘microscopic’ has a different meaning for a physicist
and for a biologist. For a biologist, a system is microscopic if it requires a microscope for
its observation. For a physicist however, a system is microscopic if its characteristic action
is of the order of the quantum of action. In short, for a physicist, a system is microscopic
if it is not visible in a (light) microscope. To increase the confusion, some quantum phys-
icists nowadays call their own class of microscopic systems ‘mesoscopic,’ whereas many

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

classical, macroscopic systems are now called ‘nanoscopic’. Both names mainly help to
attract attention and funding.
There is another way to characterize the difference between a microscopic or quantum
system on one side and a macroscopic or classical system on the other. A minimum action
implies that the difference of action values S between two successive observations of the
same system, spaced by a time ∆t, is limited. Therefore one has

S(t + ∆t) − S(t) = (E + ∆E)(t + ∆t) − Et = E∆t + t∆E + ∆E∆t E

ħ
. (474)
2
Now the value of the energy E and of the time t – but not that of ∆E or of ∆t – can be set
to zero if we choose a suitable observer; thus the existence of a quantum of action implies
that in any system the evolution is constrained by

∆E∆t E
ħ
, (475)
2
where E is the energy of the system and t its age. In other words, ∆E is the change of
Challenge 1171 ny energy and ∆t the time between two successive observations. By a similar reasoning we
find that for any system the position and momentum values are constrained by Copyright © Christoph Schiller November 1997–May 2006

∆x∆p E
ħ
, (476)
2
where ∆p is the indeterminacy in momentum and ∆x the indeterminacy in position.
These two famous relations were called indeterminacy relations by their discoverer,
Werner Heisenberg.* The name is often incorrectly translated into English as ‘uncertainty

* One often hears the myth that the indeterminacy relation for energy and time has another weight than
the one for momentum and position. That is wrong; it is a myth propagated by the older generation of
physicists. This myth survived through many textbooks for over 70 years; just forget it, as it is incorrect. It
is essential to remember that all four quantities appearing in the inequalities are quantities describing the
internal properties of the system. In particular, it means that t is some time variable deduced from changes

Dvipsbugw
minimum action – quantum theory for poets and l aw yers 709

TA B L E 56 Some small systems in motion and the observed action values for their changes

System & change Action Motion

Light
Smallest amount of light absorbed by a coloured surface 1ħ quantum
Smallest hit when light reflects from mirror 2ħ quantum
Smallest visible amount of light c. 5 ħ quantum
Smallest amount of light absorbed in flower petal c. 1 ħ quantum Dvipsbugw
Blackening of photographic film c. 3 ħ quantum
Photographic flash c. 1017 ħ classical
Electricity
Electron ejected from atom c. 1 − 2 ħ quantum
Electron added to molecule c. 1 − 2 ħ quantum
Electron extracted from metal c. 1 − 2 ħ quantum

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Electron motion inside microprocessor c. 2 − 6 ħ quantum
Signal transport in nerves, from one molecule to the next c. 5 ħ quantum
Current flow in lighting bolt c. 1038 ħ classical
Materials science
Tearing apart two neighbouring iron atoms c. 1 − 2 ħ quantum
Breaking a steel bar c. 1035 ħ classical
Basic process in superconductivity 1ħ quantum
Basic process in transistors 1ħ quantum
Basic process in magnetic effects 1ħ quantum
Chemistry
Atom collisions in liquids at room temperature c. 1 ħ quantum
Shape oscillation of water molecule c. 1 − 5 ħ quantum
Shape change of molecule, e.g. in chemical reaction c. 1 − 5 ħ quantum
Single chemical reaction curling a hair c. 2 − 6 ħ quantum
Tearing apart two mozzarella molecules c. 300 ħ quantum
Smelling one molecule c. 10 ħ quantum
Burning fuel in a cylinder in an average car engine explosion c. 1037 ħ classical
Life
Copyright © Christoph Schiller November 1997–May 2006

Air molecule hitting ear drum c. 2 ħ quantum

Smallest sound signal detectable by the ear Challenge 1170 ny
DNA duplication step in cell division c. 100 ħ quantum
Ovule fecundation c. 1014 ħ classical
Smallest step in molecular motor c. 5 ħ quantum
Sperm motion by one cell length c. 1015 ħ classical
Cell division c. 1019 ħ classical
Fruit fly’s wing beat c. 1024 ħ classical
Person walking one body length c. 2 ë 1036 ħ classical
Nuclei and stars
Nuclear fusion reaction in star c. 1 − 5 ħ quantum
Particle collision in accelerator c. 1 ħ quantum
Explosion of gamma ray burster c. 1080 ħ classical

Dvipsbugw
 710 v quanta of light and matter • 18. minimum action

relations’. However, this latter name is wrong: the quantities are not uncertain, but undeter-
mined. Due to the quantum of action, system observables have no definite value. There
is no way to ascribe a precise value to momentum, position and other observables of a
quantum system.
Any system whose indeterminacy is of the order of ħ is a quantum system; if the inde-
terminacy product is much larger, the system is classical, and classical physics is sufficient
for its description. In other words, even though classical physics assumes that there are no
measurement indeterminacies in nature, a system is classical only if its indeterminacies
are large compared to the minimum possible ones. As a result, quantum theory is neces- Dvipsbugw
sary in all those cases in which one tries to measure some quantity as precisely as possible.
The indeterminacy relations again show that motion cannot be observed to infinite pre-
cision. In other words, the microscopic world is fuzzy. This strange result has many im-
portant and many curious consequences. For example, if motion cannot be observed with
infinite precision, the very concept of motion needs to be used with great care, as it can-
not be applied in certain situations. In a sense, the rest of our quest is an exploration of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the implications of this result. In fact, as long as space-time is flat, it turns out that we
can keep motion as a concept describing observations, provided we remain aware of the
limitations of the quantum principle.

Quantum surprises
The quantum of action implies short-time deviations from energy, momentum and angu-
lar momentum conservation in microscopic systems. Now, in the first part of our moun-
Page 154 tain ascent we realized that any type of nonconservation implies the existence of surprises
in nature. Well, here are some of them.
Since uniform motion does not exist in the precise meaning of the term, a system mov-
ing in one dimension only, such as the hand of a clock, always has a possibility to move
a bit in the opposite direction, thus leading to incorrect readings. Indeed, quantum the-
ory predicts that clocks have limits, and that perfect clocks do not exist. In fact, quantum
theory implies that strictly speaking, one-dimensional motion does not exist.
Challenge 1172 ny Obviously, the limitations apply also to metre bars. Thus the quantum of action is
responsible on one hand that the possibility to perform measurements exists, and on the
other hand for the limitations of measurements.
In addition, it follows from the quantum of action that any observer must be large to
be inertial or freely falling, as only large systems approximate inertial motion. An observer
Copyright © Christoph Schiller November 1997–May 2006

cannot be microscopic. If humans were not macroscopic, they could neither observe nor
study motion.

observed inside the system and not the external time coordinate measured by an outside clock, in the same
Ref. 681 way that the position x is not the external space coordinate, but the position characterizing the system.
Werner Heisenberg (1901–1976) was an important German theoretical physicist and an excellent table
tennis and tennis player. In 1925, as a young man, he developed, with some help by Max Born and Pas-
cual Jordan, the first version of quantum theory; from it he deduced the indeterminacy relations. For these
achievements he received the Nobel Prize for physics in 1932. He also worked on nuclear physics and on tur-
bulence. During the second world war, he worked in the German nuclear fission program. After the war, he
published several successful books on philosophical questions in physics and he unsuccessfully tried, with
some half-hearted help by Wolfgang Pauli, to find a unified description of nature based on quantum theory,
the ‘world formula’.

Dvipsbugw
 minimum action – quantum theory for poets and l aw yers 711

Due to the finite accuracy with which microscopic motion can be observed, faster
than light motion should be possible in the microscopic domain. Quantum theory thus
predicts tachyons, at least over short time intervals. For the same reason, also motion
backwards in time should be possible over microscopic times and distances. In short, a
quantum of action implies the existence of microscopic time travel.
But there is more: the quantum of action implies that there is no permanence in nature.
Imagine a moving car suddenly disappearing for ever. In such a situation neither mo-
mentum nor energy would be conserved. The action change for such a disappearance
is large compared to ħ, so that its observation would contradict even classical physics, Dvipsbugw
Challenge 1173 ny as you might want to check. However, the quantum of action allows that a microscopic
particle, such as an electron, disappears for a short time, provided it reappears afterwards.
The quantum of action also implies that the vacuum is not empty. If one looks at empty
space twice in a row, the two observations being spaced by a tiny time interval, some en-
ergy will be observed the second time. If the time interval is short enough, due to the
quantum of action, matter particles will be observed. Indeed, particles can appear any-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where from nowhere, and disappear just afterwards, as the action limit requires it. In
other words, classical physics’ idea of an empty vacuum is correct only when observed
over long time scales. In summary, nature shows short time appearance and disappear-
ance of matter.
The quantum of action implies that compass needles cannot work. If we look twice in
a row at a compass needle or even at a house, we usually observe that they stay oriented
Challenge 1174 ny in the same direction. But since physical action has the same unit as angular momentum,
a minimum value for action also means a minimum value for angular momentum. There-
fore, every macroscopic object has a minimum value for its rotation. In other words,
quantum theory predicts that in everyday life, everything rotates. Lack of rotation exists
only approximately, when observations are spaced by long time intervals.
For microscopic systems, the situation is more involved. If their rotation angle can
be observed, such as for molecules, they behave like macroscopic objects: their position
and their orientation are fuzzy. But for those systems whose rotation angle cannot be ob-
served, the quantum of action turns out to have somewhat different consequences. Their
angular momentum is limited to values which are multiples of ħ~2. As a result, all mi-
croscopic bound systems, such as molecules, atoms, or nuclei, contain rotational motion
and rotating components.
But there is more to come. A minimum
action implies that cages in zoos are dan-
Copyright © Christoph Schiller November 1997–May 2006

gerous and banks are not safe. A cage is a

feature requiring a lot of energy to be over- E
come. Mathematically, the wall of a cage m p
is an energy hill, similar to the one shown
in Figure 284. If a particle on one side of 0 ∆x
the hill has momentum p, it is simple to F I G U R E 284 Hills are never high enough
show that the particle can be observed on
the other side of the hill, at position ∆x, even if its kinetic energy p2 ~2m is smaller than
the height E of theºhill. In everyday life this is impossible. But imagine that the missing
momentum ∆p = 2mE − p to overcome the hill satisfies ∆x∆p E ħ~2. The quantum of

Dvipsbugw
 712 v quanta of light and matter • 18. minimum action

action thus implies that a hill of width

∆x D º
ħ~2
(477)
2mE − p

is not an obstacle to a particle of mass m. But this is not all. Since the value of the particle
momentum p is itself undetermined, a particle can overcome the hill even if the hill is
wider than value (477), though the broader it is the smaller the probability is. As a result,
Dvipsbugw
any particle can overcome any obstacle. This effect, for obvious reasons, is called the tun-
nelling effect. In short, the minimum action principle implies that there are no safe boxes
in nature. Due to tunnelling, matter is not impenetrable, in contrast to everyday, classical
Challenge 1175 ny observation. Can you explain why lion cages work despite the quantum of action?
By the way, the quantum of action also implies that a particle with a kinetic energy
larger than the energy height of a hill can get reflected by the hill. Classically this is im-
Challenge 1176 ny possible. Can you explain the observation?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The minimum action principle also
implies that book shelves are dangerous.
E1
Shelves are obstacles to motion. A book
in a shelf is in the same situation as the
mass in Figure 285; the mass is surroun- m E2
ded by energy hills hindering its escape to
the outer, lower energy world. Now, due to
the tunnelling effect, escape is always pos- F I G U R E 285 Leaving enclosures
sible. The same picture applies to a branch
of a tree, a nail in a wall, or to anything attached to anything else. Fixing things to each
other is never for ever. We will find out that every example of light emission, even radio-
activity, results from this effect. The quantum of action thus implies that decay is part of
nature. In short, there are no stable excited systems in nature. For the same reason by the
Challenge 1177 ny way, no memory can be perfect. (Can you confirm the deduction?) Note that decay often
appears in everyday life, where it just has a different name: breaking. In fact, all cases in
Ref. 682 which something breaks require the quantum of action for their description. Obviously,
the cause of breaking is often classical, but the mechanism of breaking is always quantum.
Only objects that follow quantum theory can break.
Taking a more general view, also ageing and death result from the quantum of action.
Death, like ageing, is a composition of breaking processes. Breaking is a form of decay,
Copyright © Christoph Schiller November 1997–May 2006

and is due to tunnelling. Death is thus a quantum process. Classically, death does not
Challenge 1178 ny exist. Might this be the reason that so many believe in immortality or eternal youth?
We will also discover that the quantum of action is the origin for the importance of
the action observable in classical physics. In fact, the existence of a minimal action is the
reason for the least action principle of classical physics.
A minimum action also implies that matter cannot be continuous, but must be com-
posed of smallest entities. Indeed, the flow of a truly continuous material would contra-
Challenge 1179 ny dict the quantum principle. Can you give the precise argument? Of course, at this point
of our adventure, the non-continuity of matter is no news any more. In addition, the
quantum of action implies that even radiation cannot be continuous. As Albert Einstein

Dvipsbugw
 minimum action – quantum theory for poets and l aw yers 713

stated clearly for the first time, light is made of quantum particles. More generally, the
quantum of action implies that in nature all flows and all waves are made of microscopic
particles. The term ‘microscopic’ or ‘quantum’ is essential, as such particles do not behave
like little stones. We have already encountered several differences and we will encounter
others shortly. For these reasons, microscopic particles should bear a special name; but
all proposals, of which quanton is the most popular, have not caught on yet.
The quantum of action has several
strange consequences for microscopic m
particles. Take two of them with the same Dvipsbugw
mass and the same composition. Ima-
gine that their paths cross and that at
the crossing they approach each other to
small distances, as shown in Figure 286.
A minimum action implies that in such
m
a situation, if the distance becomes small

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

enough, the two particles can switch role F I G U R E 286 Identical objects with crossing paths
without anybody being able to avoid or to
ever notice it. For example, in a gas it is impossible, due to the quantum of action, to fol-
low particles moving around and to say which particle is which. Can you confirm this
Challenge 1180 ny deduction and specify the conditions using the indeterminacy relations? In summary, in
nature it is impossible to distinguish identical particles. Can you guess what happens in the
Challenge 1181 n case of light?
But matter deserves still more atten-
tion. Imagine two particles, even two dif- M
ferent ones, approaching each other to m1
small distances, as shown in Figure 287.
We know that if the approach distance
gets small, things get fuzzy. Now, if some-
thing happens in that small domain in m2
such a way that the resulting outgoing
m
products have the same total momentum m3
and energy as the incoming ones, the
minimum action principle makes such F I G U R E 287 Transformation through reaction
processes possible. Indeed, ruling out
such processes would imply that arbitrary small actions could be observed, thus elim-
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1182 ny inating nature’s fuzziness, as you might want to check by yourself. In short, a minimum
action allows transformation of matter. One also says that the quantum of action allows
particle reactions. In fact, we will discover that all kinds of reactions in nature, including
chemical and nuclear ones, are only due to the existence of the quantum of action.
But there is more. Due to the indeterminacy relations, it is impossible to give a def-
inite value to both the momentum and the position of a particle. Obviously, this is also
impossible for all the components of a measurement set-up or an observer. This implies
that initial conditions – both for a system and for the measurement set-up – cannot be
exactly duplicated. A minimum action thus implies that whenever an experiment on a
microscopic system is performed twice, the outcome will be different. The result would
be the same only if both the system and the observer would be in exactly the same condi-

Dvipsbugw
 714 v quanta of light and matter • 18. minimum action

Dvipsbugw

F I G U R E 288 How do train F I G U R E 289 A particle and a

windows manage to show two screen with two nearby slits
superimposed images?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tion in both situations. This turns out to be impossible, both due to the second principle
of thermodynamics and due to the quantum principle. Therefore, microscopic systems
behave randomly. Obviously, there will be some average outcome; nevertheless, micro-
scopic observations are probabilistic. Albert Einstein found this conclusion of quantum
theory the most difficult to swallow, as this randomness implies that the behaviour of
quantum systems is strikingly different from that of classical systems. But the conclusion
is unavoidable: nature behaves randomly.
A good example is given by trains. Einstein used trains to develop and explain relativ-
ity. But trains are also important for quantum physics. Everybody knows that one can use
a train window to look either at the outside landscape or, by concentrating on the reflec-
ted image, to observe some interesting person inside the carriage. In other words, glass
reflects some of the light particles and lets some others pass through. More precisely, glass
reflects a random selection of light particles, yet with constant average. Partial reflection
is thus similar to the tunnelling effect. Indeed, the partial reflection of glass for photons is
a result of the quantum of action. Again, the average situation can be described by classical
physics, but the precise amount of partial reflection cannot be explained without quantum
theory. Without the quantum of action, train trips would be much more boring.

Waves
The quantum of action implies a central result about the path of particles. If a particle
Copyright © Christoph Schiller November 1997–May 2006

travels from a point to another, there is no way to say which path it has taken in between.
Indeed, in order to distinguish between the two possible, but only slightly different paths,
actions smaller than ħ~2 would have to be measured. In particular, if a particle is sent
through a screen with two sufficiently close slits, it is impossible to say through which slit
the particle passed to the other side. The impossibility is fundamental. Matter is predicted
to show interference.
We know already a moving phenomenon for which it is not possible to say with preci-
sion which path it takes when crossing two slits: waves. All waves follow the indetermin-
Page 210 acy relations
∆ω∆t E and ∆k∆x E .
1 1
(478)
2 2

Dvipsbugw
 minimum action – quantum theory for poets and l aw yers 715

We saw above that quantum systems follow

∆E∆t E ∆p∆x E
ħ ħ
and (479)
2 2
As a result, one is lead to ascribe a frequency and a wavelength to quantum systems:

E = ħω p = ħk = ħ
2π
and (480)
λ Dvipsbugw

The energy–frequency relation was deduced by Albert Einstein in 1905; it is found to be

valid in particular for all examples of light emission. The more spectacular momentum–
wavelength relation was first predicted by Louis de Broglie* in 1923 and 1924. (This is thus
another example of a discovery that was made about twenty years too late.) In short, the
quantum of action implies that matter particles behave as waves.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Information
In computer science, a smallest change is called a ‘bit’. The existence of a smallest change in
nature thus means that computer language or information science can be used to describe
nature, and in particular quantum theory. However, computer language can describe only
the software side; the hardware side of nature is also required. The hardware of nature
enters the description whenever the actual value ħ of the quantum of action must be
introduced.
Exploring the analogy between nature and information science, we will discover that
the quantum of action implies that macroscopic physical systems cannot be copied or
‘cloned’, as quantum theorists like to say. Nature does not allow to copy objects. Copying
machines do not exist. The quantum of action makes it impossible to gather and use all
information in a way that allows to produce a perfect copy. As a result, we will deduce that
the precise order in which measurements are performed does play a role in experiments.
When the order is important, physicists speak of lack of ‘commutation’. In short physical
observables do not commute.
We will also find out that the quantum of action implies that systems are not always in-
Page 800 dependent, but can be ‘entangled’. This term, introduced by Erwin Schrödinger, describes
the most absurd consequences of quantum theory. Entanglement makes everything in
nature connected to everything else. Entanglement produces effects which look (but are
Copyright © Christoph Schiller November 1997–May 2006

not) faster than light. Entanglement produces a (fake) form of non-locality. Entanglement
Ref. 683 also implies that trustworthy communication does not exist.
Don’t all these deductions look wrong or at least crazy? In fact, if you or your lawyer
made any of these statements in court, maybe even under oath, you would be likely to

* Louis de Broglie (b. 1892 Dieppe, d. 1987 Paris) French physicist and professor at the Sorbonne. The
energy–frequency relation had earned Albert Einstein his Nobel prize already in 1921. De Broglie expan-
ded it to the prediction of the wave nature of the electron (and of all other quantum particles); this was the
essential part of his PhD. The prediction was confirmed experimentally a few years later, in 1927. For the
prediction of the wave nature of matter, de Broglie received the Nobel Prize in physics in 1929. Being an
aristocrat, de Broglie never did anything else in research after that. For example, it was Schrödinger who
then wrote down the wave equation, even though de Broglie could equally have done it.

Dvipsbugw
 716 v quanta of light and matter • 18. minimum action

end up in prison! However, all above statements are correct, as they are all confirmed by
experiment. And the surprises are by far not finished. You might have noticed that, in
the preceding examples, no situation related to electricity, to the nuclear interactions or
to gravity was included. In these domains the surprises are even more astonishing; the
observation of antimatter, of electric current flow without resistance, of the motion inside
muscles, of vacuum energy, of nuclear reactions in stars and maybe soon of boiling empty
space will fascinate you as much as they have fascinated and still fascinate thousands of
researchers.
In particular, the consequences of the quantum of action on the early universe are Dvipsbugw
Challenge 1183 ny simply mind-boggling. Just try to explore for yourself its consequences for the big bang.
Together, all these topics will lead us a long way towards the top of Motion Mountain.
The topics are so strange, so incredible and at the same time so numerous that quantum
physics can be rightly called the description of motion for crazy scientists. In a sense,
this is the generalization of the previous definition, when we called quantum physics the
description of motion related to pleasure.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Sometimes it is heard that ‘nobody understands quantum theory.’ This is wrong. In
fact it is worse than wrong: it is disinformation, a habit found only in dictatorships. It
is used there to prevent people from making up their own mind and from enjoying life.
Quantum theory is the set of consequences that follows from the existence of a minimal
action. These consequences can be understood and enjoyed by everybody. In order to do
so, our first task on our way towards the top of Motion Mountain will be the study of our
classical standard of motion: the motion of light.
Nie und nirgends hat es Materie ohne

“ Bewegung gegeben, oder kann es sie geben.

Friedrich Engels, Anti–Dühring.*

”
Curiosities and fun challenges about the quantum of action
Even if we accept that experiments so far do not contradict the minimum action, we still
have to check that the minimum action does not contradict reason. In particular, the
minimum action must also appear in all imagined experiments. This is not evident.

**
Angular momentum has the same unit as action. A smallest action implies that there is a
smallest angular momentum in nature. How can this be, given that some particles have
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1184 n spin zero, i.e., have no angular momentum?

**
Could we have started the whole discussion of quantum theory by stating that there is a
Challenge 1185 n minimum angular momentum instead of a minimum action?
**
Niels Bohr, besides propagating the idea of minimum action, was also a fan of the comple-
mentarity principle. This is the idea that certain pairs of observables of a system – such as
Ref. 684 * ‘Never and nowhere has matter existed, or can it exist without motion.’ Friedrich Engels (1820–1895) was
one of the theoreticians of Marxism, often also called Communism.

Dvipsbugw
 light – the strange consequences of the quantum of action 717

position and momentum – have linked precision: if one observable of the pair is known
to high precision, the other is necessarily known with low precision. Can you deduce the
Challenge 1186 ny principle from the minimum action?
**
When electromagnetic fields play a role, the value of the action (usually) depends on the
choice of the vector potential, and thus on the gauge choice. We found out in the section
of electrodynamics that a suitable gauge choice can change the action value by adding or
subtracting any desired amount. Nevertheless, there is a smallest action in nature. This Dvipsbugw
is possible, because in quantum theory, physical gauge changes cannot add or subtract
any amount, but only multiples of twice the minimum value. The addition property thus
does not help to go below the minimum action.
**
(Adult) plants stop to grow in the dark. Plant needs light to grow. Without light, the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

reactions necessary for growth stop. Can you deduce that this is a quantum effect, not
Challenge 1187 ny explainable by classical physics?

In short, besides experiment also all imagined system confirm that nature shows a
minimum action.

19. light – the strange consequences of the

quantum of action
Alle Wesen leben vom Lichte,

“ jedes glückliche Geschöpfe.

Friedrich Schiller, Wilhelm Tell.*

”
What is colour?
If all the colours of materials are quantum effects, as just argued, it becomes especially
Ref. 686 interesting to study the properties of light in the light of the quantum of action. If in
nature there is a minimum change, there should also be a minimum illumination. This
had been already predicted by Epicurus (341–271 bce ) in ancient Greece. He stated that
Copyright © Christoph Schiller November 1997–May 2006

light is a stream of little particles, so that the smallest illumination would be that of a
single light particle.
Ref. 685 In the 1930s, Brumberg and Vavilov found a beautiful way to check this prediction
using the naked eye, despite our inability to detect single photons. In fact, the experi-
ment is so simple that it could have been performed at least a century before that; but
nobody had a sufficiently daring imagination to try it. The two researchers constructed
a small shutter that could be opened for time intervals of 0.1 s. From the other side, in
a completely dark room, they illuminated the opening with extremely weak green light

* ‘All beings live of light, every happy creature.’ Friedrich Schiller (b. 1759 Marbach, d. 1805 Weimar), im-
portant German poet, playwright and historian.

Dvipsbugw
718 v quanta of light and matter • 19. light

– 20 aW at 505 nm. At that intensity, whenever the shutter opens, on average about 50
photons can pass, which is just the sensitivity threshold of the eye. To perform the exper-
iment, they simply looked into the shutter repeatedly. The result is surprising but simple.
Sometimes they observed light, sometimes they did not. The result is completely random.
Brumberg and Vavilov also gave the simple explanation: due to fluctuations, half of the
time the number of photons is above eye threshold, half of the time below. The fluctu-
ations are random, and thus the detection is as well. This would not happen if light would
be a continuous stream; in that case, the eye would detect light at every opening of the
shutter. (At higher light intensities, the percentage of non-observations quickly decreases, Dvipsbugw
in accordance with the explanation.) Nobody knows what would have happened to the
description of light if this simple experiment had been performed 100 years earlier.
The experiment becomes clearer when we
use devices to help us. A simple way is to start photographic
glass
with a screen behind a prism illuminated with film
white light. The light is split into colours. When

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the screen is put further and further away, the white red
illumination intensity cannot become infinitely
small, as that would contradict the quantum of violet

action. To check this prediction, we only need

some black and white photographic film. Every- F I G U R E 290 Illumination by pure-colour
body knows that film is blackened by daylight light
of any colour; at medium light intensities it be-
comes dark grey and at lower intensities light grey. Looking at an extremely light grey film
under the microscope, we discover that even under uniform illumination the grey shade
actually is a more or less dense collections of black spots. Exposed film does not show a
homogeneous colour; on the contrary, it reacts as if light is made of small particles.
This is a general observation: whenever sensitive light detectors are constructed with
the aim to ‘see’ as accurately as possible, thus in environments as dark as possible, one
always finds that light manifests itself as a stream of light quanta. Nowadays they are
usually called photons, a term that appeared in 1926. A low or high light intensity is simply
a small or high number of photons.
Another weak source of light
are single atoms. Atoms are tiny
spheres; when they radiate light
or X-rays, the radiation should
Copyright © Christoph Schiller November 1997–May 2006

be emitted as spherical waves.

Figure to be inserted
But in all experiments it is found
that each atom emits only one
‘blob’ of light. One never finds F I G U R E 291 Observation of photons
that the light emitted by atoms
forms a spherical wave, as is suggested by everyday physics. If a radiation emitting atom
is surrounded by detectors, there is always only a single detector that is triggered.
Experiments in dim light thus show that the continuum description of light is not
correct. More precise measurements confirm the role of the quantum of action: every
photon leads to the same amount of change in the film. This amount of change is the
minimal amount of change that light can produce. Indeed, if a minimum action would not

Dvipsbugw
 light – the strange consequences of the quantum of action 719

exist, light could be packaged into arbitrary small amounts. In other words, the classical
Page 552 description of light by a continuous state function A(t, x) or F(t, x), whose evolution
is described by a principle of least action, is wrong, as it does not describe the observed
particle effects. Another, modified description is required. The modification has to be
important only at low light intensities, since at high intensities the classical Lagrangian
accurately describes all experimental observations.*
At which intensities does light cease to behave as a continuous wave? Our eye can
help us to find a limit. Human eyesight does not allow to consciously distinguish single
Ref. 687 photons, even though experiments show that the hardware of the eye is able to do this. Dvipsbugw
The faintest stars which can be seen at night produce a light intensity of about 0.6 nW~m2 .
Since the pupil of the eye is quite small, and as we are not able to see individual photons,
photons must have energies smaller than 10−16 J.
In today’s laboratory experiments, recording and counting individual photons is stand-
ard practice. Photon counters are part of many spectroscopy set-ups, such as those used
to measure smallest concentration of materials. For example, they help to detect drugs in

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

human hair. All these experiments thus prove directly that light is a stream of particles,
as Epicurus had advanced in ancient Greece.
This and many other experiments show that a beam of light of frequency f or angular
frequency ω, which determines its colour, is accurately described as a stream of photons,
each with the same energy E given by

E = ħ 2π f = ħ ω . (481)

This shows that for light, the smallest measurable action is given by the quantum of action
ħ. This is twice the smallest action observable in nature; the reasons and implications
will unfold during the rest of our walk. In summary, colour is a property of photons. A
coloured light beam is a hailstorm of the corresponding photons.
The value of Planck’s constant can be determined from measurements of black bodies
Page 612 or other light sources. The result

ħ = 1.1 ë 10−34 Js (482)

is so small that we understand why photons go unnoticed by humans. Indeed, in normal

light conditions the photon numbers are so high that the continuum approximation for
the electromagnetic field is of high accuracy. In the dark, the insensitivity of the signal pro-
Copyright © Christoph Schiller November 1997–May 2006

cessing of the human eye, in particular the slowness of the light receptors, makes photon
Ref. 687 counting impossible. The eye is not far from maximum possible sensitivity though. From
the numbers given above about dim stars we can estimate that humans are able to see
Challenge 1188 ny consciously flashes of about half a dozen detected photons.
In the following, we will systematically deduce the remaining properties of photons,
using the data collected in classical physics, while taking the quantum of action firmly
into account. For example, photons have no (rest) mass and no electric charge. Can you
Challenge 1189 ny confirm this? In fact, experiments can only give an upper limit for both quantities. The

* This transition from the classical case to the quantum case used to be called quantization. The concept and
the ideas behind it are only of historical interest today.

Dvipsbugw
 720 v quanta of light and matter • 19. light

Ref. 688 present experimental upper limit for the (rest) mass of a photon is 10−51 kg.
We know that light can hit objects. Since the energy and the speed of photons is known,
Challenge 1190 ny we guess that the photon momentum obeys

p= =ħ or p = ħ k .
E 2π
(483)
c λ
In other words, if light is made of particles, we should be able to play billiard with them.
Ref. 689 This is indeed possible, as Arthur Compton showed in a famous experiment in 1923. He Dvipsbugw
directed X-rays, which are high energy photons, onto graphite, a material in which elec-
trons move almost freely. He found that whenever the electrons in the material get hit by
the X-ray photons, the deflected X-rays change colour. As expected, the strength of the hit
depends on the deflection angle of the photon. From the colour change and the reflection
angle, Compton confirmed that the photon momentum indeed obeys the above expres-
sion. All other experiments agree that photons have momentum. For example, when an

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

atom emits light, the atom feels a recoil; the momentum again turns out to be given by the
same value (483). In short, every photon has momentum.
The value of photon momentum respects the indeterminacy principle. In the same way
that it is impossible to measure exactly both the wavelength of a wave and the position
of its crest, it becomes impossible to measure both the momentum and the position of a
Challenge 1191 ny photon. Can you confirm this? In other words, the value of the photon momentum is a
direct consequence of the quantum of action.
From our study of classical physics we know that light has a property beyond its col-
our: light can be polarized. That is only a complicated way to say that light can turn ob-
Page 573 jects it shines on. In other words, light has an angular momentum oriented along the
axis of propagation. What about photons? Measurements consistently find that each light
particle carries an angular momentum given by L = ħ. It is called its helicity; to make it
more clear that the quantity is similar to that found for massive particles, one also speaks
of the spin of photons. Photons somehow ‘turn’ – in a sense either parallel or antiparallel
to the direction of motion. Again, the magnitude of the photon helicity or spin is not
Challenge 1192 ny a surprise; it confirms the classical relation L = E~ω between energy and angular mo-
mentum that we found in the section on classical electrodynamics. Note that in contrast
to intuition, the angular momentum of a single photon is fixed, and thus independent of
its energy. Even the photons with the highest energy have L = ħ. Of course, the value of
the helicity also respects the limit given by the quantum of action. The helicity value ħ –
Copyright © Christoph Schiller November 1997–May 2006

twice the minimum ħ~2 – has important consequences which will become clear shortly.

What is light? – Again

La lumière est un mouvement luminaire de

“ corps lumineux.*
Blaise Pascal

”
* Light is the luminary movement of luminous bodies.

Dvipsbugw
 light – the strange consequences of the quantum of action 721

In the seventeenth century, Blaise Pascal* used this sentence to make fun about certain
physicists. He ridiculed (rightly so) the blatant use of a circular definition. Of course,
he was right; in his time, the definition was indeed circular, as no meaning could be
given to any of the terms. But as usual, whenever an observation is studied with care
by physicists, they give philosophers a beating. All those originally undefined terms now
have a definite meaning. Light is indeed a type of motion; this motion can rightly be called
luminary because in opposition to motion of material bodies, it has the unique property
v = c; the luminous bodies, today called photons, are characterized and differentiated
from all other particles by their dispersion relation E = c p, their energy E = ħω, their Dvipsbugw
spin L = ħ, the vanishing value of all other quantum numbers, and by being the quanta
of the electromagnetic field.
In short, light is a stream of photons. It is indeed a luminary movement of luminous
bodies. The existence of photons is the first example of a general property of the world
on small scales: all waves and all flows in nature are made of quantum particles. Large
numbers of (coherent) quantum particles – or quantons – do indeed behave as waves. We

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

will see shortly that this is the case even for matter. The fundamental constituents of all
waves are quantons. There is no exception. The everyday, continuum description of light
is thus similar in many aspects to the description of water as a continuous fluid; photons
are the atoms of light, and continuity is an approximation valid for large particle numbers.
Small numbers of quantons often behave like classical particles. In the old days of physics,
books used to discuss at length a so-called wave–particle duality. Let us be clear from the
start: quantons, or quantum particles, are neither classical waves nor classical particles. In
the microscopic world, quantons are the fundamental objects.
However, a lot is still unclear. Where inside matter do these monochromatic photons
come from? Even more interestingly, if light is made of quantons, all electromagnetic
fields, even static ones, must be made of photons as well. However, in static fields nothing
is flowing. How is this apparent contradiction solved? And what effects does the particle
aspect have on these static fields? What is the difference between quantons and classical
particles? The properties of photons thus require some more careful study. Let us go on.

The size of photons

First of all, we might ask: what are these photons made of? All experiments so far, per-
formed down to the present limit of about 10−20 m, give the same answer: ‘we can’t find
anything’. That is consistent both with a vanishing mass and a vanishing size of photons;
Copyright © Christoph Schiller November 1997–May 2006

indeed, one intuitively expects any body with a finite size to have a finite mass. Thus, even
though experiments give only an upper limit, it is consistent to claim that a photon has
no size.
A particle with no size cannot have any constituents. A photon thus cannot be divided
into smaller entities. For this reason people refer to photons as elementary particles. We
Challenge 1193 ny will give some strong additional arguments for this deduction soon. (Can you find one?)
This is a strange result. How can a photon have vanishing size, have no constituents, and
still be something? This is a hard questions; the answer will appear only later on. At the

* Blaise Pascal (b. 1623 Clermont, d. 1662 Paris) important French mathematician and physicist up to the
age of twenty-six; he then turned theologian and philosopher.

Dvipsbugw
722 v quanta of light and matter • 19. light

moment we simply have to accept the situation as it is. We therefore turn to an easier
issue.

Are photons countable? – Squeezed light

Also gibt es sie doch.

“
Above we saw that in order to count photons, the simplest way is to distribute them across
a large screen and then to absorb them. But this method is not fully satisfying, as it des-
Max Planck*

” Dvipsbugw
troys the photons. How can one count photons without destroying them?
One way is to reflect photons on a mirror and to measure the recoil of the mirror.
This seems almost unbelievable, but nowadays the effect is becoming measurable even
for small number of photons. For example, this effect has to be taken into account in the
mirrors used in gravitational wave detectors, where the position of laser mirrors has to
be measured to high precision.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Another way of counting photons without destroying them uses special high quality
laser cavities. Using smartly placed atoms inside such a cavity, it is possible to count the
number of photons by the effect they have on these atoms.
In other words, light intensity can indeed be measured without absorption. However,
the next difficulty appears straight away. Measurements show that even the best light
beams, from the most sophisticated lasers, fluctuate in intensity. There are no steady
beams. This does not come as a surprise: if a light beam would not fluctuate, observing it
twice in a row would yield a vanishing value for the action. However, there is a minimum
action in nature, namely ħ~2. Thus any beam and any flow in nature fluctuates. But there
is more.
A light beam is described by its intensity and its phase. The change – or action – oc-
curring while a beam moves is given by the variation in the product of intensity and
phase. Experiments confirm the obvious deduction: intensity and phase of beams behave
like momentum and position of particles: they obey an indeterminacy relation. You can
deduce it yourself, in the same way we deduced Heisenberg’s relations. Using as charac-
teristic intensity I = E~ω the energy per circular frequency and calling the phase φ, we
get**
∆I ∆φ E .
ħ
(485)
2
Copyright © Christoph Schiller November 1997–May 2006

* ‘Thus they do exist after all.’ Max Planck, in later years, said this after standing silently, for a long time, in
front of an apparatus which counted single photons by producing a click for each photon it detected. It is
not a secret that for a large part of his life, Planck was not a friend of the photon concept, even though his
own results were the starting point for its introduction.
** A large photon number is assumed in the expression; this is obvious, as ∆φ cannot grow beyond all
bounds. The exact relations are

∆I ∆ cos φ E S`sin φeS

1
2
∆I ∆ sin φ E S`cos φeS
1
(484)
2
where `xe denotes the expectation value of the observable x.

Dvipsbugw
 light – the strange consequences of the quantum of action 723

Thermal light Laser light Nonclassical light

Photon sequence: bunching little or no bunching anti-bunching

time
Intensity I(t)

Dvipsbugw
time
Probability P(I)

Sub-
Bose-Einstein Poisson Poisson

intensity

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Intensity correlations

2 2 2
1 1 1

coherence time
time
Amplitude-
phase
diagram

F I G U R E 292 Various types of light

For light emitted from usual lamps, the product of the left side is much larger than the
quantum of action. On the other hand, laser beams can (almost) reach the limit. Laser
light beams in which the two indeterminacies strongly differ from each other are called
non-classical light or squeezed light; they are used in many modern research applications.
Copyright © Christoph Schiller November 1997–May 2006

Such light beams have to be treated carefully, as the smallest disturbances transform them
back into usual laser beams, where the two indeterminacies have the same value. An
extreme example of non-classical light are those beams with a given, fixed photon number,
thus with an extremely large phase indeterminacy.
The observation of non-classical light highlights a strange consequence valid even for
classical light: the number of photons in a light beam is not a defined quantity. In general
it is undetermined, and it fluctuates. The number of photons at the beginning of a beam is
not necessarily the same as at the end of the beam. Photons, in contrast to stones, cannot
be counted precisely – as long as they move and are not absorbed. In flight it is only
possible to determine an approximate number, within the limit set by indeterminacy.
The most extreme example are those light beams with an (almost) fixed phase. In such

Dvipsbugw
 724 v quanta of light and matter • 19. light

source detectors
mirrors
beam beam
splitter splitter
possible
two identical light
photons paths

F I G U R E 293 An interferometer
Dvipsbugw

beams, the photon number fluctuates from zero to infinity. In other words, in order to
produce a coherent laser beam that approximates a pure sine wave as perfectly as possible
one must build a source in which the photon number is as undetermined as possible.
The other extreme is a beam with a fixed number of photons; in such a beam, the phase

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fluctuates erratically. Most daily life situations, such as the light from incandescent lamps,
lie somewhere in the middle: both phase and intensity indeterminacies are of similar
magnitude.
As an aside, it turns out that in deep, dark intergalactic space, far from every star, there
still are about 400 photons per cubic centimetre. But also this number, like the number
Challenge 1194 ny of photons in a light beam, has its measurement indeterminacy. Can you estimate it?
In summary, unlike little stones, photons are not countable. And this it not the last
difference between photons and stones.

The position of photons

Where is a photon when it moves in a beam of light? Quantum theory gives a simple
answer: nowhere in particular. The proof is given most spectacularly by experiments with
interferometers; they show that even a beam made of a single photon can be split, be led
along two different paths and then be recombined. The resulting interference shows that
the single photon cannot be said to have taken either of the two paths. If one of the two
paths is blocked, the pattern on the screen changes. In other words, the photons somehow
must have taken both paths at the same time. Photons cannot be localized; they have no
position.*
This impossibility of localization can be specified. It is impossible to localize photons
in the direction transverse to the motion. It is less difficult to localize photons along the
Copyright © Christoph Schiller November 1997–May 2006

motion direction. In the latter case, the quantum of action implies that the longitudinal
position is uncertain within a value given by the wavelength of the corresponding colour.
Challenge 1195 ny Can you confirm this?
In particular, this means that photons cannot be simply visualized as short wave trains.
Photons are truly unlocalizable entities specific to the quantum world.
Now, if photons can almost be localized along their motion, we can ask the following
question: How are photons lined up in a light beam? Of course, we just saw that it does not

* This conclusion cannot be avoided by saying that photons are split at the beam splitter: if one puts a detector
into each arm, one finds that they never detect a photon at the same time. Photons cannot be divided.

Dvipsbugw
 light – the strange consequences of the quantum of action 725

D1

coincidence
counter

D2

F I G U R E 294 How to measure photon statistics:

Dvipsbugw
with an electronic coincidence counter and the
variation by varying the geometrical position of a
detector

make sense to speak of their precise position. But are photons in a perfect beam arriving
in almost regular intervals or not?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

To the shame of physicists, the study of this question was started by two astronomers,
Ref. 690 Robert Hanbury Brown and Richard Twiss, in 1956. They used a simple method to meas-
ure the probability that a second photon in a light beam arrives at a given time after a
first one. They simply split the beam, put one detector in the first branch and varied the
position of a second detector in the other branch.
Hanbury Brown and Twiss found that for coherent light the clicks in the two counters,
and thus the photons, are correlated. This result is in complete contrast with classical
electrodynamics. Photons are indeed necessary to describe light. In more detail, their
experiment showed that whenever a photon hits, the probability that a second one hits
just afterwards is highest. Photons in beams are thus bunched.
Every light beam shows an upper limit time to bunching, called the coherence time. For
times larger than the coherence time, the probability for bunching is low and independ-
ent of the time interval, as shown in Figure 294. The coherence time characterizes every
light beam, or better, every light source. In fact, it is more intuitive to use the concept of
coherence length, as it gives a clearer image for a light beam. For thermal lamps, the co-
herence length is only a few micrometers, a small multiple of the wavelength. The largest
coherence lengths, up to over 100 000 km, are realized in research lasers. Interestingly, co-
herent light is even found in nature; several special stars have been found to emit coherent
Ref. 691 light.
Even though the intensity of a good laser light is almost constant, laser beam photons
Copyright © Christoph Schiller November 1997–May 2006

still do not arrive in regular intervals. Even the best laser light shows bunching, though
with different statistics and to a lower degree than lamp light. Light for which photons ar-
rive regularly, thus showing so-called (photon) anti-bunching, is obviously non-classical
in the sense defined above; such light can be produced only by special experimental ar-
rangements. The most extreme example is pursued at present by several research groups;
they aim to construct light sources which emit one photon at a time, at regular time in-
tervals, as reliably as possible.
In summary, experiments force us to conclude that light is made of photons, but that
photons cannot be localized in light beams. It makes no sense to talk about the position
of a photon in general; the idea makes only sense in some special situations, and then
only approximately and as a statistical average.

Dvipsbugw
 726 v quanta of light and matter • 19. light

Ekin kinetic energy of

emitted electrons
lamp
electrons

threshold

Dvipsbugw
metal plate
frequency of lamp light ω
in vacuum

F I G U R E 295 The kinetic energy of electrons emitted in the

photoelectric effect

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Are photons necessary?
In light of the results uncovered so far, the answer to the title question is obvious. But
the issue is tricky. In school books, the photoelectric effect is usually cited as the first and
most obvious experimental proof of the existence of photons. In 1887, Heinrich Hertz
observed that for certain metals, such as lithium or caesium, incident ultraviolet light
leads to charging of the metal. Later it was shown that the light leads to the emission
of electrons, and that that the energy of the ejected electrons is not dependent on the
intensity of the light, but only dependent on the difference between ħ times its frequency
Ref. 693 and a material dependent threshold energy. In 1905, Albert Einstein predicted this result
from the assumption that light is made of photons of energy E = ħω. He imagined that
this energy is used partly to extract the electron over the threshold, and partly to give it
kinetic energy. More photons only lead to more electrons, not to faster ones.
Einstein received the Nobel price for this explanation. But Einstein was a genius; that
means he deduced the correct result by a somewhat incorrect reasoning. The (small) mis-
take was the prejudice that a classical, continuous light beam would produce a different
effect. It does not take a lot to imagine that a classical, continuous electromagnetic field
interacting with discrete matter, made of discrete atoms containing discrete electrons,
leads to exactly the same result, if the motion of electrons is described by quantum the-
Ref. 694 ory. Several researchers confirmed this point already early in the twentieth century. The
photoelectric effect by itself does not require photons for its explanation.
Copyright © Christoph Schiller November 1997–May 2006

Many researchers were unconvinced by the photoelectric effect. Historically, the most
important argument for the necessity of light quanta was given by Henri Poincaré. In 1911
and 1912, at age 57 and only a few months before his death, he published two influential
papers proving that the radiation law of black bodies, the one in which the quantum
Ref. 695 of action had been discovered by Max Planck, requires the introduction of photons. He
also showed that the amount of radiation emitted by a hot body is finite only due to the
quantum nature of the processes leading to light emission. A description of the processes
by classical electrodynamics would lead to infinite amounts of radiated energy. These
two influential papers convinced most of the sceptic physics researchers at the time that
it was worthwhile to study quantum phenomena in more detail. Poincaré did not know
about the action limit S E ħ~2; yet his argument is based on the observation that light of a

Dvipsbugw
 light – the strange consequences of the quantum of action 727

given frequency always has a minimum intensity, namely one photon. Splitting such a one
photon beam into two beams, e.g. using a half-silvered mirror, does produce two beams.
However, there is no way to find more than one photon in those two beams together.
Another interesting experiment requiring photons is the observation of ‘molecules of
Ref. 696 photons’. In 1995, Jacobson et al. predicted that the de Broglie wavelength of a packet of
photons could be observed. Following quantum theory it is given by the wavelength of
a single photon divided by the number of photons in the packet. The team argued that
the packet wavelength could be observable if one would be able to split and recombine
such packets without destroying the cohesion within the packet. In 1999, this effect was Dvipsbugw
indeed observed by de Pádua and his brazilian research group. They used a careful set-up
with a nonlinear crystal to create what they call a biphoton, and observed its interference
properties, finding a reduction of the effective wavelength by the predicted factor of two.
In the meantime, packages with three and even four entangled photons have been created
Ref. 697 and observed.
Still another argument for the necessity of photons is the mentioned recoil felt by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

atoms emitting light. The recoil measured in these cases is best explained by the emis-
sion of a photon in a particular direction. In contrast, classical electrodynamics predicts
the emission of a spherical wave, with no preferred direction.
Obviously, the observation of non-classical light, also called squeezed light, also argues
for the existence of photons, as squeezed light proves that photons indeed are an intrinsic
Ref. 698 aspect of light, necessary even when no interactions with matter play a role. The same is
true for the Hanbury Brown–Twiss effect.
Finally, the spontaneous decay of excited atomic states also requires the existence of
photons. A continuum description of light does not explain the observation.
In summary, the concept of photon is indeed necessary for a precise description of
light, but the details are often subtle, as the properties of photons are unusual and require a
change in thinking habits. To avoid these issues, all school books stop discussing photons
after the photoelectric effect. That is a pity; things get interesting only after that. To savour
the fascination, ponder the following issue. Obviously, all electromagnetic fields are made
of photons. Photons can be counted for gamma rays, X-rays, ultraviolet light, visible light
and infrared light. However, for lower frequencies, photons have not been detected yet.
Can you imagine what would be necessary to count the photons emitted from a radio
Challenge 1196 ny station?
The issue directly leads to the most important question of all: Copyright © Christoph Schiller November 1997–May 2006

How can a wave be made up of particles?

Fünfzig Jahre intensiven Nachdenkens haben

“ mich der Antwort auf die Frage ‘Was sind

Lichtquanten?’ nicht näher gebracht. Natürlich
bildet sich heute jeder Wicht ein, er wisse die
Antwort. Doch da täuscht er sich.
Albert Einstein, 1951 *

”
* ‘Fifty years of intense reflection have not brought me nearer to the answer of the question ‘What are light
quanta?’ Of course nowadays every little mind thinks he knows the answer. But he is wrong.’

Dvipsbugw
728 v quanta of light and matter • 19. light

Dvipsbugw

lasers or other
pocket lamps coherent sources

F I G U R E 296 Light crossing light

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

If a light wave is made of particles, one must be able to explain each and every wave prop-
erties with the help of photons. The experiments mentioned above already hinted that
this is possible only because photons are quantum particles. Let us take a more detailed
look at this connection.
Light can cross other light undisturbed. This observation is not hard to explain with
photons; since photons do not interact with each other, and since they are point-like,
they ‘never’ hit each other. In fact, there indeed is an extremely small probability for their
interaction, as will be found below, but this effect is not observable in everyday life.
But the problems are not finished yet. If two light beams of identical frequencies and
fixed phase relation cross, we observe alternating bright and dark regions, so-called in-
terference fringes. How do these interference fringes appear? Obviously, photons are not
detected in the dark regions. How can this be? There is only one possible way to answer:
the brightness gives the probability for a photon to arrive at that place. The fringes imply
that photons behave like little arrows. Some additional thinking leads to the following
description:
(1) the probability of a photon arriving somewhere is given by the square of an arrow;
(2) the final arrow is the sum of all arrows getting there, taking all possible paths;
(3) the arrow’s direction stays fixed in space when photons move;
(4) the length of an arrow shrinks with the square of the travelled distance;
(5) photons emitted by one-coloured sources are emitted with arrows of constant
Copyright © Christoph Schiller November 1997–May 2006

length pointing in direction ω t; in other words, such monochromatic sources spit out
photons with a rotating mouth.
(6) photons emitted by thermal sources, such as pocket lamps, are emitted with arrows
of constant length pointing in random directions.
With this model* we can explain the stripes seen in laser experiments, such as those of
Figure 296 and Figure 297. You can check that in some regions, the two arrows travelling
through the two slits add up to zero for all times. No photons are detected there. In other
regions, the arrows always add up to the maximal value. These regions are always bright.
In between regions give in between shades. Obviously, for the case of pocket lamps the

* The model gives a correct description of light with the exception that it neglects polarization.

Dvipsbugw
 light – the strange consequences of the quantum of action 729

brightness also behaves as expected: the averages then simply add up, as in the common
region in the left case of Figure 296.
You might want to calculate the distance of
the lines when the source distance, the colour
Challenge 1197 ny and the distance to the screen is given. two lasers or screen
Obviously, the photon model implies that in- point sources
terference patterns are built up as the sum of a
large number of one-photon hits. Using low in- S1
tensity beams, we should therefore be able to see Dvipsbugw
how these little spots slowly build up an interfer-
ence pattern by accumulating at the bright spots
and never hitting the dark regions. That is indeed S2
the case. All experiments have confirmed this de-
scription.
It is important to stress that interference of the arrow model:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

two light beams is not the result of two different
photons cancelling out or adding each other up. t1
The cancelling would contradict energy and mo-
mentum conservation. Interference is an effect t2
valid for each photon separately, because each
photon is spread out over the whole set-up; each t3
photon takes all possible paths and interferes. As
Paul Dirac said, each photon interferes only with F I G U R E 297 Interference and the
Ref. 699 itself. Interference only works because photons description of light with arrows (at one
are quantons, and not at all classical particles. particular instant of time)
Dirac’s widely cited statement leads to a fam-
ous paradox: if a photon can interfere only with itself, how can two laser beams from two
different lasers show interference? The answer of quantum physics is simple but strange:
in the region where the beams interfere, there is no way to say from which source a photon
is arriving. Photons are quantons; the photons in the crossing region cannot be said to
come from a specific source. Photons in the interference region are quantons on their
own right, which indeed interfere only with themselves. In that region, one cannot hon-
estly say that light is a flow of photons. Despite regular claims of the contrary, Dirac’s
statement is correct. That is the strange result of the quantum of action.
Waves also show diffraction. To understand this phenomenon with photons, let us
Copyright © Christoph Schiller November 1997–May 2006

start with a simple mirror and study reflection first. Photons (like any quantum particle)
move from source to detector in all ways possible. As the discoverer of this explanation,
Richard Feynman,* likes to stress, the term ‘all’ has to be taken literally. This was not a

* Richard (‘Dick’) Phillips Feynman (b. 1918 New York City, d. 1988), US American physicist. One of the
founders of quantum electrodynamics, he discovered the ‘sum-over-histories’ reformulation of quantum
theory, made important contributions to the theory of the weak interaction and of quantum gravity, and
co-authored a famous physics textbook, the Feynman Lectures on Physics. He is one of those theoretical
physicists who made career mainly by performing complex calculations, a fact he tried to counter at the end
of his life. Though he tried to surpass the genius of Wolfgang Pauli throughout his whole life, he failed in this
endeavour. He was famously arrogant, disrespectful of authorities, as well as deeply dedicated to physics and
to enlarging knowledge in his domain. He also was a well known collector of surprising physical explanations

Dvipsbugw
730 v quanta of light and matter • 19. light

screen

source image

mirror
Dvipsbugw

arrow sum

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 298 Light reflected by a mirror and the corresponding
arrows (at one particular instant of time)

big deal in the explanation of interference. But in order to understand a mirror we have
to include all possibilities, as crazy as they seem, as shown in Figure 298.
As said above a source emits rotating arrows. To determine the probability that light
arrives at a certain image location, we have to add up all the arrows arriving at the same
time at that location. For each path, the arrow orientation at the image is shown – for
convenience only – below the corresponding segment of the mirror. Depending on the
length of the path, the arrow arrives with a different angle and a different length. One
notes that the sum of all arrows does not vanish: light does indeed arrive at the image.
The sum also shows that the largest part of the contribution is from those paths near
the middle one. Moreover, if we were to perform the same calculation for another image
location, (almost) no light would get there. In other words, the rule that reflection occurs
with incoming angle equal to the outgoing angle is an approximation; it follows from the
arrow model of light.
In fact, a detailed calculation, with more arrows, shows that the approximation is quite
precise; the errors are much smaller than the wavelength of the light used.
Copyright © Christoph Schiller November 1997–May 2006

The proof that light does indeed take all these strange paths is given by a more special-
ized mirror. As show in Figure 299, one can repeat the experiment with a mirror which
reflects only along certain stripes. In this case, the stripes were carefully chosen such that
the corresponding path lengths lead to arrows with a bias to one direction, namely to
the left. The arrow addition now shows that such a specialized mirror, usually called a
grating, allows light to be reflected in unusual directions. And indeed, this behaviour is
standard for waves: it is called diffraction. In short, the arrow model for photons does al-
low to describe this wave property of light, provided that photons follow the mentioned

and an author of several popular texts on his work and his life. He shared the 1965 Nobel Prize in physics
for his work on quantum electrodynamics.

Dvipsbugw
 light – the strange consequences of the quantum of action 731

crazy probability scheme. Do not get upset; as said before, quantum theory is the theory
of crazy people.
You may want to check that the
arrow model, with the approxima- source point
tions it generates by summing over
all possible paths, automatically en-
sures that the quantum of action is
indeed the smallest action that can
Challenge 1198 ny be observed.
arrow sum Dvipsbugw
at point
All waves have a signal velocity. usual vanishes
As a consequence, waves show re- mirror
fraction when they move from one
medium into another with different
signal velocity. Interestingly, the na-
screen
ive particle picture of photons as

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

source image
little stones would imply that light
is faster in materials with high in-
dices of refraction, the so-called
Challenge 1199 ny dense materials. Just try it. How-
ever, experiments show that light in arrow sum
dense materials moves slowly. The at image
striped
wave picture has no difficulties ex- mirror
plaining this observation. (Can you
Challenge 1200 ny confirm it?) Historically, this was
F I G U R E 299 The light reflected by a badly placed mirror
one of the arguments against the and by a grating
particle theory of light. However,
Challenge 1201 ny the arrow model of light presented above is able to explain refraction properly. It is not
difficult doing so; try it.
Waves also reflect partially from materials such as glass. This
is one of the toughest properties of waves to be explained with
photons. The issue is important, as it is one of the few effects that
is not explained by a classical wave theory of light. However, it is air
explained by the arrow model, as we will find out shortly. Partial water
reflection confirms the description of the rules (1) and (2) of the
arrow model. Partial reflection shows that photons indeed behave
Copyright © Christoph Schiller November 1997–May 2006

randomly: some are reflected and other are not, without any selec-
tion criterion. The distinction is purely statistical. More about this
issue shortly.
F I G U R E 300 If light
In waves, the fields oscillate in time and space. One way to show were made of little
how waves can be made of particles is to show once for all how to stones, they would
build up a sine wave using a large number of photons. A sine wave move faster inside
Ref. 700 is a coherent state of light. The way to build them up was explained water
by Glauber. In fact, to build a pure sine wave, one needs a superpos-
ition of a beam with one photon, a beam with two photons, a beam with three photons,
continuing up to a beam with an infinite number of them. Together, they give a perfect
sine wave. As expected, its photon number fluctuates to the highest degree possible.

Dvipsbugw
 732 v quanta of light and matter • 19. light

If we repeat the calculation for non-ideal beams, we find that the indeterminacy rela-
tion for energy and time is respected; every emitted wave will possess a certain spectral
width. Purely monochromatic light does not exist. Similarly, no system which emits a
wave at random can produce a monochromatic wave. All experiments confirm these res-
ults.
Waves can be polarized. So far, we disregarded this property. In the photon picture,
polarization is the result of carefully superposing beams of photons spinning clockwise
and anticlockwise. Indeed, we know that linear polarization can be seen as a result of
superposing circularly polarized light of both signs, using the proper phase. What seemed Dvipsbugw
a curiosity in classical optics turns out to be the fundamental explanation of quantum
theory.
Photons are indistinguishable. When two photons of the same colour cross, there is no
way to say, after the crossing, which of the two is which. The quantum of action makes
this impossible. The indistinguishability of photons has an interesting consequence. It is
impossible to say which emitted photon corresponds to which arriving photon. In other

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

words, there is no way to follow the path of a photon in the way we are used to follow
the path of a billiard ball. Particles which behave in this way are called bosons. We will
discover more details about the indistinguishability of photons in the next chapter.
In summary, we find that light waves can indeed be built of particles. However, this is
only possible under the condition that photons are not precisely countable, that they are
not localizable, that they have no size, no charge and no mass, that they carry an (approx-
imate) phase, that they carry spin, that they are indistinguishable bosons, that they can
take any path whatsoever, that one cannot pinpoint their origin and that their probability
to arrive somewhere is determined by the square of the sum of amplitudes for all possible
paths. In other words, light can be made of particles only under the condition that these
particles have extremely special, quantum properties. Only these quantum properties al-
low them to behave like waves, in the case that they are present in large numbers.
Quantons are thus quite different from usual particles. In fact, one can argue that the
only (classical) particle aspects of photons are their quantized energy, momentum and
spin. In all other aspects photons are not like little stones. It is more honest to say that
Ref. 701 photons are calculating devices to precisely describe observations about light. Often these
calculating devices are called quantons. In summary, all waves are streams of quantons. In
fact, all waves are correlated streams of quantons. That is true both for light, for any other
form of radiation, as well as for matter, in all its forms.
The strange properties of quantons are the reason that earlier attempts to describe light
Copyright © Christoph Schiller November 1997–May 2006

as a stream of (classical) particles, such as the one by Newton, failed miserably, under the
rightly deserved ridicule of all other scientists. Indeed, Newton upheld his idea against all
experimental evidence, especially that on light’s wave properties, something a physicist
should never do. Only when people accepted that light is a wave and then discovered and
understood that quantum particles are different from classical particles was the approach
successful.
The indeterminacy relations show that even a single quanton can be seen as a wave;
however, whenever it interacts with the rest of the world, it behaves as a particle. In fact
it is essential that all waves are made of quantons; if not, interactions would not be local,
and objects, in contrast to experience, could not be localized at all. To separate between
wave and particle descriptions, we can use the following criterion. Whenever matter and

Dvipsbugw
 light – the strange consequences of the quantum of action 733

light interact, it is more appropriate to describe electromagnetic radiation as a wave if the

wavelength λ obeys
λQ
ħc
, (486)
kT

where k = 1.4 ë 10−23 J~K is Boltzmann’s constant. If the wavelength is much smaller than
the right hand side, the particle description is most appropriate. If the two sides are of
the same order of magnitude, both effects play a role.
Dvipsbugw
Can light move faster than light? – Virtual photons
Light can move faster than c in vacuum, as well as slower than c. The quantum principle
provides the details. As long as the quantum principle is obeyed, the speed of a short light
flash can differ a bit from the official value, though only a tiny bit. Can you estimate the
Challenge 1202 ny allowed difference in arrival time for a light flash from the dawn of times?
The little arrow explanation gives the same result. If one takes into account the crazy

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

possibility that photons can move with any speed, one finds that all speeds very different
from c cancel out. The only variation that remains, translated in distances, is the inde-
Challenge 1203 ny terminacy of about one wavelength in the longitudinal direction which we mentioned
already above.
However, the most absurd results of the quantum of action appear when one stud-
ies static electric fields, such as the field around a charged metal sphere. Obviously, such
a field must also be made of photons. How do they move? It turns out that static elec-
tric fields are built of virtual photons. In the case of static electric fields, virtual photons
are longitudinally polarized, do not carry energy away, and cannot be observed as free
particles. Virtual photons are photons who do not appear as free particles; they only have
extremely short-lived appearances before they disappear again. In other words, photons,
like any other virtual particle, are ‘shadows’ of particles that obey

∆x∆p D ħ~2 . (487)

Virtual particles do not obey the indeterminacy relation, but its opposite; the opposite
relation expresses their short-lived appearance. Despite their intrinsically short life, and
the impossibility to detect them directly, they have important effects. We will explore
Page 759 virtual particles in detail shortly.
In fact, the vector potential A allows four polarizations, corresponding to the four
Copyright © Christoph Schiller November 1997–May 2006

coordinates (t, x, y, z). It turns out that for the photons one usually talks about, the free
or real photons, the polarizations in t and z direction cancel out, so that one observes
only the x and y polarizations in actual experiments.
For bound or virtual photons, the situation is different.

– CS – more to be written – CS –

In short, static electric and magnetic fields are continuous flows of virtual photons.
Virtual photons can have mass, can have spin directions not pointing along the motion
path, and can have momentum opposite to their direction of motion. All these proper-

Dvipsbugw
 734 v quanta of light and matter • 19. light

ties are different from real photons. In this way, exchange of virtual photons leads to the
attraction of bodies of different charge. In fact, virtual photons necessarily appear in any
description of electromagnetic interactions; more about their effects, such as the famous
attraction of neutral bodies, will be discussed later on.
In summary, light can indeed move faster than light, though only in amounts allowed
by the quantum of action. For everyday situations, i.e. for cases with a high value of the
action, all quantum effects average out, including light velocities different from c.
A different topic also belongs into this section. Not only the position, but also the en-
Ref. 702 ergy of a single photon can be undefined. For example, certain materials split one photon Dvipsbugw
of energy ħω into two photons, whose two energies sum up to the original one. Quantum
mechanics implies that the energy partioning is known only when the energy of one of
the two photons is measured. Only at that very instant the energy of the second photon is
known. Before that, both photons have undefined energies. The process of energy fixing
Page 801 takes place instantaneously, even if the second photon is far away. We will explain below
the background of this and similar strange effects, which seem to be faster than light but

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

which are not. Indeed, such effects do not transmit energy or information faster than
Challenge 1204 ny light.

Indeterminacy of electric fields

We saw that the quantum of action implies an indeterminacy for light intensity. That
implies a similar limit for electric and magnetic fields. This conclusion was first drawn in
Ref. 703 1933 by Bohr and Rosenfeld. They started from the effects of the fields on a test particle
of mass m and charge q, which are described by

ma = q(E + v  B) . (488)

Since it is impossible to measure momentum and position of a particle, they deduced an

Challenge 1205 ny indeterminacy for the electrical field given by

∆E =
ħ
, (489)
q ∆x T

where T is the measurement time and ∆x is the position uncertainty. Every value of an
electric field, and similarly that of every magnetic field, is thus affected with an indeterm-
inacy. The physical state of the electromagnetic field behaves like the state of matter in
Copyright © Christoph Schiller November 1997–May 2006

this aspect. This is the topic we explore now.

Curiosities and fun challenges about photons

**
Can diffraction be explained with photons? Newton was not able to do so. Today we can
do so. Figure 301 is translationally invariant along the horizontal direction; therefore, the
momentum component along this direction is also conserved: p 1 sin α 1 = p 2 sin α 2 . The
photon energy E = E 1 = E 2 is obviously conserved.

Dvipsbugw
 motion of matter – beyond cl assical physics 735

p1
α1 air
α
water
p2

α2
axis Dvipsbugw
F I G U R E 301 Diffraction and F I G U R E 302 A
photons falling pencil

The index of refraction n is defined with momentum and energy as

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

n=
cp
. (490)
E
As a result, the ‘law’ of refraction follows.
There is an important issue here. The velocity of a photon v = δE~δp in a light ray is
not the same as the phase velocity u = E~p that enters in the calculation.

**
A typical effect of the quantum ‘laws’ is the yellow colour of the lamps used for street illu-
mination in most cities. They emit pure yellow light of one frequency; that is the reason
that no other colours can be distinguished in their light. Following classical electrodyna-
mics, harmonics of that light frequency should also be emitted. Experiments show how-
ever that this is not the case; classical electrodynamics is thus wrong. Is this argument
Challenge 1206 ny correct?

20. motion of mat ter – beyond cl assical physics

All great things begin as blasphemies.

“ George Bernard Shaw

”
Copyright © Christoph Schiller November 1997–May 2006

The existence of a smallest action has numerous effects on the motion of matter. We start
with a few experimental results that show that the quantum of action is indeed the smal-
lest action.

Wine glasses and pencils

A simple consequence of the quantum of action is the impossibility of completely filling
a glass of wine. If we call ‘full’ a glass at maximum capacity (including surface tension
effects, to make the argument precise), we immediately see that the situation requires
Challenge 1207 n complete rest of the liquid’s surface; however, the quantum of action forbids this. Indeed,
a completely quiet surface would allow two subsequent observations which differ by less

Dvipsbugw
 736 v quanta of light and matter • 20. motion of matter

than ħ~2. There is no rest in nature. In other words, the quantum of action proves the old
truth that a glass of wine is always partially empty and partially full.
The quantum of action has many similar consequences for everyday life. For example,
a pencil on its tip cannot stay vertical, even if it is isolated from all disturbances, such as
Challenge 1208 ny vibrations, air molecules and thermal motion. Are you able to confirm this? In fact, it is
even possible to calculate the time after which a pencil must have fallen over.*

Cool gas
Dvipsbugw
Rest is impossible in nature. Even at lowest temperatures, particles inside matter are in
motion. This fundamental lack of rest is said to be due to the so-called zero-point fluc-
tuations. A good example are the recent measurements of Bose–Einstein condensates,
systems with a small number of atoms (between ten and a few million) at lowest tem-
peratures (around 1 nK). These cool gases can be observed with high precision. Using
elaborate experimental techniques, Bose–Einstein condensates can be put into states for

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

which ∆p∆x is almost exactly equal to ħ~2, though never lower than this value.
That leads to an interesting puzzle. In a normal object, the distance between the atoms
Challenge 1211 ny is much larger than their de Broglie wavelength. (Are you able to confirm this?) But today
it is possible to cool objects to very low temperatures. At extremely low temperatures,
Ref. 715 less than 1 nK, the wavelength of the atoms may be larger than their separation. Can you
Challenge 1212 ny imagine what happens in such cases?

No rest
Otium cum dignitate.**

“ Cicero, De oratore.

The impossibility of rest, like all other unexplained effects of classical physics, is most ”
apparent in domains where the action is near the minimum observable one. To make the
effects most obvious, we study the smallest amount of matter that can be isolated: a single
particle. Later on we will explore situations that cover higher numbers of particles.
Experiments show that perfect rest is never observed. The quantum of action prevents
this in a simple way. Whenever the position of a system is determined to high precision,
we need a high energy probe. Indeed, only a high energy probe has a wavelength small
enough to allow a high precision for position measurements. As a result of this high en-
ergy however, the system is disturbed. Worse, the disturbance itself is also found to be
Copyright © Christoph Schiller November 1997–May 2006

imprecisely measurable. There is thus no way to determine the original position even by
taking the disturbance itself into account. In short, perfect rest cannot be observed. All

Ref. 704 * That is not easy, but neither too difficult. For an initial orientation close to the vertical, the fall time T turns
out to be
1 8
T= T0 ln (491)
2π α
where α is the starting angle, and a fall by π is assumed. Here T0 is the oscillation time of the pencil for small
Challenge 1209 ny angles. (Can you determine it?)
The indeterminacy relation for the tip of the pencil yields a minimum starting angle, because the mo-
Challenge 1210 ny mentum indeterminacy cannot be made as large as wanted. You should be able to provide an upper limit.
Once the angle is known, you can calculate the maximum time.
** ‘Rest with dignity.’

Dvipsbugw
 motion of matter – beyond cl assical physics 737

systems who have ever been observed with high precision confirm that perfect rest does
not exist. Among others, this result has been confirmed for electrons, neutrons, protons,
ions, atoms, molecules and crystals.

Flows and the quantization of matter

Die Bewegung ist die Daseinsform der Materie.

“ Friedrich Engels, Anti–Dühring.*

Not only is rest made impossible by the quantum of action; the same impossibility ap- ” Dvipsbugw

plies to any situation which does not change in time, like any constant velocity. The most
important example are flows. The quantum of action implies that no flow can be station-
ary. More precisely, a smallest action implies that all flows are made of smallest entities.
All flows are made of quantum particles. Two flows ask for direct confirmation: flows of
electricity and flows of liquids.
If electrical current would be a continuous

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

flow, it would be possible to observe action val- Current
ues as small as desired. The simplest confirm-
ation of the discontinuity of current flow was
discovered only in the 1990s: take two metal
wires on the table, laying across each other. It is preliminary
not hard to let a current flow from one wire to graph
the other, via the crossover, and to measure the
voltage. A curve like the one shown in Figure 303
is found: the current increases with voltage in
Voltage
regular steps.
Many other experiments confirm the result
and leave only one conclusion: there is a smal- F I G U R E 303 Steps in the flow of electricity
lest charge in nature. This smallest charge has the in metals
same value as the charge of an electron. Indeed,
electrons turn out to be part of every atom, in a complex way to be explained shortly.
In metals, quite a a number of electrons can move freely; that is the reason that metal
conduct electricity so well.
Also the flow of matter shows smallest units. We mentioned in the first part that a
consequence of the particle structure of liquids is that even in the smoothest of pipes,
even oil or any other smooth liquid still produces noise when it flows through the pipe.
Copyright © Christoph Schiller November 1997–May 2006

We mentioned that the noise we hear in our ears in situations of absolute silence, such as
in a snowy landscape in the mountains, is due to the granularity of matter. Depending
on the material, the smallest units of matter are called atoms, ions or molecules.

Quantons
Electrons, ions, atoms and molecules are quantum particles or quantons. Like photons,
they show some of the aspects of everyday particles, but show many other aspects which
are different from what is expected from little stones. Let us have a rapid tour.

Ref. 684 * ‘Motion is matter’s way of being.’

Dvipsbugw
 738 v quanta of light and matter • 20. motion of matter

Figure to be included

F I G U R E 304 Matter diffracts and interferes Dvipsbugw

Everyday matter has mass, position and momentum, orientation and angular mo-
mentum, size, shape, structure and colour. What about matter quantons? First of all, mat-
ter quantons do have mass. Single particles can be slowed down or accelerated; in addi-
tion, hits by single electrons, atoms or molecules can be detected. Experiments also show

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that (composed) quantons have structure, size, shape and colour. We will discuss their
details below. How do they move while respecting the quantum of action?

The motion of quantons – matter as waves

Ref. 708 In 1923 and 1924, the French physicist Louis de Broglie pondered over the concept of
photon and the possible consequences of the quantum of action for matter particles.
It dawned to him that like light quanta, streams of matter particles with the same mo-
mentum should also behave as waves. The quantum of action implies wave behaviour.
This, de Broglie reasoned, should also apply to matter. He predicted that constant matter
flows should have a wavelength and angular frequency given by

λ= ω=
2π ħ E
and . (492)
p ħ

where p and E are the momentum and the energy of the single particles. Soon after
the prediction, experiments started to provide the confirmation of the idea. It is indeed
found that matter streams can diffract, refract and interfere. Due to the small value of
the wavelength, one needs careful experiments to detect the effects. Nevertheless, one
after the other, all experiments which proved the wave properties of light have been re-
peated for matter beams. For example, in the same way that light diffracts when passing
Copyright © Christoph Schiller November 1997–May 2006

around an edge or through a slit, matter has been found to diffract in these situations.
Similarly, researchers inspired by light interferometers built matter interferometers; they
Ref. 709 work with beams of electrons, nucleons, nuclei, atoms and even large molecules. In the
Page 560 same way that the observations of interference of light proves the wave property of light,
the interference patterns observed with these instruments show the wave properties of
matter.
Like light, matter is also made of particles; like light, matter behaves as a wave when
large numbers of particles with the same momentum are involved. Even though beams
of large molecules behave as waves, for everyday objects, such as cars on a highway, one
never makes such observations. There are two main reasons. First, for cars on highways
the involved wavelength is extremely small. Second, the speeds of cars vary too much;

Dvipsbugw
 motion of matter – beyond cl assical physics 739

Figure to be included

F I G U R E 305 Trying to measure position and momentum Dvipsbugw

streams of objects with the same speed for all objects – only such streams have a chance
to be coherent – are extremely rare in nature.
If matter behaves like a wave, we can draw a strange conclusion. For every type of wave,
the position X of its maximum and the wavelength λ cannot both be sharply defined
simultaneously; on the contrary, their indeterminacies follow the relation

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

∆λ∆X =
1
. (493)
2
Similarly, for every wave the frequency ω and the instant T of its peak amplitude cannot
both be sharply defined. Their indeterminacies are related by

∆ω∆T =
1
. (494)
2
Using the wave properties of matter we get

∆p∆X E ∆E∆T E
ħ ħ
and . (495)
2 2
These famous relations are called Heisenberg’s indeterminacy relations. They were dis-
covered by the German physicist Werner Heisenberg in 1925. They state that there is no
way to ascribe a precise momentum and position to a material system, nor a precise en-
ergy and age. The more accurately one quantity is known, the less accurately the other
is.* Matter quantons – like stones, but in contrast to photons – can be localized, but only
approximately.
Copyright © Christoph Schiller November 1997–May 2006

Both indeterminacy relations have been checked experimentally in great detail. The
limits are easily experienced in experiments. Some attempts are shown in Figure 305. In
fact, every experiment proving that matter behaves like a wave is a confirmation of the
indeterminacy relation, and vice versa.
As a note, Niels Bohr called the relation between two variables linked in this way com-
plementarity. He then explored systematically all such possible pairs. You should search
Challenge 1213 ny for such observable pairs yourself. Bohr was deeply fascinated by the existence of a com-
plementarity principle. Bohr later extended it also to philosophical aspects. In a fam-

* The policeman stops the car being driven by Werner Heisenberg. ‘Do you know how fast you were driving?’
‘No, but I know exactly where I am!’

Dvipsbugw
 740 v quanta of light and matter • 20. motion of matter

Dvipsbugw
source

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 306 On the quantization of angular momentum

ous story, somebody asked him what was the quantity complementary to precision. He
answered: ‘Clarity’.
In summary, we conclude that the quantum of action prevents position and mo-
mentum values to be exactly defined for microscopic systems. Their values are fuzzy. Like
Bohr, we will explore some additional limits on motion that follow from the quantum of
action.

Rotation and the lack of North Poles

Tristo quel discepolo che non avanza il suo

“ maestro.

The quantum of action also has important consequences for rotational motion. Action
Leonardo da Vinci*

”
and angular momentum have the same physical dimensions. It only takes a little thought
to show that if matter or radiation has a momentum and wavelength related by the
quantum of action, then angular momentum is fixed in multiples of the quantum of ac-
tion; angular momentum is thus quantized.
Copyright © Christoph Schiller November 1997–May 2006

Ref. 718 The argument is due to Dicke and Wittke. Just imagine a source at the centre of a
circular fence, made of N steel bars spaced by a distance a = 2πR~N, as shown in
Figure 306. In the centre of the fence we imagine a source of matter or radiation that
emits particles towards the fence in any chosen direction. The linear momentum of the
particle is p = ħk = 2πħ~λ. Outside the fence, the direction of the particle is given by the
condition of positive interference. In other words, the angle θ is given by a sin θ = Mλ,
where M is an integer. In this process, the fence receives a linear momentum p sin θ, or an
angular momentum L = pR sin θ. Inserting all expressions one finds that the transferred

* ‘Sad is that disciple who does not surpass his master.’ The statement is painted in large letters in the Aula
Magna of the University of Rome.

Dvipsbugw
 motion of matter – beyond cl assical physics 741

angular momentum is
L = N Mħ . (496)

In other words, the angular momentum of the fence is an integer multiple of ħ. Of course,
this argument is only a hint, not a proof. Nevertheless, the argument is correct. The angu-
lar momentum of bodies is always a multiple of ħ. Quantum theory thus states that every
object rotates in steps.
But rotation has more interesting aspects. Due to the quantum of action, in the same
way that linear momentum is fuzzy, angular momentum is so as well. There is an inde- Dvipsbugw
Ref. 710 terminacy relation for angular momentum L. The complementary variable is the phase
Ref. 711 angle φ of the rotation. The indeterminacy relation can be expressed in several ways. The
Page 722 simplest, but also the less precise is

∆L ∆φ E
ħ
. (497)
2

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(The approximation is evident: the relation is only valid for large angular momenta; the
relation cannot be valid for small values, as ∆φ by definition cannot grow beyond 2π. In
particular, angular momentum eigenstates have ∆L = 0.*)
The quantization and indeterminacy of angular momentum has important con-
sequences. Classically speaking, the poles of the Earth are spots which do not move when
observed by a non-rotating observer. Therefore at those spots matter would have a defined
position and a defined momentum. However, the quantum of action forbids this. There
cannot be a North Pole on Earth. More precisely, the idea of a rotation axis is an approx-
imation not valid in general.
Even more interesting are the effects of the quantum of action on microscopic particles,
such as atoms, molecules or nuclei. To begin with, we note that action and angular mo-
mentum have the same units. The precision with which angular momentum can be meas-
ured depends on the precision of the rotation angle. But if a microscopic particle rotates
by an angle, this rotation might be unobservable, a situation in fundamental contrast
with the case of macroscopic objects. Experiments indeed confirm that many microscopic
particles have unobservable rotation angles. For example, in many, but not all cases, an
atomic nucleus rotated by half a turn cannot be distinguished from the unrotated nucleus.
If a microscopic particle has a smallest unobservable rotation angle, the quantum of
action implies that the angular momentum of that particle cannot be zero. It must always
Copyright © Christoph Schiller November 1997–May 2006

be rotating. Therefore we need to check for each particle what its smallest unobservable
angle of rotation is. Physicists have checked experimentally all particles in nature and
have found – depending on the particle type – the following smallest unobservable angle
values: 0, 4π, 2π, 4π~3, π, 4π~5, 2π~3, etc.

* An exact way to state the indeterminacy relation for angular momentum is

S1 − 2πP(π)S ,
ħ
∆L ∆φ E (498)
2

º
where P(π) is the normalized probability that the angular position has the value π. For an angular mo-
Ref. 712 mentum eigenstate, one has ∆φ = π~ 3 and P(π) = 1~2π. This expression has been tested experimentally.

Dvipsbugw
 742 v quanta of light and matter • 20. motion of matter

Figure to be included

F I G U R E 307 The Stern–Gerlach experiment Dvipsbugw

Let us take an example. Certain nuclei have a smallest unobservable rotation angle
of half a turn. That is the case for a prolate nucleus – one that looks like a rugby ball –
turning around its short axis. Both the largest observable rotation and the indeterminacy
are thus a quarter turn. Since the change or action produced by a rotation is the number of
turns times the angular momentum, we find that the angular momentum of this nucleus

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is 2 ë ħ.
As a general result we deduce from the smallest angle values that the angular mo-
mentum of a microscopic particle can be 0, ħ~2, ħ, 3ħ~2, 2ħ, 5ħ~2, 3ħ, etc. In other words,
the intrinsic angular momentum of particles, usually called their spin, is an integer mul-
tiple of ħ~2. Spin describes how a particle behaves under rotations. (It turns out that all
spin 0 particles are composed and contain other particles; the quantum of action thus
remains the limit for rotational motion in nature.)
How can a particle rotate? At this point we do not know yet how to picture the rotation.
But we can feel it. This is done in the same way we showed that light is made of rotating
entities: all matter, including electrons, can be polarized. This was shown most clearly by
the famous Stern–Gerlach experiment.

Silver, Stern and Gerlach

In 1922, Otto Stern and Walter Gerlach* found that a beam of silver atoms that is extracted
from an oven splits into two separate beams when it passes through an inhomogeneous
Ref. 705 magnetic field. There are no atoms that leave the magnetic field in intermediate locations.
The split into two is an intrinsic property of silver atoms. The split is due to the spin
value of the atoms. Silver atoms have spin ħ~2, and depending on their orientation in
space, they are deflected either in direction of the field or against it. The splitting of the
Copyright © Christoph Schiller November 1997–May 2006

beam is a pure quantum effect: there are no intermediate options. This result is so peculiar
that it was studied in great detail.
When one of the two beams is selected – say the ‘up’ beam – and passed through a
second set-up, all atoms end up in the ‘up’ beam. The other exit, for the ‘down’ beam,
remains unused in this case. The up and down beams, in contrast to the original beam,
cannot be split further. This is not surprising.
But if the second set-up is rotated by π~2 with respect to the first, again two beams –
‘right’ and ‘left’ – are formed; it plays no role whether the incoming beam was from the

* Otto Stern (1888–1969) and Walter Gerlach (1889–1979), both German physicists, worked together at
the University in Frankfurt.

Dvipsbugw
motion of matter – beyond cl assical physics 743

oven or an ‘up’ beam. A partially rotated set-up yields a partial, uneven split. The number
ratio depends on the angle.
We note directly that if a beam from the oven is split first vertically and then hori-
zontally, the result differs from the opposite order. Splitting processes do not commute.
(Whenever the order of two operations is important, physicists speak of ‘lack of commut-
ation’.) Since all measurements are processes as well, we deduce that measurements in
quantum systems do not commute in general.
Beam splitting is direction dependent. Matter beams behave almost in the same way as
polarized light beams. The inhomogeneous magnetic field acts somewhat like a polarizer. Dvipsbugw
The up and down beams, taken together, behave like a fully polarized light beam. In fact,
the polarization direction can be rotated (with the help of a homogeneous magnetic field).
Indeed, a rotated beam behaves in a unrotated magnet like an unrotated beam in a rotated
magnet.
The ‘digital’ split forces us to rethink the description of motion. In special relativity,
the existence of a maximum speed forced us to introduce the concept of space-time, and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

then to refine the description of motion. In general relativity, the maximum force obliged
us to introduce the concepts of horizon and curvature, and then to refine the description
of motion. At this point, the existence of the quantum of action forces us to take two
similar steps. We will introduce the new concept of Hilbert space, and then we will refine
the description of motion.

The language of quantum theory and its description of motion

In classical physics, a physical system is said to have momentum, position, and a axis of
rotation. The quantum of action makes it impossible to continue using this language. In
classical physics, the state and the measurement result coincide, because measurements
can be imagined to disturb the system as little as possible. But due to a smallest action in
nature, the interaction necessary to perform the measurement cannot be made arbitrarily
small. For example, the Stern–Gerlach experiment shows that the measured spin orienta-
tion values – like those of any other observable – are not intrinsic values, but result from
the measurement process itself.
Therefere, the quantum of action forces us to distinguish three entities:
— the state of the system;
— the operation of measurement;
— the result of the measurement.
Copyright © Christoph Schiller November 1997–May 2006

A general state of a quantum system is thus not described by the outcomes of a measure-
ment. The simplest case showing this is the system made of a single particle in the Stern–
Gerlach experiment. The experiment shows that a spin measurement on a general (oven)
particle state sometimes gives ‘up’, sometimes ‘down’ (‘up’ might be +1, ‘down’ might be
−1) showing that a general state has no intrinsic properties. It was also found that feeding
‘up’ into the measurement apparatus gives ‘up’ states; thus certain special states (‘eigen-
states’) do remain unaffected. Finally, the experiment shows that states can be rotated by
applied fields; they have an abstract direction.
These details can be formulated in a straightforward way. Since measurements are
operations that take a state as input and produce as output a measurement result and an

Dvipsbugw
744 v quanta of light and matter • 20. motion of matter

output state, we can say:

— States are described by small rotating arrows; in other words they are complex vectors
in an abstract space. The space of all possible states is called a Hilbert space.
— Measurements are operations on the state vectors. Measurements are said to be de-
scribed by (self-adjoint or) Hermitean operators (or matrices).
— Measurement results are real numbers.
— Changes of viewpoint are described by unitary operators (or matrices) that transform
states and measurement operators. Dvipsbugw
As required by quantum experiments, we have thus distinguished the quantities that are
not distinguished in classical physics. Once this step is completed, quantum theory fol-
lows quite simply, as we shall see.
Quantum theory describes observables as operators, thus as transformations in Hilbert
space, because any measurement is an interaction with a system and thus a transforma-
tion of its state.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Quantum mechanical experiments also show that the measurement of an observable
can only give as result one of the possible eigenvalues of this transformation. The resulting
states are eigenvectors – the ‘special’ states just mentioned. Therefore every expert on mo-
tion must know what an eigenvalue and an eigenvector is. For any linear transformation
T, those special vectors ψ that are transformed into multiples of themselves,

Tψ = λψ (499)

are called eigenvectors, and the multiplication factor λ is called the associated eigenvalue.
Experiments show that the state of the system after the measurement is given by the ei-
genvector of the measured eigenvalue.
In summary, the quantum of action obliges us to distinguish between three concepts
that are all mixed up in classical physics: the state of a system, the measurement on a sys-
tem and the measurement result. The quantum of action forces us to change the vocabu-
lary with which we describe nature and obliges to use more differentiated concepts. Now
follows the main step: we describe motion with these concepts. This is the description
that is usually called quantum theory.
In classical physics, motion is given by the path that minimizes the action. Motion
takes place in such a way that the action variation δS vanishes when paths with fixed end
points are compared. For quantum systems, we need to redefine the concept of action
Copyright © Christoph Schiller November 1997–May 2006

and to find a description of its variation that are not based on paths, as the concept of
‘path’ does not exist for quantum systems.
Instead of defining action variations for changing paths between start and end points,
one defines it for given initial and final states. In detail, the action variation δS between
an initial and a final state is defined as

δS = `ψ i S δ ∫ LdtSψ e ,
f (500)

where L is the Lagrangian (operator). The variation of the action is defined in the same
way as in classical physics, except that the momentum and position variables are replaced

Dvipsbugw
 motion of matter – beyond cl assical physics 745

by the corresponding operators.*

In the classical principle of least action, the path is varied while keeping the end points
fixed. This variation must be translated into the language of quantum theory. In quantum
theory, paths do not exist, because position is not a well-defined observable. The only
variation that we can use is
δ`ψ i Sψ f e . (501)

This complex number describes the variation of the temporal evolution of the system.
The variation of the action must be as small as possible when the temporal evolution Dvipsbugw
is varied, while at the same time it must be impossible to observe actions below ħ~2. This
double condition is realized by the so-called quantum action principle

`ψ i Sδ ∫ LdtSψ e = −iħδ`ψ Sψ e .
f i f (502)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

This principle describes all quantum motion in nature. Classically, the right hand side is
zero – since ħ can then be neglected – and we then recover the minimum action prin-
ciple of classical physics. In quantum theory however, the variation of the action is pro-
portional to the variation at the end points. The intermediate situations – the ‘paths’ – do
not appear. In the quantum action principle, the factor −i plays an important role. We
recall that states are vectors. The factor −i implies that in the complex plane, the complex
variation on the right hand side is rotated by a right angle; in this way, even if the variation
at the end points is small, no action change below ħ~2 can be observed.
To be convinced about the correctness of the quantum action principle, we proceed in
the following way. We first deduce evolution equations, we then deduce all experimental
effects given so far, and finally we deduce new effects that we compare to experiments.

The state – or wave function – and its evolution

We can also focus on the change of states with time. The quantum action principle then
gives
iħ Sψe = HSψe .
∂
(503)
∂t
Ref. 713 This famous equation is Schrödinger’s equation of motion.** In fact, Erwin Schrödinger
had found his equation in two slightly different ways. In his first paper, he used a vari-
Copyright © Christoph Schiller November 1997–May 2006

Ref. 714 ational principle slightly different from the one just given. In the second paper he simply
asked: how does the state evolve? He imagined the state of a quanton to behave like a

* More precisely, there is also a condition for ordering of operators in mixed products, so that the lack of
commutation between operators is taken into account. We do not explore this issue here.
** Erwin Schrödinger (b. 1887 Vienna, d. 1961 Vienna) was famous for being a physicien bohémien, and
always lived in a household with two women. In 1925 he discovered the equation which brought him in-
ternational fame and the Nobel prize for physics in 1933. He was also the first to show that the radiation
discovered by Victor Hess in Vienna was indeed coming from the cosmos. He left Germany and then again
Austria out of dislike of national socialism, and was for many years professor in Dublin. There he published
the famous and influential book What is life?. In it, he comes close to predicting the then unknown nuclear
acid DNA from theoretical insight alone.

Dvipsbugw
 746 v quanta of light and matter • 20. motion of matter

wave and like a particle at the same time. If the state behaves ψ like a wave, it must be
described by a function (hence he called it ‘wave function’) with amplitude W multiplied
by a phase factor e ikx−ωt . The state can thus be written as

Sψe = ψ(t, x) = W(t, x)e ikx−ωt . (504)

At the same time, the state must also behave like a a particle. In particular, the non-
relativistic particle relation between energy and momentum E = p2 ~2m + V (x) must
remain valid for these waves. Using the two relations for matter wavelength and frequency, Dvipsbugw
we thus must have
= + V (x)ψ = Hψ .
∂ψ ∆ψ
iħ (505)
∂t 2m
This ‘wave’ equation for the complex field ψ became instantly famous in 1926 when
Schrödinger, by inserting the potential felt by an electron near a proton, explained the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

energy levels of the hydrogen atom. In other words, the equation explained the discrete
Page 754 colours of all radiation emitted by hydrogen. We will do this below. The frequency of the
light emitted by hydrogen gas was found to be in agreement with the prediction of the
equation to five decimal places. The aim of describing motion of matter had arrived at a
new high point.
The most precise description of matter is found when the re-
lativistic energy–momentum relation is taken into account. We
explore this approach below. Even today, predictions of atomic
spectra are the most precise and accurate in the whole study of
nature. No other description of nature has achieved a higher ac-
curacy.
We delve a bit into the details of the description with the
Schrödinger equation (505). The equation expresses a simple con-
nection: the classical speed of matter is the group velocity of the
field ψ. We know from classical physics that the group velocity
is not always well defined; in cases where the group dissolves in
several peaks the concept of group velocity is not of much use;
these are the cases in which quantum motion is much different Erwin Schrödinger
from classical motion, as we will soon discover.
As an example, the left corner pictures in these pages show the evolution of a wave
function – actually its modulus SΨS2 – for the case of an exploding two-particle system.
Copyright © Christoph Schiller November 1997–May 2006

The Schrödinger equation makes another point: velocity and position of matter are
not independent variables and cannot be chosen at leisure. Indeed, the initial condition
of a system is given by the initial value of the wave function alone. No derivatives have
to or can be specified. In other words, quantum systems are described by a first order
evolution equation, in strong contrast to classical physics.
We note for completeness that in the Schrödinger equation the wave function is in-
deed a vector, despite the apparent differences. The scalar product of two wave func-
tions/vectors is the spatial integral of the product between complex conjugate of the first
function and the second function. In this way, all concepts of vectors, such as unit vectors,
null vectors, basis vectors, etc. can be reproduced.

Dvipsbugw
 motion of matter – beyond cl assical physics 747

Why are atoms not flat? Why do shapes exist?

In 1901, Jean Perrin, and in 1904 the Japanese physicist Nagaoka Hantaro proposed that
Ref. 706 atoms are small solar systems. In 1913, Niels Bohr used this idea and based his epoch-
Ref. 707 making atomic calculations on it. All thus somehow assumed that hydrogen atoms are
flat. However, this is not observed.
Atoms, in contrast to solar systems, are quantum systems. Atoms, like protons and
many other quantum systems, do have sizes and shapes. Atoms are spherical, molecules
have more complex shapes. Quantum theory gives a simple recipe for the calculation: the Dvipsbugw
shape of an atom or a molecule is due to the probability distribution of its electrons. The
probability distribution is given by the square modulus SΨS2 of the wave function.
In other words, Schrödinger’s equation defines the shape of molecules. That is why
it was said that the equation contains all of chemistry and biology. The precise shape of
matter is determined by the interactions of electrons and nuclei. We come back to the
Page 837 issue later.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In short, the wave aspect of quantons is responsible for all shapes in nature. For ex-
ample, only the wave aspect of matter, and especially that of electrons, allows to under-
stand the shapes of molecules and therefore indirectly the shapes of all bodies around us,
from flowers to people.
Obviously, the quantum of action also implies that shapes fluctuate. If a long molecule
is held fixed at its two ends, the molecule cannot remain at rest in between. Such exper-
iments are common today; they confirm that rest does not exist, as it would contradict
the existence of a minimum action in nature.

Rest – spread and the quantum Z eno effect

In special relativity, anything moving inertially is at rest. However, the quantum of action
implies that no particle can ever be at rest. Therefore, no quantum system can be at in
inertial motion. That is the reason that any wave function spreads out in time. In this way,
a particle is never at rest, whatever the observer may be.
Only if a particle is bound, not freely moving, one can have the situation that the
density distribution is stationary in time.
Another apparent case of rest in quantum theory is called the quantum Zeno effect.
Usually, observation changes the system. However, for certain systems, observation can
have the opposite effect.
The quantum Zeno effect was partially observed by Wayno Itano and his group in 1990,
Copyright © Christoph Schiller November 1997–May 2006

Ref. 716 and definitively observed by Mark Raizen and his group, in 2001.

– CS – more to be told – CS –

Ref. 717 In an fascinating prediction, Saverio Pascazio and his team have predicted that the
quantum Zeno effect can be used to realize X-ray tomography of objects with the lowest
radiation levels imaginable.

Dvipsbugw
 748 v quanta of light and matter • 20. motion of matter

Tunnelling , hills and limits on memory

A slow ball cannot roll over a high hill, says everyday experience. More precisely, classical
physics says that if the kinetic energy T is smaller than the potential energy V the ball
would have at the top of the hill, the ball cannot roll over the hill. Quantum theory simply
states the opposite. There is a probability to pass the hill for any energy of the ball.
Since hills in quantum theory are de-
scribed by potential barriers, and objects
by wave functions, the effect that an object Dvipsbugw
can pass the hill is called the tunnelling ef- V
fect. For a potential barrier of finite height, m T
any initial wave function will spread bey-
ond the barrier. The wave function will 0 ∆x
even be non-vanishing at the place of the F I G U R E 308 Climbing a hill
barrier. All this is different from everyday

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

experience and thus from classical mechanics.
Something new is contained in this description of hills: the assumption that all
obstacles in nature can be overcome with a finite effort. No obstacle is infinitely difficult
to surmount. (Only for a potential of infinite height the wave function would vanish and
not spread on the other side.)
Challenge 1214 ny How large is the effect? A simple calculation shows that the transmission probability

16T(V − T) −2w º2m(V −T)~ħ 2

P is given by
P e (506)
V2
where w is the width of the hill. For a system of many particles, the probability is the
product of the probabilities for each particle. In the case of a car in a garage, assuming
it is made of 1028 atoms of room temperature, and assuming that a garage wall has a
thickness of 0.1 m and a potential height of 1 keV= 160 aJ for the passage of an atom, one
gets a probability of finding the car outside the garage of

P  Š10−(10 )   10−(10
12 (10 28 ) 40
)
. (507)

Challenge 1215 ny This rather small value – just try to write it down to be convinced – is the reason why it
is never taken into account by the police when a car is missing. (Actually, the probability
Copyright © Christoph Schiller November 1997–May 2006

is considerably smaller; can you name at least one effect that has been forgotten in this
Challenge 1216 ny simple calculation?) Obviously, tunnelling can be important only for small » systems, made
of a few particles, and for thin barriers, with a thickness of the order of ħ 2 ~2m(V − T) .
Tunnelling of single atoms is observed in solids at high temperature, but is not of import-
ance in daily life. For electrons the effect is larger; the formula gives
º º
w  0.5 nm aJ V −T . (508)

At room temperature, kinetic energies are of the order of 6 zJ; increasing temperature
obviously increases tunnelling. As a result, electrons or other light particles tunnel quite

Dvipsbugw
 motion of matter – beyond cl assical physics 749

easily. Indeed, every tv tube uses tunnelling at high temperature to generate the electron
beam producing the picture. The heating is the reason that TV tubes take time to switch
on.
For example, the tunnelling of electrons limits the ability to reduce the size of computer
memories, and thus makes it impossible to produce silicon integrated circuits with one
Challenge 1217 ny terabyte (TB) of random access memory (RAM). Are you able to imagine why? In fact,
tunnelling limits the working of any type of memory, also that of our brain. If we would
be much hotter than 37°C, we could not remember anything!
By the way, light, being made of particles, can also tunnel through potential barri- Dvipsbugw
ers. The best potential barriers for light are called mirrors; they have barrier heights of
the order of one aJ. Tunnelling implies that light can be detected behind a mirror. These
so-called evanescent waves have indeed been detected. They are used in several high-
precision experiments.

Spin and motion

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Everything turns.

“
Spin describes how a particle behaves under rotations. The full details of spin of electrons
Anonymous

”
were deduced from experiments by two Dutch students, George Uhleneck and Samuel
Ref. 721 Goudsmit. They had the guts to publish what also Ralph Kronig had suspected: that elec-
trons rotate around an axis with an angular momentum of ħ~2. In fact, this value is correct
for all elementary matter particles. (In contrast, radiation particles have spin values that
are integer multiples of ħ.) In particular, Uhlenbeck and Goudsmit proposed a g-value
of 2 for the electron in order to explain the optical spectra. The factor was explained by
Ref. 722 Llewellyn Thomas as a relativistic effect a few months afterwards.
In 2004, experimental techniques became so sensitive that the magnetic effect of a
single electron spin attached to an impurity (in an otherwise unmagnetic material) has
been detected. Researchers now hope to improve these so-called magnetic resonance
force microscopes until they reach atomic resolution.
In 1927, the Austrian physicist Wolfgang Pauli* discovered how to include spin in a
quantum mechanical description; instead of a state function with a single component,
one needs a state function with two components. Nowadays, Pauli’s equation is mainly of
conceptual interest, because like the one by Schrödinger, it does not comply with special
relativity. However, the idea to double the necessary components was taken up by Dirac
Copyright © Christoph Schiller November 1997–May 2006

* Wolfgang Ernst Pauli (b. 1900 Vienna, d. 1958 Zürich), when 21 years old, wrote one of the best texts on
special and general relativity. He was the first to calculate the energy levels of hydrogen with quantum theory,
discovered the exclusion principle, included spin into quantum theory, elucidated the relation between spin
and statistics, proved the CPT theorem and predicted the neutrino. He was admired for his intelligence
and feared for his biting criticisms, which lead to his nickname ‘conscience of physics’. Despite this habit
he helped many people in their research, such as Heisenberg with quantum theory, without claiming any
Ref. 723 credit for himself. He was seen by many, including Einstein, as the greatest and sharpest mind of twentieth
century physics. He was also famous for the ‘Pauli effect’, i.e. his ability to trigger disasters in laboratories,
machines and his surroundings by his mere presence. As we will see shortly, one can argue that Pauli got
the Nobel Prize in physics in 1945 (officially ‘for the discovery of the exclusion principle’) for finally settling
the question on the number of angels that can dance on the tip of a pin.

Dvipsbugw
 750 v quanta of light and matter • 20. motion of matter

when he introduced the relativistic description of the electron, and the idea is used for all
other particle equations.

Relativistic wave equations

A few years after Max Planck had discovered the quantum of action, Albert Einstein pub-
lished the theory of special relativity. The first question Planck asked himself was whether
the value of the quantum of action would be independent of the observer. For this reason,
he invited Einstein to Berlin. By doing this, he made the then unknown patent office clerk Dvipsbugw
famous in the world of physicis.
The quantum of action is indeed independent of the speed of the observer. All observ-
ers find the same minimum value. To include special relativity into quantum theory, we
only need to find the correct quantum Hamiltonian operator.
Given that the classical Hamiltonian of a free particle is given by
»
H = β c 4 m 2 + c 2 p2 with p = γmv ,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(509)

one might ask: what is the corresponding Hamilton operator? A simple answer was given,
Ref. 729 only in 1950, by L.L. Foldy and S.A. Wouthuysen. The operator is almost the same one:

’ 1 0 0 0 “
» – —
H =β c m +c p β=– —
0 1 0 0
– —
4 2 2 2
with (510)
0 0 −1 0
” 0 0 0 −1 •

The signs of the operator β distinguishes particles and antiparticles; it has two 1s and
two -1s to take care of the two possible spin directions. With this Hamiltonian operator,
a wave function for a particle has vanishing antiparticle components, and vice versa. The
Hamilton operator yields the velocity operator v through the same relation that is valid
in classical physics:
v = x = β»
d p
. (511)
dt c m + c 2 p2
4 2

» minus to plus the speed of

This velocity operator shows a continuum of eigenvalues from
light. The velocity v is a constant of motion, as are p and c 4 m 2 + c 2 p2 .
Copyright © Christoph Schiller November 1997–May 2006

The orbital angular momentum l is also defined as in classical physics through

l=xp . (512)

Ref. 731 Both the orbital angular momentum l and the spin σ are separate constants of motion. A
particle (or antiparticle) with positive (or negative) component has a wave function with
only one non-vanishing component; the other three components vanish.
But alas, the representation of relativistic motion given by Foldy and Wouthuysen is
not the most simple for a generalization to particles when electromagnetic interactions
are present. The simple identity between classical and quantum-mechanical description
is lost when electromagnetism is included. We give below the way to solve the problem.

Dvipsbugw
 motion of matter – beyond cl assical physics 751

Maximum acceleration
Combining quantum theory with special relativity leads to a maximum acceleration value
for microscopic particles. Using the time–energy indeterminacy relation, you can deduce
Challenge 1218 n that
2mc 3
aD . (513)
ħ
Up to the present, no particle has ever been observed with a higher acceleration than
Dvipsbugw
Ref. 719 this value. In fact, no particle has ever been observed with accelerations approaching this
value. We note that the acceleration limit is different from the acceleration limit due to
general relativity:
c4
aD . (514)
4Gm
In particular, the quantum limit (513) applies to microscopic particles, whereas the gen-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

eral relativistic limit applies to macroscopic systems. Can you confirm that in each do-
Challenge 1219 e main the respective limit is the smaller of the two?

– CS – the rest of quantum theory will appear in soon – CS –

Curiosities and fun challenges about quantum theory

Quantum theory is so full of strange results that all of it could be titled ‘curiosities’. A few
of the prettier cases are given here.

**
The quantum of action implies that there are no fractals in nature. Can you confirm this
Challenge 1220 ny result?
**
Can atoms rotate? Can an atom that falls on the floor roll under the table? Can atoms
be put into high speed rotation? The answer is no to all questions, because angular mo-
Ref. 734 mentum is quantized and because atoms are not solid objects. The macroscopic case of
an object turning slower and slower until it stops does not exist in the microscopic world.
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1221 ny Can you explain how this follows from the quantum of action?
**
Do hydrogen atoms exist? Most types of atoms have been imaged with microscopes, pho-
tographed under illumination, levitated one by one, and even moved with needles, one
by one, as the picture shows. Others have moved single atoms using laser beams to push
Ref. 737 them. However, not a single of these experiments measured hydrogen atoms. Is that a
Challenge 1222 n reason to doubt the existence of hydrogen atoms? Taking seriously this not-so-serious
discussion can be a lot of fun.
**

Dvipsbugw
 752 v quanta of light and matter • 20. motion of matter

Light is refracted when entering dense matter. Do matter waves behave similarly? Yes,
they do. In 1995, David Pritchard showed this for sodium waves entering helium and
Ref. 736 xenon gas.
**
Two observables can commute for two different reasons: either they are very similar, such
as the coordinate x and x 2 , or they are very different, such as the coordinate x and the
Challenge 1223 ny momentum p y . Can you give an explanation?
Dvipsbugw
**
Space and time translations commute. Why then do the momentum operator and the
Challenge 1224 ny Hamiltonian not commute in general?
**
Ref. 738 With two mirrors and a few photons, it is possible to capture an atom and keep it floating

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

between the two mirrors. This feat, one of the several ways to isolate single atoms, is now
Challenge 1225 ny standard practice in laboratories. Can you imagine how it is realized?
**
For a bound system in a non-relativistic state with no angular momentum, one has the
Ref. 739 relation
9ħ 2
`rre `Te E , (515)
8m
where m is the reduced mass and T the kinetic energy of the components, and r the size
Challenge 1226 n of the system. Can you deduce the result and check it for hydrogen?

**
Electrons don’t like high magnetic fields. When a magnetic field is too high, electrons are
squeezed into a small space, in the direction transversal to their motion. If this spacing
becomes smaller than the Compton wavelength, something special happens. Electron-
positron pairs appear from the vacuum and move in such a way as to reduce the applied
magnetic field. The corresponding field value is called the quantum critical magnetic field.
Physicists also say that the Landau levels spacing then becomes larger than the electron
Challenge 1227 n rest energy. Its value is about 4.4 GT. Nevertheless, in magnetars, fields over 20 times as
Copyright © Christoph Schiller November 1997–May 2006

high have been measured. How is this possible?

**
Often one reads that the universe might have been born from a quantum fluctuation.
Challenge 1228 ny Does this statement make sense?

The examples so far have shown how quantons move that are described only by mass.
Now we study the motion of quantum systems that are electrically charged.

Dvipsbugw
colours and other interactions between light and matter 753

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 309 The spectrum of daylight: a stacked section of the rainbow (© Nigel
Sharp (NOAO), FTS, NSO, KPNO, AURA, NSF)

21. colours and other interactions bet ween light

and mat ter
Rem tene; verba sequentur.

“
After the description of the motion of matter and radiation, the next step is the descrip-
tion of their interactions. In other words, how do charged particles react to electromag-
Cato

”
netic fields and vice versa? Interactions lead to surprising effects, most of which appear
when the problem is treated taking special relativity into account.

What are stars made of ?

In the beginning of the eighteenth century the English physicist William Wollaston and
again the Bavarian instrument maker Joseph Fraunhofer* noted that the rainbow lacks
certain colours. These colours appear as black lines when the rainbow is spread out suf-
Copyright © Christoph Schiller November 1997–May 2006

ficiently. Figure 309 shows them in detail; the lines are called Fraunhofer lines today. In
1860, Gustav Kirchhoff and Robert Bunsen showed that the missing colours were exactly
* Born as Joseph Fraunhofer (b. 1787 Straubing, d. 1826 München). Bavarian, orphan at 11, he learned lens
polishing at that age; autodidact, he studied optics from books. He entered an optical company at age 19,
ensuring the success of the business, by producing the best available lenses, telescopes, micrometers, optical
gratings and optical systems of his time. He invented the spectroscope and the heliometer. He discovered
and counted 476 lines in the spectrum of the Sun, today named after him. Up to this day, Fraunhofer lines
are used as measurement standards. Physicists across the world would buy their equipment from him, visit
him and ask for copies of his publications. Even after his death, his instruments remain unsurpassed. With
his telescopes, in 1837 Bessel was able to measure the first parallax of a star and in 1846 Johann Gottfried
Galle discovered Neptune. Fraunhofer became professor in 1819; he died young, from the consequences of
the years spent working with lead and glass powder.

Dvipsbugw
 754 v quanta of light and matter • 21. colours and other interactions

those colours that certain elements emitted when heated. With a little of experimenting
they managed to show that sodium, calcium, barium, nickel, magnesium, zinc, copper
and iron existed on the Sun. However, they were unable to attribute 13 of the 476 lines
they observed. In 1868, Jules Janssen and Joseph Lockyer independently predicted that
the unknown lines were from a new element; it was eventually found also on Earth, in an
uranium mineral called cleveite, in 1895. Obviously it was called ‘helium’, from the Greek
word ‘helios’ – Sun. Today we know that it is the second ingredient of the Sun, in order of
frequency, and of the universe, after hydrogen. Helium, despite being so common, is rare
on Earth because it is a light noble gas that does not form chemical components. Helium Dvipsbugw
thus tends to rise in the atmosphere until it leaves the Earth.
Understanding the colour lines produced by each element had started to become of
interest already before the discovery of helium; the interest rose even more afterwards,
due to the increasing applications of colours in chemistry, physics, technology, crystallo-
graphy, biology and lasers. It is obvious that classical electrodynamics cannot explain the
sharp lines. Only quantum theory can explain colours.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What determines the colour of atoms?
The simplest atom to study is the atom of hydrogen. Hydrogen gas emits light consisting
of a number of sharp spectral lines. Already in 1885 century the Swiss school teacher
Johann Balmer (1828–1898) had discovered that the wavelengths of visible lines follow
from the formula
= R‹ − 2 .
1 1 1
(516)
λ mn 4 m

This expression was generalized by Johannes Rydberg (1854–1919) to include the ultravi-
olet and infrared colours
= R‹ 2 − 2 ,
1 1 1
(517)
λ mn n m

where n and m A n are positive integers, and the so-called Rydberg constant R has the
value 10.97 µm−1 . Thus quantum theory has a clearly defined challenge here: to explain
the formula and the value of R.
By the way, the transition λ 21 for hydrogen – the shortest wavelength possible – is
called the Lyman-alpha line. Its wavelength, 121.6 nm, lies in the ultraviolet. It is easily
observed with telescopes, since most of the visible stars consist of excited hydrogen. The
Copyright © Christoph Schiller November 1997–May 2006

Lyman-alpha line is regularly used to determine the speed of distant stars or galaxies,
since the Doppler effect changes the wavelength when the speed is large. The record so
Ref. 724 far is a galaxy found in 2004 with a Lyman-alpha line shifted to 1337 nm. Can you calculate
Challenge 1229 ny the speed with which it moves away from the Earth?
There are many ways to deduce Balmer’s formula from the minimum action. In
1926, Schrödinger solved his equation of motion for the electrostatic potential V (r) =
e 2 ~4πε 0 r of a point-like proton; this famous calculation however, is long and complex.
In order to understand hydrogen colours, it is not necessary to solve an equation of mo-
tion; it is sufficient to compare the initial and final state. This can be done most easily by
noting that a specific form of the action must be a multiple of ħ~2. This approach was
developed by Einstein, Brillouin and Keller and is now named after them. It states that

Dvipsbugw
 colours and other interactions between light and matter 755

Ref. 725 the action S of any quantum system obeys

S= y dq i p i = (n i + )ħ
1 µi
2π ∫ 4
(518)

for every coordinate q i and its conjugate momentum p i . Here, n i can be 0 or any positive
integer and µ i is the so-called Maslov index (an even integer) that in the case of atoms
has the value 2 for the radial and azimuthal coordinates r and θ, and 0 for the rotation
angle φ. Dvipsbugw
Any rotational motion in a spherical potential V (r) is characterized by a constant
energy E, and constant angular momenta L and L z . Therefore the conjugate momenta
Challenge 1230 ny for the coordinates r, θ and φ are
¾
L2
pr = 2m(E − V (r)) − 2
¾
r

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2
p θ = L2 −
Lz
sin2 θ
pφ = Lz . (519)

Using these expressions in equation (518) and setting n = n r + n θ + n φ + 1 yields* the

result
−me 4 −R 2.19 aJ 13.6 eV
En = 2 = 2  
1
.
n 2(4πε 0 ) ħ
(522)
2 2 n n2 n2

Using the idea that a hydrogen atom emits a single photon when its electron changes from
Challenge 1232 e state E n to E m , one gets the formula found by Balmer and Rydberg. (This whole discussion
assumes that the electrons in hydrogen atoms are in eigenstates. Can you argue why this
Challenge 1233 ny is the case?)
The effective radius of the electron orbit in hydrogen is given by

ħ 2 4πε 0
rn = n2 = n 2 a 0  n 2 53 pm . (523)
πme 2
The smallest value 53 pm for n = 1 is called the Bohr radius and is abbreviated a 0 .
Quantum theory thus implies that a hydrogen atom excited to the level n = 500 is
Copyright © Christoph Schiller November 1997–May 2006

about 12 µm in size, larger than many bacteria! This feat has indeed been achieved, even

* The calculation is straightforward. After insertion of V (r) = e~4πε 0 r into equation (519) one needs to
Challenge 1231 ny perform the (tricky) integration. Using the general result

dz » 2 º
Az + 2Bz − C = − C + º
1 B
2π y∫ z −A
(520)

¾
one gets
e2
(n r + )ħ + L = nħ =
1 m
. (521)
2 4πε 0 −2E
This leads to the energy formula (522).

Dvipsbugw
 756 v quanta of light and matter • 21. colours and other interactions

Figure to be included

F I G U R E 310 The energy levels of hydrogen Dvipsbugw

though such blown-up atoms, usually called Rydberg atoms, are extremely sensitive to
Ref. 727 perturbations.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The orbital frequency of electrons in hydrogen is

1 e4 m
fn = (524)
n 3 4ε 20 h 3

and the electron speed is

e2
vn =  
2.2 Mm~s 0.007 c
(525)
2nε 0 h n n

As expected, the further electrons orbit the nucleus, the slower they move. This result
can be checked by experiment. Exchanging the electron by a muon allows to measure
Ref. 726 the time dilation of its lifetime. The measurements coincide with the formula. We note
that the speeds are slightly relativistic. However, this calculation did not take into account
relativistic effects. Indeed, precision measurements show slight differences between the
calculated energy levels and the measured ones.

Relativistic hydrogen
Also in the relativistic case, the EBK action has to be a multiple of ħ~2. From the relativistic
expression of energy
Copyright © Christoph Schiller November 1997–May 2006

»
Ref. 725
e2
E + mc 2 = p2 c 2 + m 2 c 4 − (526)
4πε 0 r

Challenge 1234 ny we get the expression

2me 2
p2r = 2mE(1 + ) (1 + ).
E E
+ (527)
2mc 2 4πε 0 r mc 2

We now use the expression for the dimensionless fine structure constant α = e2
=
» 4πε 0 ħc
Challenge 1235 ny 2R~mc 2  1~137.036. The radial EBK action then implies that

Dvipsbugw
colours and other interactions between light and matter 757

¿
Á 2Á α2
(E n l + mc ) = mc Á
2
À1 + ¼ . (528)
(n − l − 21 + (l + 21 )2 − α 2 )2

This result is correct for point-like electrons. In reality, the electron has spin 1/2; the cor-
rect relativistic energy levels thus appear when we set l = j  1~2 in the above formula.
The result can be approximated by

−R α2 n Dvipsbugw
En j = (1 ( − ) + ...)
3
+ (529)
n2 n 2 j + 21 4

It reproduces the hydrogen spectrum to an extremely high accuracy. Only the introduc-
tion of virtual particle effects yields an even better result. We will present this point later
on.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Relativistic wave equations – again
The equation was more intelligent than I was.

“ Paul Dirac about his equation, repeating

a statement made by Heinrich Hertz.

Unfortunately, the representation of relativistic motion given by Foldy and Wouthuysen ”

is not the most simple for a generalization to particles in the case that electromagnetic
interactions are present. The simple identity between classical and quantum-mechanical
description is lost if electromagnetism is included.
Charged particles are best described by another, equivalent representation of the
Hamiltonian, which was discovered much earlier, in 1926, by the British physicist Paul
Dirac.* Dirac found a neat trick to take the square root appearing in the relativistic en-
ergy operator. In this representation, the Hamilton operator is given by

H Dirac = βm + α ë p (530)

Its position operator x is not the position of a particle, but has additional terms; its velo-
city operator has only the eigenvalues plus or minus the velocity of light; the velocity op-
erator is not simply related to the momentum operator; the equation of motion contains
the famous ‘Zitterbewegung’ term; orbital angular momentum and spin are not separate
Copyright © Christoph Schiller November 1997–May 2006

constants of motion.

* Paul Adrien Maurice Dirac (b. 1902 Bristol, d. 1984 Tallahassee), British physicist, born as son of a French-
speaking Swiss immigrant. He studied electrotechnics in Bristol, then went to Cambridge, where he later
became professor on the chair Newton had held before. In the years from 1925 to 1933 he published a stream
of papers, of which several were worth a Nobel Prize, which he received in 1933. He unified special relativ-
ity and quantum theory, he predicted antimatter, he worked on spin and statistics, he predicted magnetic
monopoles, he speculated on the law of large numbers etc. His introversion, friendliness and shyness, his
deep insights into nature, combined with a dedication to beauty in theoretical physics, made him a legend
all over the world already during his lifetime. For the latter half of his life he tried, unsuccessfully, to find an
alternative to quantum electrodynamics, of which he was the founder, as he was repelled by the problems of
infinities. He died in Florida, where he lived and worked after his retirement from Cambridge.

Dvipsbugw
 758 v quanta of light and matter • 21. colours and other interactions

So why use this horrible Hamiltonian? It is the only Hamilto-

nian that can be easily used for charged particles. Indeed, it is
transformed to the one coupled to the electromagnetic field by
Page 551 the so-called minimal coupling, i.e. by the substitution

p p − qA . (531)

that treats electromagnetic momentum like particle momentum.

With this prescription, Dirac’s Hamiltonian describes the mo- Dvipsbugw
tion of charged particles interacting with an electromagnetic field Paul Dirac
A. This substitution is not possible in the Foldy–Wouthuysen
Hamiltonian. In the Dirac representation, particles are pure, point-like, structureless elec-
tric charges; in the Foldy–Wouthuysen representation they acquire a charge radius and
Ref. 730 a magnetic moment interaction. (We come back to the reasons below, in the section on
QED.)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In more detail, the simplest description of an electron (or any other elementary, stable,
electrically charged particle of spin 1~2) is given by the equations

= [H, ρ]
dρ
dt
H Dirac = βmc 2 + α ë (p − qA(x, t))c + qφ(x, t) with
’ 0 0 0 1 “ ’ 0 0 0 −i “ ’ 0 0 1 0 “
– — – 0 0 i 0 — – 0 0 0 −1 —
α1 = – — α2 = – — α3 = – —
0 0 1 0
– 0 1 0 0 — – 0 −i 0 0 — – 1 0 0 0 —
” 1 0 0 0 • ” i 0 0 0 • ” 0 −1 0 0 •
’ 1 0 0 0 “
– —
β=– —
0 1 0 0
– 0 0 −1 0 —
” 0 0 0 −1 •
H Maxwell = . (532)

The first Hamiltonian describes how charged particles are moved by electromagnetic
fields, and the second describes how fields are moved by charged particles. Together, they
form what is usually called quantum electrodynamics or QED for short.
Copyright © Christoph Schiller November 1997–May 2006

As far as is known today, the relativistic description of the motion of charged mat-
ter and electromagnetic fields given by equation (532) is perfect: no differences between
theory and experiment have ever been found, despite intensive searches and despite a
high reward for anybody who would find one. All known predictions completely corres-
pond with the measurement results. In the most spectacular cases, the correspondence
between theory and measurement is more than fourteen digits. But the precision of QED
is less interesting than those of its features that are missing in classical electrodynamics.
Let’s have a quick tour.

– CS – more to come here – CS –

Dvipsbugw
 colours and other interactions between light and matter 759

Antimatter
Antimatter is now a household term. Interestingly, the concept was formed before any ex-
perimental evidence for it was known. Indeed, the antimatter companion of the electron
was predicted in 1926 by Paul Dirac from his equation. Without knowing this prediction,
Carl Anderson discovered it in 1932 and called it positron, even though ‘positon’, without
the ‘r’, would have been the correct name. Anderson was studying cosmic rays and noticed
that some ‘electrons’ were turning the wrong way in the magnetic field he had applied to
his apparatus. He checked everything in his machine and finally deduced that he found Dvipsbugw
a particle with the same mass as the electron, but with positive electric charge.
The existence of positrons has many strange implications. Already in 1928, before their
discovery, the swedish theorist Oskar Klein had pointed out that Dirac’s equation for
electrons makes a strange prediction: when an electron hits a sufficiently steep potential
wall, the reflection coefficient is larger than unity. Such a wall will reflect more than what
is thrown at it. In 1935, after the discovery of the positron, Werner Heisenberg and Hans

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 732 Euler explained the paradox. They found that the Dirac equation predicts a surprising
effect: if an electric field exceeds the critical value of

m e c 2 m e2 c 3
Ec = = = 1.3 EV~m , (533)
eλ e eħ

the vacuum will spontaneously generate electron–positron pairs, which then are separ-
ated by the field. As a result, the original field is reduced. This so-called vacuum polar-
ization is also the reason for the reflection coefficient greater than unity found by Klein,
since steep potentials correspond to high electric fields.
Truly gigantic examples of vacuum polarization, namely around charged black holes,
Page 875 will be described later on.
We note that such effects show that the number of particles is not a constant in the
microscopic domain, in contrast to everyday life. Only the difference between particle
number and antiparticle number turns out to be conserved. This topic will be expanded
in the chapter on the nucleus.
Of course, the generation of electron–positron pairs is not a creation out of nothing,
but a transformation of energy into matter. Such processes are part of every relativistic
description of nature. Unfortunately, physicists have the habit to call this transformation
‘creation’ and thus confuse this issue somewhat. Vacuum polarization is a process trans-
forming, as we will see, virtual photons into matter. That is not all: the same can also be
Copyright © Christoph Schiller November 1997–May 2006

done with real photons.

Virtual particles and QED diagrams

Contrary to what was said so far, there is a case where actions smaller than the minimal
one do play a role. We already encountered an example: in the collision between two
Page 733 electrons, there is an exchange of virtual photons. We know that the exchanged virtual
photon cannot be observed. Indeed, the action value S for this exchange obeys

SD
ħ
. (534)
2

Dvipsbugw
 760 v quanta of light and matter • 21. colours and other interactions

In short, virtual particles are those particles that appear only as mediators in interactions;
they cannot be observed. Virtual particles are intrinsically short-lived; they are the oppos-
ite of free or real particles. In a certain sense, virtual particles are particles bound both in
space and time.

– CS – more to come – CS –

In summary, all virtual matter and radiation particle-antiparticles pairs together form
what we call the vacuum; in addition, virtual radiation particles form static fields. Vir- Dvipsbugw
tual particles are needed for a full description of interactions, and in particular, they are
responsible for every decay process.
We will describe a few more successes of quantum theory shortly. Before we do that,
we settle one important question.

Compositeness

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

When is an object composite? Quantum theory gives several pragmatic answers. The first
one is somewhat strange: an object is composite when its gyromagnetic ratio is different
than the one predicted by QED. The gyromagnetic ratio γ is defined as the ratio between
the magnetic moment M and the angular momentum L. In other terms,

M = γL . (535)

The gyromagnetic ratio γ is measured in s−1 T−1 =C~kg and determines the energy levels
of magnetic spinning particles in magnetic fields; it will reappear later in the context of
Page 897 magnetic resonance imaging. All candidates for elementary particles have spin 1~2. The
gyromagnetic ratio for spin 1~2 particles of mass m can be written as

γ= =д
M e
. (536)
ħ~2 2m

(The expression eħ~2m is often called the magneton of the particle; the dimensionless
factor g/2 is often called the gyromagnetic ratio as well; this sometimes leads to confu-
sion.) The criterion of elementarity thus can be reduced to a criterion on the value of
the dimensionless number д, the so-called д-factor. If the д-factor differs from the value
predicted by QED for point particles, about 2.0, the object is composite. For example, a
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1236 ny
4
He+ helium ion has a spin 1~2 and a д value of 14.7 ë 103 . Indeed, the radius of the he-
lium ion is 3 ë 10−11 m, obviously finite and the ion is a composite entity. For the proton,
one measures a д-factor of about 5.6. Indeed, experiments yield a finite proton radius of
about 0.9 fm.
Also the neutron, which has a magnetic moment despite being neutral, must therefore
be composite. Indeed, its radius is approximately that of the proton. Similarly, molecules,
mountains, stars and people must be composite. Following this first criterion, the only
elementary particles are leptons – i.e. electrons, muons, tauons and neutrinos –, quarks
and intermediate bosons – i.e. photons, W-bosons, Z-bosons and gluons. More details on
Page 897 these particles will be uncovered in the chapter on the nucleus.

Dvipsbugw
 colours and other interactions between light and matter 761

Another simple criterion for compositeness has just been mentioned: any object with a
measurable size is composite. This criterion produces the same list of elementary particles
as the first. Indeed, this criterion is related to the previous one. The simplest models for
Ref. 733 composite structures predicts that the д-factor obeys

д−2=
R
(537)
λC

where R is the radius and λ C = h~mc the Compton wavelength of the system. The expres- Dvipsbugw
sion is surprisingly precise for helium 4 ions, helium 3, tritium ions and protons, as you
Challenge 1237 e might want to check.
A third criterion for compositeness is more general: any object larger than its Compton
length is composite. The background idea is simple. An object is composite if one can
detect internal motion, i.e. motion of some components. Now the action of any part with
mass m part moving inside a composed system of size r follows

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

S part < 2π r m part c < π r m c (538)

where m is the mass of the composite object. On the other hand, following the principle
of quantum theory, this action, to be observable, must be larger than ħ~2. Inserting this
condition, we find that for any composite object*

rA
ħ
. (539)
2π m c

The right hand side differs only by a factor 4π 2 from the so-called Compton (wave)length

λ=
h
. (540)
mc
of an object. Any object larger than its own Compton wavelength is thus composite. Any
object smaller than the right hand side of expression (539) is thus elementary. Again, only
leptons, including neutrinos, quarks and intermediate bosons pass the test. All other ob-
jects are composite, as the tables in Appendix C make clear. This third criterion produces
Challenge 1239 ny the same list as the previous ones. Can you explain the reason?
Copyright © Christoph Schiller November 1997–May 2006

Interestingly, the topic is not over yet. Even stranger statements about compositeness
will appear when gravity is taken into account. Just be patient; it is worth it.

Curiosities and fun challenges about colour

Colours are at least as interesting in quantum theory as they are in classical electrodyna-
mics.
**

Challenge 1238 ny * Can you find the missing factor of 2? And is the assumption valid that the components must always be
lighter than the composite?

Dvipsbugw
 762 v quanta of light and matter • 21. colours and other interactions

If atoms contain orbiting electrons, the rotation of the Earth, via the Coriolis acceleration,
should have an effect on their motion. This beautiful prediction is due to Mark Silverman;
Ref. 726 the effect is so small however, that is has not been measured yet.
**
Light is diffracted by material gratings. Can matter be diffracted by light gratings? Sur-
prisingly, it actually can, as predicted by Dirac and Kapitza in 1937. In 1986, this was
Ref. 735 accomplished with atoms. For free electrons the feat is more difficult; the clearest con-
firmation came in 2001, when the technology advances for lasers were used to perform Dvipsbugw
a beautiful measurement of the typical diffraction maxima for electrons diffracted by a
light grating.
**
Light is totally reflected when it is directed to a dense material under an angle so large that
it cannot enter it any more. Interestingly, in the case that the material is excited, the totally

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 726 reflected beam can be amplified. This has been shown by several Russian physicists.

**
Where is the sea bluest? Sea water is blue because it absorbs red and green light. Sea water
can also be of bright colour if the sea floor reflects light. Sea water is often also green,
because it often contains small particles that scatter or absorb blue light. Most frequently,
this is soil or plankton. The sea is thus especially blue if it is deep, clear and cold, so that it
is low in plankton content. (Satellites determine plankton content from the ‘greenness’ of
the sea.) There is a place where the sea is deep, cold and quiet for most parts of the year:
the Sargasso sea. It is often called the bluest spot of the oceans.

The strength of electromagnetism

The great Wolfgang Pauli used to say that after his death, the first question he would ask
the devil would be an explanation of Sommerfeld’s fine structure constant. (People used
to comment that after the devil will have explained it to him, he would think a little, and
then snap ‘Wrong!’) The name fine structure constant was given by Arnold Sommerfeld
Page 578 to the dimensionless constant of nature given by

e2
α=   0.007 297 352 533(27) .
1
(541)
Copyright © Christoph Schiller November 1997–May 2006

4πε 0 ħc 137.035 999 76(50)

This number first appeared in explanations for the fine structure of certain atomic colour
spectra, hence its name. Sommerfeld was the first to understand its general importance.
The number is central to quantum electrodynamics for several reasons. First of all, it
describes the strength of electromagnetism. Since all charges are multiples of the electron
charge, a higher value would mean a stronger attraction or repulsion between charged
bodies. The value of α thus determines the size of atoms, and thus the size of all things,
as well as all colours.
Secondly, only because this number is quite a bit smaller than unity are we able to
talk about particles at all. The argument is somewhat involved; it will be detailed later

Dvipsbugw
 colours and other interactions between light and matter 763

on. In any case, only the small value of the fine structure constant makes it possible
to distinguish particles from each other. If the number were near or larger than one,
particles would interact so strongly that it would not be possible to observe or to talk
about particles at all.
This leads to the third reason for the importance of the fine structure constant. Since it
is a dimensionless number, it implies some yet unknown mechanism that fixes its value.
Uncovering this mechanism is one of the challenges remaining in our adventure. As long
as the mechanism remains unknown, we do not understand the colour and size of a single
thing around us. Dvipsbugw
Small changes in the strength of electromagnetic attraction between electrons and pro-
tons would have numerous important consequences. Can you describe what would hap-
pen to the size of people, to the colour of objects, to the colour of the Sun or to the work-
ings of computers if the strength would double? And if it would drop to half the usual
Challenge 1240 ny value over time?
Explaining the number is the most famous and the toughest challenge of modern phys-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ics since the issue appeared in the 1920s. It is the reason for Pauli’s request to the devil. In
1946, during his Nobel Prize lecture, he repeated the statement that a theory that does not
Ref. 740 determine this number cannot be complete. The challenge is so tough that for the first
50 years there were only two classes of physicists: those who did not even dare to take on
the challenge, and those who had no clue. This fascinating story still awaits us.
The topic of the fine structure constant is so deep that it leads many astray. For example,
it is often heard that in physics it is impossible to change physical units in such a way that
ħ, c and e are equal to 1 at the same time; these voices suggest that doing so would not
allow that the number 1~137.036... would keep its value. Can you show that the argument
Challenge 1241 n is wrong, and that doing so does not affect the fine structure constant?
To continue with the highest efficiency on our path across quantum theory, we first
look at two important topics: the issue of indistinguishability and the issue of interpreta-
tion of its probabilities.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
764 v quanta of light and matter

Biblio graphy

677 Giuseppe Fumagalli, Chi l’ha detto?, Hoepli, Milano, 1983. Cited on page 704.

678 The quantum of action is introduced in Max Pl anck, Über irreversible Strahlungsvor-
gänge, Sitzungsberichte der Preußischen Akademie der Wissenschaften, Berlin, pp. 440–
480, 1899. In this paper, Planck used the letter b for what nowadays is called h. Cited on
page 705.
Dvipsbugw
679 Bohr explained the indivisibilty of the quantum of action in his famous Como lecture. N.
B ohr, Atomtheorie und Naturbeschreibung, Springer, Berlin, 1931. On page 16 he writes:
“No more is it likely that the fundamental concepts of the classical theories will ever become
superfluous for the description of physical experience. The recognition of the indivisibility
of the quantum of action, and the determination of its magnitude, not only depend on an
analysis of measurements based on classical concepts, but it continues to be the application
of these concepts alone that makes it possible to relate the symbolism of the quantum theory

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

to the data of experience.” He also writes: “...the fundamental postulate of the indivisibility
of the quantum of action is itself, from the classical point of view, an irrational element
which inevitably requires us to forgo a causal mode of description and which, because of
the coupling between phenomena and their observation, forces us to adopt a new mode
of description designated as complementary in the sense that any given application of
classical concepts precludes the simultaneous use of other classical concepts which in a
different connection are equally necessary for the elucidation of the phenomena...” and
“...the finite magnitude of the quantum of action prevents altogether a sharp distinction
being made between a phenomenon and the agency by which it is observed, a distinction
which underlies the customary concept of observation and, therefore, forms the basis of
the classical ideas of motion.” Other statements about the indivisibility of the quantum of
action can be found in N. B ohr, Atomic Physics and Human Knowledge, Science Editions,
New York, 1961. See also Max Jammer, The Philosophy of Quantum Mechanics, Wiley,
first edition, 1974, pp. 90–91.
For some of the rare modern publications emphasizing the quantum of action see M.B.
Mensky, Physics Letters A The action uncertainty principle and quantum gravity, 162,
p. 219, 1992, and M.B. Mensky, The action uncertainty principle in continuous quantum
measurements, Physics Letters A 155, pp. 229–235, 1991.
Schwinger’s quantum action principle is also used in R ichard F.W. Bader, Atoms
in Molecules - A Quantum Theory, Oxford University Press, 1994.
There is a large number of general textbooks on quantum theory. The oldest textbooks
are obviously German or English. The choice should be taken by the reader, considering
Copyright © Christoph Schiller November 1997–May 2006

his own preferences.

A well-known conceptual introduction is Jean-Marc Lév y-Leblond &
Françoise Balibar, Quantique – Rudiments, Masson, 1997, translated into English
as Quantics, North-Holland, 1990.
One of the most beautiful books is Julian Schwinger, Quantum Mechanics - Sym-
bolism of Atomic Measurements, edited by Berthold-Georg Englert, Springer Verlag, 2001.
A modern approach with a beautiful introduction is Max Schubert & Gerhard
Weber, Quantentheorie – Grundlagen und Anwendungen, Spektrum Akademischer Verlag,
1993.
A standard beginner’s text is C. Cohen-Tannoudji, B. Diu & F. L aloë, Mécanique
quantique I et II, Hermann, Paris, 1977. It is also available in several translations.
A good text is Asher Peres, Quantum theory – Concepts and Methods, Kluwer, 1995.

Dvipsbugw
bibliography 765

For a lively approach, see Vincent Icke, The Force of Symmetry, Cambridge Univer-
sity Press/ 1994.
Other textbooks regularly appear around the world. Cited on page 705.
680 Max B orn, Zur Quantenmechanik der Stoßvorgänge (vorläufige Mitteilung), Zeitschrift
für Physik 37, pp. 863–867, 1926, Max B orn, Quantenmechanik der Stoßvorgänge, Zeits-
chrift für Physik 38, pp. 803–827, 1926. Cited on page 707.
681 See for example the papers by Jan Hilgevoord, The uncertainty principle for energy
and time, American Journal of Physics 64, pp. 1451–1456, 1996, and by Paul Busch, On the
time–energy uncertainty reaction, part 1 & 2, Foundations of Physics 20, pp. 1–43, 1990. A Dvipsbugw
classic is the paper by Eugene P. Wigner, On the time–energy uncertainty reaction, in
Abdus Sal am & Eugene P. Wigner, editors, Aspects of Quantum Theory, Cambridge
University Press, 1972. Cited on page 710.
682 See also the booklet by Cl aus Mattheck, Warum alles kaputt geht - Form und Versagen
in Natur und Technik, Forschungszentrum Karlsruhe, 2003. Cited on page 712.
683 R. Clifton, J. Bub & H. Halvorson, Characterizing quantum theory in terms of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

information-theoretic constraints, http://www.arxiv.org/abs/quant-ph/0211089. Cited on
page 715.
684 The quotes on motion are found in chapter VI of F. Engels, Herrn Eugen Dührings Um-
wälzung der Wissenschaft, Verlag für fremdsprachliche Literatur, Moskau, 1946. The book is
commonly called Anti–Dühring. Cited on pages 716 and 737.
685 E.M. Brumberg & S.I. Vavilov, Izvest. Akad. Nauk. Omen Ser. 7, p. 919, 1933. Cited
on page 717.
686 Rodney Loud on, The Quantum Theory of Light, Oxford University Press, 2000. Cited on
page 717.
687 F. R ieke & D.A. Baylor, Single-photon detection by rod cells of the retina, Reviews of
Modern Physics 70, pp. 1027–1036, 1998. They also mention that the eye usually works at
photon fluxes between 108 ~µm2 s (sunlight) and 10−2 ~µm2 s (starlight). The cones in the
retina detect, in colour, light intensities in the uppermost 7 or 8 decades, whereas the rods
detect, in black and white, the lower light intensities. Cited on page 719.
688 E. Fischbach, H. Kloor, R.A. L angel, A.T.Y. Lui & M. Pered o, New geomag-
netic limit on the photon mass and on long-range forces coexisting with electromagnetism,
Physical Review Letters 73, pp. 514–517, 1994. Cited on page 720.
689 A.H. Compton, The scattering of X-rays as particles, American Journal of Physics 29,
pp. 817–820, 1961. Cited on page 720.
690 The famous paper is R. Hanbury Brown & R.Q. Twiss, Nature 178, p. 1046, 1956.
Copyright © Christoph Schiller November 1997–May 2006

They got the idea to measure light in this way from their earlier work, which used the same
method with radiowaves: R. Hanbury Brown & R.Q. Twiss, Nature 177, p. 27, 1956,
Cited on page 725.
691 Science 8 September 1995 Cited on page 725.
692 L. Mandel, Configuration-space photon number operators in quantum optics, Physical
Review 144, pp. 1071–1077, 1966. No citations.
693 A. Einstein, Über einen die Erzeugung und Umwandlung des Lichtes betreffenden heur-
istischen Standpunkt, Annalen der Physik 17, pp. 132–184, 1905. Cited on page 726.
694 See the summary by P.W. Milonni, Answer to question 45: What (if anything) does the
photoelectric effect teach us?, American Journal of Physics 65, pp. 11–12, 1997. Cited on page
726.

Dvipsbugw
766 v quanta of light and matter

695 For a detailed account, See Jeffrey J. Prentis, Poincaré’s proof of the quantum discon-
tinuity of nature, American Journal of Physics 63, pp. 339–350, 1995. The original papers are
Henri Poincaré, Sur la théorie des quanta, Comptes Rendus de l’Académie des Sciences
(Paris) 153, pp. 1103–1108, 1911, as well as Henri Poincaré, Sur la théorie des quanta,
Journal de Physique (Paris) 2, pp. 5–34, 1912. Cited on page 726.
696 J. Jacobson, G. Björk, I. Chang & Y. Yamamoto, Photonic de Broglie waves, Phys-
ical Review Letters 74, pp. 4835–4838, 1995. The first measurement was published by E.J.S.
Fonseca, C.H. Monken & S. de Pádua, Measurement of the de Broglie wavelength of
a multiphoton wave packet, Physical Review Letters 82, pp. 2868–2671, 1995. Cited on page Dvipsbugw
727.
697 For the three photon state, see M.W. Mitchell, J.S. Lundeen & A.M. Steinberg,
Super-resolving phase measurements with a multiphoton entangled state, Nature 429,
pp. 161–164, 2004, and for the four-photon state see, in the same edition, P. Walther,
J.-W. Pan, M. Aspelmeyer, R. Ursin, S. Gasparoni & A. Zeilinger, De Broglie
wavelength of a non-local four-photon state, Nature 429, p. 158-161, 2004. Cited on page

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

727.
698 For an introduction to spueezed light, see L. Mandel, Non-classical states of the electro-
magnetic field, Physica Scripta T 12, pp. 34–42, 1986. Cited on page 727.
699 The famous quote on single photon interference is found on page 9 of P.A.M. Dirac,
The Principles of Quantum Mechanics, Clarendon Press, Oxford, 1930. It is also discussed,
somewhat confusely, in the otherwise informative article by H. Paul, Interference between
independent photons, Reviews of Modern Physics 58, pp. 209–231, 1986. Cited on page 729.
700 The original papers on coherent states are three R.J. Gl auber, The quantum theory of
optical coherence, Physical Review 130, pp. 2529–2539, 1963, J.R. Kl auder, Continuous-
representation theory, I and II, Journal of Mathematical Physics 4, pp. 1055–1058, 1963, and
E.C.G. Sudarshan, Equivalence of semiclassical and quantum mechanical descriptions
of statistical light beams, Physical Review Letters 10, p. 227, 1963. Cited on page 731.
701 R. Kidd, J. Aedini & A. Anton, Evolution of the modern photon, American Journal of
Physics 57, pp. 27–35, 1989, Cited on page 732.
702 Wolf gang Tittel, J. Brendel, H. Zbinden & N. Gisin, Violation of bell inequalit-
ies by photons more than 10 km apart, Physical Review Letters 81, pp. 3563–3566, 26 October
1998. Cited on page 734.
703 N. B ohr & L. Rosenfeld, Zur Frage der Meßbarkeit der elektromagnetischen Feld-
größen, Mat.-fys. Medd. Danske Vid. Selsk. 12, p. 8, 1933. The results were later published
in English language as N. B ohr & L. Rosenfeld, Field and charge measurements in
Copyright © Christoph Schiller November 1997–May 2006

quantum electrodynamics, Physical Review 78, pp. 794–798, 1950. Cited on page 734.
704 E.I. Bu tikov, The rigid pendulum – an antique but evergreen physical model, European
Journal of Physics 20, pp. 429–441, 1999. Cited on page 736.
705 W. Gerl ach & O. Stern, Der experimentelle Nachweis des magnetischen Moments des
Silberatoms, Zeitschrift für Physik 8, p. 110, 1921. See also the pedagogical explanation by
M. Hannou t, S. Hoyt, A. Kryowonos & A. Wid om, Quantum measurement and
the Stern–Gerlach experiment, American Journal of Physics 66, pp. 377–379, 1995. Cited on
page 742.
706 J. Perrin, Nobel Prize speech, http://www.nobel.se. H. Nagaoka, Kinetics of a system
of particles illustrating the line and the band spectrum and the phenomena of radioactivity,
Philosophical Magazine S6, 7, pp. 445–455, March 1904. Cited on page 747.

Dvipsbugw
bibliography 767

707 N. B ohr, On the constitution of atoms and molecules: Introduction and Part I – binding of
electrons by positive nuclei, Philosophical Magazine 26, pp. 1–25, 1913, On the constitution
of atoms and molecules: Part II – systems containing only a single nucleus, ibid., pp. 476–
502, On the constitution of atoms and molecules: Part III, ibid., pp. 857–875. Cited on page
747.
708 L. de Broglie, Comptes rendus de l’Académie des Sciences 177, pp. 507–510, 1923. Cited
on page 738.
709 M. Arndt, O. Nairz, J. Vos–Andreae, C. Keller, G. van der Z ouw & A.
Zeilinger, Wave–particle duality of C60 molecules, Nature 401, pp. 680–682, 14 Octo- Dvipsbugw
ber 1999. See also the observation for tetraphenyleprophyrin and C 60 F48 by the same team,
published as L. Hackermüller & al., Wave nature of biomolecules and fluorofullerenes,
Physical Review Letters 91, p. 090408, 2003.
No experiment of quantum theory has been studied as much as quantum intereference.
Also the transition from interference to no interference has been epxlored, as in P. Facchi,
A. Mariano & S. Pascazio, Mesoscopic interference, Recent Developments in Physics 3,
pp. 1–29, 2002. Cited on page 738.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

710 P. Carru thers & M.M. Nieto, Phase and angle variables in quantum mechanics, Re-
view of Modern Physics 40, pp. 411–440, 1968. Cited on page 741.
711 The indeterminacy relation for rotational motion is well explained by W.H. Louisell,
Amplitude and phase uncertainty relations, Physics Letters 7, p. 60, 1963. Cited on page 741.
712 S. Franke-Arnold, S.M. Barnett, E. Yao, J. Leach, J. Courtial & M. Pad-
gett, Uncertainty principle for angular position and angular momentum, New Journal of
Physics 6, p. 103, 2004. This is a freely accessible online journal. Cited on page 741.
713 Erwin Schrödinger, Quantisierung als Eigenwertproblem I, Annalen der Physik 79,
pp. 361–376, 1926, and Quantisierung als Eigenwertproblem II, Annalen der Physik 79,
pp. 489–527, 1926. Cited on page 745.
714 C.G. Gray, G. Karl & V.A. Novikov, From Maupertius to Schrödinger. Quantization
of classical variational principles, American Journal of Physics 67, pp. 959–961, 1999. Cited
on page 745.
715 The whole bunch of atoms behaves as one single molecule; one speaks of a Bose–Einstein
condensate. The first observations, worth a Nobel prize, were by M.H. Anderson & al.,
Observation of Bose–Einstein condensation in a dilute atomic vapour, Science 269, pp. 198–
201, 1995, C.C. Bradley, C.A. Sackett, J.J. Tollett & R.G. Hulet, Evidence of
bose-einstein condensation in an atomic gas with attractive interactions, Physical Review
Letters 75, pp. 1687–1690, 1995, K.B. Davis, M.-O. Mewes, M.R. Andrews, N.J. van
Dru ten, D.S. Durfee, D.M. Kurn & W. Ketterle, Bose-Einstein Condensation in
Copyright © Christoph Schiller November 1997–May 2006

a Gas of Sodium Atoms, Physical Review Letters 75, pp. 3969–3973, 1995. For a simple intro-
duction, see W. Ketterle, Experimental studies of Bose–Einstein condensation, Physics
Today pp. 30–35, December 1999. Cited on page 736.
716 W.M. Itano, D.J. Heinzen, J.J. B ollinger & D.J. Winel and, Quantum Zeno effect,
Physical Review A 41, pp. 2295–2300, 1990. M.C. Fischer, B. Gu tiérrez-Medina &
M.G. R aizen, Observation of the Quantum Zeno and Anti-Zeno effects in an unstable sys-
tem, Physical Review Letters 87, p. 040402, 2001, also http://www-arxiv.org/abs/quant-ph/
0104035. Cited on page 747.
717 See P. Facchi, Z. Hradil, G. Krenn, S. Pascazio &J. Řeháček Quantum Zeno
tomography, Physical Review A, 66, p. 012110, 2002. Cited on page 747.
718 Robert H. Dicke & James P. Wittke, Introduction to quantum theory, Addison–

Dvipsbugw
768 v quanta of light and matter

Wesley, Reading, Massachusetts, 1960. See also Stephen Gasiorowicz, Quantum Phys-
ics, Jojn Wiley & Sons, 1974. Cited on page 740.
719 G. Papini, Shadows of a maximal acceleration, http://www.arxiv.org/abs/gr-qc/0211011.
Cited on page 751.
720 Y. Aharonov & D. B ohm, Significance of electromagnetic potentials in the quantum
theory, Physical Review 115, pp. 485–491, 1959. No citations.
721 G.E. Uhlenbeck & S. Gousmit, Ersetzung der Hypothese vom unmechanischen
Zwang durch eine Forderung bezüglich des inneren Verhaltens jedes einzelnen Elektrons,
Naturwissenschaften 13, pp. 953–954, 1925. Cited on page 749.
Dvipsbugw
722 L. Thomas, The motion of the spinning electron, Nature 117, p. 514, 1926. Cited on page
749.
723 K. von Meyenn & E. Schucking, Wolfgang Pauli, Physics Today pp. 43–48, February
2001. Cited on page 749.
724 R. Pello, D. Schaerer, J. R ichard, J.-F. Le B orgne & J.-P. Kneib, ISAAC/VLT
observations of a lensed galaxy at z=10.0, Astronomy and Astrophysics 416, p. L35, 2004.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Cited on page 754.
725 A pedagogical introduction is given by L.J. Curtis & D.G. Ellis, Use of the Einstein–
Brillouin–Keller action quantization, American journal of Physics 72, pp. 1521–1523, 2004.
Cited on pages 755 and 756.
726 Mark P. Silverman, And yet it moves: strange systems and subtle questions in physics,
Cambridge University Press 1993. A beautiful book from an expert of motion. Cited on
pages 756 and 762.
727 J. Neukammer & al., Spectroscopy of Rydberg atoms at n  500, Physical Review Letters
59, pp. 2947–2950, 1987. Cited on page 756.
728 This is explained by J.D. Hey, Mystery error in Gamow’s Tompkins reappears, Physics
Today pp. 88–89, May 2001. No citations.
729 L.L. Foldy & S.A. Wou thuysen, On the Dirac theory of spin 1/2 particles and its non-
relativistic limit, Physical Review 78, pp. 29–36, 1950. Cited on page 750.
730 L.L. Foldy, The electromagnetic properties of Dirac particles, Physical Review 83, pp. 688–
693, 1951. L.L. Foldy, The electron–neutron interaction, Physical Review 83, pp. 693–696,
1951. L.L. Foldy, Electron–neutron interaction, Review of Modern Physics 30, pp. 471–481,
1952. Cited on page 758.
731 John P. Costell a & Bruce H.J. Mc Kell ar, The Foldy–Wouthuysen transforma-
tion, American Journal of Physics 63, pp. 1119–1121, 1995. Cited on page 750.
732 H. Euler & B. Kockel, Naturwissenschaften 23, p. 246, 1935, H. Euler, Annalen der
Copyright © Christoph Schiller November 1997–May 2006

Physik 26, p. 398, 1936, W. Heisenberg & H. Euler, Folgerung aus der Diracschen The-
orie des Electrons, Zeitschrift für Physik 98, pp. 714–722, 1936. Cited on page 759.
733 The g-factors for composite nuclei are explained by ... See also Hans Dehmelt, Is the
electron a composite particle?, Hyperfine Interactions 81, pp. 1–3, 1993. Cited on page 761.
734 J.P. Woerdman, G. Nienhuis & I. Kuščer, Is it possible to rotate an atom?, Optics
Communications 93, pp. 135–144, 1992. We are talking about atoms rotating around their
centre of mass; atoms can of course rotate around other bodies, as discussed by M.P. Sil-
verman, Circular birefringence of an atom in uniform rotation: the classical perspective,
American Journal of Physics 58, pp. 310–317, 1990. Cited on page 751.
735 For the atomic case, see P.L. Gould, G.A. Ruff & D.E. Pritchard, Diffraction of
atoms by light: the near resonant Kapitza–Dirac effect, Physical Review Letters 56, pp. 827–

Dvipsbugw
bibliography 769

830, 1986. Many ealry experimental attempts, in particular those performed in the 1980s,
for the diffraction of electrons by light were controversial; most showed only the deflec-
tion of electrons, as explained by H. Batel aan, Contemporary Physics 41, p. 369, 2000.
Later on, he and his group performed the newest and most spectacular experiment, demon-
strating real diffraction, including interference effects; it is described in D.L. Freimund,
K. Afl atooni & H. Batel aan, Observation of the Kapitza–Dirac effect, Nature 413,
pp. 142–143, 2001. Cited on page 762.
736 J. Schmiedmayer, M.S. Chapman, C.R. Ekstrom, T.D. Hammond, S. We-
hinger & D.E. Pritchard, Index of refraction of various gases for sodium matter Dvipsbugw
waves, Physical Review Letters 74, p. 1043-1046, 1995. Cited on page 752.
737 The nearest to an image of a hydrogen atom is found in A. Yazdani, Watching an atom
tunnel, Nature 409, pp. 471–472, 2001. The experiments on Bose–Einstein condensates are
also candidates for images of hydrogen atoms. The company Hitachi made a fool of itself in
1992 by claiming in a press release that its newest electron microscope could image hydrogen
atoms. Cited on page 751.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

738 An introduction is given by P. Pinkse & G. R empe, Wie fängt man ein Atom mit einem
Photon?, Physikalische Blätter 56, pp. 49–51, 2000. Cited on page 752.
739 A.M. Wolsky, Kinetic energy, size, and the uncertainty principle, American Journal of
Physics 42, pp. 760–763, 1974. Cited on page 752.
740 Wolf gang Pauli, Exclusion principle and quantum mechanics, Nobel lecture, 13 Decem-
ber 1946, in Nobel Lectures, Physics, Volume 3, 1942-1962, Elsevier, 1964. Cited on page 763.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 C h a p t e r VI

PER MUTATION OF PARTIC LES

Dvipsbugw

W
hy are we able to distinguish twins from each other? Why can we distinguish
hat looks alike, such as a copy from an original? Most of us are convinced that
henever we compare an original with a copy, we can find a difference. This con-
viction turns out to be correct, even though it is a quantum effect that is in contrast with
classical physics.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Indeed, quantum theory has a lot to say about copies and their differences. Think about
any method that allows to distinguish objects: you will find that it runs into trouble for
Challenge 1242 n point-like particles. Therefore in the quantum domain something must change about our
ability to distinguish particles and objects. Let us explore the issue.

22. are particles like gloves?

Some usually forgotten properties of objects are highlighted by studying a pretty combin-
Ref. 741 atorial puzzle: the glove problem. It asks:

How many surgical gloves (for the right hand) are necessary if m doctors
need to operate w patients in a hygienical way, so that nobody gets in contact
with the body fluids of anybody else?

The same problem also appears in other settings. For example, it also applies to condoms,
men and women – and is then (officially) called the condom problem – or to computers,
interfaces and computer viruses. In fact, the term ‘condom problem’ is the term used in
Copyright © Christoph Schiller November 1997–May 2006

the books that discuss it. Obviously, the optimal number of gloves is not the product wm.
In fact, the problem has three subcases.
Challenge 1243 e — The simple case m = w = 2 already provides the most important ideas needed. Are
you able to find the optimal solution and procedure?
— In the case w = 1 and m odd or the case m = 1 and w odd, the solution is (m + 1)~2
Challenge 1244 e gloves. This is the optimal solution, as you can easily check yourself.
Ref. 742 — A solution with a simple procedure for all other cases is given by 2w~3 + m~2 gloves,
where x means the smallest integer greater than or equal to x. For example, for two
men and three women this gives only three gloves. (However, this formula does not
Challenge 1245 e always give the optimal solution; better values exist in certain subcases.)

Dvipsbugw
 772 vi permu tation of particles • 22. are particles like gloves?

Two basic properties of gloves determine the solution to the puzzle. First, gloves have
two sides, an interior and an exterior one. Secondly, gloves can be distinguished from
each other. Do these two properties also apply to particles? We will discuss the issue of
Page 1051 double-sidedness in the third part of the mountain ascent. In fact, the question whether
particles can be turned inside out will be of great importance for their description and
their motion. In the present chapter we concentrate on the second issue, namely whether
objects and particles can always be distinguished. We will find that elementary particles
do not behave like gloves in these but in an even more surprising manner. (In fact, they
do behave like gloves in the sense that one can distinguish right-handed from left-handed Dvipsbugw
ones.)
In everyday life, distinction of objects can be achieved in two ways. We are able to
distinguish objects – or people – from each other because they differ in their intrinsic
properties, such as their mass, colour, size or shape. In addition, we are also able to distin-
guish objects if they have the same intrinsic properties. Any game of billiard suggests that
by following the path of each ball, we can distinguish it from the others. In short, objects

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

with identical properties can also be distinguished using their state.
The state of a billiard ball is given by its position and momentum. In the case of billiard
balls, the state allows distinction because the measurement error for the position of the
ball is much smaller than the size of the ball itself. However, in the microscopic domain
this is not the case. First of all, atoms or other microscopic particles of the same type have
the same intrinsic properties. To distinguish them in collisions, we would need to keep
track of their motion. But we have no chance to achieve this. Already in the nineteenth
century it was shown experimentally that even nature itself is not able to do this. This
result was discovered studying systems which incorporate a large number of colliding
atoms of the same type: gases.
Page 238 The calculation of the entropy of a simple gas, made of N simple particles of mass m
moving in a volume V , gives

mkT 3~2 N 3
S = kln V ( )  + kN + klnα (542)
2πħ 2 2

where k is the Boltzmann constant, T the temperature and ln the natural logarithm. In
this formula, the pure number α is equal to 1 if the particles are distinguishable, and equal
to 1~N! if they are not. Measuring the entropy thus allows us to determine α and therefore
whether particles are distinguishable. It turns out that only the second case describes
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1246 e nature. This can be checked with a simple test: only in the second case does the entropy
of two volumes of identical gas add up.* The result, often called Gibbs’ paradox,** thus
Ref. 743 proves that the microscopic components of matter are indistinguishable: in a system of

Challenge 1247 ny * Indeed, the entropy values observed by experiment are given by the so-called Sackur–Tetrode formula

S = kNln  ( )  + kN
V mkT 3~2 5
(543)
N 2πħ 2 2
which follows when α = 1~N! is inserted above.
** Josiah Willard Gibbs (1839–1903), US-American physicist who was, with Maxwell and Planck, one of the
three founders of statistical mechanics and thermodynamics; he introduced the concepts of ensemble and
of phase.

Dvipsbugw
are particles like gloves? 773

F I G U R E 311 Identical objects with crossing paths Dvipsbugw

microscopic particles, there is no way to say which particle is which. Indistinguishability

is an experimental property of nature.*
The properties of matter would be completely different without indistinguishability.
For example, we will discover that without it, knifes and swords would not cut. In addition,
the soil would not carry us; we would fall right through it. To illuminate the issue in more

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

detail, we explore the next question.

Why does indistinguishability appear in nature?

Take two microscopic particles with the same mass, the same composition and the same
shape, such as two atoms. Imagine that their paths cross, and that they approach each
other to small distances at the crossing, as shown in Figure 311. In a gas, both the col-
lision of atoms or a near miss are examples. Experiments show that at small distances
it is impossible to say whether the two particles have switched roles or not. This is the
main reason that makes it impossible in a gas to follow particles moving around and to
determine which particle is which. This impossibility is a consequence of the quantum
of action.
For a path that brings two approaching particles very close to each other, a role switch
requires only a small amount of change, i.e. only a small (physical) action. However, we
know that there is a smallest observable action in nature. Keeping track of each particles
at small distances would require action values smaller than the minimal action observed
in nature. The existence of a smallest action makes it thus impossible to keep track of
microscopic particles when they come too near to each other. Any description of several
particles must thus take into account that after a close encounter, it is impossible to say
which is which.
Copyright © Christoph Schiller November 1997–May 2006

In short, indistinguishability is a consequence of the existence of a minimal action in

nature. This result leads straight away to the next question:

Can particles be counted?

In everyday life, objects can be counted because they can be distinguished. Since quantum
particles cannot be distinguished, we need some care in determining how to count them.
The first step is the definition of what is meant by a situation without any particle at all.

* When radioactivity was discovered, people thought that it contradicted the indistinguishability of atoms,
as decay seems to single out certain atoms compared to others. But quantum theory then showed that this
is not the case and that atoms do remain indistinguishable.

Dvipsbugw
 774 vi permu tation of particles • 22. are particles like gloves?

This seems an easy thing to do, but later on we will encounter situations where already this
step runs into difficulties. In any case, the first step is thus the specification of the vacuum.
Any counting method requires that situations without particles be clearly separated from
situations with particles.
The second step is the specification of an observable useful for determining particle
number. The easiest way is to chose one of those quantum numbers which add up under
composition, such as electric charge.* Counting is then performed by measuring the total
charge and dividing by the unit charge.
This method has several advantages. First of all, it is not important whether particles Dvipsbugw
are distinguishable or not; it works in all cases. Secondly, virtual particles are not coun-
ted. This is a welcome state of affairs, as we will see, because for virtual particles, i.e. for
particles for which E 2 x p2 c 2 + m 2 c 4 , there is no way to define a particle number anyway.
The side effect of the counting method is that antiparticles count negatively! Also this
consequence is a result of the quantum of action. We saw above that the quantum of action
implies that even in vacuum, particle–antiparticle pairs are observed at sufficiently high

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

energies. As a result, an antiparticle must count as minus one particle. In other words,
any way of counting particles can produce an error due to this effect. In everyday life this
limitation plays no role, as there is no antimatter around us. The issue does play a role
at higher energies, however. It turns out that there is no general way to count the exact
number of particles and antiparticles separately; only the sum can be defined. In short,
quantum theory shows that particle counting is never perfect.
In summary, nature does provide a way to count particles even if they cannot be distin-
guished, though only for everyday, low energy conditions; due to the quantum of action,
antiparticles count negatively, and provide a limit to the counting of particles at high en-
ergies.

What is permutation symmetry?

Since particles are countable but indistinguishable, there exists a symmetry of nature for
systems composed of several identical particles. Permutation symmetry, also called ex-
change symmetry, is the property of nature that observations are unchanged under ex-
change of identical particles. Together with space-time symmetry, gauge symmetry and
the not yet encountered renormalization symmetry, permutation symmetry forms one
of the four pillars of quantum theory. Permutation symmetry is a property of composed
systems, i.e. of systems made of many (identical) subsystems. Only for such systems does
Copyright © Christoph Schiller November 1997–May 2006

indistinguishability play a role.

In other words, ‘indistinguishable’ is not the same as ‘identical’. Two particles are not
the same; they are more like copies of each other. On the other hand, everyday life ex-
perience shows us that two copies can always be distinguished under close inspection,
so that the term is not fully appropriate either. In the microscopic domain, particles are
countable and completely indistinguishable.** Particles are perfect copies of each other.

* In everyday life, the weight or mass is commonly used as observable. However, it cannot be used in the
quantum domain, except for simple cases. Can you give at least two reasons, one from special relativity and
Challenge 1248 n one from general relativity?
** The word ‘indistinguishable’ is so long that many physicists sloppily speak of ‘identical’ particles never-
theless. Take care.

Dvipsbugw
 are particles like gloves? 775

We will discover shortly that permutation is partial rotation. Permutation symmetry

Challenge 1249 e thus is a symmetry under partial rotations. Can you find out why?

Indistinguishability and symmetry

The indistinguishability of particles leads to important conclusions about the description
of their state of motion. This happens because it is impossible to formulate a description
of motion that includes indistinguishability right from the start. Are you able to confirm
Challenge 1250 n this? As a consequence we describe a n-particle state with a state Ψ1... i ... j...n which as- Dvipsbugw
sumes that distinction is possible, as expressed by the ordered indices in the notation,
and we introduce the indistinguishability afterwards. Indistinguishability means that the
exchange of any two particles results in the same physical system.* Now, two quantum
states have the same physical properties if they differ at most by a phase factor; indistin-
guishability thus requires
Ψ1... i ... j...n = e i α Ψ1... j...i ...n (544)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

for some unknown angle α. Applying this expression twice, by exchanging the same
couple of indices again, allows us to conclude that e2i α = 1. This implies that

Ψ1... i ... j...n =  Ψ1... j...i ...n , (545)

in other words, a wave function is either symmetric or antisymmetric under exchange of

indices. Quantum theory thus predicts that particles are indistinguishable in one of two
distinct ways.** Particles corresponding to symmetric wave functions are called bosons,
those corresponding to antisymmetric wave functions are called fermions.***
Experiments show that the behaviour depends on the type of particle. Photons are bo-
sons. On the other hand, electrons, protons and neutrons are found to be fermions. Also
about half of the atoms are found to behave as bosons (at moderate energies). In fact, a
composite of an even number of fermions (at moderate energies) – or of any number of
bosons (at any energy) – turns out to be a boson; a composite of an odd number of fer-
mions is (always) a fermion. For example, almost all of the known molecules are bosons
(electronically speaking). Fermionic molecules are rather special and even have a special
name in chemistry; they are called radicals and are known for their eagerness to react
and to form normal bosonic molecules. Inside the human body, too many radicals can Copyright © Christoph Schiller November 1997–May 2006

* We therefore have the same situation that we encountered already several times: an overspecification of
the mathematical description, here the explicit ordering of the indices, implies a symmetry of this description,
which in our case is a symmetry under exchange of indices, i.e., under exchange of particles.
** This conclusion applies to three-dimensional space only. In two dimensions there are more possibilities.
*** The term ‘fermion’ is derived from the name of the Italian physicist and Nobel Prize winner Enrico Fermi
(b. 1901 Roma, d. 1954 Chicago) famous for his all-encompassing genius in theoretical and experimental
physics. He mainly worked on nuclear and elementary particle physics, on spin and on statistics. For his
experimental work he was called ‘quantum engineer’. He is also famous for his lectures, which are still pub-
lished in his own hand-writing, and his brilliant approach to physical problems. Nevertheless, his highly
deserved Nobel Prize was one of the few cases in which the prize was given for a discovery which turned
out to be incorrect.
‘Bosons’ are named after the Indian physicist Satyenra Nath Bose (b. 1894 Calcutta, d. 1974 Calcutta)
Ref. 744 who first described the statistical properties of photons. The work was later expanded by Albert Einstein.

Dvipsbugw
 776 vi permu tation of particles • 22. are particles like gloves?

detectors
mirrors
beam
source splitter
possible
two identical light
photons paths

F I G U R E 312 Two photons and interference

Dvipsbugw

have adverse effects on health; it is well known that vitamin C is important because it is
effective in reducing the number of radicals.
To which class of particles do mountains, trees, people and all other macroscopic ob-
Challenge 1251 ny jects belong?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The behaviour of photons
A simple experiment allows to determine the behaviour of photons. Take a source that
emits two photons of identical frequency and polarization at the same time, as shown
in Figure 312. In the laboratory, such a source can be realized with a down-converter, a
material that converts a photon of frequency 2ω into two photons of frequency ω. Both
photons, after having travelled exactly the same distance, are made to enter the two sides
of a beam splitter (for example, a half-silvered mirror). At the two exits of the beam splitter
Ref. 745 are two detectors. Experiments show that both photons are always detected together on
the same side, and never separately on opposite sides. This result shows that photons are
bosons. Fermions behave in exactly the opposite way; two fermions are always detected
separately on opposite sides, never together on the same side.

The energy dependence of permutation symmetry

If experiments force us to conclude that nobody, not even nature, can distinguish any
two particles of the same type, we deduce that they do not form two separate entities, but
some sort of unity. Our naive, classical sense of particle as a separate entity from the rest
of the world is thus an incorrect description of the phenomenon of ‘particle’. Indeed, no
experiment can track particles with identical intrinsic properties in such a way that they
Copyright © Christoph Schiller November 1997–May 2006

can be distinguished with certainty. This impossibility has been checked experimentally
with all elementary particles, with nuclei, with atoms and with numerous molecules.
How does this fit with every-
day life, i.e. with classical physics?
Photons do not worry us much here.
Let us focus the discussion on mat-
ter particles. We know to be able to
distinguish electrons by pointing to F I G U R E 313 Particles as localized excitations
the wire in which they flow, and we
can distinguish our fridge from that of our neighbour. While the quantum of action
makes distinction impossible, everyday life allows it. The simplest explanation is to ima-

Dvipsbugw
 are particles like gloves? 777

gine a microscopic particle, especially an elementary one, as a bulge, i.e. as a localized

excitation of the vacuum. Figure 313 shows two such bulges representing two particles.
It is evident that if particles are too near to each other, it makes no sense to distinguish
them; we cannot say any more which is which.
The bulge image shows that either for large distances or for high potential walls sep-
arating them, distinction of identical particles does become possible. In such situations,
measurements allowing to track them independently do exist. In other words, we can
specify a limit energy at which permutation symmetry of objects or particles separated
by a distance d becomes important. It is given by Dvipsbugw

E=
cħ
. (546)
d
Challenge 1252 ny Are you able to confirm the expression? For example, at everyday temperatures we can
distinguish atoms inside a solid from each other, since the energy so calculated is much

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

higher than the thermal energy of atoms. To have fun, you might want to determine at
what energy two truly identical human twins become indistinguishable. Estimating at
what energies the statistical character of trees or fridges will become apparent is then
Challenge 1253 ny straightforward.
The bulge image of particles thus purveys the idea that distinguishability exists for
objects in everyday life but not for particles in the microscopic domain. To sum up, in
daily life we are able to distinguish objects and thus people for two reasons: because they
are made of many parts, and because we live in a low energy environment.
The energy issue immediately adds a new aspect to the discussion. How can we de-
scribe fermions and bosons in the presence of virtual particles and of antiparticles?

Indistinguishability in quantum field theory

Quantum field theory, as we will see shortly, simply puts the bulge idea of Figure 313 into
mathematical language. A situation with no bulge is called vacuum state. Quantum field
theory describes all particles of a given type as excitations of a single fundamental field.
Particles are indistinguishable because each particle is an excitation of the same basic
substrate and each excitation has the same properties. A situation with one particle is
then described by a vacuum state acted upon by a creation operator. Adding a second
particle is described by adding a second creation operator, and subtracting a particle by
adding a annihilation operator; the latter turns out to be the adjunct of the former.
Copyright © Christoph Schiller November 1997–May 2006

Quantum field theory then studies how these operators must behave to describe ob-
servations.* It arrives at the following conclusions:
— Fields with half-integer spin are fermions and imply (local) anticommutation.
* Whenever the relation
[b, b ] = bb − b b = 1
† † †
(547)
holds between the creation operator b † and the annihilation operator b, the operators describe a boson. If
the operators for particle creation and annihilation anticommute
˜d, d †  = dd † + d † d = 1 (548)
they describe a fermion. The so defined bracket is called the anticommutator bracket.

Dvipsbugw
 778 vi permu tation of particles • 22. are particles like gloves?

— Fields with integer spin are bosons and imply (local) commutation.
— For all fields at spacelike separations, the commutator – respectively anticommutator
– vanishes.
— Antiparticles of fermions are fermions, and antiparticles of bosons are bosons.
— Virtual particles behave like their real counterparts.
These connections are at the basis of quantum field theory. They describe how particles
are identical. But why are they? Why are all electrons identical? Quantum field theory
describes electrons as identical excitations of the vacuum, and as such as identical by Dvipsbugw
construction. Of course, this answer is only partially satisfying. We will find a better one
only in the third part of our mountain ascent.

How accurately is permutation symmetry verified?

A simple but effective experiment testing the fermion behaviour of electrons was carried
Ref. 746 out by Ramberg and Snow. They sent an electric current of 30 A through a copper wire

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

for one month and looked for X-ray emission. They did not find any. They concluded that
electrons are always in an antisymmetric state, with a symmetric component of less than

2 ë 10−26 (549)

of the total state. Electrons are always in an antisymmetric state: thus they are fermions.
The reasoning behind this elegant experiment is the following. If electrons would not
always be fermions, every now and then an electron could fall into the lowest energy
level of a copper atom, leading to X-ray emission. The lack of such X-rays implies that
electrons are fermions to a very high accuracy. X-rays could be emitted only if they were
bosons, at least part of the time. Indeed, two electrons, being fermions, cannot be in the
same state: this restrition is called the Pauli exclusion principle. It applies to all fermions
and is our next topic.

Copies, clones and gloves

Can classical systems be indistinguishable? They can: large molecules are examples –
provided they are made of exactly the same isotopes. Can large classical systems, made
of a mole or more particles be indistinguishable? This simple question effectively asks
whether a perfect copy, or (physical) clone of a system is possible.
Copyright © Christoph Schiller November 1997–May 2006

It could be argued that any factory for mass-produced goods, such as one producing
shirt buttons or paper clips, shows that copies are possible. But the appearance is deceiv-
ing. On a microscope there is usually some difference. Is this always the case? In 1982, the
Dutch physicist Dennis Dieks and independently, the US-American physicists Wootters
Ref. 748 and Zurek, published simple proofs that quantum systems cannot be copied. This is the
famous no-cloning theorem.
A copying machine is a machine that takes an original, reads out its properties and
produces a copy, leaving the original unchanged. This seems definition seems straight-
forward. However, we know that if we extract information from an original, we have to
interact with it. As a result, the system will change at least by the quantum of action. We
thus expect that due to quantum theory, copies and originals can never be identical.

Dvipsbugw
 are particles like gloves? 779

Quantum theory proves this in detail. A copying machine is described by an operator

that maps the state of an original system to the state of the copy. In other words, a copying
machine is linear. This linearity leads to a problem. Simply stated, if a copying machine
were able to copy originals either in state SAe or in state SBe, it could not decide what to do
if the state of the original were SAe+SBe. On one hand, the copy should be SAe+SBe; on the
other hand, the linearity of the copier forbids this. Indeed, a copier is a device described
by an operator U that changes the starting state Ssec of the copy in the following way:
— If the original is in state SAe, a copier acts as Dvipsbugw

USAeSsec = SAeSAec . (550)

— If the original is in state SBe, a copier acts as

USBeSsec = SBeSBec . (551)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

As a result of these two requirements, an original in the state SA + Be is treated by the
copier as
USA + BeSsec = SAeSAec + SBeSBec . (552)

This is in contrast to what we want, which would be

U wanted SA + BeSsec = (SAe + SBe)(SAec + SBec ) . (553)

In other words, a copy machine cannot copy a state completely.* This is the no-cloning
theorem.
The impossibility of copying is implicit in quantum theory. If we were able to clone
systems, we could to measure a variable of a system and a second variable on its copy. We
would be thus able to beat the indeterminacy relation. This is impossible. Copies are and
always must be imperfect.
Other researchers then explored how near to perfection a copy can be, especially in the
Ref. 749 case of classical systems. To make a long story short, these investigations show that also
the copying or cloning of macroscopic systems is impossible. In simple words, copying
machines do not exist. Copies can always be distinguished from originals if observations
are made with sufficient care. In particular, this is the case for biological clones; biological
Copyright © Christoph Schiller November 1997–May 2006

clones are identical twins born following separate pregnancies. They differ in their finger
prints, iris scans, physical and emotional memories, brain structures, and in many other
Challenge 1254 n aspects. (Can you specify a few more?) In short, biological clones, like identical twins, are
not copies of each other.
The lack of quantum mechanical copying machines is disappointing. Such machines,
or teleportation devices, could be fed with two different inputs, such as a lion and a goat,

* The no-cloning theorem puts severe limitations on quantum computers, as computations often need cop-
ies of intermediate results. It also shows that faster-than-light communication is impossible in EPR experi-
ments. In compensation, quantum cryptography becomes possible – at least in the laboratory. Indeed, the
no-cloning theorem shows that nobody can copy a quantum message without being noticed. The specific
ways to use this result in cryptography are the 1984 Bennett–Brassard protocol and the 1991 Ekert protocol.

Dvipsbugw
 780 vi permu tation of particles • rotations and statistics

and produce a superposition: a chimaera. Quantum theory shows that all these imaginary
beings cannot be realized.
In summary, everyday life objects such as photocopies, billiard balls or twins are always
distinguishable. There are two reasons: first, quantum effects play no role in everyday life,
so that there is no danger of unobservable exchange; secondly, perfect clones of classical
systems do not exist anyway, so that there always are tiny differences between any two
objects, even if they look identical at first sight. Gloves can always be distinguished.
Dvipsbugw
23. rotations and statistics – visualizing spin
We saw above that spin is the observation that matter rays, like light rays, can be polarized.
Spin thus describes how particles behave under rotations, and it proves that particles
are not simple spheres shrunk to points. We also saw that spin describes a fundamental
difference between quantum systems and gloves: spin specifies the indistinguishability of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

quantum systems. Let us explore this connection in more detail.
Ref. 750 The general background for the appearance of spin was clarified by Eugene Wigner in
1939.* He started by recapitulating that any quantum mechanical particle, if elementary,
must behave like an irreducible representation of the set of all viewpoint changes. This
set forms the symmetry group of flat space-time, the so-called inhomogeneous Lorentz
group. We have seen in the chapter on classical mechanics how this connection between
Page 192 elementarity and irreducibility arises. To be of physical relevance for quantum theory,
representations have to be unitary. The full list of irreducible unitary representations of
viewpoint changes thus provides the range of possibilities for any particle that wants to
be elementary.
Cataloguing the possibilities, one finds first of all that every elementary particle is de-
scribed by four-momentum – no news so far – and by an internal angular momentum, the
spin. Four-momentum results from the translation symmetry of nature, and spin from its
rotation symmetry. The momentum value describes how a particle behaves under trans-
lation, i.e. under position and time shift of viewpoints. The spin value describes how an
object behaves under rotations in three dimensions, i.e. under orientation change of view-
points.** As is well known, the magnitude of four-momentum is an invariant property,
given by the mass, whereas its orientation in space-time is free. Similarly, the magnitude
of spin is an invariant property, and its orientation has various possibilities with respect
to the direction of motion. In particular, the spin of massive particles behaves differently
Copyright © Christoph Schiller November 1997–May 2006

from that of massless particles.

» the inhomogeneous Lorentz group implies that the invariant
For massive particles,
magnitude of spin is J(J + 1) ħ, often simply written J. Since the value specifies the
magnitude of the angular momentum, it gives the representation under rotations of a
given particle type. The spin magnitude J can be any multiple of 1~2, i.e. it can take the

* Eugene Wigner (b. 1902 Budapest, d. 1995 Princeton), Hungarian–US-American theoretical physicist, re-
ceived the Nobel Prize for physics in 1993. He wrote over 500 papers, many about symmetry in physics. He
was also famous for being the most polite physicist in the world.
** The group of physical rotations is also called SO(3), since mathematically it is described by the group of
Special Orthogonal 3 by 3 matrices.

Dvipsbugw
 rotations and statistics – visualizing spin 781

Dvipsbugw

F I G U R E 314 An argument showing why rotations by

4π are equivalent to no rotation at all

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

values 0, 1~2, 1, 3~2, 2, 5~2, etc. Experiments show that electrons, protons and neutrons
have spin 1~2, the W and Z particles spin 1 and helium atoms spin 0. In addition, the
representation of spin J is 2J + 1 dimensional, meaning that the spatial orientation of
the spin has 2J + 1 possible values. For electrons there are thus two possibilities; they are
usually called ‘up’ and ‘down’.
Spin thus only takes discrete values. This is in contrast with linear momentum, whose
representations are infinite dimensional and whose possible values form a continuous
range.
Also massless particles are characterized by the value of their spin. It can take the
same values as in the massive case. For example, photons and gluons have spin 1. For
massless particles, the representations are one-dimensional, so that massless particles are
completely described by their helicity, defined as the projection of the spin onto the direc-
tion of motion. Massless particles can have positive or negative helicity, often also called
right-handed and left-handed. There is no other freedom for the orientation of spin in
the massless case.
The symmetry investigations lead to the classification of particles by their mass, their
momentum and their spin. To complete the list, the remaining symmetries must be in-
cluded. These are motion inversion parity, spatial parity and charge inversion parity. Since
these symmetries are parities, each elementary particle has to be described by three addi-
tional numbers, called T, P and C, each of which can take values of either +1 or −1. Being
Copyright © Christoph Schiller November 1997–May 2006

parities, they must be multiplied to yield the value for a composed system.
Page 1174 A list of the values observed for all elementary particles in nature is given in
Appendix C. Spin and parities together are called quantum numbers. As we will discover
later on, additional interaction symmetries will lead to additional quantum numbers. But
let us return to spin.
The main result is that spin 1/2 is a possibility in nature, even though it does not appear
in everyday life. Spin 1/2 means that only a rotation of 720 degrees is equivalent to one of
0 degrees, while one of 360 degrees is not, as explained in Table 57. The mathematician
Page 49 Hermann Weyl used a simple image explaining this connection.
Take two cones, touching each other at their tips as well as along a line. Hold one cone
and roll the other around it, as shown in Figure 314. When the rolling cone, after a full

Dvipsbugw
 782 vi permu tation of particles • rotations and statistics

TA B L E 57 Particle spin as representation of the rotation group

Spin System Massive examples Massless examples

[ħ] unchanged after elementary composite elementary
rotation by
a,b b
0 any angle none mesons, nuclei, none
atoms
1/2 2 turns e, µ, τ, q, nuclei, atoms, none, as neutrinos have a tiny mass
νe , ν µ , ντ molecules Dvipsbugw
1 1 turn W, Z mesons, nuclei, д, γ
atoms, molecules,
toasters
b b
3/2 2/3 turn none baryons, nuclei, none
atoms

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

c
2 1/2 turn none nuclei ‘graviton’
5/2 2/5 turn none nuclei none
d
3 1/3 turn none nuclei none
etc.d etc.d etc.d etc.d etc.d

a. Whether the Higgs particle is elementary or not is still unknown.

b. Supersymmetry predicts particles in these and other boxes.
c. The graviton has not yet been observed.
d. Nuclei exist with spins values up to at least 101~2 and 51 (in units of ħ). Ref. 752

turn around the other cone, has come back to the original position, it has rotated by some
angle. If the cones are wide, the rotation angle is small. If the cones are very thin, almost
like needles, the moving cone has rotated by almost 720 degrees. A rotation of 720 degrees
is thus similar to one by 0 degrees. If we imagine the cone angle to vary continuously, this
Copyright © Christoph Schiller November 1997–May 2006

visualization also shows that a 720 degree rotation can be continuously deformed into a
0 degree one, whereas a 360 degree rotation cannot.
To sum up, the list of possible representations thus shows that rotations require the
existence of spin. But why then do experiments show that all fermions have half-integer
spin and that all bosons have integer spin? Why do electrons obey the Pauli exclusion
principle? At first, it is not clear what the spin has to do with the statistical properties of
a particle.
In fact, there are several ways to show that rotations and statistics are connected. His-
torically, the first proof used the details of quantum field theory and was so complicated
Ref. 751 that its essential ingredients were hidden. It took quite some years to convince everybody
that a simple observation about belts was the central part of the proof.

Dvipsbugw
 rotations and statistics – visualizing spin 783

F I G U R E 315 A belt visualizing two spin 1/2 F I G U R E 316 Another model for two spin
particles 1/2 particles

Dvipsbugw
α=0 atpi α = 4π

F I G U R E 317 The human arm as spin 1/2 model

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The belt trick
The well-known belt trick (also called scissor trick) was often used by Dirac to explain
Ref. 753 the features of spin 1/2. Taking Figure 313, which models particles as indistinguishable
excitations, it is not difficult to imagine a sort of sheet connecting them, similar to a
belt connecting the two parts of the buckle, as shown in Figure 315. If one end of the
belt is rotated by 2π along any axis, a twist is inserted into the belt. If the end is rotated
for another 2π, bringing the total to 4π, the ensuing double twist can easily be undone
without moving or rotating the ends. You need to experience this yourself in order to
Challenge 1255 e believe it.
In addition, if you take the two ends and simply swap positions, a twist is introduced
into the belt. Again, a second swap will undo the twist.
In other words, if we take each end to represent a particle and a twist to mean a factor
−1, the belt exactly describes the phase behaviour of spin 1~2 wave functions under ex-
change and under rotations. In particular, we see that spin and exchange behaviour are
related.
The human body has such a belt built in: the arm. Just take your hand, put an object
on it for clarity such as a cup, and turn the hand and object by 2π by twisting the arm.
After a second rotation the whole system will be untangled again.
The trick is even more impressive when many arms are used. You can put your two
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1256 e hands (if you chose the correct starting position) under the cup or you can take a friend
or two who each keep a hand attached to the cup. The feat can still be performed: the
Challenge 1257 e whole system untangles after two full turns.
Another demonstration is to connect two buckles with many bands or threads, like in
Figure 316. Both a rotation by 2π of one end or an exchange of the two ends produces
quite a tangle, even if one takes paths that ‘in between’ the bands; nevertheless, in both
cases a second rotation leads back to the original situation.
There is still another way to show the connection between rotation and exchange. Just
glue any number of threads or bands, say half a metre long, to an asymmetric object.
Like the arm of a human being, the bands are supposed to go to infinity and be attached
there. If any of the objects, which represent the particles, is rotated by 2π, twists appear

Dvipsbugw
 784 vi permu tation of particles • rotations and statistics

rotating the buckle

either by 4π

or simply rearranging
the bands gives the Dvipsbugw
other situation

F I G U R E 318 The extended belt trick, modelling a

spin 1/2 particle: independently of the number of
bands attached, the two situations can be

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

transformed into each other, either by rotating the
central object by 4π or by keeping the central object
fixed and moving the bands around it

in its strings. If the object is rotated by an additional turn, to a total of 4π, as shown in
Figure 318, all twists and tangles can be made to disappear, without moving or turning
the object. You really have to experience this in order to believe it. And the trick really
works with any number of bands glued to the object.
Even more astonishing is the exchange part of the experiment. Take two particles of
the type shown on the left side of Figure 318. If you exchange the positions of two such
spin 1/2 particles, always keeping the ends at infinity fixed, a tangled mess is created. But
incredibly, if you exchange the objects a second time, everything untangles neatly, inde-
Challenge 1258 e pendently of the number of attached strings. You might want to test yourself that the
behaviour is also valid with sets of three or more particles.
All these observations together form the spin statistics theorem for spin 1/2 particles:
spin and exchange behaviour are related. Indeed, these almost ‘experimental’ arguments
Ref. 754 can be put into exact mathematical language by studying the behaviour of the configura-
tion space of particles. These investigations result in the following statements:

U Objects of spin 1/2 are fermions.*

Copyright © Christoph Schiller November 1997–May 2006

U Exchange and rotation of spin 1/2 particles are similar processes.

Note that all these arguments require three dimensions, because there are no tangles (or
knots) in fewer dimensions.** And indeed, spin exists only in three or more spatial di-
mensions.
* A mathematical observable behaving like a spin 1/2 particle is neither a vector nor a tensor, as you may
Challenge 1259 e want to check. An additional concept is necessary; such an observable is called a spinor. We will introduce
Page 757 it later on.
** Of course, knots and tangles do exist in higher dimensions. Instead of considering knotted one-
dimensional lines, one can consider knotted planes or knotted higher-dimensional hyperplanes. For ex-
ample, deformable planes can be knotted in four dimensions and deformable 3-spaces in five dimensions.

Dvipsbugw
 rotations and statistics – visualizing spin 785

Here is a challenge. A spin 1/2 object can be modelled with one belt attached to it. If
you want to model the spin behaviour with attached one-dimensional strings instead of
Challenge 1260 n attached bands, what is the minimum number required?

The Pauli exclusion principle and the hardness of matter

Why are we able to knock on a door? Why can stones not fly through tree trunks? How
does the mountain we are walking on carry us? In classical physics, we avoided this issue,
by taking solidity as a defining property of matter. But doing so, we cheated: we have seen Dvipsbugw
that matter consists mainly of empty space, so that we have to study the issue without
any sneaky way out. The answer is now clear: penetration is made impossible by Pauli’s
exclusion principle between the electrons inside atoms.
Why do electrons and other fermions obey the Pauli exclusion principle? The answer
Ref. 755 can be given with a beautifully simple argument. We know that exchanging two fermions
produces a minus sign. Imagine these two fermions being, as a classical physicist would

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

say, located at the same spot, or as a quantum physicist would say, in the same state. If that
could be possible, an exchange would change nothing in the system. But an exchange of
fermions must produce a minus sign for the total state. Both possibilities – no change at
all as well as a minus sign – cannot be realized at the same time. There is only one way
out: two fermions must avoid to ever be in the same state.
The exclusion principle is the reason that two pieces of matter in everyday life cannot
penetrate each other, but have to repel each other. For example, bells only work because
of the exclusion principle. Bells would not work if the colliding pieces that produce the
sound would interpenetrate. But in any example of two interpenetrating pieces the elec-
trons in the atoms would have to be in similar states. This is forbidden. For the same
reason we do not fall through the floor, even though gravity pulls us down, but remain
on the surface. In other words, the exclusion principle implies that matter cannot be com-
pressed indefinitely, as at a certain stage an effective Pauli pressure appears, so that a com-
pression limit ensues. For this reason for example, planets or neutron stars do not collapse
under their own gravity.
The exclusion principle also answers the question about how many angels can dance
on the top of a pin. (Note that angels must be made of fermions, as you might want to
Challenge 1261 n deduce from the information known about them.) Both theory and experiment confirm
Ref. 747 the answer already given by Thomas Aquinas in the Middle Ages: only one. The fermion
exclusion principle could also be called ‘angel exclusion principle’. To stay in the topic,
Copyright © Christoph Schiller November 1997–May 2006

the principle also shows that ghosts cannot be objects, as ghosts are supposed to be able
to traverse walls.
Whatever the interpretation, the exclusion principle keeps things in shape; without
it, there would be no three-dimensional objects. Only the exclusion principle keeps the
cloudy atoms of nature from merging, holding them apart. Shapes are a direct con-
sequence of the exclusion principle. As a result, when we knock on a table or on a door,
we show that both objects are made of fermions.
Since permutation properties and spin properties of fermions are so well described
by the belt model, we could be led to the conclusion that these properties might really
be consequence of such belt-like connections between particles and the outside world.
Maybe for some reason we only observe the belt buckles, not the belts themselves. In the

Dvipsbugw
 786 vi permu tation of particles • rotations and statistics

J=0 J = 1/2 J=1

F I G U R E 319 Some visualizations of spin Dvipsbugw

representations

third part of this walk we will discover whether this idea is correct.
So far, we have only considered spin 1~2 particles. We will not talk much about systems
with odd spin of higher value, such as 3~2 or 5~2. Such systems can be seen as being

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1262 ny composed of spin 1~2 entities. Can you confirm this?
We did not talk about lower spins than 1~2 either. A famous theorem states that a
positive spin value below 1~2 is impossible, because the largest angle that can be measured
in three dimensions is 4π. There is no way to measure a larger angle;* The quantum of
action makes this impossible. Thus there cannot be any spin value between 0 and 1~2.

Integer spin
Under rotations, integer spin particles behave differently from half-integer particles. In-
teger spin particles do not show the strange sign changes under rotations by 2π. In the
belt imagery, integer spin particles need no attached strings. The spin 0 particle obviously
corresponds to a sphere. Models for other spin values are shown in Figure 319. Exploring
their properties in the same way as above, we arrive at the so-called spin-statistics theorem:

U Exchange and rotation of objects are similar processes.

U Objects of half-integer spin are fermions. They obey the Pauli exclusion prin-
ciple.
U Objects of integer spin are bosons.

You might prove by yourself that this suffices to show the following:
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1263 ny

U Composites of bosons, as well as composites of an even number of fermions

(at low energy), are bosons; composites of an uneven number of fermions are
fermions.**

* This is possible in two dimensions though.

** This sentence implies that spin 1 and higher can also be achieved with tails; can you find such a represent-
Challenge 1264 ny ation?
Note that composite fermions can be bosons only up to that energy at which the composition breaks
down. Otherwise, by packing fermions into bosons, we could have fermions in the same state.

Dvipsbugw
 rotations and statistics – visualizing spin 787

These connections express basic characteristics of the three-dimensional world in which

we live.

Is spin a rotation about an axis?

The spin of a particle behaves experimentally like an intrinsic angular momentum, adds
up like angular momentum, is conserved as part of angular momentum, is described
like angular momentum and has a name synonymous with angular momentum. Despite
all this, for many decades a strange myth was spread in physics courses and textbooks Dvipsbugw
around the world, namely that spin 1~2 is not a rotation about an axis. The myth maintains
that any rotating object must have integer spin. Since half integer spin is not possible in
classical physics, it is argued that such spin is not due to rotation. It is time to finish with
this example of muddled thinking.
Electrons do have spin 1~2 and are charged. Electrons and all other charged particles
with spin 1~2 do have a magnetic moment.* A magnetic moment is expected for any rotat-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ing charge. In other words, spin 1~2 does behave like rotation. However, assuming that a
particle consists of a continuous charge distribution in rotational motion gives the wrong
value for the magnetic moment. In the early days of the twentieth century, when physicists
were still thinking in classical terms, they concluded that spin 1~2 particles thus cannot be
rotating. This myth has survived through many textbooks. The correct deduction is that
the assumption of continuous charge distribution is wrong. Indeed, charge is quantized;
nobody today expects that elementary charge is continuously spread over space, as that
would contradict its quantization.
Let us remember what rotation is. Both the belt trick for spin 1/2 as well as the integer
spin case remind us: a rotation of one body around another is a fraction or a multiple of an
exchange. What we call a rotating body in everyday life is a body continuously exchanging
the positions of its parts. Rotation and exchange are the same.
Above we found that spin is exchange behaviour. Since rotation is exchange and spin
is exchange, it follows that spin is rotation. Since we deduced, like Wigner, spin from
rotation invariance, this consequence is not a surprise.
The belt model of a spin 1/2 particle tells us that such a particle can rotate continuously
Challenge 1266 ny without any hindrance. In short, we are allowed to maintain that spin is rotation about an
axis, without any contradiction to observations, even for spin 1/2. The belt model helps
to keep two things in mind: we must assume that in the belt model only the buckles can
Ref. 756 be observed and do interact, not the belts, and we must assume that elementary charge
Copyright © Christoph Schiller November 1997–May 2006

is pointlike and cannot be distributed.**

Why is fencing with laser beams impossible?

When a sword is approaching dangerously, we can stop it with a second sword. Many old
films use such scenes. When a laser beam is approaching, it is impossible to fend it off

Challenge 1265 ny * This can easily be measured in a an experiment; however, not one of the Stern–Gerlach type. Why?
** Obviously, the detailed structure of the electron still remains unclear at this point. Any angular mo-
mentum S is given classically by S = Θω; however, neither the moment of inertia Θ, connected to the
rotation radius and electron mass, nor the angular velocity ω are known at this point. We have to wait quite
a while, until the third part of our adventure, to find out more.

Dvipsbugw
 788 vi permu tation of particles • rotations and statistics

Dvipsbugw

F I G U R E 320 Equivalence of exchange

and rotation in space-time

with a second beam, despite all science fiction films showing so. Banging two laser beams

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

against each other is impossible.
The above discussion shows why. The electrons in the swords are fermions and obey
the Pauli exclusion principle. Fermions make matter impenetrable. On the other hand,
photons are bosons. Bosons can be in the same state; they allow penetration. Matter is
impenetrable because at the fundamental level it is composed of fermions. Radiation is
composed of bosons. The distinction between fermions and bosons thus explains why
objects can be touched while images cannot. In the first part of our mountain ascent we
Page 74 started by noting this difference; now we know its origin.

Rotation requires antiparticles

The connection between rotation and antiparticles may be the most astonishing conclu-
sion from the experiments showing the existence of spin. So far, we have seen that rota-
tion requires the existence of spin, that spin appears when relativity is introduced into
Page 315 quantum theory, and that relativity requires antimatter. Taking these three statements
together, the conclusion of the title is not surprising any more. Interestingly, there is a
simple argument making the same point directly, without any help of quantum theory,
when the belt model is extended from space alone to full space-time.
To learn how to think in space-time, let us take a particle spin 1, i.e. a particle looking
like a detached belt buckle in three dimensions. When moving in a 2+1 dimensional space-
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1267 ny time, it is described by a ribbon. Playing around with ribbons in space-time, instead of
belts in space, provides many interesting conclusions. For example, Figure 320 shows that
wrapping a rubber ribbon around the fingers can show that a rotation of a body by 2π
in presence of a second one is the same as exchanging the positions of the two bodies.*
Both sides of the hand transform the same initial condition, at one border of the hand, to
the same final condition at the other border. We have thus successfully extended a known
result from space to space-time. Interestingly, we can also find a smooth sequence of steps
realizing this equivalence.

* Obviously, the next step would be to check the full spin 1/2 model of Figure 318 in four-dimensional
Challenge 1268 ny space-time. But this is not an easy task; there is no generally accepted solution yet.

Dvipsbugw
 rotations and statistics – visualizing spin 789

t t t t t

x x x x x
Dvipsbugw
F I G U R E 321 Belts in space-time: rotation and antiparticles

If you think that Figure 320 is not a satisfying explanation, you are right. A more com-
plete (yet equivalent) explanation is given by Figure 321. We assume that each particle is
described by a segment; in the figure, they lie horizontally. The leftmost diagram shows

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

two particles: one at rest and one being rotated by 2π. The deformation of the ribbons
shows that this process is equivalent to the exchange in position of two particles, which is
shown in the rightmost diagram. Again, one notes that the sequence that shows the equi-
valence between rotation and exchange requires the use of a particle–antiparticle pair.
Without antiparticles, the equivalence of rotation and exchange would not hold. Rota-
tion in space-time indeed requires antiparticles.

Limits and open questions of quantum statistics

The topic of statistics is an important research field in theoretical and experimental phys-
ics. In particular, researchers have searched and still are searching for generalizations of
the possible exchange behaviours of particles.
In two spatial dimensions, the result of an exchange of the wave function is not de-
scribed by a sign, but by a continuous phase. Two-dimensional objects behaving in this
way, called anyons because they can have ‘any’ spin, have experimental importance, since
in many experiments in solid state physics the set-up is effectively two-dimensional. The
fractional quantum Hall effect, perhaps the most interesting discovery of modern exper-
Page 849 imental physics, has pushed anyons onto the stage of modern research.
Other theorists generalized the concept of fermions in other ways, introducing par-
Ref. 757 afermions, parabosons, plektons and other hypothetical concepts. O.W. Greenberg has
spent most of his professional life on this issue. His conclusion is that in 3 + 1 space-time
Copyright © Christoph Schiller November 1997–May 2006

dimensions, only fermions and bosons exist. (Can you show that this implies that the
Challenge 1269 n ghosts appearing in scottish tales do not exist?)
From a different viewpoint, the above belt model invites to study the behaviour of
braids and knots. (In mathematics, a braid is a knot extending to infinity.) This fascinat-
ing part of mathematical physics has become important with the advent of string theory,
which states that particles, especially at high energies, are not point-like, but extended
entities.
Still another generalization of statistical behaviour at high energies is the concept of
quantum group, which we will encounter later on. In all of these cases, the quest is to un-
derstand what happens to permutation symmetry in a unified theory of nature. A glimpse
of the difficulties appears already above: how can Figures 313, 318 and 321 be reconciled

Dvipsbugw
790 vi permu tation of particles • rotations and statistics

and combined? We will settle this issue in the third part of our mountain ascent.

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
bibliography 791

Biblio graphy

741 See the paper by Martin Gardner, Science fiction puzzle tales, Clarkson Potter, 67,
pp. 104–105, 1981, or his book Aha! Insight, Scientific American & W.H. Freeman, 1978. The
rabbit story is from A. Hajnal & P. Lovász, An algorithm to prevent the propagation
of certain diseases at minimum cost, in Interfaces Between Computer Science and Opera-
tions Research, edited by J.K. Lenstra, A.H.G. R innooy Kan & P. Van Emde B oas,
Mathematisch Centrum, Amsterdam 1978, whereas the computer euphemism is used by A.
Orlitzky & L. Shepp, On curbing virus propagation, Technical memorandum, Bell Labs Dvipsbugw
1989. Cited on page 771.
742 A complete discussion of the problem can be found in chapter 10 of Il an Vardi, Computa-
tional Recreations in Mathematica, Addison Wesley, Redwood City, California, 1991. Cited
on page 771.
743 On Gibbs’ paradox, see your favourite text on thermodynamics or statistical mechanics. See
also W.H. Zurek, Algorithmic randomness and physical entropy, Physical Review A 40,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pp. 4731–4751, 1989. Zurek shows that the Sackur–Tetrode formula can be derived from
algorithmic entropy considerations. Cited on page 772.
744 S.N. B ose, Plancks Gesetz und Lichtquantenhypothese, Zeitschrift für Physik 26, pp. 178–
181, 1924. The theory was then expanded in A. Einstein, Quantentheorie des einatomigen
idealen Gases, Sitzungsberichte der Preussischen Akademie der Wissenschaften zu Berlin 22,
pp. 261–267, 1924, A. Einstein, Quantentheorie des einatomigen idealen Gases. Zweite
Abhandlung, Sitzungsberichte der Preussischen Akademie der Wissenschaften zu Berlin 23,
pp. 3–14, 1925, A. Einstein, Zur Quantentheorie des idealen Gases, Sitzungsberichte der
Preussischen Akademie der Wissenschaften zu Berlin 23, pp. 18–25, 1925. Cited on page 775.
745 C.K. Hong, Z.Y. Ou & L. Mandel, Measurement of subpicosecond time intervals
between two photons by interference, Physical Review Letters 59, pp. 2044–2046, 1987. Cited
on page 776.
746 The experiment is described in E. R amberg & G.A. Snow, Experimental limit on a small
violation of the Pauli principle, Physics Letters B 238, pp. 438–441, 1990. Other experimental
tests are reviewed in O.W. Greenberg, Particles with small violations of Fermi or Bose
statistics, Physical Review D 43, pp. 4111–4120, 1991. Cited on page 778.
747 The issue is treated in his Summa Theologica, in question 52 of the first part. The complete
text, several thousand pages, can be found on the http://www.newadvent.org website. Cited
on page 785.
748 The original no-cloning theorem is by D. Dieks, Communication by EPR devices, Phys-
Copyright © Christoph Schiller November 1997–May 2006

ics Letters A 92, pp. 271–272, 1982, and by W.K. Wootters & W.H. Zurek, A single
quantum cannot be cloned, Nature 299, pp. 802–803, 1982. For a discussion of photon and
multiparticle cloning, see N. Gisin & S. Massar, Optimal quantum cloning machines,
Physics Review Letters 79, pp. 2153–2156, 1997. The whole topic has been presented in detail
by V. Buzek & M. Hillery, Quantum cloning, Physics World 14, pp. 25–29, November
2001. Cited on page 778.
749 The most recent experimental and theoretical results on physical cloning are described in
A. L amas-Linares, C. Simon, J.C. Howell & D. B ouwmeester, Experimental
quantum cloning of single photons, Science 296, pp. 712 – 714, 2002, D. Collins &
S. Popescu, A classical analogue of entanglement, preprint http://www.arxiv.org/abs/
quant-ph/0107082, 2001, and A. Daffertshofer, A.R. Pl astino & A. Pl astino,
Classical no-cloning theorem, Physical Review Letters 88, p. 210601, 2002. Cited on page

Dvipsbugw
792 vi permu tation of particles

779.
750 E. Wigner, On unitary representations of the inhomogeneous Lorentz group, Annals of
Mathematics 40, pp. 149–204, 1939. This famous paper summarises the work which later
brought him the Nobel Prize for physics. Cited on page 780.
751 Wolf gang Pauli, The connection between spin and statistics, Physical Review 58,
pp. 716– 722, 1940. Cited on page 782.
752 For a full list of isotopes, see R.B. Firestone, Table of Isotopes, Eighth Edition, 1999 Up-
date, with CDROM, John Wiley & Sons, 1999. Cited on page 782.
Dvipsbugw
753 This famous explanation is given, for example, on page 1148 in C.W. Misner, K.S.
Thorne & J.A. Wheeler, Gravitation, Freeman, 1973. It is called the scissor trick on
page 43 of volume 1 of R. Penrose & W. R indler, Spinors and Spacetime, 1984. It is also
cited and discussed by R. Gould, Answer to question #7, American Journal of Physics 63,
p. 109, 1995. Cited on page 783.
754 M.V. Berry & J.M. Robbins, Indistinguishability for quantum particles: spin, statistics
and the geometric phase, Proceedings of the Royal Society in London A 453, pp. 1771–1790,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1997. See also the comments to this result by J. Twamley, Statistics given a spin, Nature
389, pp. 127–128, 11 September 1997. Their newer results are M.V. Berry & J.M. Robbins,
Quantum indistinguishability: alternative constructions of the transported basis, Journal of
Physics A (Letters) 33, pp. L207–L214, 2000, and M.V. Berry & J.M. Robbins, in Spin-
Statistics, eds. R. Hilborn & G. Tino, (American Institute of Physics), 2000, pp. 3–15.
See also Michael Berry’s home page on the http://www.phy.bris.ac.uk/staff/berry_mv.html
website. Cited on page 784.
755 R.W. Hartung, Pauli principle in Euclidean geometry, American Journal of Physics 47,
pp. 900–910, 1979. Cited on page 785.
756 The point that spin can be seen as a rotation was already made by F.J. Belinfante, On the
spin angular momentum of mesons, Physica 6, p. 887, 1939, and taken up again by Hans C.
Ohanian, What is spin?, American Journal of Physics 54, pp. 500–505, 1986. See also E.
Duran & A. Erschow, Physikalische Zeitschrift der Sowjetunion 12, p. 466, 1937. Cited
on page 787.
757 Generalizations of bosons and fermions are reviewed in the (serious!) paper by
O.W. Greenberg, D.M. Greenberger & T.V. Greenbergest, (Para)bosons,
(para)fermions, quons and other beasts in the menagerie of particle statistics, at http://
www.arxiv.org/abs/hep-th/9306225. A newer summary is O.W. Greenberg, Theories
of violation of statistics, electronic preprint available at http://www.arxiv.org/abs/hep-th/
0007054, 2000. Cited on page 789.
Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 C h a p t e r VII

DETAI LS OF QUANTUM THEORY AND

ELEC TROMAGNETI SM
Dvipsbugw

The fact that an adequate philosophical

Ref. 758
“ presentation has been so long delayed is no
doubt caused by the fact that Niels Bohr
brainwashed a whole generation of theorists
into thinking that the job was done fifty years
ago.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Murray Gell-Mann

”
W
hy is this famous physical issue arousing such strong emotions? In particular,
ho is brainwashed, Gell-Mann, the discoverer of the quarks, or most of the
orld’s physicists working on quantum theory who follow Niels Bohr’s* opinion?
In the twentieth century, quantum mechanics has thrown many in disarray. Indeed, it rad-
ically changed the two most basic concepts of classical physics: state and system. The state
is not described any more by the specific values taken by position and momentum, but
by the specific wave function ‘taken’ by the position and momentum operators.** In ad-
dition, in classical physics a system was described as a set of permanent aspects of nature;
permanence was defined as negligible interaction with the environment. Quantum mech-
anics shows that this definition has to be modified as well.

24. superp ositions and probabilities – quantum

theory withou t ideolo gy
In order to clarify the implications of superpositions, we take a short walk around the
strangest aspects of quantum theory. The section is essential if we want to avoid getting
lost on our way to the top of Motion Mountain, as happened to quite a number of people
Copyright © Christoph Schiller November 1997–May 2006

since quantum theory appeared.

* Niels Bohr (b. 1885 Copenhagen, d. 1962 Copenhagen) made Copenhagen University into one of the
centres of quantum theory, overshadowing Göttingen. He developed the description of the atom with
quantum theory, for which he received the 1922 Nobel Prize in physics. He had to flee Denmark in 1943
after the German invasion, because of his Jewish background, but returned there after the war.
** It is equivalent, but maybe conceptually clearer, to say that the state is described by a complete set of
commuting operators. In fact, the discussion is somewhat simplified in the Heisenberg picture. However,
here we study the issue in the Schrödinger picture, using wave functions.

Dvipsbugw
 794 vii details of quantum theory • superpositions and probabilities

Why are people either dead or alive?

The evolution equation of quantum mechanics is linear in the wave function; thus we can
imagine and try to construct systems where the state ψ is a superposition of two radically
distinct situations, such as those of a dead and of a living cat. This famous fictional animal
is called Schrödinger’s cat after the originator of the example. Is it possible to produce it?
How would it evolve in time? We can ask the same questions about a superposition of a
state where a car is inside a closed garage with a state where the car is outside.
Such strange situations are not usually observed in everyday life. The reason for this Dvipsbugw
rareness is an important aspect of what is often called the ‘interpretation’ of quantum
mechanics. In principle, such strange situations are possible, and the superposition of
macroscopically distinct states has actually been observed in a few cases, though not for
cats, people or cars. To get an idea of the constraints, let us specify the situation in more
detail.* The object of discussion are linear superpositions of the type ψ = aψ a + bψ b ,
where ψ a and ψ b are macroscopically distinct states of the system under discussion, and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where a and b are some complex coefficients. States are called macroscopically distinct
when each state corresponds to a different macroscopic situation, i.e. when the two states
can be distinguished using the concepts or measurement methods of classical physics.
In particular, this means that the physical action necessary to transform one state into
the other must be much larger than ħ. For example, two different positions of any body
composed of a large number of molecules are macroscopically distinct.
A ‘strange’ situation is thus
a superposition of macro-
scopic distinct states. Let us
work out the essence of mac-
roscopic superpositions more
clearly. Given two macroscop- no figure yet
ically distinct states ψ a and
ψ b , a superposition of the type F I G U R E 322 Artist’s impression of a macroscopic
ψ = aψ a + bψ b is called a pure superposition
state. Since the states ψ a and
ψ b can interfere, one also talks about a (phase) coherent superposition. In the case of a
superposition of macroscopically distinct states, the scalar product ψ †a ψ b is obviously
vanishing. In case of a coherent superposition, the coefficient product a ‡ b is different
from zero. This fact can also be expressed with the help of the density matrix ρ of the
system, defined as ρ = ψ a ψ † . In the present case it is given by
Copyright © Christoph Schiller November 1997–May 2006

ρ pure = ψ a ψ † = SaS2 ψ a a ψ †a + SbS2 ψ b a ψ †b + a b‡ ψ a a ψ †b + a ‡ b ψ b a ψ †a

SaS2 a b‡
= (ψ a , ψ b ) Œ 2 ‘Œ ‘ .
ψ †a
a b SbS
(554)
‡
ψ †b

We can then say that whenever the system is in a pure state, its density matrix, or density
functional, contains off-diagonal terms of the same order of magnitude as the diagonal
* Most what can be said about this topic has been said by two people: John von Neumann, who in the
Ref. 759 nineteen-thirties stressed the differences between evolution and decoherence, and by Hans Dieter Zeh, who
Ref. 760 in the nineteen seventies stressed the importance of baths and the environment in the decoherence process.

Dvipsbugw
 why are people either dead or alive? 795

ones.* Such a density matrix corresponds to the above-mentioned strange situations that
we do not observe in daily life.
We now have a look at the opposite situation. In contrast to the case just mentioned, a
density matrix for macroscopic distinct states with vanishing off-diagonal elements, such
as the two state example

ρ = SaS2 ψ a a ψ †a + SbS2 ψ b a ψ †b
SaS2 0
= (ψ a , ψ b ) Œ 2 ‘Œ ‘
ψ †a Dvipsbugw
0 SbS
(556)
ψ †b

describes a system which possesses no phase coherence at all. (Here, a denotes the non-
commutative dyadic product or tensor product which produces a tensor or matrix start-
ing from two vectors.) Such a diagonal density matrix cannot be that of a pure state; it
describes a system which is in the state ψ a with probability SaS2 and which is in the state
ψ b with probability SbS2 . Such a system is said to be in a mixed state, because its state is not

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

known, or equivalently, to be in a (phase) incoherent superposition, because interference
effects cannot be observed in such a situation. A system described by a mixed state is al-
ways either in the state ψ a or in the state ψ b . In other words, a diagonal density matrix for
macroscopically distinct states is not in contrast, but in agreement with everyday experi-
ence. In the picture of density matrices, the non-diagonal elements contain the difference
between normal, i.e. incoherent, and unusual, i.e. coherent, superpositions.
The experimental situation is clear: for macroscopically distinct states, only diagonal
density matrices are observed. Any system in a coherent macroscopic superposition
somehow loses its off-diagonal matrix elements. How does this process of decoherence**
take place? The density matrix itself shows the way.
Indeed, the density matrix for a large system is used, in thermodynamics, for the
Ref. 761 definition of its entropy and of all its other thermodynamic quantities. These studies show
Challenge 1270 ny that
S = −k tr(ρ ln ρ) (557)

where tr denotes the trace, i.e. the sum of all diagonal elements. We also remind ourselves
that a system with a large and constant entropy is called a bath. In simple physical terms,
a bath is a system to which we can ascribe a temperature. More precisely, a (physical) bath,
or (thermodynamic) reservoir, is any large system for which the concept of equilibrium can
be defined. Experiments show that in practice, this is equivalent to the condition that a
Copyright © Christoph Schiller November 1997–May 2006

bath consists of many interacting subsystems. For this reason, all macroscopic quantities
describing the state of a bath show small, irregular fluctuations, a fact that will be of central
importance shortly.
* Using the density matrix, we can rewrite the evolution equation of a quantum system:
dρ
= − [H, ρ] .
i
ψ̇ = −iHψ becomes (555)
dt ħ
Both are completely equivalent. (The new expression is sometimes also called the von Neumann equation.)
We won’t actually do any calculations here. The expressions are given so that you recognize them when you
encounter them elsewhere.
** In many settings, decoherence is called disentanglement, as we will see below.

Dvipsbugw
 796 vii details of quantum theory • superpositions and probabilities

Obviously, an everyday bath is also a bath in the physical sense: a thermodynamic bath
is similar to an extremely large warm water bath, one for which the temperature does not
change even if one adds some cold or warm water to it. Examples of physical baths are an
intense magnetic field, a large amount of gas, or a large solid. (The meanings of ‘intense’
and ‘large’ of course depend on the system under study.) The physical concept of bath is
thus an abstraction and a generalization of the everyday concept of bath.
It is easy to see from the definition (557) of entropy that the loss of off-diagonal ele-
Challenge 1271 n ments corresponds to an increase in entropy. And it is known that any increase in entropy
of a reversible system, such as the quantum mechanical system in question, is due to an Dvipsbugw
interaction with a bath.
Where is the bath interacting with the system? It obviously must be outside the system
one is talking about, i.e. in its environment. Indeed, we know experimentally that any
environment is large and characterized by a temperature. Some examples are listed in
Table 58. Any environment therefore contains a bath. We can even go further: for every
experimental situation, there is a bath interacting with the system. Indeed, every system

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

which can be observed is not isolated, as it obviously interacts at least with the observer;
and every observer by definition contains a bath, as we will show in more detail shortly.
Usually however, the most important baths we have to take into consideration are the
atmosphere around a system, the radiation attaining the system or, if the system itself is
large enough to have a temperature, those degrees of freedom of the system which are
not involved in the superposition under investigation.
Since every system is in contact with baths, every density matrix of a macroscopic
superposition will lose its diagonal elements evenually. At first sight, this direction of
thought is not convincing. The interactions of a system with its environment can be made
extremely small by using clever experimental set-ups; that would imply that the time
for decoherence can be made extremely large. Thus we need to check how much time a
superposition of states needs to decohere. It turns out that there are two standard ways to
estimate the decoherence time: either by modelling the bath as large number of colliding
particles, or by modelling it as a continuous field.

If the bath is described as a set of particles randomly hitting the microscopic system,
it is best characterized by the effective wavelength λ eff of the particles and by the average
Challenge 1272 ny interval t hit between two hits. A straightforward calculation shows that the decoherence
time t d is in any case smaller than this time interval, so that Copyright © Christoph Schiller November 1997–May 2006

t d D t hit =
1
, (558)
φσ

where φ is the flux of particles and σ the cross-section for the hit.* Typical values are given
* The decoherence time is derived by studying the evolution of the density matrix ρ(x, x ′ ) of ob-
jects localized at two points x and x ′ . One finds that the off-diagonal elements follow ρ(x, x ′ , t) =
′ 2
ρ(x, x ′ , 0)e−Λt(x−x ) , where the localization rate Λ is given by
2
Λ = k φσeff (559)
where k is the wave number, φ the flux and σeff the cross-section of the collisions, i.e. usually the size of the
Ref. 762 macroscopic object.

Dvipsbugw
 why are people either dead or alive? 797

TA B L E 58 Common and less common baths with their main properties

Bat h t y p e T e m p e r - Wa v e - Pa r - H i t t i m e t hit = 1~σ φ

at u r e length ticle for
flux
T λ eff φ a t o ma o b j e c ta

matter baths
solid, liquid 300 K 10 pm 1031 ~m2 s 10−12 s 10−25 s Dvipsbugw
air 300 K 10 pm 1028 ~m2 s 10−9 s 10−22 s
laboratory vacuum 50 mK 10 µm 1018 ~m2 s 10 s 10−12 s
photon baths
sunlight 5800 K 900 nm 1023 ~m2 s 10−4 s 10−17 s
‘darkness’ 300 K 20 µm 1021 ~m2 s 10−2 s 10−15 s
cosmic microwaves 2.7 K 2 mm 1017 ~m2 s 102 s 10−11 s

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

terrestrial radio waves .. K
Casimir effect .. K
Unruh radiation of Earth 40 zK
nuclear radiation baths
radioactivity 10 pm 10 .. s 10 .. s
cosmic radiation A1000 K 10 pm 10 .. s 10 .. s
solar neutrinos  10 MK 10 pm 1015 ~m2 s 10 .. s 10 .. s
cosmic neutrinos 2.0 K 3 mm 1017 ~m2 s 10 .. s 10 .. s
gravitational baths
gravitational radiation 5 ë 1031 K 10−35 m A10 .. s A10 .. s

a. The cross-section σ in the case of matter and photon baths was assumed to be 10−19 m2 for
atoms; for the macroscopic object a size of 1 mm was used as example. For neutrino baths, ...

in Table 58. We easily note that for macroscopic objects, decoherence times are extremely
short. Scattering leads to fast decoherence. However, for atoms or smaller systems, the
situation is different, as expected.
A second method to estimate the decoherence time is also common. Any interaction
Copyright © Christoph Schiller November 1997–May 2006

of a system with a bath is described by a relaxation time t r . The term relaxation designates
any process which leads to the return to the equilibrium state. The terms damping and
friction are also used. In the present case, the relaxation time describes the return to equi-
librium of the combination bath and system. Relaxation is an example of an irreversible
evolution. A process is called irreversible if the reversed process, in which every compon-
ent moves in opposite direction, is of very low probability.* For example, it is usual that

One also finds the surprising result that a system hit by a particle of energy E hit collapses the density
Ref. 763 matrix roughly down to the de Broglie (or thermal de Broglie) wavelength of the hitting particle. Both results
together give the formula above.
* Beware of other definitions which try to make something deeper out of the concept of irreversibility, such as

Dvipsbugw
 798 vii details of quantum theory • superpositions and probabilities

a glass of wine poured into a bowl of water colours the whole water; it is very rarely ob-
served that the wine and the water separate again, since the probability of all water and
wine molecules to change directions together at the same time is rather low, a state of
affairs making the happiness of wine producers and the despair of wine consumers.
Now let us simplify the description of the bath. We approximate it by a single, un-
specified, scalar field which interacts with the quantum system. Due to the continuity
of space, such a field has an infinity of degrees of freedom. They are taken to model the
many degrees of freedom of the bath. The field is assumed to be in an initial state where
its degrees of freedom are excited in a way described by a temperature T. The interac- Dvipsbugw
tion of the system with the bath, which is at the origin of the relaxation process, can be
described by the repeated transfer of small amounts of energy E hit until the relaxation
process is completed.
The objects of interest in this discussion, like the mentioned cat, person or car, are
described by a mass m. Their main characteristic is the maximum energy E r which can
be transferred from the system to the environment. This energy describes the interac-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tions between system and environment. The superpositions of macroscopic states we are
interested in are solutions of the Hamiltonian evolution of these systems.
The initial coherence of the superposition, so disturbingly in contrast with our every-
Ref. 764 day experience, disappears exponentially within a decoherence time t d given by*

E hit eE hit ~k T − 1
td = tr (562)
E r eE hit ~k T + 1

where k is the Boltzmann constant and like above, E r is the maximum energy which can
be transferred from the system to the environment. Note that one always has t d D t r .
After the decoherence time t d is elapsed, the system has evolved from the coherent to
the incoherent superposition of states, or, in other words, the density matrix has lost its
off-diagonal terms. One also says that the phase coherence of this system has been des-
troyed. Thus, after a time t d , the system is found either in the state ψ a or in the state ψ b ,
respectively with the probability SaS2 or SbS2 , and not any more in a coherent superpos-
ition which is so much in contradiction with our daily experience. Which final state is
selected depends on the precise state of the bath, whose details were eliminated from the
calculation by taking an average over the states of its microscopic constituents.
claims that ‘irreversible’ means that the reversed process is not at all possible. Many so-called ‘contradictions’
Copyright © Christoph Schiller November 1997–May 2006

between the irreversibility of processes and the reversibility of evolution equations are due to this mistaken
interpretation of the term ‘irreversible’.
* This result is derived as in the above case. A system interacting with a bath always has an evolution given
Ref. 765 by the general form
dρ
= − [H, ρ] − Q[Vj ρ, Vj ] + [Vj , ρVj ] ,
i 1 † †
(560)
dt ħ 2t o j
where ρ is the density matrix, H the Hamiltonian, V the interaction, and t o the characteristic time of the
Challenge 1273 ny interaction. Are you able to see why? Solving this equation, one finds for the elements far from the diagonal
ρ(t) = ρ 0 e−t~t 0 . In other words, they disappear with a characteristic time t o . In most situations one has a
relation of the form
E hit
t0 = tr = t hit (561)
Er
or some variations of it, as in the example above.

Dvipsbugw
 why are people either dead or alive? 799

The important result is that for all macroscopic objects, the decoherence time t d is
extremely small. In order to see this more clearly, we can study a special simplified case.
A macroscopic object of mass m, like the mentioned cat or car, is assumed to be at the
same time in two locations separated by a distance l, i.e. in a superposition of the two
corresponding states. We further assume that the superposition is due to the object mov-
ing as a quantum mechanical oscillator with frequency ω between the two locations; this
is the simplest possible system that shows superpositions of an object located in two dif-
ferent positions. The energy of the object is then given by E r = mω 2 l 2 , and the smallest
transfer energy E hit = ħω is the difference between the oscillator levels. In a macroscopic Dvipsbugw
situation, this last energy is much smaller than kT, so that from the preceding expression
Ref. 766 we get
2
ħ2 λ 2T
td = tr = tr =
E hit
t r (563)
2E r kT 2mkT l 2 l2
º
in which the frequency ω has disappeared. The quantity λ T = ħ~ 2mkT is called the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

thermal de Broglie wavelength of a particle.
It is straightforward to see that for practically all macroscopic objects the typical deco-
herence time t d is extremely short. For example, setting m = 1 g, l = 1 mm and T = 300 K
we get t d ~t r = 1.3 ë 10−39 . Even if the interaction between the system and the environment
would be so weak that the system would have as relaxation time the age of the universe,
which is about 4 ë 1017 s, the time t d would still be shorter than 5 ë 10−22 s, which is over
a million times faster than the oscillation time of a beam of light (about 2 fs for green
light). For Schrödinger’s cat, the decoherence time would be even shorter. These times
are so short that we cannot even hope to prepare the initial coherent superposition, let
alone to observe its decay or to measure its lifetime.
For microscopic systems however, the situation is different. For example, for an elec-
tron in a solid cooled to liquid helium temperature we have m = 9.1 ë 10−31 kg, and typic-
ally l = 1 nm and T = 4 K; we then get t d  t r and therefore the system can stay in a co-
herent superposition until it is relaxed, which confirms that for this case coherent effects
can indeed be observed if the system is kept isolated. A typical example is the behaviour
Ref. 767 of electrons in superconducting materials. We will mention a few more below.
In 1996 the first actual measurement of decoherence times was published by the Paris
Ref. 768 team around Serge Haroche. It confirmed the relation between the decoherence time
and the relaxation time, thus showing that the two processes have to be distinguished
at microscopic scale. In the meantime, other experiments confirmed the decoherence
Copyright © Christoph Schiller November 1997–May 2006

Ref. 769 process with its evolution equation, both for small and large values of t d ~t r . A particularly
Ref. 770 beautiful experiment has been performed in 2004, where the disappearance of two-slit
interference for C 70 molecules was observed when a bath interacts with them.

Conclusions on decoherence, life and death

In summary, both estimates of decoherence times tell us that for most macroscopic ob-
jects, in contrast to microscopic ones, both the preparation and the survival of superposi-
tions of macroscopically different states is made practically impossible by the interaction
with any bath found in their environment, even if the usual measure of this interaction,
given by the friction of the motion of the system, is very small. Even if a macroscopic

Dvipsbugw
 800 vii details of quantum theory • superpositions and probabilities

system is subject to an extremely low friction, leading to a very long relaxation time, its
decoherence time is still vanishingly short.
Our everyday environment is full of baths. Therefore, coherent superpositions of mac-
roscopically distinct states never appear in everyday life. In short, we cannot be dead and
alive at the same time.
We also arrive at a second conclusion: decoherence results from coupling to a bath in
the environment. Decoherence is a statistical or thermodynamic effect. We will return to
this issue below.
Dvipsbugw
What is a system? What is an object?
In classical physics, a system is a part of nature which can be isolated from its environ-
ment. However, quantum mechanics tells us that isolated systems do not exist, since in-
teractions cannot be made vanishingly small. The results above allow us to define the
concept of system with more accuracy. A system is any part of nature which interacts in-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

coherently with its environment. In other words, an object is a part of nature interacting
with its environment only through baths.
In particular, a system is called microscopic or quantum mechanical and can described
by a wave function ψ whenever
— it is almost isolated, with t evol = ħ~∆E < t r , and
Ref. 771 — it is in incoherent interaction with its environment.
In short, a microscopic system interacts incoherently and weakly with its environment.
In contrast, a bath is never isolated in the sense just given, because its evolution time
is always much larger than its relaxation time. Since all macroscopic bodies are in con-
tact with baths – or even contain one – they cannot be described by a wave function. In
particular, it is impossible to describe any measuring apparatus with the help of a wave
function.
We thus conclude that a macroscopic system is a system with a decoherence time much
shorter than any other evolution time of its constituents. Obviously, macroscopic systems
also interact incoherently with their environment. Thus cats, cars and television news
speakers are all macroscopic systems.
A third possibility is left over by the two definitions: what happens in the situation in
which the interactions with the environment are coherent? We will encounter some ex-
amples shortly. Following this definition, such situations are not systems and cannot be
Copyright © Christoph Schiller November 1997–May 2006

described by a wave function. For example, it can happen that a particle forms neither a
macroscopic nor a microscopic system! In these situations, when the interaction is coher-
ent, one speaks of entanglement; such a particle or set of particles is said to be entangled
with its environment.
Entangled, coherently interacting systems are separable, but not divisible. In quantum
theory, nature is not found to be made of isolated entities, but is still made of separable
entities. The criterion of separability is the incoherence of interaction. Coherent super-
positions imply the surprising consequence that there are systems which, even though
they look divisible, are not. Entanglement poses a limit to divisibility. All surprising prop-
erties of quantum mechanics, such as Schrödinger’s cat, are consequences of the classical
prejudice that a system made of two or more parts must necessarily be divisible into two

Dvipsbugw
 what is a system? what is an object? 801

space
collapse

Dvipsbugw

t1 t2 t3 t4

slit screen

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

space

F I G U R E 323 Quantum mechanical motion:

an electron wave function (actually its
module squared) from the moment it
passes a slit until it hits a screen

subsystems. But coherent superpositions, or entangled systems, do not allow division.

Whenever we try to divide indivisible systems, we get strange or incorrect conclusions,
such as apparent faster-than-light propagation, or, as one says today, non-local behaviour.
Let us have a look at a few typical examples.

Is quantum theory non-local? – A bit about the

Einstein-Podolsky-Rosen paradox
[Mr. Duffy] lived a little distance away from his

Page 494
“ body ...
James Joyce, A Painful Case

After we explored non-locality in general relativity, we now study it in quantum mech- ”

anics. We first look at the wave function collapse for an electron hitting a screen after
Copyright © Christoph Schiller November 1997–May 2006

passing a slit. Following the description just deduced, the process proceeds schematically
as depicted in Figure 323. A film of the same process can be seen in the lower left corners
on the pages following page 703. The situation is surprising: due to the short decoher-
ence time, in a wave function collapse the maximum of the function changes position at
extremely high speed. In fact, the maximum moves faster than light. But is it a problem?
A situation is called acausal or non-local if energy is transported faster than light. Using
Challenge 1274 n Figure 323 you can determine the energy velocity involved, using the results on signal
Page 577 propagation. The result is a value smaller than c. A wave function maximum moving
faster than light does not imply energy moving faster than light.
In other words, quantum theory has speeds greater than light, but no energy speeds
Ref. 772 greater than light. In classical electrodynamics, the same happens with the scalar and the

Dvipsbugw
 802 vii details of quantum theory • superpositions and probabilities

space

detector 2

Dvipsbugw

detector 1

collapse

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

time

F I G U R E 324 Bohm’s Gedanken experiment

vector potentials if the Coulomb gauge is used. We have also encountered speeds faster
Page 306 than that of light in the motion of shadows and in many other observations. Any physicist
now has two choices: he can be straight, and say that there is no non-locality in nature;
or he can be less straight, and claim there is. In the latter case, he has to claim that even
classical physics is non-local. However, this never happens. On the other hand, there is a
danger in this more provoking usage: a small percentage of those who say that the world
is non-local after a while start to believe that there really are faster-than-light effects in
nature. These people become prisoners of their muddled thinking; on the other hands,
muddled thinking helps to get more easily into newspapers. In short, even though the
definition of non-locality is not unanimous, here we stick to the stricter one, and define
non-locality as energy transport faster than light.
An often cited Gedanken experiment that shows the pitfalls of non-locality was
proposed by Bohm* in the discussion around the so-called Einstein–Podolsky–Rosen
Ref. 773, Ref. 774 paradox. In the famous EPR paper the three authors try to find a contradiction between
quantum mechanics and common sense. Bohm translated their rather confused paper
Copyright © Christoph Schiller November 1997–May 2006

into a clear Gedanken experiment. When two particles in a spin 0 state move apart,
measuring one particle’s spin orientation implies an immediate collapse also of the other
particle’s spin, namely in the exactly opposite direction. This happens instantaneously
over the whole separation distance; no speed limit is obeyed. In other words, entangle-
ment seems to lead to faster-than-light communication.
Again, in Bohm’s experiment, no energy is transported faster than light. No non-
locality is present, despite numerous claims of the contrary by certain authors. The two

* David Joseph Bohm (1917–1992) American–British physicist. He codiscovered the Aharonov–Bohm ef-
fect; he spent a large part of his later life investigating the connections between quantum physics and philo-
sophy.

Dvipsbugw
 what is a system? what is an object? 803

entangled electrons belong to one system: assuming that they are separate only because
the wave function has two distant maxima is a conceptual mistake. In fact, no signal
can be transmitted with this method; the decoherence is a case of prediction which
looks like a signal without being one. We already discussed such cases in the section
Page 580 on electrodynamics.
Bohm’s experiment has actually been performed. The first and most famous realization
Ref. 775 was realized in 1982 by Alain Aspect; he used photons instead of electrons. Like all latter
tests, it has fully confirmed quantum mechanics.
In fact, experiments such as the one by Aspect confirm that it is impossible to treat Dvipsbugw
either of the two particles as a system by itself; it is impossible to ascribe any physical
property, such as a spin orientation, to either of them alone. (The Heisenberg picture
would express this restriction even more clearly.)
The mentioned two examples of apparent non-locality can be dismissed with the re-
mark that since obviously no energy flux faster than light is involved, no problems with
causality appear. Therefore the following example is more interesting. Take two identical

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

atoms, one in an excited state, one in the ground state, and call l the distance that separ-
ates them. Common sense tells that if the first atom returns to its ground state emitting a
photon, the second atom can be excited only after a time t = l~c has been elapsed, i.e. after
the photon has travelled to the second atom.
Surprisingly, this conclusion is wrong. The atom in its ground state has a non-zero
probability to be excited at the same moment in which the first is de-excited. This has
Ref. 776 been shown most simply by Gerhard Hegerfeldt. The result has even been confirmed
experimentally.
More careful studies show that the result depends on the type of superposition of the
two atoms at the beginning: coherent or incoherent. For incoherent superpositions, the
intuitive result is correct; the counter-intuitive result appears only for coherent superpos-
itions. Again, a careful discussion shows that no real non-locality of energy is involved.

Curiosities and fun challenges about superpositions

Coherent superposition, or entanglement, is such a surprising phenomenon that many
aspects have been and still are being explored.
**
In a few cases, the superposition of different macroscopic states can actually be observed
Copyright © Christoph Schiller November 1997–May 2006

by lowering the temperature to sufficiently small values and by carefully choosing suit-
ably small masses or distances. Two well-known examples of coherent superpositions
are those observed in gravitational wave detectors and in Josephson junctions. In the
Ref. 766 first case, one observes a mass as heavy as 1000 kg in a superposition of states located
at different points in space: the distance between them is of the order of 10−17 m. In the
second case, in superconducting rings, superpositions of a state in which a macroscopic
current of the order of 1 pA flows in clockwise direction with one where it flows in counter-
Ref. 784 clockwise direction have been produced.

**
Ref. 778 Superpositions of magnetization in up and down direction at the same time have also be

Dvipsbugw
 804 vii details of quantum theory • superpositions and probabilities

observed for several materials.

**
Some people wrongly state that an atom that is in a superposition of states centred at
different positions has been photographed. (This lie is even used by some sects to attract
Challenge 1275 n believers.) Why is this not true?
**
Since the 1990s, the sport of finding and playing with new systems in coherent macro- Dvipsbugw
Ref. 780 scopic superpositions has taken off across the world. The challenges lie in the clean exper-
iments necessary. Experiments with single atoms in superpositions of states are among
Ref. 781 the most popular ones.
**
Ref. 782 In 1997, coherent atom waves were extracted from a cloud of sodium atoms.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Macroscopic objects usually are in incoherent states. This is the same situation as for
light. The world is full of ‘macroscopic’, i.e. incoherent light: daylight, and all light from
lamps, from fire and from glow-worms is incoherent. Only very special and carefully
constructed sources, such as lasers or small point sources, emit coherent light. Only these
allow to study interference effects. In fact, the terms ‘coherent’ and ‘incoherent’ originated
in optics, since for light the difference between the two, namely the capacity to interfere,
had been observed centuries before the case of matter.
Coherence and incoherence of light and of matter manifest themselves differently,
since matter can stay at rest but light cannot and because light is made of bosons, but
Page 787 matter is made of fermions. Coherence can be observed easily in systems composed of
bosons, such as light, sound in solids, or electron pairs in superconductors. Coherence
is less easily observed in systems of fermions, such as systems of atoms with their elec-
tron clouds. However, in both cases a decoherence time can be defined. In both cases
coherence in many particle systems is best observed if all particles are in the same state
(superconductivity, laser light) and in both cases the transition from coherent to incoher-
ent is due to the interaction with a bath. A beam is thus incoherent if its particles arrive
randomly in time and in frequency. In everyday life, the rarity of observation of coherent
matter superpositions has the same origin as the rarity of observation of coherent light.
Copyright © Christoph Schiller November 1997–May 2006

**
We will discuss the relation between the environment and the decay of unstable systems
later on. The phenomenon is completely described by the concepts given here.
**
Challenge 1276 ny Can you find a method to measure the degree of entanglement? Can you do so for a system
made of many particles?

**
The study of entanglement leads to a simple conclusion: teleportation contradicts correla-

Dvipsbugw
 what is all the fuss abou t measurements in quantum theory? 805

Challenge 1277 ny tion. Can you confirm the statement?

**
Some people say that quantum theory could be used for quantum computing, by using
Ref. 786 coherent superpositions of wave functions. Can you give a general reason that makes this
Challenge 1278 n aim very difficult, even without knowing how such a quantum computer might work?

What is all the fuss about measurements in quantum

Dvipsbugw
theory?
Measurements in quantum mechanics are disturbing. They lead to statements in which
probabilities appear. That is puzzling. For example, we speak about the probability of find-
ing an electron at a certain distance from the nucleus of an atom. Statements like this
belong to the general type ‘when the observable A is measured, the probability to find the
outcome a is p.’ In the following we will show that the probabilities in such statements
are inevitable for any measurement, because, as we will show, any measurement and any

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

observation is a special case of decoherence or disentanglement process. (Historically
however, the process of measurement was studied before the more general process of
decoherence. That explains in part why the topic is so confused in many peoples’ minds.)
Page 638 What is a measurement? As already mentioned in the intermezzo a measurement is
any interaction which produces a record or a memory. (In everyday life, any effect is a re-
Challenge 1279 ny cord; but this is not true in general. Can you give some examples and counter-examples?)
Measurements can be performed by machines; when they are performed by people, they
are called observations. In quantum theory, the action of measurement is not as straight-
forward as in classical physics. This is seen most strikingly when a quantum system, such
as a single electron, is first made to pass a diffraction slit, or better – in order to make its
wave aspect become apparent – a double slit and then is made to hit a photographic plate,
in order to make also its particle aspect appear. Experiment shows that the blackened
dot, the spot where the electron has hit the screen, cannot be determined in advance.
(The same is true for photons or any other particle.) However, for large numbers of elec-
trons, the spatial distribution of the black dots, the so-called diffraction pattern, can be
calculated in advance with high precision.
The outcome of experiments on microscopic systems thus forces us to use probabilities
for the description of microsystems. We find that the probability distribution p(x) of the
spots on the photographic plate can be calculated from the wave function ψ of the electron
at the screen surface and is given by p(x) = Sψ † (x)ψ(x)S2 . This is in fact a special case of
Copyright © Christoph Schiller November 1997–May 2006

the general first property of quantum measurements: the measurement of an observable A

for a system in a state ψ gives as result one of the eigenvalues a n , and the probability Pn
to get the result a n is given by
Pn = Sφ†n ψS2 , (564)

Page 1207 where φ n is the eigenfunction of the operator A corresponding to the eigenvalue a n .
Experiments also show a second property of quantum measurements: after the meas-
urement, the observed quantum system is in the state φ n corresponding to the measured
eigenvalue a n . One also says that during the measurement, the wave function has col-
Ref. 783 lapsed from ψ to φ n . By the way, both properties can also be generalized to the more

Dvipsbugw
 806 vii details of quantum theory • superpositions and probabilities

general cases with degenerate and continuous eigenvalues.

At first sight, the sort of probabilities encountered in
quantum theory are different from the probabilities we en-
counter in everyday life. Roulette, dice, pachinko machines, gravity
the direction in which a pencil on its tip falls, have been meas-
ured experimentally to be random (assuming no cheating by
the designer or operators) to a high degree of accuracy. These
systems do not puzzle us. We unconsciously assume that the
random outcome is due to the small, but uncontrollable vari- Dvipsbugw
ations of the starting conditions every time the experiment
F I G U R E 325 A system
is repeated.* showing probabilistic
But microscopic systems seem to be different. The two behaviour
measurement properties just mentioned express what phys-
icists observe in every experiment, even if the initial conditions are taken to be exactly
the same every time. But why then is the position for a single electron, or most other

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

observables of quantum systems, not predictable? In other words, what happens during
the collapse of the wave function? How long does it take? In the beginning of quantum
theory, there was the perception that the observed unpredictability is due to the lack of
information about the state of the particle. This lead many to search for so-called ‘hidden
variables’. All these attempts were doomed to fail, however. It took some time for the sci-
entific community to realize that the unpredictability is not due to the lack of information
about the state of the particle, which is indeed described completely by the state vector ψ.
In order to uncover the origin of probabilities, let us recall the nature of a measure-
ment, or better, of a general observation. Any observation is the production of a record.
The record can be a visual or auditive memory in our brain, or a written record on pa-
Page 637 per, or a tape recording, or any such type of object. As explained in the intermezzo, an
object is a record if it cannot have arisen or disappeared by chance. To avoid the influ-
ence of chance, all records have to be protected as much as possible from the outer world;
e.g. one typically puts archives in earthquake safe buildings with fire protection, keeps
documents in a safe, avoids brain injury as much as possible, etc.
On top of this, records have to be protected from their internal fluctuations. These in-
ternal fluctuations are due to the many components any recording device is made of. But
if the fluctuations were too large, they would make it impossible to distinguish between
the possible contents of a memory. Now, fluctuations decrease with increasing size of a
system, typically with the square root of the size. For example, if a hand writing is too
Copyright © Christoph Schiller November 1997–May 2006

small, it is difficult to read if the paper gets brittle; if the magnetic tracks on tapes are
too small, they demagnetize and loose the stored information. In other words, a record is
rendered stable against internal fluctuations by making it of sufficient size. Every record
thus consists of many components and shows small fluctuations.
Therefore, every system with memory, i.e. every system capable of producing a record,
contains a bath. In summary, the statement that any observation is the production of a
record can be expressed more precisely as: Any observation of a system is the result of an

* To get a feeling for the limitations of these unconscious assumptions, you may want to read the already
mentioned story of those physicists who built a machine that could predict the outcome of a roulette ball
Ref. 81 from the initial velocity imparted by the croupier.

Dvipsbugw
 what is all the fuss abou t measurements in quantum theory? 807

the quantum apparatus, e.g. eye, ear,

mechanical or machine, with memory,
system i.e. coupled to a bath

H H int tr

description determined friction, Dvipsbugw

of its possible by the type e.g. due
states of measurement to heat flow

F I G U R E 326 The concepts used in the description

of measurements

interaction between that system and a bath in the recording apparatus.*

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

But we can say more. Obviously, any observation measuring a physical quantity uses
an interaction depending on that same quantity. With these seemingly trivial remarks,
we can describe in more detail the process of observation, or, as it is usually called in the
quantum theory, the measurement process.
Any measurement apparatus, or detector, is characterized by two main aspects: the
interaction it has with the microscopic system, and the bath it contains to produce the
Ref. 779 record. Any description of the measurement process thus is the description of the evolu-
tion of the microscopic system and the detector; therefore one needs the Hamiltonian for
the particle, the interaction Hamiltonian, and the bath properties, such as the relaxation
time. The interaction specifies what is measured and the bath realizes the memory.
We know that only classical thermodynamic systems can be irreversible; quantum
systems are not. We therefore conclude: a measurement system must be described clas-
sically: otherwise it has no memory and is not a measurement system: it produces no
record! Memory is a classical effect. (More precisely, it is an effect that only appears in
the classical limit.) Nevertheless, let us see what happens if one describes the measure-
ment system quantum mechanically. Let us call A the observable which is measured in
the experiment and its eigenfunctions φ n . We describe the quantum mechanical system
under observation – often a particle – by a state ψ. This state can always be written as
ψ = ψ p ψ other = Pn c n φ n ψ other , where ψ other represents the other degrees of freedom of
the particle, i.e. those not described – spanned, in mathematical language – by the oper-
Copyright © Christoph Schiller November 1997–May 2006

ator A corresponding to the observable we want to measure. The numbers c n = Sφ†n ψ p S

give the expansion of the state ψ p , which is taken to be normalized, in terms of the basis
φ n . For example, in a typical position measurement, the functions φ n would be the pos-
ition eigenfunctions and ψ other would contain the information about the momentum, the
spin and all other properties of the particle.
How does the system–detector interaction look like? Let us call the state of the appar-
atus before the measurement χ start ; the measurement apparatus itself, by definition, is
a device which, when it is hit by a particle in the state φ n ψ other , changes from the state
χ start to the state χ n . One then says that the apparatus has measured the eigenvalue a n
* Since baths imply friction, we can also say: memory needs friction.

Dvipsbugw
 808 vii details of quantum theory • superpositions and probabilities

corresponding to the eigenfunction φ n of the operator A. The index n is thus the record
of the measurement; it is called the pointer index or variable. This index tells us in which
state the microscopic system was before the interaction. The important point, taken from
our previous discussion, is that the states χ n , being records, are macroscopically distinct,
precisely in the sense of the previous section. Otherwise they would not be records, and
the interaction with the detector would not be a measurement.
Of course, during measurement, the apparatus sensitive to φ n changes the part ψ other
of the particle state to some other situation ψ other,n , which depends on the measurement
and on the apparatus; we do not need to specify it in the following discussion.* Let us have Dvipsbugw
an intermediate check of our reasoning. Do apparatuses as described here exist? Yes, they
do. For example, any photographic plate is a detector for the position of ionizing particles.
A plate, and in general any apparatus measuring position, does this by changing its mo-
mentum in a way depending on the measured position: the electron on a photographic
plate is stopped. In this case, χ start is a white plate, φ n would be a particle localized at
spot n, χ n is the function describing a plate blackened at spot n and ψ other,n describes the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

momentum and spin of the particle after it has hit the photographic plate at the spot n.
Now we are ready to look at the measurement process itself. For the moment, let us
disregard the bath in the detector. In the time before the interaction between the particle
and the detector, the combined system was in the initial state ψ i given simply by

ψ i = ψ p χ start = Q c n φ n ψ other χ start . (567)

n

After the interaction, using the just mentioned characteristics of the apparatus, the com-
bined state ψ a is
ψ a = Q c n φ n ψ other,n χ n . (568)
n

This evolution from ψ i to ψ a follows from the evolution equation applied to the particle
detector combination. Now the state ψ a is a superposition of macroscopically distinct
states, as it is a superposition of distinct macroscopic states of the detector. In our example
ψ a could correspond to a superposition of a state where a spot on the left upper corner
is blackened on an otherwise white plate with one where a spot on the right lower corner
of the otherwise white plate is blackened. Such a situation is never observed. Let us see
Copyright © Christoph Schiller November 1997–May 2006

* How does the interaction look like mathematically? From the description we just gave, we specified the
final state for every initial state. Since the two density matrices are related by
†
ρf = T ρi T (565)

Challenge 1280 ny we can deduce the Hamiltonian from the matrix T. Are you able to see how?
By the way, one can say in general that an apparatus measuring an observable A has a system interaction
Hamiltonian depending on the pointer variable A, and for which one has

[H + H int , A] = 0 . (566)

Dvipsbugw
 what is all the fuss abou t measurements in quantum theory? 809

why. The density matrix ρ a of this situation, given by

ρ a = ψ a a ψ †a = Q c n c ‡m (φ n ψ other,n χ n ) a (φ m ψ other,m χ m )† , (569)

n,m

contains non-diagonal terms, i.e. terms for n x m, whose numerical coefficients are dif-
ferent from zero. Now let’s take the bath back in.
From the previous section we know the effect of a bath on such a macroscopic super-
position. We found that a density matrix such as ρ a decoheres extremely rapidly. We as- Dvipsbugw
sume here that the decoherence time is negligibly small, in practice thus instantaneous,*
so that the off-diagonal terms vanish, and only the final, diagonal density matrix ρ f , given
by
ρ f = Q Sc n S2 (φ n ψ other,n χ n ) a (φ n ψ other,n χ n )† (570)
n

has experimental relevance. As explained above, such a density matrix describes a mixed

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

state and the numbers Pn = Sc n S2 = Sφ†n ψ p S2 give the probability of measuring the value
a n and of finding the particle in the state φ n ψ other,n as well as the detector in the state χ n .
But this is precisely what the two properties of quantum measurements state.
We therefore find that describing a measurement as an evolution of a quantum system
interacting with a macroscopic detector, itself containing a bath, we can deduce the two
properties of quantum measurements, and thus the collapse of the wave function, from
the quantum mechanical evolution equation. The decoherence time t d of the previous
section becomes the time of collapse in the case of a measurement:

t collapse = t d < t r (571)

We thus have a formula for the time the wave function takes to collapse. The first experi-
Ref. 785 mental measurements of the time of collapse are appearing and confirm these results.

Hidden variables
Obviously a large number of people are not satisfied with the arguments just presented.
They long for more mystery in quantum theory. The most famous approach is the idea
that the probabilities are due to some hidden aspect of nature which is still unknown to
humans. But the beautiful thing about quantum mechanics is that it allows both concep-
Copyright © Christoph Schiller November 1997–May 2006

tual and experimental tests on whether such hidden variables exist without the need of
knowing them.
Clearly, hidden variables controlling the evolution of microscopic system would con-
tradict the result that action values below ħ~2 cannot be detected. This minimum observ-
able action is the reason for the random behaviour of microscopic systems.
Historically, the first argument against hidden variables was given by John von Neu-

* Note however, that an exactly vanishing decoherence time, which would mean a strictly infinite number
of degrees of freedom of the environment, is in contradiction with the evolution equation, and in particular
with unitarity, locality and causality. It is essential in the whole argument not to confuse the logical con-
sequences of a extremely small decoherence time with those of an exactly vanishing decoherence time.

Dvipsbugw
 810 vii details of quantum theory • superpositions and probabilities

mann.*

– CS – to be written – CS –

An additional no-go theorem for hidden variables was published by Kochen and
Ref. 787 Specker in 1967, (and independently by Bell in 1969). It states that non-contextual hid-
den variables are impossible, if the Hilbert space has a dimension equal or larger than
three. The theorem is about non-contextual variables, i.e. about hidden variables inside
the quantum mechanical system. The Kochen–Specker theorem thus states that there is Dvipsbugw
no non-contextual hidden variables model, because mathematics forbids it. This result
essentially eliminates all possibilities, because usual quantum mechanical systems have
dimensions much larger than three.
But also common sense eliminates hidden variables, without any recourse to mathem-
atics, with an argument often overlooked. If a quantum mechanical system had internal
hidden variables, the measurement apparatus would have zillions of them.** And that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

would mean that it could not work as a measurement system.
Of course, one cannot avoid noting that about contextual hidden variables, i.e. vari-
ables in the environment, there are no restricting theorems; indeed, their necessity was
shown earlier in this section.
Obviously, despite these results, people have also looked for experimental tests on hid-
den variables. Most tests are based on the famed Bell’s equation, a beautifully simple rela-
tion published by John Bell*** in the 1960s.
The starting idea is to distinguish quantum theory and locally realistic theories using
hidden variables by measuring the polarizations of two correlated photons. Quantum
theory says that the polarization of the photons is fixed only at the time it is measured,
whereas local realistic theories say that it is fixed already in advance. The correct descrip-
tion can be found by experiment.
Imagine the polarization is measured at two distant points A and B, each observer can
measure 1 or −1 in each of his favourite direction. Let each observer choose two directions,
1 and 2, and call their results a 1 , a 2 , b 1 and b 2 . Since the measurement results all are either
1 or −1, the value of the specific expression (a 1 + a 2 )b 1 + (a 2 − a 1 )b 2 has always the value
2.
Ref. 788 Imagine you repeat the experiment many times, assuming that the hidden variables
appear statistically. You then can deduce (a special case of) Bell’s inequality for two hidden
Challenge 1281 e variables
S(a 1 b 1 ) + (a 2 b 1 ) + (a 2 b 2 ) − (a 1 b 2 )S D 2
Copyright © Christoph Schiller November 1997–May 2006

(572)

where the expressions in brackets are the averages of the measurement products over a

* János von Neumann (b. 1903 Budapest, d. 1957 Washington DC) Hungarian mathematician. One of the
greatest and clearest minds of the twentieth century, he settled already many questions, especially in applied
mathematics and quantum theory, that others still struggle with today. He worked on the atomic and the
hydrogen bomb, on ballistic missiles, and on general defence problems. In another famous project, he build
the first US-American computer, building on his extension of the ideas of Konrad Zuse.
** Which leads to the definition: one zillion is 1023 .
*** John Stewart Bell (1928–1990), theoretical physicist who worked mainly on the foundations of quantum
theory.

Dvipsbugw
conclusions on probabilities and determinism 811

large number of samples. This result holds independently of the directions of the involved
polarizers.
On the other hand, if the polarizers 1 and 2 at position A and the corresponding ones
at position B are chosen with angles of π~4, quantum theory predicts that the result is
º
S(a 1 b 1 ) + (a 2 b 1 ) + (a 2 b 2 ) − (a 1 b 2 )S = 2 2 A 2 (573)

which is in complete contradiction with the hidden variable result.

So far, all experimental checks of Bell’s equation have confirmed standard quantum Dvipsbugw
mechanics. No evidence for hidden variables has been found. This is not really surprising,
since the search for such variables is based on a misunderstanding of quantum mechanics
or on personal desires on how the world should be, instead of relying on experimental
evidence.
Another measurable contradiction between quantum theory and locally realistic the-
ories has been predicted by Greenberger, Horn and Zeilinger. Experiments trying to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

check the result are being planned. No deviation from quantum theory is expected.

Conclusions on probabilities and determinism

Geometric demonstramus quia facimus; si

“ physics demonstrare possemus, faceremus.

From the arguments presented here we draw a number of conclusions which we need
Giambattista Vico*

”
for the rest of our mountain ascent. Note that these conclusions are not yet shared by all
physicists! The whole topic is still touchy.
— Probabilities do not appear in measurements because the state of the quantum system
is unknown or fuzzy, but because the detailed state of the bath in the environment
is unknown. Quantum mechanical probabilities are of statistical origin and are due to
baths in the environment (or the measurement apparatus). The probabilities are due to
the large number of degrees of freedom contained in any bath. These large numbers
make the outcome of experiments unpredictable. If the state of the bath were known,
the outcome of an experiment could be predicted. The probabilities of quantum theory
are ‘thermodynamic’ in origin.
In other words, there are no fundamental probabilities in nature. All probabilities
in nature are due to decoherence; in particular, all probabilities are due to the statistics
Copyright © Christoph Schiller November 1997–May 2006

of the many particles – some of which may be virtual – that are part of the baths in the
environment. Modifying well-known words by Albert Einstein, ‘nature really does not
play dice.’ We therefore called ψ the wave function instead of ‘probability amplitude’, as
is often done. State function would be an even better name.
— Any observation in everyday life is a special case of decoherence. What is usually called
the ‘collapse of the wave function’ is a decoherence process due to the interaction with

* ‘We are able to demonstrate geometrical matters because we make them; if we could prove physical mat-
ters we would be able to make them.’ Giovanni Battista Vico (b. 1668 Napoli, d. 1744 Napoli) important
Italian philosopher and thinker. In this famous statement he points out a fundamental distinction between
mathematics and physics.

Dvipsbugw
 812 vii details of quantum theory • superpositions and probabilities

the baths present in the environment or in the measuring apparatus. Because humans
are warm-blooded and have memory, humans themselves are thus measurement ap-
paratuses. The fact that our body temperature is 37°C is thus the reason that we see
Challenge 1282 n only a single world, and no superpositions. (Actually, there are more reasons; can you
name a few?)
— A measurement is complete when the microscopic system has interacted with the bath
in the measuring apparatus. Quantum theory as a description of nature does not re-
quire detectors; the evolution equation describes all examples of motion. However,
measurements do require the existence of detectors. Detectors, being machines that re- Dvipsbugw
cord observations, have to include a bath, i.e. have to be classical, macroscopic objects.
In this context one speaks also of a classical apparatus. This necessity of the measure-
ment apparatus to be classical had been already stressed in the very early stages of
quantum theory.
— All measurements, being decoherence processes that involve interactions with baths,
are irreversible processes and increase entropy.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

— A measurement is a special case of quantum mechanical evolution, namely the evolu-
tion for the combination of a quantum system, a macroscopic detector and the environ-
ment. Since the evolution equation is relativistically invariant, no causality problems
appear in measurements; neither do locality problems and logical problems appear.
— Since both the evolution equation and the measurement process does not involve
quantities other than space-time, Hamiltonians, baths and wave-functions, no other
quantity plays a role in measurement. In particular, no human observer nor any con-
sciousness are involved or necessary. Every measurement is complete when the micro-
scopic system has interacted with the bath in the apparatus. The decoherence inherent
in every measurement takes place even if nobody is looking. This trivial consequence
is in agreement with the observations of everyday life, for example with the fact that
the Moon is orbiting the Earth even if nobody looks at it.* Similarly, a tree falling in
the middle of a forest makes noise even if nobody listens. Decoherence is independent
of human observation, of the human mind and of human existence.
— In every measurement the quantum system interacts with the detector. Since there is a
minimum value for the magnitude of action, every observation influences the observed.
Therefore every measurement disturbs the quantum system. Any precise description
of observations must also include the description of this disturbance. In the present
section the disturbance was modelled by the change of the state of the system from
ψ other to ψ other,n . Without such a change of state, without a disturbance of the quantum
Copyright © Christoph Schiller November 1997–May 2006

system, a measurement is impossible.

— Since the complete measurement is described by quantum mechanics, unitarity is
and remains the basic property of evolution. There are no non-unitary processes in
quantum mechanics.
— The description of the collapse of the wave function as a decoherence process is an
explanation exactly in the sense in which the term ‘explanation’ was defined in the
Page 687 intermezzo; it describes the relation between an observation and all the other aspects
of reality, in this case the bath in the detector or the environment. The collapse of the

* The opposite view is sometimes falsely attributed to Niels Bohr. The Moon is obviously in contact with
Challenge 1283 n many radiation baths. Can you list a few?

Dvipsbugw
 conclusions on probabilities and determinism 813

wave function has been both calculated and explained. The collapse is not a question
of ‘interpretation’, i.e. of opinion, as unfortunately often is suggested.*
— It is not useful to speculate whether the evolution for a single quantum measure-
ment could be determined if the state of the environment around the system were
known. Measurements need baths. But a bath, being irreversible, cannot be described
by a wave function, which behaves reversibly.** Quantum mechanics is deterministic.
Baths are probabilistic.
— In summary, there is no irrationality in quantum theory. Whoever uses quantum the-
ory as argument for superstitions, irrational behaviour, new age beliefs or ideologies is Dvipsbugw
guilty of disinformation. The statement by Gell-Mann at the beginning of this chapter
is thus such an example. Another is the following well-known but incorrect statement
by Richard Feynman:
Ref. 790 ...nobody understands quantum mechanics.
Nobel Prizes in physics obviously do not prevent infection by ideology. On the other

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

hand, the process of decoherence allows a clear look on various interesting issues.

What is the difference between space and time?

Space and time differ. Objects are localized in space but not in time. Why is this the case?
Most bath–system interactions are mediated by a potential. All potentials are by defin-
ition position dependent. Therefore, every potential, being a function of the position x,
commutes with the position observable (and thus with the interaction Hamiltonian). The
decoherence induced by baths – except if special care is taken – thus first of all destroys
the non-diagonal elements for every superposition of states centred at different locations.
In short, objects are localized because they interact with baths via potentials.
For the same reason, objects also have only one spatial orientation at a time. If the
system–bath interaction is spin-dependent, the bath leads to ‘localization’ in the spin
variable. This happens for all microscopic systems interacting with magnets. As a result,
macroscopic superpositions of magnetization are almost never observed. Since electrons,
protons and neutrons have a magnetic moment and a spin, this conclusion can even be
extended: everyday objects are never seen in superpositions of different rotation states
because their interactions with baths are spin-dependent.
As a counter-example, most systems are not localized in time, but on the contrary exist
for very long times, because practically all system–bath interactions do not commute with
time. In fact, this is the way a bath is defined to begin with. In short, objects are permanent
Copyright © Christoph Schiller November 1997–May 2006

because they interact with baths.

Are you able to find an interaction which is momentum-dependent? What is the con-
Challenge 1284 n sequence for macroscopic systems?
In other words, in contrast to general relativity, quantum theory produces a distinction
between space and time. In fact, we can define position as the observable that commutes

* This implies that the so-called ‘many worlds’ interpretation is wishful thinking. The conclusion is con-
Ref. 789 firmed when studying the details of this religious approach. It is a belief system, not based on facts.
** This very strong type of determinism will be very much challenged in the last part of this text, in which
it will be shown that time is not a fundamental concept, and therefore that the debate around determinism
looses most of its interest.

Dvipsbugw
 814 vii details of quantum theory • superpositions and probabilities

with interaction Hamiltonians. This distinction between space and time is due to the
properties of matter and its interactions. We could not have deduced this distinction in
general relativity.

Are we good observers?

Are humans classical apparatuses? Yes, they are. Even though several prominent physi-
cists claim that free will and probabilities are related, a detailed investigation shows that
Ref. 791 this in not the case. Our senses are classical machines in the sense described above: they Dvipsbugw
record observations by interaction with a bath. Our brain is also a classical apparatus: the
neurons are embedded in baths. Quantum probabilities do not play a determining role
in the brain.
Any observing entity needs a bath and a memory to record its observations. That
means that observers have to be made of matter; an observer cannot be made of radi-
ation. Our description of nature is thus severely biased: we describe it from the stand-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

point of matter. That is a bit like describing the stars by putting the Earth at the centre
of the universe. Can we eliminate this basic anthropomorphism? We will find out as we
Page 1016 continue.

What connects information theory, cryptology and quantum

theory?
Physics means talking about observations of nature. Like any observation, also measure-
ments produce information. It is thus possible to translate much (but not all) of quantum
theory into the language of information theory. In particular, the existence of a minimal
change in nature implies that the information about a physical system can never be com-
plete, that information transport has its limits and that information can never be fully
trusted. The details of these studies form a fascinating way to look at the microscopic
world. The studies become even more interesting when the statements are translated into
Ref. 792 the language of cryptology. Cryptology is the science of transmitting hidden messages
that only the intended receiver can decrypt. In our modern times of constant surveillance,
cryptology is an important tool to protect personal freedom.*
The quantum of action implies that messages can be sent in an (almost) safe way.
Listening to a message is a measurement process. Since there is a smallest action, one
can detect whether somebody has tried to listen to a sent message. A man in the middle
attack – somebody who pretends to be the receiver and then sends a copy of the message
Copyright © Christoph Schiller November 1997–May 2006

to the real, intended receiver – can be avoided by using entangled systems as signals to
transmit the information. Quantum cryptologists therefore usually use communication
systems based on entangled photons.
The major issue of quantum cryptology is the key distribution problem. All secure
communication is based on a secret key that is used to decrypt the message. Even if the
communication channel is of the highest security – like entangled photons – one still has

* Cryptology consists of the field of cryptography, the art of coding messages, and the field of cryptoana-
lysis, the art of deciphering encrypted messages. For a good introduction to cryptology, see the text by
Albrecht Beu telspacher, Jörg S chwenk & Klaus-Dieter Wolfenstätter, Moderne Ver-
fahren der Kryptographie, Vieweg 1995.

Dvipsbugw
 conclusions on probabilities and determinism 815

to find a way to send the communication partner the secret key necessary for the decryp-
tion of the messages. Finding such methods is the main aspect of quantum cryptology, a
large research field. However, close investigation shows that all key exchange methods are
limited in their security. In short, due to the quantum of action, nature provides limits on
the possibility of sending encrypted messages. The statement of these limits is (almost)
equivalent to the statement that change in nature is limited by the quantum of action.
The quantum of action provides a limit to secure information exchange. This connec-
tion also allows to brush aside several incorrect statements often found in the media.
Page 1048 Stating that ‘the universe is information’ or that ‘the universe is a computer’ is as devoid Dvipsbugw
of reason as saying that the universe is an observation, a measurement apparatus, a clock-
work or a chewing-gum dispenser. Any expert of motion should beware of these and sim-
ilarly fishy statements; people who use them either deceive themselves or try to deceive
others.

Does the universe have a wave function? And initial conditions?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The wave function of the universe is is frequently invoked in discussions about quantum
theory. Many deduce conclusions from it, for example on the irreversibility of time, on
the importance of initial conditions, on changes required to quantum theory and much
more. Are these arguments correct?
Page 445 The first thing to clarify is the meaning of ‘universe’. As explained above, the term can
have two meanings: either the collection of all matter and radiation, or this collection plus
all of space-time. Then we recall the meaning of ‘wave function’: it describes the state of a
system. The state distinguishes two otherwise identical systems; for example, position and
velocity distinguish two otherwise identical ivory balls on a billiard table. Alternatively
and equivalently, the state describes changes in time.
Does the universe have a state? If we take the wider meaning of universe, obviously
Page 39 it does not. Talking about the state of the universe is a contradiction: by definition, the
concept of state, defined as the non-permanent aspects of an object, is applicable only to
parts of the universe.
We then can take the narrower sense of ‘universe’ – the sum of all matter and radiation
only – and ask the question again. To determine its state, we need a possibility to measure
it: we need an environment. But the environment of matter and radiation is space-time
only; initial conditions cannot be determined since we need measurements to do this,
and thus an apparatus. An apparatus is material system with a bath attached to it; there
Copyright © Christoph Schiller November 1997–May 2006

is no such system outside the universe.

In short, quantum theory does not allow for measurements of the universe; therefore
the universe has no state. Beware of anybody who claims to know something about the
wave function of the universe. Just ask him: If you know the wave function of the universe,
why aren’t you rich?
Despite this conclusion, several famous physicists have proposed evolution equations
Ref. 793 for the wave function of the universe. (The best-known is the Wheeler–DeWitt equation.)
It seems a silly point, but the predictions of these equations have not been compared
to experiments; the arguments just given even make this impossible in principle. The
pursuits in this direction, so interesting they are, must therefore be avoided if we want to
safely reach the top of Motion Mountain.

Dvipsbugw
 816 vii details of quantum theory • applied quantum mechanics

There are many more twists to this story. One possibility is that space-time itself, even
without matter, is a bath. This speculation will be shown to be correct later on and seems
to allow speaking of the wave function of all matter. But then again, it turns out that time
is undefined at the scales where space-time would be an effective bath; this means that
the concept of state is not applicable there.
A lack of ‘state’ for the universe is a strong statement. It also implies a lack of initial
conditions! The arguments are precisely the same. This is a tough result. We are so used
to think that the universe has initial conditions that we never question the term. (Even in
this text the mistake might appear every now and then.) But there are no initial conditions Dvipsbugw
of the universe.
We can retain as result, valid even in the light of the latest results of physics: the uni-
verse has no wave function and no initial conditions, independently of what is meant by
‘universe’. But before we continue to explore the consequences of quantum theory for the
whole universe, we study in more detail the consequences for our everyday observations.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

25. applied quantum mechanics – life, pleasure and
the means to achieve them
Homo sum, humani nil a me alienum puto.*

“
Now that we can look at quantum effects without ideological baggage, let us have some
serious fun in the world of quantum theory. The quantum of action has important con-
Terence

”
sequences for biology, chemistry, technology and science fiction. We will only explore a
cross-section of these topics, but it will be worth it.

Biolo gy
A special form of electromagnetic motion is of importance to humans: life. We mentioned
at the start of quantum theory that life cannot be described by classical physics. Life is a
quantum effect. Let us see why.
Living beings can be described as objects showing metabolism, information pro-
cessing, information exchange, reproduction and motion. Obviously, all these properties
follow from a single one, to which the others are enabling means:
U Living beings are objects able to reproduce.**
Copyright © Christoph Schiller November 1997–May 2006

From your biology lessons you might remember the some properties of heredity. Repro-
duction is characterized by random changes from one generation to the next. The stat-
istics of mutations, for example Mendel’s ‘laws’ of heredity, and the lack of intermedi-
ate states, are direct consequences of quantum theory. In other words, reproduction and
growth are quantum effects.

* ‘I am a man and nothing human is alien to me.’ Terence is Publius Terentius Afer (c. 190–159 bce ), the
important roman poet. He writes this in his play Heauton Timorumenos, verse 77.
** However, there are examples of objects which reproduce and which nobody would call living. Can you
Challenge 1285 n find some examples, together with a sharper definition?

Dvipsbugw
 biology 817

In order to reproduce, living beings must be able to move in self-directed ways. An ob-
ject able to perform self-directed motion is called a machine. All self-reproducing beings
are machines.
Since reproduction and growth is simpler the smaller the system is, most living be-
ings are extremely small machines for the tasks they perform, especially when compared
to human made machines. This is the case even though the design of human machines
has considerably fewer requirements: human-built machines do not need to be able to
reproduce; as a result, they do not need to be made of a single piece of matter, as all living
beings have to. But despite all the restrictions nature has to live with, living beings hold Dvipsbugw
many miniaturization world records:
— The brain has the highest processing power per volume of any calculating device so
far. Just look at the size of chess champion Gary Kasparov and the size of the computer
against which he played.
— The brain has the densest and fastest memory of any device so far. The set of com-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pact discs (CDs) or digital versatile discs (DVDs) that compare with the brain is many
Page 639 thousand times larger.
— Motors in living beings are many orders of magnitude smaller than human-built ones.
Just think about the muscles in the legs of an ant.
— The motion of living beings beats the acceleration of any human-built machine by
orders of magnitude. No machine moves like a grasshopper.
— Living being’s sensor performance, such as that of the eye or the ear, has been sur-
passed by human machines only recently. For the nose this feat is still far away. Never-
theless, the sensor sizes developed by evolution – think also about the ears or eyes of
a common fly – is still unbeaten.
— Living beings that fly, swim or crawl – such as fruit flies, plankton or amoebas – are
still thousands of times smaller than anything built by humans.
Challenge 1286 n — Can you spot more examples?
The superior miniaturization of living beings is due to their continuous strife for efficient
construction. In the structure of living beings, everything is connected to everything:
each part influences many others. Indeed, the four basic processes in life, namely meta-
bolic, mechanical, hormonal and electrical, are intertwined in space and time. For ex-
ample, breathing helps digestion; head movements pump liquid through the spine; a
single hormone influences many chemical processes. Furthermore, all parts in living sys-
tems have more than one function. For example, bones provide structure and produce
Copyright © Christoph Schiller November 1997–May 2006

blood; fingernails are tools and shed chemical waste.

The miniaturization, the reproduction, the growth and the functioning of living beings
all rely on the quantum of action. Let us see how.

Reproduction
Life is a sexually transmitted disease.

“ Anonymous

All the astonishing complexity of life is geared towards reproduction. Reproduction is the ”
ability of an object to build other objects similar to itself. Quantum theory told us that

Dvipsbugw
 818 vii details of quantum theory • applied quantum mechanics

Dvipsbugw

F I G U R E 327 A quantum machine (© Elmar Bartel)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 778 only a similar object is possible, as an exact copy would contradict the quantum of action,
as we found out above.
Since reproduction requires mass increase, reproducing objects show both metabol-
ism and growth. In order that growth leads to an object similar to the original, a con-
struction plan is necessary. This plan must be similar to the plan used by the previous
generation. Organizing growth with a construction plan is only possible if nature is made
of smallest entities which can be assembled following that plan.
We can thus deduce that reproduction implies that matter is made of smallest entities.
If matter were not made of smallest entities, there would be no way to realize reproduc-
tion. Reproduction thus requires quantum theory. Indeed, without the quantum of action
there would be no DNA molecules and there would be no way to inherit our own proper-
ties – our own construction plan – to children.
Passing on a plan requires that living beings have ways to store information. Living
beings must have some built-in memory. We know already that a system with memory
must be made of many particles. There is no other way to store information. The large
number of particles is mainly necessary to protect the information from the influences
of the outside world.
Our own construction plan, made of what biologists call genes, is stored in DNA mo-
lecules. Reproduction is thus first of all a transfer of parent’s genes to the next generation.
Page 838 We will come back to the details below. We first have a look on how our body moves itself
Copyright © Christoph Schiller November 1997–May 2006

and its genes around.

Quantum machines
Living beings are machines. How do these machines work? From a physical point of view,
we need only a few sections of our walk so far to describe them: universal gravity and QED.
Simply stated, life is an electromagnetic process taking place in weak gravity.*But the
details of this statement are tricky and interesting. Table 59 gives an overview of motion

* In fact, also the nuclear interactions play some role for life: cosmic radiation is one source for random
mutations, which are so important in evolution. Plant growers often use radioactive sources to increase
Ref. 797 mutation rates. But obviously, radioactivity can also terminate life.

Dvipsbugw
biology 819

TA B L E 59 Motion and motors in living beings

Motion type Examples M a i n i n vo lv e d d e v i c e s

Growth collective molecular processes in cell ion pumps

growth
gene turn-on and turn-off linear molecular motors
ageing linear molecular motors
Construction materialtypes and properties (poly- material transport through muscles Dvipsbugw
saccharides, lipids, proteins, nucleic
acids, others)
forces and interactions between bio- cell membrane pumps
molecules
Functioning details of metabolism (respiration, di- muscles, ion pumps
gestion)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

energy flow in biomolecules
thermodynamics of whole living sys- muscles
tem and of its parts
muscle working linear molecular motors
nerve signalling ion motion, ion pumps
brain working ion pumps
illnesses cell motility, chemical pumps
viral infection of a cell rotational molecular motors for
RNA transport

Defence the immune system cell motility, linear molecular mo-

tors
Reproduction information storage and retrieval linear molecular motors inside cells,
sometimes rotational motors, as in
viruses
cell division linear molecular motors inside cells
sperm motion rotational molecular motors
courting muscles, brain, linear molecular mo-
tors
Copyright © Christoph Schiller November 1997–May 2006

evolution muscles, linear molecular motors

processes in living beings. Interestingly, all motion in living beings can be summarized
in a few classes by asking for the motor driving it.

The nuclear interactions are also implicitly involved in several other ways. They were necessary to form
the materials – carbon, oxygen, etc. – required for life. Nuclear interactions are behind the main mechanism
for the burning of the Sun, which provides the energy for plants, for humans and for all other living beings
(except a few bacteria in inaccessible places).
Summing up, the nuclear interactions play a role in the appearance and in the in destruction of life; but
they play no (known) role for the actions of particular living beings.

Dvipsbugw
 820 vii details of quantum theory • applied quantum mechanics

Nature only needs few small but powerful devices to realize all motion types used by
living beings. Given the long time that living systems have been around, these devices
are extremely efficient. In fact, ion pumps, chemical pumps, rotational and linear mo-
lecular motors are all specialized molecular motors. Ion and chemical pumps are found
in membranes and transport matter. Rotational and linear motor move structures against
membranes. In short, all motion in living beings is due to molecular motors. Even though
there is still a lot to be learned about them, what is known already is spectacular enough.

How do we move? – Molecular motors Dvipsbugw

How do our muscles work? What is the underlying motor? One of the beautiful results of
modern biology is the elucidation of this issue. It turns out that muscles work because they
contain molecules which change shape when supplied with energy. This shape change is
repeatable. A clever combination and repetition of these molecular shape changes is then
used to generate macroscopic motion. There are three basic classes of molecular motors:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

linear motors, rotational motors and pumps.
Linear motors are at the basis of muscle motion; other linear motors separate genes
during cell division. They also move organelles inside cells and displace cells through the
body during embryo growth, when wounds heal, or in other examples of cell motility. A
typical molecular motor consumes around 100 to 1000 ATP molecules per second, thus
about 10 to 100 aW. The numbers are small; however, we have to take into account that
Challenge 1287 n the power white noise of the surrounding water is 10 nW. In other words, in every mo-
lecular motor, the power of the environmental noise is eight to nine orders of magnitude
higher than the power consumed by the motor. The ratio shows what a fantastic piece of
machinery such a motor is.
Page 74 We encountered rotational motors already above. Nature uses them to rotate the cilia
Ref. 800 of many bacteria as well as sperm tails. Researchers have also discovered that evolution
produced molecular motors which turn around DNA helices like a motorized bolt would
turn around a screw. Such motors are attached at the end of some viruses and insert the
DNA into virus bodies when they are being built by infected cells, or extract the DNA from
Ref. 798 the virus after it has infected a cell. Another rotational motor, the smallest known so far
– 10 nm across and 8 nm high – is ATP synthase, a protein that synthesizes most ATP in
cells.
The ways molecules produce movement in linear motors was uncovered during the
Ref. 799 1990s. The results then started a wave of research on all other molecular motors found
Copyright © Christoph Schiller November 1997–May 2006

in nature. All molecular motors share a number of characteristic properties. There are
no temperature gradients involved, as in car engines, no electrical currents, as in elec-
trical motors, and no concentration gradients, as found in chemically induced motion.
The central part of linear molecular motors is a combination of two protein molecules,
namely myosin and actin. Myosin changes between two shapes and literally walks along
actin. It moves in regular small steps. The motion step size has been measured with beau-
tiful experiments to always be an integer multiple of 5.5 nm. A step, usually forward, but
Ref. 799 sometimes backwards, results whenever an ATP (adenosine triphosphate) molecule, the
standard biological fuel, hydrolyses to ADP (adenosine diphosphate), thus releasing its
energy. The force generated is about 3 to 4 pN; the steps can be repeated several times
a second. Muscle motion is the result of thousand of millions of such elementary steps

Dvipsbugw
 biology 821

Dvipsbugw

Figure missing

F I G U R E 328 Myosin and actin: the building bricks of a simple linear

molecular motor

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

taking place in concert.
How do molecular motors work? These motors are are so small that the noise due to
the molecules of the liquid around them is not negligible. But nature is smart: with two
tricks it takes advantage of Brownian motion and transforms it into macroscopic molecu-
lar motion. Molecular motors are therefore also called Brownian motors. The transform-
ation of disordered molecular motion into ordered macroscopic motion is one of the
great wonders of nature. The first trick of nature is the use of an asymmetric, but periodic
potential, a so-called ratchet.* The second trick of nature is a temporal variation of the po-
tential, together with an energy input to make it happen. The most important realizations
Ref. 801 are shown in Figure 329.
The periodic potential variation allows that for a short time the Brownian motion of
the moving molecule – typically 1 µm~s – affects its position. Then the molecule is fixed
again. In most of these short times of free motion, the position will not change. But if the
position does change, the intrinsic asymmetry of the ratchet shape ensures that in most
cases the molecule advances in the preferred direction. Then the molecule is fixed again,
waiting for the next potential change. On average, the myosin molecule will thus move in
one direction. Nowadays the motion of single molecules can be followed in special exper-
imental set-ups. These experiments confirm that muscles use such a ratchet mechanism.
Copyright © Christoph Schiller November 1997–May 2006

The ATP molecule adds energy to the system and triggers the potential variation through
the shape change it induces in the myosin molecule. That is how our muscles work.
Another well-studied linear molecular motor is the kinesin–microtubule system
which carries organelles from one place to the other within a cell. As in the previous
example, also in this case chemical energy is converted into unidirectional motion. Re-
searchers were able to attach small silica beads to single molecules and to follow their
motion. Using laser beams, they could even apply forces to these single molecules. Kin-
esin was found to move with around 800 nm~s, in steps lengths which are multiples of
8 nm, using one ATP molecule at a time, and exerting a force of about 6 pN.

* It was named by Walt Disney after by Ratchet Gearloose, the famous inventor from Duckburg.

Dvipsbugw
 822 vii details of quantum theory • applied quantum mechanics

Fixed position

U(t 1 )

U(t 2 )
Brownian motion
can take place Dvipsbugw

U(t 3 )

Most probable next fixed position

if particle moved

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 329 Two types of Brownian motors: switching potential (left) and tilting potential
(right)

Quantum ratchets also exist as human built systems, such as electrical ratchets for elec-
tron motion or optical ratchets that drive small particles. Extensive experimental research
is going on in the field.

Curiosities and fun challenges about biology

The physics of life is still not fully explored yet.
**
Challenge 1288 n How would you determine which of two identical twins is the father of a baby?

**
Can you give at least five arguments to show that a human clone, if there will ever be one,
Challenge 1289 n is a completely different person that the original? In fact, the first cloned cat, born in 2002,
looked completely different from the ‘original’ (in fact, its mother). The fur colour and its
patch pattern were completely different from that of the mother. Analogously, identical
Copyright © Christoph Schiller November 1997–May 2006

human twins have different finger prints, iris scans, blood vessel networks, intrauterine
experiences, among others.

**
Many molecules found in living beings, such as sugar, have mirror molecules. However,
in all living beings only one of the two sorts is found. Life is intrinsically asymmetric.
Challenge 1290 n How can this be?
**
How is it possible that the genetic difference between man and chimpanzee is regularly
given as about 1 %, whereas the difference between man and woman is one chromosome

Dvipsbugw
 biology 823

TA B L E 60 Approximate number of living species

Life group Described species E s t i m at e d s p e c i e s

min. max.
Viruses 4 000 50 ë 103 1 ë 106
Prokaryotes (‘bacteria’) 4000 50 ë 103 3 ë 106
Fungi 72 000 200 ë 103 2.7 ë 106
Protozoa 40 000 60 ë 103 200 ë 103
Algae 40 000 150 ë 103 1 ë 106 Dvipsbugw
Plants 270 000 300 ë 103 500 ë 103
Nematodes 25 000 100 ë 103 1 ë 106
Crustaceans 40 000 75 ë 103 200 ë 103
Arachnids 75 000 300 ë 103 1 ë 106
Insects 950 000 2 ë 106 100 ë 106
Molluscs 70 000 100 ë 103 200 ë 103

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Vertebrates 45 000 50 ë 103 55 ë 103
Others 115 000 200 ë 103 800 ë 103
Total 1.75 ë 106 3.6 ë 106 112 ë 106

Challenge 1291 n in 46, in other words, about 2.2 %?

**
What is the longest time a single bacterium has survived? Not 5000 years as the bacteria
found in Egyptian mummies, not even 25 million years as the bacteria resurrected from
the intestines in insects enclosed in amber. It is much longer, namely an astonishing 250
million years. This is the time that bacteria discovered in the 1960s by Hans Pflug in
(low-radioactivity) salt deposits in Fulda (Germany) have hibernated there before being
brought back to life in the laboratory. The result has been recently confirmed by the dis-
covery of another bacterium in a North-American salt deposit in the Salado formation.
**
Copyright © Christoph Schiller November 1997–May 2006

In 1967, a TV camera was deposited on the Moon. Unknown to everybody, it contained

a small patch of Streptococcus mitis. Three years later, the camera was brought back to
Ref. 410 Earth. The bacteria were still alive. They had survived for three years without food, water
or air. Life can be resilient indeed.
**
In biology classifications are extremely useful. This is in full contrast to the situation in
Ref. 802 physics. Table 60 gives an overview of the magnitude of the task. This wealth of material
can be summarized in one graph, shown in Figure 330.
Newer research seems to suggest some slight changes to the picture. So far however,
there still is only a single root to the tree.

Dvipsbugw
 824 vii details of quantum theory • applied quantum mechanics

Bacteria Archaea Eucarya

Green
non-sulfur Animals Ciliates
bacteria Green plants
Methano-
Gram-positive microbiales Fungi
Methano- extreme
bacteria bacteriales Flagellates
Halophiles
Purple bacteria Thermo-
proteus
Cyanobacteria Microsporidia
Flavobacteria Pyro-
dictum Methanococcales
and relatives Thermococcales Dvipsbugw

Thermotogales

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 330 A modern version of the evolutionary tree

**
Challenge 1292 r How did life start?
**
Challenge 1293 n Could life have arrived to Earth from outer space?
**
Life is not a clearly defined concept. The definition used above, the ability to reproduce,
has its limits when applied to old animals, to a hand cut off by mistake, to sperm or to
ovules. It also gives problems when trying to apply it to single cells. Is the definition of
life as ‘self-determined motion in the service of reproduction’ more appropriate? Or is
Challenge 1294 ny the definition of living beings as ‘what is made of cells’ more precise?
**
Also growth is a type of motion. Some is extremely complex. Take the growth of acne.
It requires a lack of zinc, a weak immune system, several bacteria, as well as the help of
Copyright © Christoph Schiller November 1997–May 2006

Demodex brevis, a mite (a small insect) that lives in skin pores. With a size of 0.3 mm,
somewhat smaller than the full stop at the end of this sentence, this and other animals
living on the human face can be observed with the help of a strong magnifying glass.
Humans have many living beings on board. For example, humans need bacteria to live.
It is estimated that 90% of the bacteria in the human mouth alone are not known yet; only
500 species have been isolated so far. These useful bacteria help us as a defence against
the more dangerous species.
**
Mammals have a narrow operating temperature. In contrast to machines, humans func-
Challenge 1295 d tion only if the internal temperature is within a narrow range. Why? Does this require-

Dvipsbugw
 the physics of pleasure 825

ment also apply to extraterrestrials – provided they exist?

**
How did the first cell arise? This question is still open. However, researchers have found
several substances that spontaneously form closed membranes in water. Such substances
also form foams. It might well be that life formed in foam.

The physics of pleasure

Dvipsbugw
What is mind but motion in the intellectual

“ sphere?
Oscar Wilde (1854–1900) The Critic as Artist.

Pleasure is a quantum effect. The reason is simple. Pleasure comes from the senses. All ”
senses measure. And all measurements rely on quantum theory. The human body, like an
expensive car, is full of sensors. Evolution has build these sensors in such a way that they

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

trigger pleasure sensations whenever we do with our body what we are made for.
Of course, no scientist will admit that he studies pleasure. So he says that he studies
the senses. The field is fascinating and still evolving; here we can only have a quick tour
of the present knowledge.
Among the most astonishing aspects of our body sensors is their sensitivity. The ear is
so sensitive and at the same time so robust against large signals that the experts are still
studying how it works. No known sensor can cover an energy range of 1013 ; the detected
intensity ranges from 1 pW~m2 (some say 50 pW~m2 ) to 10 W~m2 , the corresponding air
pressure variations from 20 µPa to 60 Pa. The lowest intensity is that of a 20 W sound
source heard at a distance of 10 000 km, if no sound is lost in between.
Audible sound wavelengths span from 17 m (20 Hz) to 17 mm (20 kHz). In this range,
the ear is able to distinguish at least 1500 pitches with its 16 000 to 20 000 hair cells. But
the ear is also able to distinguish 400 from 401 Hz using a special pitch sharpening mech-
anism.
The eye is a position dependent photon detector. Each eye contains around 126 million
separate detectors on the retina. Their spatial density is the highest possible that makes
sense, given the diameter of the lens of the eye. They give the eye a resolving power of 1′
and the capacity to detect down to 60 incident photons in 0.15 s, or 4 absorbed photons
Page 586 in the same time. There are about 6 million less sensitive colour detectors, the cones,
whose distribution we have seen earlier on. The different chemicals in the three cone types
Copyright © Christoph Schiller November 1997–May 2006

(red, green, blue) lead to different sensor speeds; this can be checked with the simple test
Ref. 803 shown in Figure 331. The images of the eye are only sharp if the eye constantly moves in
small random motions. If this motion is stopped, for example with chemicals, the images
produced by the eye become unsharp.
The eye also contains 120 million highly sensitive general light intensity detectors, the
rods. This sensitivity difference is the reason that at night all cats are grey. Until recently,
human built light sensors with the same sensitivity as rods had to be helium cooled, be-
cause technology was not able to build sensors at room temperature as sensitive as the
human eye.
The touch sensors are distributed over the skin, with a surface density which varies
from one region to the other. It is lowest on the back and highest in the face and on the

Dvipsbugw
 826 vii details of quantum theory • applied quantum mechanics

Dvipsbugw

F I G U R E 331 The different speed of the eye’s colour sensors, the cones, lead
to a strange effect when this picture (in colour version) is shaken right to left
in weak light

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tongue. There are separate sensors for pressure, for deformation, for vibration, for tickling,
for heat, for coldness, and for pain. Some react proportionally to the stimulus intensity,
some differentially, giving signals only when the stimulus changes.
The taste mechanisms of tongue are only partially known. Since 2005, the tongue is
known to produces six taste signals* – sweet, salty, bitter, sour, proteic and fat – and the
mechanisms are just being unravelled. (The sense for proteic, also called umami, has been
Ref. 804 discovered in 1907, by Ikeda Kikunae; the sense for ‘fat’ has been discovered only in 2005.)
Democritus imagined that taste depends on the shape of atoms. Today it is known that
sweet taste is connected with certain shape of molecules. Despite all this, no sensor with
a distinguishing ability of the same degree as the tongue has yet been built by humans.

The nose has about 350 different smell receptors; through combinations it is estimated
Ref. 805 that it can smell about 10 000 different smells. Together with the five signals that the sense
of taste can produce, the nose also produces a vast range of taste sensations. It protects
against chemical poisons, such as smoke, and against biological poisons, such as faecal
matter. In contrast, artificial gas sensors exist only for a small range of gases. Good artifi-
cial taste and smell sensors would allow to check wine or cheese during their production,
Challenge 1297 r thus making its inventor extremely rich.
Copyright © Christoph Schiller November 1997–May 2006

The human body also contains orientation sensors in the ear, extension sensors in each
muscle, pain sensors almost all over the skin and inside the body, heat sensors and cold-
ness sensors on the skin and in other places. Other animals feature additional types of
Page 528 sensors. Sharks can feel electrical fields, snakes have sensors for infrared; both are used
to locate prey. Pigeons, trout and sharks can feel magnetic fields, and use this sense for
navigation. Many birds and certain insects can see UV light. Bats are able to hear ultra-
sound up to 100 kHz and more. Whales and elephants can detect and localize infrasound
signals.

* Taste sensitivity is not separated on the tongue into distinct regions; this is an incorrect idea that has been
copied from book to book for over a hundred years. You can perform a falsification by yourself, using sugar
Challenge 1296 n or salt grains.

Dvipsbugw
 the physics of pleasure 827

In summary, the sensors with which nature provides us are state of the art; their sens-
itivity and ease of use is the highest possible. Since all sensors trigger pleasure or help to
Ref. 806 avoid pain, nature obviously wants us to enjoy life with the most intense pleasure possible.
Studying physics is one way to do this.
There are two things that make life worth living:

“ Mozart and quantum mechanics.

Victor Weisskopf*

” Dvipsbugw
The nerves and the brain
There is no such thing as perpetual tranquillity

“ of mind while we live here; because life itself is

but motion, and can never be without desire,
nor without fear, no more than without sense.
Thomas Hobbes (1588–1679) Leviathan.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The main unit processing all these signals, the brain, is another of the great wonders of
nature. The human brain has the highest complexity of all brains known;** its processing
power and speed is orders of magnitude larger than any device build by man.
Page 212 We saw already how electrical signals from the sensors are transported into the brain.
In the brain, the arriving signals are classified and stored, sometimes for a short time,
sometimes for a long time. The details of the various storage mechanisms, essentially
taking place in the structure and the connection strength between brain cells, were elu-
cidated by modern neuroscience. The remaining issue is the process of classification. For
certain low level classifications, such as colours or geometrical shapes for the eye or sound
harmonies for the ear, the mechanisms are known. But for high-level classifications, such
as the ones used in conceptual thinking, the aim is not yet achieved. It is not well known
how to describe the processes of reading, understanding and talking in terms of signal
motions. Research is still in full swing and will remain so for the largest part of the twenty-
first century.
In the following we have a look at a few abilities of our brain, of our body and of other
bodies which are important for the study of motion.

Clocks in quantum mechanics

L’horologe fait de la réclame pour le temps.***

“ Georges Perros

”
Copyright © Christoph Schiller November 1997–May 2006

* Victor Friedrich Weisskopf (b. 1908 Vienna, d. 2002 Cambridge), acclaimed theoretical physicist who
worked with Einstein, Born, Bohr, Schrödinger and Pauli. He catalysed the development of quantum elec-
trodynamics and nuclear physics. He worked on the Manhattan project but later in life intensely campaigned
against the use of nuclear weapons. During the cold war he accepted the membership in the Soviet Academy
of Sciences. He was professor at MIT and for many years director of CERN, in Geneva. He wrote several suc-
cessful physics textbooks. The author heard him making the above statement in Geneva, in 1981, during
one of his lectures.
** This is not in contrast with the fact that one or two whale species have brains with a slightly larger mass.
The larger mass is due to the protection these brains require against the high pressures which appear when
whales dive (some dive to depths of 1 km). The number of neurons in whale brains is considerably smaller
than in human brains.
*** Clocks are ads for time.

Dvipsbugw
 828 vii details of quantum theory • applied quantum mechanics

Challenge 1298 n Most clocks used in everyday life are electromagnetic. (Do you know an exception?) Any
clock on the wall, be it mechanical, quartz controlled, radio or solar controlled, or of any
other type, is based on electromagnetic effects. There are even clocks of which we do not
even know how they work. Just look at singing. We know from everyday experience that
Ref. 807 humans are able to keep the beat to within a few per cent for a long time. Also when
we sing a musical note we reproduce the original frequency with high accuracy. In many
movements humans are able to keep time to high accuracy, e.g. when doing sport or when
dancing. (For shorter or longer times, the internal clocks are not so precise.) Dvipsbugw
In addition, all clocks are limited by quantum mechanics, including the simple pendu-
lum. Let us explore the topic.

Do clocks exist?
Die Zukunft war früher auch besser.*

“ Karl Valentin, German writer.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In general relativity, we found that clocks do not exist, because there is no unit of time
that can be formed using the constants c and G. Clocks, like any measurement standard,
Page 492 need matter and non-gravitational interactions to work. This is the domain of quantum
theory. Let us see what the situation is in this case.
First of all, the time operator, or any operator proportional to it, is not an observable.
Indeed, the time operator is not Hermitean, as any observable must be. In other words,
there is no physical observable whose value is proportional to time. On the other hand,
clocks are quite common; for example, the Sun or Big Ben work to everybody’s satisfac-
tion. Nature thus encourages us to look for an operator describing the position of the
hands of a clock. However, if we look for such an operator we find a strange result. Any
quantum system having a Hamiltonian bounded from below – having a lowest energy –
lacks a Hermitean operator whose expectation value increases monotonically with time.
Challenge 1299 d This result can be proven rigorously. In other words, quantum theory states that time
cannot be measured.
That time cannot be measured is not really a surprise. The meaning of this statement is
that every clock needs to be wound up after a while. Take a mechanical pendulum clock.
Only if the weight driving it can fall forever, without reaching a bottom position, can the
clock go on working. However, in all clocks the weight has to stop when the chain end is
reached or when the battery is empty. In other words, in all real clocks the Hamiltonian
Copyright © Christoph Schiller November 1997–May 2006

is bounded from below.

In short, quantum theory says that any clock can only be approximate. Quantum the-
ory shows that exact clocks do not exist in nature. Obviously, this result can be of import-
ance only for high precision clocks. What happens if we try to increase the precision of a
clock as much as possible?
High precision implies high sensitivity to fluctuations. Now, all clocks have a motor
inside that makes them work. A high precision clock thus needs a high precision motor.
In all clocks, the position of the motor is read out and shown on the dial. The quantum

* Also the future used to be better in the past.

Dvipsbugw
 the physics of pleasure 829

of action implies that a precise clock motor has a position indeterminacy. The clock pre-
cision is thus limited. Worse, like any quantum system, the motor has a small, but finite
probability to stop or to run backwards for a while.
You can check this prediction yourself. Just have a look at a clock when its battery is
almost empty, or when the weight driving the pendulum has almost reached the bottom
position. It will start doing funny things, like going backwards a bit or jumping back
and forward. When the clock works normally, this behaviour is only strongly reduced in
amount; however, it is still possible, though with low probability. This is true even for a
Challenge 1300 ny sundial. Dvipsbugw
In other words, clocks necessarily have to be macroscopic in order to work properly.
A clock must be large as possible, in order to average out its fluctuations. Astronomical
systems are examples. A good clock must also be well-isolated from the environment,
such as a freely flying object whose coordinate is used as time variable, as is done in
certain optical clocks.
How big is the problem we have thus discovered? What is the actual error we make

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

when using clocks? Given the various limitations due to quantum theory, what is the
ultimate precision of a clock?
To start with, the indeterminacy relation provides the limit that the mass M of a clock
Challenge 1301 ny must be larger than
MA 2
ħ
(574)
c τ
Challenge 1302 e which is obviously always fulfilled in everyday life. But we can do better. Like for a pendu-
lum, we can relate the accuracy τ of the clock to its maximum reading time T. The idea
Ref. 808 was first published by Salecker and Wigner. They argued that

MA
ħ T
(575)
c2 τ τ
where T is the time to be measured. You might check that this directly requires that any
Challenge 1303 e clock must be macroscopic.
Let us play with the formula by Salecker and Wigner. One way to rephrase it is the
following. They showed that for a clock which can measure a time t, the size l is connected
to the mass m by ¾
lA
ħt
. (576)
Copyright © Christoph Schiller November 1997–May 2006

Ref. 810 How close can this limit be achieved? It turns out that the smallest clocks known, as well
as the clocks with most closely approach this limit are bacteria. The smallest bacteria,
the mycoplasmas, have a mass of about 8 ë 10−17 kg, and reproduce every 100 min, with a
precision of about 1 min. The size predicted from expression (576) is between 0.09 µm and
0.009 µm. The observed size of the smallest mycoplasmas is 0.3 µm. The fact that bacteria
can come so close to the clock limit shows us again what a good engineer evolution has
been.
Note that the requirement by Salecker and Wigner is not in contrast with the possibil-
ity to make the oscillator of the clock very small; people have built oscillators made of a

Dvipsbugw
 830 vii details of quantum theory • applied quantum mechanics

Ref. 809 single atom. In fact, such oscillations promise to be the most precise human built clocks.
In the real world, the expression can be stated even more strictly. The whole mass M
cannot be used in the above limit. For clocks made of atoms, only the binding energy
between atoms can be used. This leads to the so-called standard quantum limit for clocks;
it limits their frequency ν by ¾
=
δν ∆E
(577)
ν E tot

where ∆E = ħ~T is the energy indeterminacy stemming from the finite measuring time
Dvipsbugw
T and E tot = NE bind is the total binding energy of the atoms in the metre bar. However,
the quantum limit has not been achieved for clocks, even though experiments are getting
near to it.
In summary, clocks exist only in the limit of ħ being negligible. In practice, the errors
made by using clocks and metre bars can be made as small as required; it suffices to make
the clocks large enough. We can thus continue our investigation into the details of matter

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

without much worry. Only in the third part of our mountain ascent, where the precision
requirements will be higher and general relativity will limit the size of physical systems,
things will get much more interesting: the impossibility to build clocks will then become
a central issue.

Living clocks
Among many things, living beings process information. Also computers do this, and like
computers, all living beings need a clock to work well. Every clock needs is made up of
the same components. It needs an oscillator determining the rhythm and a mechanism to
feed the oscillator with energy. A clock also needs an oscillation counter, i.e. a mechanism
that reads out the clock signal; a means of signal distribution throughout the system is
required, synchronizing the processes attached to it. In addition, a clock needs a reset
mechanism. If the clock has to cover many time scales, it needs several oscillators with
different oscillation frequencies and a way to reset their relative phases.
Even though physicists know the details of technical clock building fairly well, we
still do not know many parts of biological clocks. Most oscillators are chemical systems;
Ref. 812 some, like the heart muscle or the timers in the brain, are electrical systems. The general
elucidation of chemical oscillators is due to Ilya Prigogine; it has earned him a Nobel Prize
for chemistry in 1977. But not all the chemical oscillators in the human body are known
Copyright © Christoph Schiller November 1997–May 2006

yet, not to speak of the counter mechanisms. For example, a 24–minute cycle inside each
human cell has been discovered only in 2003, and the oscillation mechanism is not yet
fully clear. (It is known that a cell fed with heavy water ticks with 27–minute instead of
Ref. 811 24–minute rhythm.) It might be that the daily rhythm, the circadian clock, is made up
of or reset by 60 of these 24–minute cycles, triggered by some master cells in the human
body. The clock reset mechanism for the circadian clock is also known to be triggered by
daylight; the cells in the eye who perform this have been pinpointed only in 2002. The
light signal is processed by the superchiasmatic nucleus, two dedicated structures in the
brain’s hypothalamus. The various cells in the human body react differently depending
on the phase of this clock.
The clock with the longest cycle in the human body controls ageing. A central mech-

Dvipsbugw
 the physics of pleasure 831

anism for this clock seems to be the number of certain molecules attached to the DNA of
the human chromosomes. At every division, one molecule is lost. When the molecules
are all lost, the cell dies. Research into the mechanisms and the exceptions to this process
(cancer cells, sexual cells) is ongoing.
The basis of the monthly period in women is equally interesting and complex.
The most fascinating clocks are those at the basis of conscious time. Of these, the
brain’s stopwatch or interval timer, has been most intensely studied. Only recently was
its mechanism uncovered by combining data on human illnesses, human lesions, mag-
Ref. 813 netic resonance studies, and effects of specific drugs. The basic mechanism takes place in Dvipsbugw
the striatum in the basal ganglia of the brain. The striatum contains thousands of timer
cells with different periods. They can be triggered by a ‘start’ signal. Due to their large
number, for small times of the order of one second, every time interval has a different
pattern across these cells. The brain can read these patterns and learn them. In this way
we can time music or specific tasks to be performed, for example, one second after a
signal.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Metre sticks
For length measurements, the situations is similar to that for time measurements. The
limit by Salecker and Wigner can also be rewritten for length measurement devices. Are
Challenge 1304 ny you able to do it?
In general relativity we found that we need matter for any length measurement.
Quantum theory, our description of matter, again shows that metre sticks are only ap-
proximately possible, but with errors which are negligible if the device is macroscopic.

Why are predictions so difficult, especially of the future?

Future: that period of time in which our affairs

“ prosper, our friends are true, and our

happpiness is assured.
Ambrose Bierce

If due to the quantum of action perfect clocks do not exist, is determinism still the correct
description of nature?
We have seen that predictions of the future are made difficult by nonlinearities and the
”
divergence of from similar conditions; we have seen that many particles make it difficult
to predict the future due to the statistical nature of their initial conditions; we have seen
Copyright © Christoph Schiller November 1997–May 2006

that quantum theory makes it often hard to fully determine initial states; we have seen
that not-trivial space-time topology can limit predictability; finally, we will discover that
black hole and similar horizons can limit predictability due to their one-way transmission
of energy, mass and signals.
Nevertheless, we also learned that all these limitations can be overcome for limited
time intervals; in practice, these time intervals can be made so large that the limitations
do not play a role in everyday life. In summary, in quantum theory both determinism and
time remain applicable, as long as we do not extend it to infinite space and time. When
extremely large dimensions and intervals need to be taken into account, quantum theory
cannot be applied alone; in those cases, general relativity needs to be taken into account.

Dvipsbugw
 832 vii details of quantum theory • applied quantum mechanics

Decay and the golden rule

I prefer most of all to remember the future.

“
The decoherence of superposition of macroscopically distinct states plays an important
Salvador Dalì

role in another common process: the decay of unstable systems or particles. Decay is any
”
spontaneous change. Like the wave aspect of matter, decay is a process with no classical
counterpart. True, decay, including the ageing of humans, can be followed in classical Dvipsbugw
physics; however, its origin is a pure quantum effect.
Experiments show that the prediction of decay, like that of scattering of particles, is
only possible on average, for a large number of particles, never for a single one. These res-
ults confirm the quantum origin of the process. In every decay process, the superposition
of macroscopically distinct states, which in this case are those of a decayed and an unde-
cayed particle, is made to decohere rapidly by the interaction with the environment, and
the ‘environment’ vacuum is sufficient to induce the decoherence. As usual, the details of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the states involved are unknown for a single system and make any prediction for a single
system impossible.
Decay is influenced by the environment, even in the case that it is ‘only’ the vacuum.
This is true for all systems, including radioactive nuclei. The statement can be confirmed
by experiment. By enclosing a part of space between two conducting plates, one can
change the degrees of freedom of the vacuum contained between them. Putting an elec-
tromagnetically unstable particle, such as an excited atom, between the plates, indeed
Ref. 814 changes the lifetime of the particle. Can you explain why this method is not useful to
Challenge 1305 ny lengthen the lifespan of humans?
What is the origin of decay? Decay is always due to tunnelling. With the language
of quantum electrodynamics, we can rephrase the answer: decay is motion induced by
the vacuum fluctuations. Vacuum fluctuations are random. The experiment between the
plates confirms the importance of the environment fluctuations for the decay process.
For a system consisting of a large number N of identical particles, the decay is de-
scribed by

Ṅ = −
N
where
τ
= S`ψ final SH int Sψ final eS .
1 2π 2
(578)
τ ħ
Copyright © Christoph Schiller November 1997–May 2006

The decay is thus essentially an exponential one, independently of the details of the
physical process. In addition, the decay time τ depends on the interaction and on the
square modulus of the transition matrix element. This result was named the golden rule
by Fermi,* because it works so well despite being an approximation whose domain of
applicability is not easy to specify.
In practice, decay follows an exponential law. Experiments failed to see a deviation
from this behaviour for over half a century. On the other hand, quantum theory shows

* Originally, the golden rule is an expression from the christian bible, namely the sentence ‘Do to others
what you want them to do to you’.

Dvipsbugw
 the physics of pleasure 833

Ref. 815 that decay is exponential only in certain special systems. A calculation that takes into
account higher order terms predicts deviations from exponential decay for completely
isolated systems: for short times, the decay rate should vanish; for long times, the de-
cay rate should follow an algebraic – not an exponential – dependence on time, in some
Ref. 816 cases even with superimposed oscillations. Only after an intense experimental search de-
viations for short times have finally been observed. The observation of deviations at long
times are rendered impossible by the ubiquity of thermal noise. Theory shows that the ex-
ponential decay so regularly found in nature results only when the environment is noisy,
the system made of many particles, or both. Since this is usually the case, the exceptional Dvipsbugw
exponential decay becomes the (golden) rule in usual observations.

Z eno and the present in quantum theory

Utere tempore.*

“ Ovidius

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

As shown by perception research, what humans call ‘present’ has a duration of a few tenths
Ref. 817 of a second. This leads us to ask whether the physical present might have a duration as
well.
Every observation, like every photograph, implies a time average: observations average
interactions over a given time. For a photograph, the duration is given by the shutter time;
for a measurement, the average is defined by the set-up used. Whatever this set-up might
be, the averaging time is never zero.
We thus need to ask whether the result of an observation will change if the observation
time is shortened as much as possible, or if the observations will simply approach some
limit situation. In everyday life, we are used to imagine that shortening the time taken
to measure the position of a point object as much as possible will approach the ideal of
a particle fixed at a given point in space. When Zeno discussed flight of an arrow, he
assumed that this is possible.
Quantum theory has brought us so many surprises that the question should be stud-
ied carefully. We already know that the quantum of action makes rest an impossibility.
However, the issue here is different: we are asking whether we can say that a system is
at a given spot at a given time. In order to determine this, we could use a photographic
camera whose shutter time can be reduced at will. What would we find? When the shut-
ter time approaches the oscillation period of light, the sharpness of the image would de-
Copyright © Christoph Schiller November 1997–May 2006

crease; in addition, the colour of the light would be influenced by the shutter motion. We
can increase the energy of the light used, but the smaller wavelengths only shift the prob-
lem. At extremely small wavelengths, matter becomes transparent, and shutters cannot be
realized any more. Quantum theory does not confirm the naive expectation that shorter
shutter times lead to sharper images. In other words, the quantum aspects of the world
show us that there is no way in principle to approach the limit that Zeno was discussing.
Whenever one reduces shutter times as much as possible, observations become unsharp.
This counter-intuitive result is due to the quantum of action: through the indeterminacy
relation, the smallest action prevents that moving objects are at a fixed position at a given

* ‘Use your time.’ Tristia 4, 3, 83

Dvipsbugw
 834 vii details of quantum theory • applied quantum mechanics

time. Zeno’s discussion was based on an extrapolation of classical physics into domains
where it is not valid any more. There is no ‘point-like’ instant of time that describes the
present. The present is always an average over a non-vanishing interval of time.

What is motion?
Zeno was thus wrong in assuming that motion is a sequence of specific positions in space.
Quantum theory implies that motion is not the change of position with time. The invest-
igation of the issue showed that this statement is only an approximation for low energies Dvipsbugw
or for long observation times.
How then can we describe motion in quantum theory? Quantum theory shows that
motion is the low energy approximation of quantum evolution. Quantum evolution as-
sumes that space and time measurements of sufficient precision can be performed. We
know that for any given observation energy, we can build clocks and metre bars with
much higher accuracy than required, so that quantum evolution is applicable in all cases.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Motion is an approximation of quantum evolution.
Obviously, this pragmatic description of motion rests on the assumption that for any
observation energy we can find a still higher energy used by the measurement instru-
ments to define space and time. We deduce that if a highest energy would exist in nature,
we would get into big trouble, as quantum theory would then break down. As long as en-
ergy has no limits, all problems are avoided, and motion remains a sequence of quantum
observables or quantum states, whichever you prefer.
The assumption of energy without limit works extremely well; it lies at the basis of the
whole second part of the mountain ascent, even though it is rather hidden. In the third
and final part, we will discover that there indeed is a maximum energy in nature, so that
we will need to change our approach. However, this energy value is so huge that it does
not bother us at all at this point of our exploration. But it will do so later on.

– CS – Several sections, on time and the quantum Zeno effect, to be added – CS –

Consciousness – a result of the quantum of action

Page 691 Consciousness is our ability to observe what is going on in our mind. This activity, like any
type of change, can itself be observed and studied. Obviously, consciousness takes place
in the brain. If it were not, there would be no way to keep it connected with a given person.
Copyright © Christoph Schiller November 1997–May 2006

We know that each brain moves with over one million kilometres per hour through the
cosmic background radiation; we also observe that consciousness moves along with it.
The brain is a quantum system; it is based on molecules and electrical currents. The
changes in consciousness that appear when matter is taken away from the brain – in
operations or accidents – or when currents are injected into the brain – in accidents,
experiments or misguided treatments – have been described in great detail by the medical
profession. Also the observed influence of chemicals on the brain – from alcohol to hard
drugs – makes the same point. The brain is a quantum system.
Magnetic resonance imaging can detect which parts of the brain work when sensing,
remembering or thinking. Not only is sight, noise and thought processed in the brain;
we can follow the processing on computer screens. The other, more questionable experi-

Dvipsbugw
 the physics of pleasure 835

mental method, positron tomography, works by letting people swallow radioactive sugar.
It confirms the findings on the location of thought and on its dependence on chemical
fuel. In addition, we already know that memory depends on the particle nature of matter.
All these observations depend on the quantum of action.
Not only the consciousness of others, also your own consciousness is a quantum pro-
Challenge 1306 ny cess. Can you give some arguments?
In short, we know that thought and consciousness are examples of motion. We are thus
in the same situation as material scientists were before quantum theory: they knew that
electromagnetic fields influence matter, but they could not say how electromagnetism Dvipsbugw
was involved in the build-up of matter. We know that consciousness is made from the
signal propagation and signal processing in the brain; we know that consciousness is an
electrochemical process. But we do not know yet the details of how the signals make up
consciousness. Unravelling the workings of this fascinating quantum system is the aim
of neurological science. This is one of the great challenges of twenty-first century science.
It is sometimes claimed that consciousness is not a physical process. Every expert of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

motion should be able to convincingly show the opposite, even though the details are not
Challenge 1307 ny clear yet. Can you add arguments to the ones given here?

Why can we observe motion?

Studying nature can be one of the most intense pleasures of life. All pleasure is based
on the ability to observe motion. Our human condition is central to this ability. In our
adventure so far we found that we experience motion only because we are of finite size,
only because we are made of a large but finite number of atoms, only because we have
a finite but moderate temperature, only because we are a mixture of liquids and solids,
only because we are electrically neutral, only because we are large compared to a black
hole of our same mass, only because we are large compared to our quantum mechanical
wavelength, only because we have a limited memory, only because our brain forces us to
approximate space and time by continuous entities, and only because our brain cannot
avoid describing nature as made of different parts. If any of these conditions were not
fulfilled we would not observe motion; we would have no fun studying physics.
In addition, we saw that we have these abilities only because our forefathers lived on
Earth, only because life evolved here, only because we live in a relatively quiet region of
our galaxy, and only because the human species evolved long after than the big bang.
If any of these conditions were not fulfilled, or if we were not humans (or animals),
Copyright © Christoph Schiller November 1997–May 2006

motion would not exist. In many ways motion is thus an illusion, as Zeno of Elea had
claimed. To say the least, the observation of motion is due to the limitations of the human
condition. A complete description of motion and nature must take this connection into
account. Before we do that, we explore a few details of this connection.

Curiosities and fun challenges about quantum experience

The fascination of quantum effects is still lasting, despite over 100 years of intense studies.
**
Challenge 1308 n Are ghost images in TV sets, often due to spurious reflections, examples of interference?

Dvipsbugw
 836 vii details of quantum theory • applied quantum mechanics

**
Challenge 1309 r What happens when two monochromatic electrons overlap?

**
The sense of smell is quite complex. For example, the substance that smells most badly
to humans is skatole or 3-methylindole. This is the molecule to which the human nose is
most sensitive. Skatole makes faeces smell bad; it is a result of haemoglobin entering the
digestive tract through the bile. In contrast to humans, skatole attracts flies; it is also used
by some plants for the same reason. Dvipsbugw
On the other hand, small levels of skatole do not smell bad to humans. It is also used
by the food industry in small quantities to give smell and taste to vanilla ice cream.
**
It is worth noting that human senses detect energies of quite different magnitudes. The
eyes can detect light energies of about 1 aJ, whereas the sense of touch can detect only

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1310 n energies as large as about 10 µJ. Is one of the two systems relativistic?

**
Compared to all primates, the human eye is special: it is white, thus allowing others to see
Ref. 818 the direction in which one looks. Comparison with primates shows that the white colour
has evolved to allow more communication between individuals.
**
The high sensitivity of the ear can be used to hear light. To do this, take an empty 750 ml
jam glass. Keeping its axis horizontal, blacken the upper half of the inside with a candle.
The lower half should remain transparent. After doing this, close the jam glass with its
lid, and drill a 2 to 3 mm hole into it. If you now hold the closed jam glass with the hole
to your ear, keeping the black side up, and shining into it from below with a 50 W light
Challenge 1311 n bulb, something strange happens: you hear a 100 Hz sound. Why?

**
Most senses work already before birth. It is well-known that playing the violin to a preg-
nant mother every day during the pregnancy has an interesting effect. Even if nothing is
told about it to the child, it will become a violin player later on. Copyright © Christoph Schiller November 1997–May 2006

**
There is ample evidence that not using the senses is damaging. People have studied what
happens when in the first years of life the vestibular sense – the one used for motion
detection and balance restoration – is not used enough. Lack of rocking is extremely
hard to compensate later in life. Also dangerous is the lack of use of the sense of touch.
Babies, like all small mammals, that are generally and systematically deprived of these
Ref. 819 experiences tend to violent behaviour during the rest of their life.
**
Nature has indeed invented pleasure as a guide for a human behaviour. All of biology
builds on chemistry and on materials science. Let’s have a short overview of both fields.

Dvipsbugw
 chemistry – from atoms to dna 837

Figure to be included

F I G U R E 332 Atoms and dangling Dvipsbugw

bonds

Chemistry – from atoms to DNA

It is an old truth that Schrödinger’s equation contains all of chemistry.* With quantum
theory, for the first time people were able to calculate the strengths of chemical bonds,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

and what is more important, the angle between them. Quantum theory thus explains the
shape of molecules and thus indirectly, the shape of all matter.
To understand molecules, the first step is to understand atoms. The early quantum
theorists, above all Niels Bohr, spent a lot of energy in understanding their structure. The
main result is that atoms are structured, though spherical electron clouds.
After more than thirty years of work by the brightest physicists in Göttingen and
Copenhagen, it was found that electrons in atoms form various layers around the cent-
ral nucleus. The layers are numbered from the inside by a number called the principal
quantum number, usually written n.
Quantum theory shows that the first layer has room for two electrons, the second for
8, the third for 18 and the general n-th shell for 2n 2 electrons. A way to picture this con-
Ref. 828 nection is shown in Figure 333. It is called the periodic table of the elements. The standard
Page 1181 way to show the table is shown in Appendix C.

– CS – more to be added – CS –

When atoms approach each other, they can form one or several bonds. The preferred
distance of these bonds, the angles between them, are due to the structure of the atomic
electron clouds.
Do you remember those funny pictures of school chemistry about orbitals and
dangling bonds? Well, dangling bonds can now be seen. Several groups were able to image
Copyright © Christoph Schiller November 1997–May 2006

Ref. 829
Ref. 830 them using scanning force or scanning tunnelling microscopes.
º
The angles between the bonds explain why the angle of tetrahedral skeletons
Challenge 1312 e (2arctan 2 = 109.47°) are so common in molecules. For example, the H-O-H angle in
water molecules is 107°.
At the centre of each atom cloud is the nucleus, which contains almost all the atomic
mass. The nucleus consists of protons and neutrons. The structure of nuclei is even more
Page 897 complex than that of electron clouds. We explore it in a separate chapter later on.

Ref. 820 * The precise statement is: the Dirac equation contains all of chemistry. The relativistic effects that distinguish
the two equations are necessary, for example, to understand why gold is yellow and does not like to react or
why mercury is liquid.

Dvipsbugw
 838 vii details of quantum theory • applied quantum mechanics

n=1 1H 2 He

7N 8O
n=2 6C 3 Li 4 Be 9F
5B 10 Ne

25 Mn 26 Fe
24 Cr 15 P 16 S 27 Co
n=3 23 V 14 Si 11 Na 12 Mg 17 Cl 28 Ni
22 Ti 13 Al 18 Ar 29 Cu
21 Sc 30 Zn Dvipsbugw
64 Gd 65 Tb
n=4 63 Eu 43 Tc 44 Ru 66 Dy
62 Sm 42 Mo 33 As 34 Se 45 Rh 67 Ho
61 Pm 41 Nb 32 Ge 19 K 20 Ca 35 Br 46 Pd 68 Er
60 Nd 40 Zr 31 Ga 36 Kr 47 Ag 69 Tm
59 Pr 39 Y 48 Cd 70 Yb
58 Ce 71 Lu

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

96 Cm 97 Bk
n=5 95 Am 75 Re 76 Os 98 Cf
94 Pu 74 W 51 Sb 52 Te 77 Ir 99 Es
93 Np 73 Ta 50 Sn 37 Rb 38 Sr 53 I 78 Pt 100 Fm
92 U 72 Hf 49 In 54 Xe 79 Au 101 Md
91 Pa 57 La 80 Hg 102 No
90 Th 103 Lr

107 Bh 108 Hs
106 Sg 83 Bi 84 Po 109 Mt
n=6 105 Ha 82 Pb 55 Cs 56 Ba 85 At 110 Ds
104 Rf 81 Tl 86 Rn 111 Uuu
89 Ac 112 Uub

115 Uup 116 Uuh

n=7 114 Uuq 87 Fr 88 Ra 117 Uus
113 Uut 118 Uuo

n=8 119 Uun 120 Udn

s p d f
shell electron number
Copyright © Christoph Schiller November 1997–May 2006

shell of last electron increases clockwise

F I G U R E 333 An unusual form of the periodic table of the elements

Ribonucleic acid and Deox yribonucleic acid

Probably the most fascinating molecule is human deoxyribonucleic acid. The nucleic
acids where discovered in 1869 by the Swiss physician Friedrich Miescher (1844–1895) in
white blood cells. In 1874 he published an important study showing that the molecule is
contained in spermatozoa, and discusses the question if this substance could be related

Dvipsbugw
 chemistry – from atoms to dna 839

Figure missing

F I G U R E 334 Several ways to picture DNA

Dvipsbugw

to heredity. With his work, Miescher paved the way to a research field that earned many
colleagues Nobel Prizes (though not for himself).
DNA is a polymer, as shown in Figure 334, and is among the longest molecules known.
Human DNA molecules, for example, can be up to 10 cm in length. It consists of a double

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

helix of sugar derivates, to which four nuclei acids are attached in irregular order.
At the start of the twentieth century it became clear that Desoxyribonukleinsäure
(DNS) or deoxyribonucleic acid (DNA) in English – was precisely what Erwin Schrödinger
had predicted to exist in his book What Is Life? As central part of the chromosomes con-
tained the cell nuclei, DNA is responsible for the storage and reproduction of the informa-
tion on the construction and functioning of Eukaryotes. The information is coded in the
ordering of the four nucleic acids. DNA is the carrier of hereditary information. DNA de-
termines in great part how the single cell we all once have been grows into the complex
human machine we are as adults. For example, DNA determines the hair colour, pre-
disposes for certain illnesses, determines the maximum size one can grow to, and much
more. Of all known molecules, human DNA is thus most intimately related to human ex-
istence. The large size of the molecules is the reason that understanding its full structure
and its full contents is a task that will occupy scientists for several generations to come.

Curiosities and fun challenges about chemistry

**
Muscles produce motion through electrical stimulation. Can technical systems do the
same? There is a candidate. So-called electroactive polymers change shape when they are
activated with electrical current or with chemicals. They are lightweight, quiet and simple
to manufacture. However, the first arm wrestling contest between human and artificial
Copyright © Christoph Schiller November 1997–May 2006

muscles held in 2005 was won by a teenage girl. The race to do better is ongoing.

**
A cube of sugar does not burn. However, if you put some cigarette ash on top of it, it
burns. Why?
**
Challenge 1313 ny Why do organic materials burn at much lower temperature than inorganic materials?

**
An important aspect of life is death. When we die, conserved quantities like our energy,

Dvipsbugw
 840 vii details of quantum theory • applied quantum mechanics

momentum, angular momentum and several other quantum numbers are redistributed.
They are redistributed because conservation means that nothing is lost. What does all
Challenge 1314 ny this imply for what happens after death?
**
Chemical reactions can be slow but still dangerous. Spilling mercury on aluminium will
lead to an amalgam that reduces the strength of the aluminium part after some time. That
is the reason that bringing mercury thermometers on aeroplanes is strictly forbidden.
Dvipsbugw
Materials science
Did you know that one cannot use a boiled egg

“ as a toothpick?
Karl Valentin

It was mentioned several times that the quantum of action explains all properties of mat-
”

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ter. Many researchers from physics, chemistry, metallurgy, engineering, mathematics and
biology have cooperated in the proof of this statement. In our mountain ascent we have
little time to explore this vast topic. Let us walk through a selection.

Why does the floor not fall?

We do not fall through the mountain we are walking on. Some interaction keeps us from
falling through. In turn, the continents keep the mountains from falling through them.
Also the liquid magma in the Earth’s interior keeps the continents from sinking. All
these statements can be summarized. Atoms do not penetrate each other. Despite being
mostly empty clouds, atoms keep a distance. All this is due to the Pauli principle between
Page 785 electrons. the fermion character of electrons avoids that atoms interpenetrate. At least on
Earth.
Not all floors keep up due to the fermion character of electrons. Atoms are not impen-
etrable at all pressures. They can collapse, and form new types of floors. Some floors are
so exciting to study that people have spent their whole life to understand why they do not
fall, or when they do, how it happens: the surfaces of stars.
In most stars, the radiation pressure of the light plays only a minor role. Light pressure
does play a role in determining the size of red giants, such as Betelgeuse; but for average
stars, light pressure is negligible.
In most stars, such as in the Sun, the gas pressure takes the role which the incompress-
Copyright © Christoph Schiller November 1997–May 2006

ibility of solids and liquids has for planets. The pressure is due to the heat produced by
the nuclear reactions.
The next star type appears whenever light pressure, gas pressure and the electronic
Pauli pressure cannot keep atoms from interpenetrating. In that case, atoms are com-
pressed until all electrons are pushed into the protons. Protons then become neut-
rons, and the whole star has the same mass density of atomic nuclei, namely about
2.3 ë 1017 kg~m3 . A drop weighs about 200 000 tons. In these so-called neutron stars, the
floor – or better, the size – is also determined by Pauli pressure; however, it is the Pauli
Page 911 pressure between neutrons, triggered by the nuclear interactions. These neutron stars are
all around 10 km in radius.

Dvipsbugw
 materials science 841

TA B L E 61 The types of rocks and stones

Ty pe Properties Subtype Example

Igneous rocks formed from volcanic or extrusive basalt (ocean floors,

(magmatites) magma, 95% of all Giant’s causeway),
rocks andesite, obsidian
plutonic or intrusive granite, gabbro
Sedimenary rocks often with fossils clastic shale, siltstone, Dvipsbugw
(sedimentites) sandstone
biogenic limestone, chalk,
dolostone
precipitate halite, gypsum
Metamorphic rocks transformed by foliated slate, schist, gneiss
(metamorphites) heat and pressure (Himalayas)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

non-foliated marble, skarn, quartzite
(grandoblastic or
hornfelsic)
Meteorites from the solar rock meteorites
system
iron meteorites

If the pressure increases still further the star becomes a black hole, and never stops
Page 476 collapsing. Black holes have no floor at all; they still have a constant size though, determ-
ined by the horizon curvature.
The question whether other star types exist in nature, with other floor forming mech-
anisms – such as quark stars – is still a topic of research.

Rocks and stones

If a geologist takes a stone his his hands, he is usually able to give, within an error of a few
percent, the age of the stone. The full story forms a large part of geology, but the general
lines should be known to every physicist.
Copyright © Christoph Schiller November 1997–May 2006

Every stone arrives in your hand through the rock cycle. The rock cycle is a process that
transforms magma from the interior of the Earth into igneous rocks, through cooling and
crystallization. Igneous rocks, such as basalt, can transform through erosion, transport
and deposition into sedimentary rocks. Either of these two rock types can be transformed
through high pressures or temperatures into metamorphic rocks, such as marble. Finally,
most rocks are generally – but not always – transformed back into magma.
The full rock cycle takes around 110 to 170 million years. For this reason, rocks that
are older that this age much less common on Earth. Any stone is the product of erosion
of one of the rock types. A geologist can usually tell, simply by looking at it, the type of
rock it belongs to; if he sees the original environment, he can also give the age, without
any laboratory.

Dvipsbugw
 842 vii details of quantum theory • applied quantum mechanics

How can one look through matter?

Quantum theory showed us that all obstacles have only finite potential heights. That leads
to a question: Is it possible to look through matter? For example, can we see what is hidden
inside a mountain? To be able to do this, we need a signal which fulfils two conditions: it
must be able to penetrate the mountain, and it must be scattered in a material-dependent
way. Table 62 gives an overview of the possibilities.
We see that many signals are able to penetrate a mountain. However, only sound or
radio waves provide the possibility to distinguish different materials, or to distinguish Dvipsbugw
solids from liquids and from air. In addition, any useful method requires a large number
of signal sources and of signal receptors, and thus a large amount of cash. Will there ever
be a simple method allowing to look into mountains as precisely as X-rays allow to study
human bodies? For example, will it ever be possible to map the interior of the pyramids?
Challenge 1315 n A motion expert like the reader should be able to give a definite answer.
One of the high points of twentieth century physics was the development of the best

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

method so far to look into matter with dimensions of about a metre or less: magnetic
Page 897 resonance imaging. We will discuss it later on.
The other modern imaging technique, ultrasound imaging, is getting more and more
criticized. It is much used for prenatal diagnostics of embryos. However, studies have
found that ultrasound produces extremely high levels of audible sound to the baby, espe-
cially when the ultrasound is switched on or off, and that babies react negatively to this
loud noise.

What is necessary to make matter invisible?

You might have already imagined what adventures would be possible if you could be
invisible for a while. Some years ago, a team of Dutch scientists found a material than
Ref. 821 can be switched from mirror mode to transparent mode using an electrical signal. This
seems a first step to realize the dream to become invisible at will.
Nature shows us how to be invisible. An object is invisible if it has no surface, no
absorption and small size. In short, invisible objects are either small clouds or composed
of them. Most atoms and molecules are examples. Homogeneous non-absorbing gases
also realize these conditions. That is the reason that air is (usually) invisible. When air is
not homogeneous, it can be visible, e.g. above hot surfaces.
In contrast to gases, solids or liquids do have surfaces. Surfaces are usually visible, even
if the body is transparent, because the refractive index changes there. For example, quartz
Copyright © Christoph Schiller November 1997–May 2006

can be made so transparent that one can look through 1 000 km of it; pure quartz is thus
more transparent than usual air. Still, objects made of pure quartz are visible to the eye
due to the index change at the surface. Quartz can be invisible only when submerged in
liquids with the same refractive index.
In other words, to become invisible, we must transform ourselves into a diffuse cloud
of non-absorbing atoms. On the way to become invisible, we would loose all memory
and all genes, in short, we would loose all our individuality. But an individual cannot be
made of gas. An individual is defined through its boundary. There is no way that we can
be invisible; a reversible way to perform the feat is also impossible. In summary, quantum
theory shows that only the dead can be invisible.

Dvipsbugw
materials science 843

TA B L E 62 Signals penetrating mountains and other matter

Signal Penet- Achie- M at e r - Use

r at i o n ved ial
depth resolu - depend -
in stone tion ence
matter
diffusion of water or c. 5 km c. 100 m medium mapping hydrosystems
liquid chemicals
Dvipsbugw
diffusion of gases c. 5 km c. 100 m medium studying vacuum systems
electromagnetism
sound, explosions, 0.1 − 10 m c. l~100 high oil and ore search, structure
seismic waves mapping in rocks
ultrasound 1 mm high medical imaging, acoustic

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

microscopy
infrasound and 100 000 km 100 km high mapping of Earth crust and
earthquakes mantle
static magnetic medium cable search, cable fault
fields localization, search for
structure inside soil and
rocks via changes of the
Earth’s magnetic field
static electric fields low no use
electrical currents soil and rock investigations,
search for tooth decay
electromagnetic soil and rock investigations in
sounding (0.2 Hz to deep water and on land
5 Hz)
radio waves 10 m 30 m to 1 mm small soil radar (up to 10 MW),
magnetic imaging, research
into solar interior
mm and THz waves below 1 mm 1 mm see through clothes,
envelopes and teeth Ref. 822
infrared c. 1 m 0.1 m medium mapping of soil over 100 m
c. 1 cm 0.1 µm
Copyright © Christoph Schiller November 1997–May 2006

visible light medium imaging of many sorts

X-rays a few metre 5 µm high medicine, material analysis,
airports
muons created by up to 0.1 m small finding caves in pyramids,
cosmic radiation c. 300 m imaging truck interiors
weak interactions
neutrino beams light years zero very weak studies of Sun
strong interactions
cosmic radiation 1 m to 1 km
radioactivity 1 mm to 1 m airports
gravitation
change of 50 m low oil & ore search
gravitational
acceleration

Dvipsbugw
 844 vii details of quantum theory • applied quantum mechanics

TA B L E 63 Matter at lowest temperatures

Phase Ty pe L ow t e m p e r at u r e b e - Example
h av i o u r
Solid conductor superconductivity lead
antiferromagnet chromium, MnO
ferromagnet iron
insulator diamagnet Dvipsbugw
Liquid bosonic Bose–Einstein condensation, i.e. su- 4 He
perfluidity
3
fermionic pairing, then BEC, i.e. superfluidity He
87
Gas bosonic Bose–Einstein condensation Rb, 7 Li, 23 Na, H,
4
He, 41 K
fermionic pairing, then Bose–Einstein con- 40 K, 6 Li

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

densation

How does matter behave at the lowest temperatures?

The low-temperature behaviour of matter has numerous experimental and theoretical
aspects. The first issue is whether matter is always solid at low temperatures. The answer
is no. All phases exist at low temperatures, as shown in Table 63.
Concerning the electric properties of matter at lowest temperatures, the present status
is that matter is either insulating or superconducting. Finally, one can ask about the mag-
netic properties of matter at low temperatures. We know already that matter can not be
paramagnetic at lowest temperatures. It seems that matter is either ferromagnetic, dia-
magnetic or antiferromagnetic at lowest temperatures.
Page 847 More about superfluidity and superconductivity will be told below.

Curiosities and fun challenges about materials science

Materials science is not a central part of this walk. A few curiosities can give a taste of it.
**
What is the maximum height of a mountain? This question is of course of interest to all
Copyright © Christoph Schiller November 1997–May 2006

Ref. 823 climbers. Many effects limit the height. The most important is the fact that under heavy
Challenge 1316 ny pressure, solids become liquid. For example, on Earth this happens at about 27 km. This
is quite a bit more than the highest mountain known, which is the volcano Mauna Kea
in Hawaii, whose top is about 9.45 km above the base. On Mars gravity is weaker, so that
mountains can be higher. Indeed the highest mountain on Mars, Olympus mons, is 80 km
Challenge 1317 n high. Can you find a few other effects limiting mountain height?
**
Do you want to become rich? Just invent something that can be produced in the factory,
is cheap and can substitute duck feathers in bed covers, sleeping bags or in badminton
Challenge 1318 r shuttlecocks. Another industrial challenge is to find an artificial substitute for latex, and a

Dvipsbugw
 materials science 845

third one is to find a substitute for a material that is rapidly disappearing due to pollution:
cork.

**
Challenge 1319 ny What is the difference between solids, liquids and gases?
**
What is the difference between the makers of bronze age knifes and the builders of the
Eiffel tower? Only their control of dislocation distributions. Dvipsbugw

**
Challenge 1320 n Quantum theory shows that tight walls do not exist. Every material is penetrable. Why?
**
Quantum theory shows that even if tight walls would exist, the lid of such a box can never

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1321 n be tightly shut. Can you provide the argument?

**
Quantum theory predicts that heat transport at a given temperature is quantized. Can
Challenge 1322 ny you guess the unit of thermal conductance?
**
Robert Full has shown that van der Waals forces are responsible for the way that geckos
Ref. 824 walk on walls and ceilings. The gecko, a small reptile with a mass of about 100 g, uses an
elaborate structure on its feet to perform the trick. Each foot has 500 000 hairs each split in
up to 1000 small spatulae, and each spatula uses the van der Waals force (or alternatively,
capillary forces) to stick to the surface. As a result, the gecko can walk on vertical glass
walls or even on glass ceilings; the sticking force can be as high as 100 N per foot.
The same mechanism is used by jumping spiders (Salticidae). For example, Evarcha
arcuata have hairs at their feet which are covered by hundred of thousands of setules.
Ref. 825 Again. the van der Waals force in each setule helps the spider to stick on surfaces.
Researchers have copied these mechanisms for the first time in 2003, using microlitho-
graphy on polyimide, and hope to make durable sticky materials in the future.
**
Copyright © Christoph Schiller November 1997–May 2006

Millimetre waves or terahertz waves are emitted by all bodies at room temperature. Mod-
ern camera systems allow to image them. In this way, it is possible to see through clothes.
This ability could be used in future to detect hidden weapons in airports. But the develop-
ment of a practical and affordable detector which can be handled as easily as a binocular
is still under way. The waves can also be used to see through paper, thus making it unne-
cessary to open letters in order to read them. Secret services are exploiting this technique.
A third application of terahertz waves might be in medical diagnostic, for example for the
search of tooth decay. Terahertz waves are almost without side effects, and thus superior
to X-rays. The lack of low-priced quality sources is still an obstacle to their application.

**

Dvipsbugw
 846 vii details of quantum theory • applied quantum mechanics

Does the melting point of water depend on the magnetic field? This surprising claim was
Ref. 826 made in 2004 by Inaba Hideaki and colleagues. They found a change of 0.9 mK~T. It is
known that the refractive index and the near infrared spectrum of water is affected by
magnetic fields. Indeed, not everything about water might be known yet.
**
Plasmas, or ionized gases, are useful for many applications. Not only can they be used
for heating or cooking and generated by chemical means (such plasmas are variously
called fire or flames) but they can also be generated electrically and used for lighting Dvipsbugw
or deposition of materials. Electrically generated plasmas are even being studied for the
disinfection of dental cavities.
**
It is known that the concentration of CO 2 in the atmosphere between 1800 and 2005
Challenge 1323 ny has increased from 280 to 380 parts per million. (How would you measure this?) It is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

known without doubt that this increase is due to human burning of fossil fuels, and not
to natural sources such as the oceans or volcanoes. There are three arguments. First of
all, there was a parallel decline of the 14 C~12 C ratio. Second, there was a parallel decline
of the 13 C~12 C ratio. Finally, there was a parallel decline of the oxygen concentration. All
three measurements independently imply that the CO 2 increase is due to the burning of
fuels, which are low in 14 C and in 13 C, and at the same time decrease the oxygen ratio.
Natural sources do not have these three effects. Since CO 2 is a major greenhouse gas, the
data implies that humans are also responsible for a large part of the temperature increase
during the same period.
**
The technologies to produce perfect crystals, without grain boundaries or dislocations,
are an important part of modern industry. Perfectly regular crystals are at the basis of the
integrated circuits used in electronic appliances, are central to many laser and telecom-
munication systems and are used to produce synthetic jewels.
Synthetic diamonds have now displaced natural diamonds in almost all applications.
In the last years, methods to produce large, white, jewel-quality diamonds of ten carats
Ref. 827 and more are being developed. These advances will lead to a big change in all the domains
that depend on these stones, such as the production of the special surgical knives used in
eye lens operation.
Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 1324 ny How can a small plant pierce through tarmac?

Quantum technolo gy
I were better to be eaten to death with a rust

“ than to be scoured to nothing with perpetual

motion.
William Shakespeare (1564–1616) King Henry
IV.

”
Dvipsbugw
 quantum technology 847

Quantum effects do not appear only in microscopic systems. Several quantum effects are
important in modern life; transistors, lasers, superconductivity and a few other effects
and systems are worth knowing.

Motion without friction – superconductivity and superfluidity

We are used to think that friction is inevitable. We even learned that friction was an
inevitable result of the particle structure of matter. I should come to the surprise of every
physicist that motion without friction is possible. Dvipsbugw
In 1911 Gilles Holst and Heike Kamerlingh Onnes discovered that at low temperatures,
electric currents can flow with no resistance, i.e., with no friction, through lead. The ob-
servation is called superconductivity. In the century after that, many metals, alloys and
ceramics have been found to show the same behaviour.
The condition for the observation of motion without friction is that quantum effects
play an essential role. That is the reason for the requirement of low temperature in such

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

experiments. Nevertheless, it took over 50 years to reach a full understanding of the ef-
Ref. 837 fect. This happened in 1957, when Bardeen, Cooper and Schrieffer published their results.
At low temperatures, electron behaviour is dominated by an attractive interaction that
makes them form pairs – today called Cooper pairs – that are effective bosons. And bo-
sons can all be in the same state, thus effectively moving without friction.
For superconductivity, the attractive interaction between electrons is due to the de-
formation of the lattice. Two electrons attract each other in the same way as two masses
attract each other due to deformation of the space-time mattress. However, in the case of
solids, the deformations are quantized. With this approach, Bardeen, Cooper and Schrief-
fer explained the lack of electric resistance of superconducting materials, their complete
diamagnetism (µ r = 0), the existence of an energy gap, the second-order transition to
normal conductivity at a specific temperature, and the dependence of this temperature
on the mass of the isotopes. Last but not least, they received the Nobel Prize in 1972.*
Another type of motion without friction is superfluidity. Already in 1937, Pyotr Kapitsa
had predicted that normal helium (4 He), below a transition observed at the temperature
of 2.17 K, would be a superfluid. In this domain, the fluid moves without friction through
tubes. (In fact, the fluid remains a mixture of a superfluid component and a normal com-
ponent.) Helium is even able, after an initial kick, to flow over obstacles, such as glass
walls, or to flow out of bottles. 4 He is a boson, so no pairing is necessary for it to flow
without friction. This research earned Kapitsa a Nobel Prize in 1978.
Copyright © Christoph Schiller November 1997–May 2006

The explanation of superconductivity also helped for fermionic superfluidity. In 1972,

Richardson, Lee, and Osheroff found that even 3 He is superfluid at temperatures be-
low 2.7 mK. 3 He is a fermion, and requires pairing to become superfluid. In fact, below
2.2 mK, 3 He is even superfluid in two different ways; one speaks of phase A and phase

* For John Bardeen (1908–1991), this was his second, after he had got the first Nobel Prize in 1956, shared
with William Shockley and Walter Brattain, for the discovery of the transistor. The first Nobel Prize was a
problem for Bardeen, as he needed time to work on superconductivity. In an example to many, he reduced
the tam-tam around himself to a minimum, so that he could work as much as possible on the problem of
superconductivity. By the way, Bardeen is topped by Frederick Sanger and by Marie Curie. Sanger first won a
Nobel Prize in chemistry in 1958 by himself and then won a second one shared with Walter Gilbert in 1980;
Marie Curie first won one with her husband and a second one by herself, though in two different fields.

Dvipsbugw
 848 vii details of quantum theory • applied quantum mechanics

Figure missing

F I G U R E 335 Nuclear magnetic resonance shows

that vortices in superfluid 3 He-B are quantized. Dvipsbugw

B.*
In this case, the theoreticians had been faster. The theory for superconductivity
through pairing had been adapted to superfluids already in 1958 – before any data were
available – by Bohr, Mottelson and Pines. This theory was then adapted again by Anthony
Leggett.** The attractive interaction between 3 He atoms turns out to be the spin-spin in-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

teraction.
In superfluids, like in ordinary fluids, one can distinguish between laminar and turbu-
lent flow. The transition between the two regimes is mediated by the behaviour of vortices.
But in superfluids, vortices have properties that do not appear in normal fluids. In the su-
perfluid 3 He-B phase, vortices are quantized: vortices only exist in integer multiples of the
Ref. 831 elementary circulation h~2m3 He . Present research is studying how these vortices behave
and how they induce the transition form laminar to turbulent flows.
In recent years, studying the behaviour of gases at lowest temperatures has become
very popular. When the temperature is so low that the de Broglie wavelength is compar-
able to the atom-atom distance, bosonic gases form a Bose-Einstein condensate. The first
one were realized in 1995 by several groups; the group around Eric Cornell and Carl Wie-
man used 87 Rb, Rand Hulet and his group used 7 Li and Wolfgang Ketterle and his group
used 23 Na. For fermionic gases, the first degenerate gas, 40 K, was observed in 1999 by
the group around Deborah Jin. In 2004, the same group observed the first gaseous fermi
condensate, after the potassium atoms paired up.

Quantized conductivity
Ref. 832 In 1996, the Spanish physicist J.L. Costa–Krämer and his colleagues performed a simple
experiment. They put two metal wires on top of each other on a kitchen table and attached
Copyright © Christoph Schiller November 1997–May 2006

a battery, a 10 kΩ resistor and a storage oscilloscope to them. Then they measured the elec-
trical current while knocking on the table. In the last millisecond before the wires detach,
the conductivity and thus the electrical current diminished in regular steps of a 7 µA, as
can easily be seen on the oscilloscope. This simple experiment could have beaten, if it
had been performed a few years earlier, a number of enormously expensive experiments
which discovered this quantization at costs of several million euro each, using complex
set-ups and extremely low temperatures.
In fact, quantization of conductivity appears in any electrical contact with a small cross-

* They received the Nobel Prize in 1996 for this discovery.

** Aage Bohr and Ben Mottelson received the Nobel Prize in 1975, Anthony Leggett in 2003.

Dvipsbugw
 quantum technology 849

section. In such situations the quantum of action implies that the conductivity can only
Challenge 1325 ny be a multiple of 2e 2 ~ħ  1~12 906 1~Ω. Can you confirm this result?
Note that electrical conductivity can be as small as required; only quantized electrical
conductivity has the minimum value of 2e 2 ~ħ.

The fractional quantum Hall effect

The fractional quantum Hall effect is one of the most intriguing discoveries of materials
science. The effect concerns the flow of electrons in a two-dimensional surface. In 1982, Dvipsbugw
Ref. 834 Robert Laughlin predicted that in this system one should be able to observe objects with
Ref. 833 electrical charge e~3. This strange and fascinating prediction was indeed verified in 1997.
The story begins with the discovery by Klaus von Klitzing of the quantum Hall ef-
Ref. 836 fect. In 1980, Klitzing and his collaborators found that in two-dimensional systems at low
temperatures – about 1 K – the electrical conductance S is quantized in multiples of the
quantum of conductance
e2
S=n

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

. (579)
ħ
The explanation is straightforward: it is the quantum analogue of the classical Hall ef-
fect, which describes how conductance varies with applied magnetic field. Von Klitzing
received the Nobel Prize for physics for the discovery, since the effect was completely
unexpected, allows a highly precise measurement of the fine structure constant and also
allows one to build detectors for the smallest voltage variations measurable so far.
Two years later, it was found that in extremely strong magnetic fields the conductance
Ref. 835 could vary in steps one third that size. Shortly afterwards, even stranger numerical frac-
Ref. 834 tions were also found. In a landmark paper, Robert Laughlin explained all these results by
assuming that the electron gas could form collective states showing quasiparticle excita-
tions with a charge e~3. This was confirmed 15 years later and earned him a Nobel price as
well. We have seen in several occasions that quantization is best discovered through noise
measurements; also in this case, the clearest confirmation came from electrical current
noise measurements. How can we imagine these excitations?

– CS – explanation to be inserted – CS –

What do we learn from this result? Systems in two dimensions have states which follow
different rules than systems in three dimensions. Can we infer something about quarks
Copyright © Christoph Schiller November 1997–May 2006

from this result? Quarks are the constituents of protons and neutrons, and have charges
e~3 and 2e~3. At this point we need to stand the suspense, as no answer is possible; we
come back to this issue later on.

L asers and other spin-one vector boson launchers

Photons are vector bosons; a lamp is thus a vector boson launcher. All lamps fall into
Ref. 853 one of three classes. Incandescent lamps use emission from a hot solid, gas discharge
lamps use excitation of atoms, ions or molecules through collision, and solid state lamps
generate (cold) light through recombination of charges in semiconductors.
Most solid state lights are light emitting diodes. The large progress in brightness of

Dvipsbugw
 850 vii details of quantum theory • applied quantum mechanics

TA B L E 64 A selection of lamps

Lamp type Highest Low - Life-

bright- est cost time
ness (2003) (2003) (2003)

Incandescent lamps
tungsten wire light bulbs, halogen lamps 25 lm~W 0.1 cent~lm 700 h
Dvipsbugw
Gas discharge lamps
oil lamps, candle
neon lamps
mercury lamps 100 lm~W ... cent~lm ... h
metal halogenide lamps (ScI3 or ‘xenon’, NaI, 110 lm~W ... cent~lm ... h
DyI3 , HoI3 , TmI5 )

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

sodium low pressure lamps 120 lm~W ... cent~lm ... h
sodium high pressure lamps 200 lm~W ... cent~lm ... h
Recombination lamps
firefly c. 500 h
light emitting diodes 100 lm~W 10 cent~lm 10 000 h
He-Ne laser 550 lm~W 2000 cent~lm 300 h
Ideal white lamp c. 300 lm~W n.a. n.a.
Ideal coloured lamp 683 lm~W n.a. n.a.

light emitting diodes could lead to a drastic reduction in future energy consumption, if
their cost is lowered sufficiently. Many engineers are working on this task. Since the cost
is a good estimate for the energy needed for production, can you estimate which lamp is
Challenge 1326 n the most friendly to the environment?
Nobody thought much about lamps, until Albert Einstein and a few other great phys-
icists came along, such as Theodore Maiman, Hermann Haken and several others that
got the Nobel Prize with their help. In 1916, Einstein showed that there are two types of
sources of light – or of electromagnetic radiation in general – both of which actually
Copyright © Christoph Schiller November 1997–May 2006

‘create’ light. He showed that every lamp whose brightness is turned up high enough
will change behaviour when a certain intensity threshold is passed. The main mechan-
ism of light emission then changes from spontaneous emission to stimulated emission.
Nowadays such a special lamp is called a laser. (The letters ‘se’ in laser are an abbreviation
of ‘stimulated emission’.) After a passionate worldwide research race, in 1960 Maiman
was the first to build a laser emitting visible light. (So-called masers emitting microwaves
were already known for several decades.) In summary, Einstein and the other physicists
showed that lasers are lamps which are sufficiently turned up. Lasers consist of some light
producing and amplifying material together with a mechanism to pump energy into it.
The material can be a gas, a liquid or a solid; the pumping process can use electrical
current or light. Usually, the material is put between two mirrors, in order to improve

Dvipsbugw
 quantum technology 851

the efficiency of the light production. Common lasers are semiconductor lasers (essen-
tially highly pumped LEDs or light emitting diodes), He–Ne lasers (highly pumped neon
lamps), liquid lasers (essentially highly pumped fire flies) and ruby lasers (highly pumped
luminescent crystals).
Lasers produce radiation in the range from microwaves and extreme ultraviolet. They
have the special property of emitting coherent light, usually in a collimated beam. There-
fore lasers achieve much higher light intensities than lamps, allowing their use as tools.
In modern lasers, the coherence length, i.e. the length over which interference can be ob-
served, can be thousands of kilometres. Such high quality light is used e.g. in gravitational Dvipsbugw
wave detectors.
People have become pretty good at building lasers. Lasers are used to cut metal sheets
up to 10 cm thickness, others are used instead of knives in surgery, others increase surface
hardness of metals or clean stones from car exhaust pollution. Other lasers drill holes in
teeth, measure distances, image biological tissue or grab living cells. Most materials can
be used to make lasers, including water, beer and whiskey.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Some materials amplify light so much that end mirrors are not necessary. This is the
case for nitrogen lasers, in which nitrogen, or simply air, is used to produce a UV beam.
Even a laser made of a single atom (and two mirrors) has been built; in this example,
Ref. 860 only eleven photons on average were moving between the two mirrors. Quite a small
lamp. Also lasers emitting light in two dimensions have been built. They produce a light
plane instead of a light beam.
Lasers have endless applications. Lasers read out data from compact discs (CDs), are
used in the production of silicon integrated circuits, and transport telephone signals; we
Page 251 already encountered lasers that work as loudspeakers. The biggest advances in recent
years came from the applications of femtosecond laser pulses. Femtosecond pulses gen-
erate high-temperature plasmas in the materials they propagate, even in air. Such short
pulses can be used to cut material without heating it, for example to cut bones in heart
operations. Femtosecond lasers have been used to make high resolution hologram of hu-
man heads within a single flash. Recently such lasers have been used to guide lightning
Ref. 854 along a predetermined path and seem promising candidates for laser ligtning rods. A
curious application is to store information in fingernails (up to 5 Mbit for a few months)
Ref. 861 using such lasers, in a way not unlike that used in recordable compact discs (CD-R).

Can two photons interfere? Copyright © Christoph Schiller November 1997–May 2006

In 1930, Dirac made the famous statement already mentioned above:*

Each photon interferes only with itself. Interference between two different photons
never occurs.

Often this statement is misinterpreted as implying that two separate photon sources can-
not interfere. It is almost unbelievable how this false interpretation has spread through
Ref. 855 the literature. Everybody can check that this statement is incorrect with a radio: two dis-
tant radio stations transmitting on the same frequency lead to beats in amplitude, i.e. to
* See the famous, beautiful but difficult textbook P.A.M. Dirac, The Principles of Quantum Mechanics,
Clarendon Press, Oxford, 1930, page 9.

Dvipsbugw
 852 vii details of quantum theory • applied quantum mechanics

wave interference. (This should not to be confused with the more common radio interfer-
ence, with usually is simply a superposition of intensities.) Radio transmitters are coher-
ent sources of photons, and any radio receiver shows that two such sources can indeed
interfere.
In 1949, interference of two different sources has been demonstrated with microwave
beams. Numerous experiments with two lasers and even with two thermal light sources
Ref. 855 have shown light interference from the fifties onwards. Most cited is the 1963 experiment
by Magyar and Mandel; they used two ruby lasers emitting light pulses and a rapid shutter
camera to produce spatial interference fringes. Dvipsbugw
However, all these experimental results do not contradict the statement by Dirac. In-
deed, two photons cannot interfere for several reasons.
— Interference is a result of space-time propagation of waves; photons appear only when
the energy–momentum picture is used, mainly when interaction with matter takes
place. The description of space-time propagation and the particle picture are mutually

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

exclusive – this is one aspect of the complementary principle. Why does Dirac seem to
mix the two in his statement? Dirac employs the term ‘photon’ in a very general sense,
as quantized state of the electromagnetic field. When two coherent beams are super-
posed, the quantized entities, the photons, cannot be ascribed to either of the sources.
Interference results from superposition of two coherent states, not of two particles.
— Interference is only possible if one cannot know where the detected photon comes
from. The quantum mechanical description of the field in a situation of interference
never allows to ascribe photons of the superposed field to one of the sources. In other
words, if you can say from which source a detected photon comes from, you cannot
observe interference.
— Interference between two beams requires a fixed phase between them, i.e. an uncertain
particle number; in other words, interference is only possible if the photon number for
each of the two beams is unknown.
A better choice of words is to say that interference is always between two (indistin-
guishable) states, or if one prefers, between two possible (indistinguishable) histories, but
never between two particles. In summary, two different electromagnetic beams can inter-
fere, but not two different photons.

Can two electron beams interfere?

Copyright © Christoph Schiller November 1997–May 2006

Do coherent electron sources exist? Yes, as it is possible to make holograms with electron
beams.* However, electron coherence is only transversal, not longitudinal. Transversal
coherence is given by the possible size of wavefronts with fixed phase. The limit of this
size is given by the interactions such a state has with its environment; if the interactions
are weak, matter wave packets of several metres of size can be produced, e.g. in particle
colliders, where energies are high and interaction with matter is low.
Actually, the term transversal coherence is a fake. The ability to interfere with oneself
is not the definition of coherence. Transversal coherence only expresses that the source
size is small. Both small lamps (and lasers) can show interference when the beam is split

Ref. 856 * In 2002, the first holograms have been produced that made use of neutron beams.

Dvipsbugw
 quantum technology 853

Figure yet to be included

Dvipsbugw
F I G U R E 336 An electron hologram

and recombined; this is not a proof of coherence. Similarly, monochromaticity is not a

proof for coherence either.
A state is called coherent if it possesses a well-defined phase throughout a given domain

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of space or time. The size of that region or of that time interval defines the degree of co-
herence. This definition yields coherence lengths of the order of the source size for small
‘incoherent’ sources. Nevertheless, the size of an interference pattern, or the distance d
between its maxima, can be much larger than the coherence length l or the source size s.
In summary, even though an electron can interfere with itself, it cannot interfere with
a second one. Uncertain electron numbers are needed to see a macroscopic interference
pattern. That is impossible, as electrons (at usual energies) carry a conserved charge.

– CS – sections on transistors and superconductivity to be added – CS –

Challenges and dreams about quantum technology

Many challenges in applied quantum physics remain, as quantum effects seem to promise
to realize many age-old technological dreams.

**
Challenge 1327 d Is it possible to make A4-size, thin and flexible colour displays for an affordable price?
**
Challenge 1328 r Will there ever be desktop laser engravers for 2000 euro?
Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 1329 r Will there ever be room-temperature superconductivity?
**
Challenge 1330 n Will there ever be teleportation of everyday objects?
**
One process that quantum physics does not allow is telephathy. An unnamed space
agency found out in the Apollo 14 mission, when, during the flight to the moon, cosmo-
naut Edgar Mitchell tested telepathy as communication means. Unsurprisingly, he found

Dvipsbugw
 854 vii details of quantum theory • quantum electrodynamics

Figure will be added in the future Dvipsbugw

F I G U R E 337 Ships in a swell

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 838 that it was useless. It is not clear why NASA spent so much money for an experiment that
could have been performed for 1000 times less cost also in other ways.
**
Challenge 1331 d Will there ever be applied quantum cryptology?

**
Will there ever be printable polymer electronic circuits, instead of lithographically pat-
Challenge 1332 d terned silicon electronics as is common now?
**
Challenge 1333 r Will there ever be radio-controlled flying toys in the size of insects?

26. quantum electrodynamics – the origin of

virtual realit y

T he central concept the quantum theory introduces in the description of nature is

he idea of virtual particles. Virtual particles are short-lived particles; they owe
heir existence exclusively to the quantum of action. Because of the quantum of ac-
tion, they do not need to follow the energy-mass relation that special relativity requires of
Copyright © Christoph Schiller November 1997–May 2006

normal, real particles. Virtual particles can move faster than light and can move backward
in time. Despite these strange properties, they have many observable effects.

Ships, mirrors and the Casimir effect

When two parallel ships roll in a big swell, without even the slightest wind blowing, they
will attract each other. This effect was well known up to the nineteenth century, when
many places still lacked harbours. Shipping manuals advised captains to let the ships be
Ref. 841 pulled apart using a well-manned rowing boat.
Waves induce oscillations of ships because a ship absorbs energy from the waves.

Dvipsbugw
 quantum electrodynamics – the origin of virtual reality 855

When oscillating, the ship also emits waves. This happens mainly towards the two sides
of the ship. As a result, for a single ship, the wave emission has no net effect on its position.
Now imagine that two parallel ships oscillate in a long swell, with a wavelength much lar-
ger than the distance between the ships. Due to the long wavelength, the two ships will
oscillate in phase. The ships will thus not be able to absorb energy from each other. As a
result, the energy they radiate towards the outside will push them towards each other.
The effect is not difficult to calculate. The energy of a rolling ship is

E = mдh α 2 ~2 (580) Dvipsbugw

where α is the roll angle amplitude, m the mass of the ship and д = 9, 8 m~s2 the accelera-
tion due to gravity. The metacentric height h is the main parameter characterizing a ship,
especially a sailing ship; it tells with what torque the ship returns to the vertical when
inclined by an angle α. Typically, one has h =1.5 m.
When a ship is inclined, it will return to the vertical by a damped oscillation. A damped

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

oscillation is characterized by a period T and a quality factor Q. The quality factor is the
number of oscillations the system takes to reduce its amplitude by a factor e = 2.718. If the
quality factor Q of an oscillating ship and its oscillation period T are given, the radiated
power W is
W = 2π
E
. (581)
QT

We saw above that radiation pressure is W~c, where c is the wave propagation velocity.
For water waves, we have the famous relation

c=
дT
. (582)
2π
Assuming that for two nearby ships each one completely absorbs the power emitted from
the other, we find that the two ships are attracted towards each other following

hα 2
ma = m2π 2 . (583)
QT 2

Inserting typical values such as Q = 2.5, T =10 s, α =0.14 rad and a ship mass of 700 tons,
we get about 1.9 kN. Long swells thus make ships attract each other. The intensity of the
Copyright © Christoph Schiller November 1997–May 2006

attraction is comparatively small and can indeed be overcome with a rowing boat. On the
other hand, even the slightest wind will damp the oscillation amplitude and have other
effects that will avoid the observation of the attraction.
Sound waves or noise in air can have the same effect. It is sufficient to suspend two
Ref. 842 metal plates in air and surround them by loudspeakers. The sound will induce attraction
(or repulsion) of the plates, depending on whether the sound wavelength cannot (or can)
be taken up by the other plate.
In 1948, the Dutch physicist Hendrik Casimir made one of the most spectacular predic-
tions of quantum theory: he predicted a similar effect for metal plates in vacuum. Casimir,
who worked at the Dutch Electronics company Philips, wanted to understand why it was

Dvipsbugw
 856 vii details of quantum theory • quantum electrodynamics

so difficult to build television tubes. Television screens are made by deposing small neut-
ral particles on glass, but Casimir observed that the particles somehow attracted each
other. Casimir got interested in understanding how neutral particles interact. During
these theoretical studies he discovered that two neutral mirrors (or metal plates) would
attract each other even in complete vacuum. This is the famous Casimir effect. Casimir
also determined the attraction strength between a sphere and a plate, and between two
spheres. In fact, all conducting bodies attract each other in vacuum, with a force depend-
ing on their geometry.
In all these situations, the role of the sea is taken by the zero-point fluctuations of the Dvipsbugw
electromagnetic field, the role of the ships by the mirrors. Casimir understood that the
space between two parallel mirrors, due to the geometrical constraints, had different zero-
point fluctuations that the free vacuum. Like two ships, the result would be the attraction
of the mirrors.
Casimir predicted that the attraction for two mirrors of mass m and surface A is given
by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ma π 3 ħc
= . (584)
A 120 d 4
The effect is a pure quantum effect; in classical electrodynamics, two neutral bodies
do not attract. The effect is small; it takes some dexterity to detect it. The first experi-
Ref. 843 mental confirmation was by Derjaguin, Abrikosova and Lifshitz in 1956; the second ex-
Ref. 844 perimental confirmation was by Marcus Sparnaay, Casimir’s colleague at Philips, in 1958.
Two beautiful high-precision measurements of the Casimir effect were performed in 1997
Ref. 845 by Lamoreaux and in 1998 by Mohideen and Roy; they confirmed Casimir’s prediction
with a precision of 5 % and 1 % respectively.*
In a cavity, spontaneous emission is suppressed, if it is smaller than the wavelength
of the emitted light! This effect has also been observed. It confirms the old saying that
spontaneous emission is emission stimulated by the zero point fluctuations.
The Casimir effect thus confirms the existence of the zero-point fluctuations of the
electromagnetic field. It confirms that quantum theory is valid also for electromagnetism.
The Casimir effect between two spheres is proportional to 1~r 7 and thus is much
weaker than between two parallel plates. Despite this strange dependence, the fascination
of the Casimir effect led many amateur scientists to speculate that a mechanism similar
to the Casimir effect might explain gravitational attraction. Can you give at least three
Challenge 1334 n arguments why this is impossible, even if the effect had the correct distance dependence?
Copyright © Christoph Schiller November 1997–May 2006

Like the case of sound, the Casimir effect can also produce repulsion instead of at-
traction. It is sufficient that one of the two materials be perfectly permeable, the other a
perfect conductor. Such combinations repel each other, as Timothy Boyer discovered in
Ref. 847 1974.
Ref. 848 The Casimir effect bears another surprise: between two metal plates, the speed of light
changes and can be larger than c. Can you imagine what exactly is meant by ‘speed of
Challenge 1335 n light’ in this context?

Ref. 846 * At very small distances, the dependence is not 1~d 4 , but 1~d 3 .

Dvipsbugw
 quantum electrodynamics – the origin of virtual reality 857

The Banach–Tarski paradox for vacuum

It implies that there is a specific energy density that can be described to the vacuum. This
seems obvious. However, the statement has a dramatic consequence: space-time cannot
be continuous!
The reasoning is simple. If the vacuum were continuous, we could make use of the
Page 55 Banach–Tarski paradox and split, without any problem, a ball of vacuum into two balls
of vacuum, each with the same volume. In other words, one ball with energy E could not
be distinguished from two balls of energy 2E. This is impossible. Dvipsbugw
The Gedanken experiment tells us something important. In the same way that we used
Page 219 the argument to show that chocolate (and any other matter) cannot be continuous, we
can now deduce that the vacuum cannot be either. However, we have no details yet. In
the same way that matter turned out to possess an intrinsic scale, we can guess that this
happens also to the vacuum. Vacuum has an intrinsic scale; it is not continuous. We will
have to wait for the third part of the text to find out more. There, the structure of the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

vacuum will turn out to be even more interesting than that of matter.

The L amb shift

In the 1947, the measurements of the spectrum of hydrogen had yielded another effect
due to virtual particles. Willis Lamb (1913–) found that the 2S 1~2 energy level in atomic
hydrogen lies slightly above the 2P1~2 level. This is in contrast to the calculation per-
Page 756 formed above, where the two levels are predicted to have the same energy. In reality, they
have an energy difference of 1057.864 MHz or 4.3 µeV. This discovery had important con-
sequences for the description of quantum theory and yielded Lamb a share of the 1955
Nobel Prize.
The reason for the difference is an unnoticed approximation performed in the simple
solution above. There are two equivalent ways to explain it. One is to say that the calcula-
tion neglects the coupling terms between the Dirac equation and the Maxwell equations.
This explanation lead to the first calculations of the Lamb shift, around the year 1950. The
other explanation is to say that the calculation neglects virtual particles. In particular, the
calculation neglects the virtual photons emitted and absorbed during the motion of the
electron around the nucleus. This is the explanation in line with the general vocabulary
of quantum electrodynamics. QED is perturbative approach to solve the coupled Dirac
and Maxwell equations. Copyright © Christoph Schiller November 1997–May 2006

The QED L agrangian

– CS – section on the QED Lagrangian to be added – CS –

Interactions and virtual particles

The electromagnetic interaction is exchange of virtual photons. So how can the interac-
tion be attractive? At first sight, any exchange of virtual photons should drive the elec-
trons from each other. However, this is not correct. The momentum of virtual photons
does not have to be in the direction of its energy flow; it can also be in opposite direc-

Dvipsbugw
 858 vii details of quantum theory • quantum electrodynamics

tion.* Obviously, this is only possible within the limits provided by the indeterminacy
principle.

Vacuum energy
The strangest result of quantum field theory is the energy density of the vacuum.

– CS – More to be written – CS –
Dvipsbugw
Moving mirrors
Mirrors also work when in motion; in contrast, walls that produce echoes do not work at
all speeds. Walls do not produce echoes if one moves faster than sound. However, mirrors
always produce an image. This observation shows that the speed of light is the same for
Challenge 1336 n any observer. Can you detail the argument?
Mirrors also differ from tennis rackets. We saw that mirrors cannot be used to change

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the speed of the light they hit, in contrast to what tennis rackets can do with balls. This
observation shows that the speed of light is also a limit velocity. In short, the simple ex-
istence of mirrors is sufficient to derive special relativity.
But there are more interesting things to be learned from mirrors. We only have to
ask whether mirrors work when they undergo accelerated motion. This issue yields a
surprising result.
In the 1970s, quite a number of researchers found that there is no vacuum for accel-
erated observers. This effect is called Fulling–Davies–Unruh effect or sometimes the dy-
namical Casimir effect. As a consequence, a mirror in accelerated motion reflects the
fluctuations it encounters and reflects them. In short, an accelerated mirror emits light.
Unfortunately, the intensity is so weak that it has not been measured up to now. We will
Page 871 explore the issue in more detail below. Can you explain why accelerated mirrors emit
Challenge 1337 n light, but not matter?

Photons hitting photons

When virtual particles are taken into account, light beams can ‘bang’ onto each other.
This result is in contrast to classical electrodynamics. Indeed, QED shows that the virtual
electron-positron pairs allow photons to hit each other. And such pairs are found in any
light beam.
Copyright © Christoph Schiller November 1997–May 2006

However, the cross-section is small. When two beams cross, most photons will pass
undisturbed. The cross-section A is approximately

A α ( ) ( )
973 4 ħ 2 ħω 6
(585)
10 125π me c me c 2

for the case that the energy ħω of the photon is much smaller than the rest energy m e c 2
of the electron. This value is about 18 orders of magnitude smaller than what was meas-
urable in 1999; the future will show whether the effect can be observed for visible light.
* One of the most beautiful booklets on quantum electrodynamics which makes this point remains the text
by Richard Feynman, QED: the Strange Theory of Light and Matter, Penguin Books, 1990.

Dvipsbugw
 quantum electrodynamics – the origin of virtual reality 859

However, for high energy photons these effects are observed daily in particle accelerators.
In these cases one observes not only interaction through virtual electron–antielectron
pairs, but also through virtual muon–antimuon pairs, virtual quark–antiquark pairs, and
much more.
Everybody who consumes science fiction knows that matter and antimatter annihilate
and transform into pure light. In more detail, a matter particle and an antimatter particle
annihilate into two or more photons. More interestingly, quantum theory predicts that
the opposite process is also possible: photons hitting photons can produce matter!
Ref. 849 In 1997, this was also confirmed experimentally. At the Stanford particle accelerator, Dvipsbugw
photons from a high energy laser pulse were bounced off very fast electrons. In this way,
the reflected photons acquired a large energy, when seen in the inertial frame of the ex-
perimenter. The original pulse, of 527 nm or 2.4 eV green light, had a peak power density
of 1022 W~m2 , about the highest achievable so far. That is a photon density of 1034 ~m3
and an electric field of 1012 V~m, both of which were record values at the time. When
this laser pulse was reflected off a 46.6 GeV electron beam, the returning photons had an

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1338 ny energy of 29.2 GeV and thus had become intense gamma rays. These gamma rays then
collided with still incoming green photons and produced electron–positron pairs by the
reaction
γ 29.2 + n γ green e+ + e− (586)

for which both final particles were detected by special apparatuses. The experiment thus
showed that light can hit light in nature, and above all, that doing so can produce matter.
This is the nearest one can get to the science fiction idea of light swords or of laser swords
banging onto each other.

Is the vacuum a bath?

If the vacuum is a sea of virtual photons and particle–antiparticle pairs, vacuum could
be suspected to act as a bath. In general, the answer is negative. Quantum field theory
works because the vacuum is not a bath for single particles. However, there is always an
exception. For dissipative systems made of many particles, such as electrical conductors,
Ref. 850 the vacuum can act as a viscous fluid. Irregularly shaped, neutral, but conducting bodies
can emit photons when accelerated, thus damping such type of motion. This is due to
the Fulling–Davies–Unruh effect, also called the dynamical Casimir effect, as described
above. The damping depends on the shape and thus also on the direction of the body’s
Copyright © Christoph Schiller November 1997–May 2006

motion.
Page 242 In 1998, Gour and Sriramkumar even predicted that Brownian motion should also
appear for an imperfect, i.e. partly absorbing mirror placed in vacuum. The fluctuations
Ref. 851 of the vacuum should produce a mean square displacement

`d 2 e = ħ~mt (587)

increasing linearly with time; however, the extremely small displacements produced this
way seem out of experimental reach so far. But the result is not a surprise. Are you able
Challenge 1339 ny to give another, less complicated explanation for it?

Dvipsbugw
 860 vii details of quantum theory • quantum electrodynamics

Renormalization – why is an electron so light?

– CS – section on renormalization to be added – CS –

Sometimes it is claimed that the infinities appearing in quantum electrodynamics in

the intermediate steps of the calculation show that the theory is incomplete or wrong.
However, this type of statement would imply that classical physics is also incomplete or
wrong, on the ground that in the definition of the velocity v with space x and time t, Dvipsbugw
namely
v= = lim = lim ∆x
dx ∆x 1
, (588)
dt ∆t 0 ∆t ∆t 0 ∆t

one gets an infinity as intermediate step. Indeed, dt being vanishingly small, one could
argue that one is dividing by zero. Both arguments show the difficulty to accept that the
result of a limit process can be a finite quantity even if infinite quantities appear in it. The

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

parallel with the definition of the velocity is closer than it seems; both ‘infinities’ stem
from the assumption that space-time is continuous, i.e. infinitely divisible. The infinities
necessary in limit processes for the definition of differentiation, of integration or for the
renormalization scheme appear only when space-time is approximated as a complete set,
or as physicists say, as a ‘continuous’ set.
On the other hand, the conviction that the appearance of an infinity might be a sign
of incompleteness of a theory was an interesting development in physics. It shows how
uncomfortable many physicists had become with the use of infinity in our description of
nature. Notably, this was the case for Dirac himself, who, after having laid in his youth the
basis of quantum electrodynamics, has tried for the rest of his life to find a way, without
Ref. 852 success, to change the theory so that infinities are avoided.*
Renormalization is a procedure that follows from the requirement that continuous
space-time and gauge theories must work together. In particular, it follows form the re-
quirement that the particle concept is consistent, i.e. that perturbation expansions are
possible.

Curiosities and fun challenges of quantum

electrodynamics
Motion is an interesting topic, and when a curious person asks a question about it, most
Copyright © Christoph Schiller November 1997–May 2006

of the time quantum electrodynamics is needed for the answer. Together with gravity,
quantum electrodynamics explains almost all of our everyday experience, including nu-
merous surprises. Let us have a look at some of them.
**
There is a famous riddle asking how far the last card (or the last brick) of a stack can hang
over the edge of a table. Of course, only gravity, no glue or any other means is allowed to
keep the cards on the table. After you solved the riddle, can you give the solution in case
Challenge 1340 n that the quantum of action is taken into account?
* Not long after his death, his wish has been fulfilled, although in a different manner that he envisaged. The
third part of this mountain ascent will show the way out of the issue.

Dvipsbugw
 curiosities and fun challenges of quantum electrodynamics 861

cards
or
bricks l

table

Dvipsbugw

F I G U R E 338 What is the maximum

possible value of h/l?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Quantum electrodynamics explains why there are only a finite number of different atom
Ref. 857 types. In fact, it takes only two lines to prove that pair production of electron–antielectron
pairs make it impossible that a nucleus has more than about 137 protons. Can you show
Challenge 1341 n this? The effect at the basis of this limit, the polarization of the vacuum, also plays a role
Page 875 in much larger systems, such as charged black holes, as we will see shortly.
**
Taking 91 of the 92 electrons off an uranium atom allows researchers to check whether the
innermost electron still is described by QED. The electric field near the uranium nucleus,
1 EV~m is near the threshold for spontaneous pair production. The field is the highest
constant field producible in the laboratory, and an ideal testing ground for precision QED
experiments. The effect of virtual photons is to produce a Lamb shift; even in these ex-
Ref. 858 tremely high fields, the value fits with the predictions.

**
Is there a critical magnetic field in nature, like there is a critical electric field, limited by
Challenge 1342 ny spontaneous pair production?
**
Copyright © Christoph Schiller November 1997–May 2006

In classical physics, the field energy of a point-like charged particle, and hence its mass,
Page 601 was predicted to be infinite. QED effectively smears out the charge of the electron over its
Compton wavelength, so that in the end the field energy contributes only a small correc-
Challenge 1343 ny tion to its total mass. Can you confirm this?
**
Microscopic evolution can be pretty slow. Light, especially when emitted by single atoms,
is always emitted by some metastable state. Usually, the decay times, being induced by
the vacuum fluctuations, are much shorter than a microsecond. However, there are meta-
stable atomic states with a lifetime of ten years: for example, an ytterbium ion in the 2 F7~2
state achieves this value, because the emission of light requires an octupole transition, in

Dvipsbugw
 862 vii details of quantum theory • quantum electrodynamics

Ref. 859 which the angular momentum changes by 3ħ; this is an extremely unlikely process.

**
Microscopic evolution can be pretty fast. Can you imagine how to deduce or to measure
Challenge 1344 n the speed of electrons inside atoms? And inside metals?
**
Take a horseshoe. The distance between the two ends is not fixed, since otherwise their
position and velocity would be known at the same time, contradicting the indeterminacy Dvipsbugw
relation. Of course, this reasoning is also valid for any other solid object. In short, both
quantum mechanics and special relativity show that rigid bodies do not exist, albeit for
different reasons.
**
Have you ever admired a quartz crystal or some other crystalline material? The beautiful

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

shape and atomic arrangement has formed spontaneously, as a result of the motion of
atoms under high temperature and pressure, during the time that the material was deep
under the Earth’s surface. The details of crystal formation are complex and interesting.
For example, are regular crystal lattices energetically optimal? This simple question
leads to a wealth of problems. We might start with the much simpler question º whether a
Challenge 1345 n regular dense packing of spheres is the most dense possible. Its density is π~ 18 , i.e. a bit
over 74 %. Even though this was conjectured to be the maximum possible value already
Ref. 862 in 1609 by Johannes Kepler, the statement was proven only in 1998 by Tom Hales. The
proof is difficult because in small volumes it is possible to pack spheres up to almost 78 %.
To show that over large volumes the lower value is correct is a tricky business.
Next, does a regular crystal of solid spheres, in which the spheres do not touch, have
the lowest possible entropy? This simple problem has been the subject of research only in
the 1990s. Interestingly, for low temperatures, regular sphere arrangements indeed show
the largest possible entropy. At low temperatures, spheres in a crystal can oscillate around
their average position and be thus more disordered than if they were in a liquid; in the
liquid state the spheres would block each other’s motion and would not allow to show
disorder at all.
This many similar results deduced from the research into these so-called entropic forces
show that the transition from solid to liquid is – at least in part – simply a geometrical
effect. For the same reason, one gets the surprising result that even slightly repulsing
Copyright © Christoph Schiller November 1997–May 2006

Ref. 863 spheres (or atoms) can form crystals and melt at higher temperatures. These are beautiful
examples of how classical thinking can explain certain material properties, using from
quantum theory only the particle model of matter.
But the energetic side of crystal formation provides other interesting questions.
Quantum theory shows that it is possible that two atoms repel each other, while three
attract each other. This beautiful effect was discovered and explained by Hans–Werner
Ref. 864 Fink in 1984. He studied rhenium atoms on tungsten surfaces and showed, as observed,
that they cannot form dimers – two atoms moving closeby – but readily form trimers.
This is an example contradicting classical physics; the effect is impossible if one pictures
atoms as immutable spheres, but becomes possible when one remembers that the electron
clouds around the atoms rearrange depending on their environment.

Dvipsbugw
 curiosities and fun challenges of quantum electrodynamics 863

F I G U R E 339 Some snow flakes (© Furukawa Yoshinori) Dvipsbugw

For an exact study of crystal energy, the interactions between all atoms have to be
included. The simplest question is to determine whether a regular array of alternatively
charged spheres has lower energy than some irregular collection. Already such simple
questions are still topic of research; the answer is still open.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Another question is the mechanism of face formation in crystals. Can you confirm
Challenge 1346 n that crystal faces are those planes with the slowest growth speed, because all fast growing
planes are eliminated? The finer details of the process form a complete research field in
Ref. 865 itself.
However, not always the slowest growing planes win out. Figure 339 shows some well-
known exceptions. Explaining such shapes is possible today, and Furukawa Yoshinori is
Ref. 866 one of the experts in the field, heading a dedicated research team. Indeed, there remains
the question of symmetry: why are crystals often symmetric, such as snowflakes, instead
of asymmetric? This issue is a topic of self-organization, as mentioned already in the
Page 255 section of classical physics. It turns out that the symmetry is an automatic result of the
way molecular systems grow under the combined influence of diffusion and nonlinear
processes. The details are still a topic of research.
**
A similar breadth of physical and mathematical problems are encountered in the study
Ref. 867 of liquids and polymers. The ordering of polymer chains, the bubbling of hot water, the
motion of heated liquids and the whirls in liquid jets show complex behaviour that can
be explained with simple models. Turbulence and self-organization will be a fascinating
research field for many years to come. Copyright © Christoph Schiller November 1997–May 2006

**
The ways people handle single atoms with electromagnetic fields is a beautiful example
of modern applied technologies. Nowadays it is possible to levitate, to trap, to excite, to
photograph, to deexcite and to move single atoms just by shining light onto them. In 1997,
the Nobel Prize in physics has been awarded to the originators of the field.
**
In 1997, a Czech group built a quantum version of the Foucault pendulum, using the
Ref. 868 superfluidity of helium. In this beautiful piece of research, they cooled a small ring of fluid
helium below the temperature of 0.28 K, below which the helium moves without friction.
In such situations it thus can behave like a Foucault pendulum. With a clever arrangement,

Dvipsbugw
 864 vii details of quantum theory • quantum electrodynamics

it was possible to measure the rotation of the helium in the ring using phonon signals, and
to show the rotation of the Earth.

**
If an electrical wire is sufficiently narrow, its electrical conductance is quantized in steps
of 2e 2 ~ħ. The wider the wire, the more such steps are added to its conductance. Can you
Challenge 1347 n explain the effect? By the way, quantized conductance has also been observed for light
Ref. 869 and for phonons.
Dvipsbugw
**
An example of modern research is the study of hollow atoms, i.e. atoms missing a number
of inner electrons. They have been discovered in 1990 by J.P. Briand and his group. They
appear when a completely ionized atom, i.e. one without any electrons, is brought in
contact with a metal. The acquired electrons then orbit on the outside, leaving the inner
shells empty, in stark contrast with usual atoms. Such hollow atoms can also be formed

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 870 by intense laser irradiation.

**
In the past, the description of motion with formulae was taken rather seriously. Before
computers appeared, only those examples of motion were studied which could be de-
scribed with simple formulae. It turned out that Galilean mechanics cannot solve the
three-body problem, special relativity cannot solve the two-body problem, general re-
lativity the one-body problem and quantum field theory the zero-body problem. It took
some time to the community of physicists to appreciate that understanding motion does
not depend on the description by formulae, but on the description by clear equations
based on space and time.
**
Challenge 1348 n Can you explain why mud is not clear?

**
Photons not travelling parallel to each other attract each other through gravitation and
thus deflect each other. Could two such photons form a bound state, a sort of atom of
light, in which they would circle each other, provided there were enough empty space for
Challenge 1349 n this to happen?
Copyright © Christoph Schiller November 1997–May 2006

**
Challenge 1350 n Can the universe ever have been smaller than its own Compton wavelength?
In fact, quantum electrodynamics, or QED, provides a vast number of curiosities and
every year there is at least one interesting new discovery. We now conclude the theme
with a more general approach.

How can one move on perfect ice? – The ultimate physics test
In our quest, we have encountered motion of many sorts. Therefore, the following test –
not to be taken too seriously – is the ultimate physics test, allowing to check your under-

Dvipsbugw
 summary of quantum electrodynamics 865

standing and to compare it with that of others.

Imagine that you are on a perfectly frictionless surface and that you want to move to
its border. How many methods can you find to achieve this? Any method, so tiny its effect
may be, is allowed.
Classical physics provided quite a number of methods. We saw that for rotating
ourselves, we just need to turn our arm above the head. For translation motion, throwing
a shoe or inhaling vertically and exhaling horizontally are the simplest possibilities. Can
Challenge 1351 n you list at least six additional methods, maybe some making use of the location of the
surface on Earth? What would you do in space? Dvipsbugw
Electrodynamics and thermodynamics taught us that in vacuum, heating one side of
the body more than the other will work as motor; the imbalance of heat radiation will
push you, albeit rather slowly. Are you able to find at least four other methods from these
Challenge 1352 n two domains?
General relativity showed that turning one arm will emit gravitational radiation un-
Challenge 1353 n symmetrically, leading to motion as well. Can you find at least two better methods?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Quantum theory offers a wealth of methods. Of course, quantum mechanics shows
that we actually are always moving, since the indeterminacy relation makes rest an im-
possibility. However, the average motion can be zero even if the spread increases with
time. Are you able to find at least four methods of moving on perfect ice due to quantum
Challenge 1354 n effects?
Materials science, geophysics, atmospheric physics and astrophysics also provide ways to
Challenge 1355 n move, such as cosmic rays or solar neutrinos. Can you find four additional methods?
Self-organization, chaos theory and biophysics also provide ways to move, when the
inner workings of the human body are taken into account. Can you find at least two
Challenge 1356 n methods?
Assuming that you read already the section following the present one, on the effects
of semiclassical quantum gravity, here is an additional puzzle: is it possible to move by
Challenge 1357 n accelerating a pocket mirror, using the emitted Unruh radiation? Can you find at least
two other methods to move yourself using quantum gravity effects? Can you find one
from string theory?
If you want points for the test, the marking is simple. For students, every working
method gives one point. Eight points is ok, twelve points is good, sixteen points is very
good, and twenty points or more is excellent.For graduated physicists, the point is given
only when a back-of-the-envelope estimate for the ensuing momentum or acceleration
is provided.
Copyright © Christoph Schiller November 1997–May 2006

Summary of quantum electrodynamics

The shortest possible summary of quantum electrodynamics is the following: matter is
made of charged particles which interact through photon exchange in the way described by
Figure 340.
No additional information is necessary. In a bit more detail, quantum electrodynamics
starts with elementary particles, characterized by their mass, their spin and their charge,
and with the vacuum, essentially a sea of virtual particle–antiparticle pairs. Interactions
between charged particles are described as the exchange of virtual photons, and decay is
described as the interaction with the virtual photons of the vacuum.

Dvipsbugw
866 vii details of quantum theory • quantum electrodynamics

radiation:
q=0
matter: s=1
q=e
s = 1/2

interaction:
α = 1/137.0359... Dvipsbugw
Σ E=0
Σ p=0
Σ s=0
Σ q=0

F I G U R E 340 QED as perturbation theory

in space-time

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

All physical results of QED can be calculated by using the single diagram of Figure 340.
As QED is a perturbative theory, the diagram directly describes the first order effects and
its composites describe effects of higher order. QED is a perturbative theory.
QED describes all everyday properties of matter and radiation. It describes the divis-
ibility down to the smallest constituents, the isolability from the environment and the
impenetrability of matter. It also describes the penetrability of radiation. All these proper-
ties are due to electromagnetic interactions of constituents and follow from Figure 340.
Matter is divisible because the interactions are of finite strength, matter is separable be-
cause the interactions are of finite range, and matter is impenetrable because interactions
among the constituents increase in intensity when they approach each other, in particu-
lar because matter constituents are fermions. Radiation is divisible into photons, and is
penetrable because photons are bosons and first order photon-photon interactions do
not exist.
Both matter and radiation are made of elementary constituents. These elementary
constituents, whether bosons or fermions, are indivisible, isolable, indistinguishable, and
point-like.
To describe observations, it is necessary to use quantum electrodynamics in all those
Copyright © Christoph Schiller November 1997–May 2006

situations for which the characteristic dimensions d are of the order of the Compton
wavelength
d  λC =
h
. (589)
mc
In situations where the dimensions are of the order of the de Broglie wavelength, or equi-
valently, where the action is of the order of the Planck value, simple quantum mechanics
is sufficient:
d  λ dB =
h
. (590)
mv
For larger dimensions, classical physics will do.

Dvipsbugw
 summary of quantum electrodynamics 867

Together with gravity, quantum electrodynamics explains almost all observations of

motion on Earth; QED unifies the description of matter and radiation in daily life. All
objects and all images are described by it, including their properties, their shape, their
transformations and their other changes. This includes self-organization and chemical
or biological. In other words, QED gives us full grasp of the effects and the variety of
motion due to electromagnetism.

Open questions in QED

Dvipsbugw
Even though QED describes motion without any discrepancy from experiment, that does
not mean that we understand every detail of every example of electric motion. For ex-
ample, nobody has described the motion of an animal with QED yet.* In fact, there is
beautiful and fascinating work going on in many branches of electromagnetism.
Atmospheric physics still provides many puzzles and regularly delivers new, previously
Ref. 872 unknown phenomena. For example, the detailed mechanisms at the origin of aurorae

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 873 are still controversial; and the recent unexplained discoveries of discharges above clouds
should not make one forget that even the precise mechanism of charge separation inside
clouds, which leads to lightning, is not completely clarified. In fact, all examples of elec-
trification, such as the charging of amber through rubbing, the experiment which gave
electricity its name, are still poorly understood.
Materials science in all its breadth, including the study of solids, fluids, and plasmas,
as well as biology and medicine, still provides many topics of research. In particular, the
twenty-first century will undoubtedly be the century of the life sciences.
The study of the interaction of atoms with intense light is an example of present re-
search in atomic physics. Strong lasers can strip atoms of many of their electrons; for
such phenomena, there are not yet precise descriptions, since they do not comply to the
weak field approximations usually assumed in physical experiments. In strong fields, new
Ref. 875 effects take place, such as the so-called Coulomb explosion.
But also the skies have their mysteries. In the topic of cosmic rays, it is still not clear
Ref. 874 how rays with energies of 1022 eV are produced outside the galaxy. Researchers are in-
tensely trying to locate the electromagnetic fields necessary for their acceleration and to
understand their origin and mechanisms.
In the theory of quantum electrodynamics, discoveries are expected by all those who
Ref. 876 study it in sufficient detail. For example, Dirk Kreimer has found that higher order inter-
action diagrams built using the fundamental diagram of Figure 340 contain relations to
Copyright © Christoph Schiller November 1997–May 2006

the theory of knots. This research topic will provide even more interesting results in the
near future.
Relations to knot theory appear because QED is a perturbative description, with the vast
richness of its nonperturbative effects still hidden. Studies of QED at high energies, where
perturbation is not a good approximation and where particle numbers are not conserved,
promise a wealth of new insights. We will return to the topic later on.
High energies provide many more questions. So far, the description of motion was
based on the idea that measurable quantities can be multiplied and added. This always

* On the other hand, there is beautiful work going on how humans move their limbs; it seems that humans
Ref. 871 move by combining a small set of fundamental motions.

Dvipsbugw
 868 vii details of quantum theory • 27. including gravitation

length l
A collapsing house

height h

figure to be inserted silicon

neutron beam mirrors

F I G U R E 341 The weakness of gravitation

Dvipsbugw

happens at one space-time point. In mathematical jargon, observables form a local al-
gebra. Thus the structure of an algebra contains, implies and follows from the idea that
local properties lead to local properties. We will discover later on that this basic assump-
tion is wrong at high energies.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We defined special relativity using v D c, general relativity using L~M E 4G~c 2 and
Challenge 1358 r quantum theory using S E ħ~2. How can we define electromagnetism in one statement?
This is not known yet.
Many other open issues of more practical nature have not been mentioned. Indeed,
by far the largest numbers of physicists get paid for some form of applied QED. However,
our quest is the description of the fundamentals of motion. So far, we have not achieved
it. For example, we still need to understand motion in the realm of atomic nuclei. But
before we do that, we take a first glimpse of the strange issues appearing when gravity
and quantum theory meet.

27. quantum mechanics with gravitation – the first

approach
Gravitation is a weak effect. Every seaman knows it: storms are the worst part of his life,
not gravity. Nevertheless, including gravity into quantum mechanics yields a list of im-
portant issues.
Only in 2004 it became possible to repeat Galileo’s experiment with atoms: indeed,
Ref. 877 single atoms fall like stones. In particular, atoms of different mass fall with the same acce-
leration, within the experimental precision of a part in 6 million.
Copyright © Christoph Schiller November 1997–May 2006

In the chapter on general relativity we already mentioned that light frequency changes
with height. But for matter wave functions, gravity also changes their phase. Can you ima-
Challenge 1359 ny gine why? The effect was first confirmed in 1975 with the help of neutron interferometers,
where neutron beams are brought to interference after having climbed some height h
at two different locations. The experiment is shown schematically in Figure 341; it fully
confirmed the predicted phase difference

δφ =
mдhl
(591)
ħv

where l is the distance of the two climbs and v and m are the speed and mass of the

Dvipsbugw
 quantum mechanics with gravitation – the first approach 869

Ref. 878 neutrons. These beautifully simple experiments have confirmed the formula within ex-
perimental errors.*
Ref. 880 In the 1990s, similar experiments have even been performed with complete atoms.
These set-ups allow to build interferometers so sensitive that local gravity д can be meas-
ured with a precision of more than eight significant digits.

Corrections to the Schrödinger equation

In 2002, the first observation of actual quantum states due to gravitational energy was Dvipsbugw
Ref. 881 performed. Any particle above the floor should feel the effect of gravity.
In a few words, one can say that because the experimenters managed to slow down
neutrons to the incredibly small value of 8 m~s, using grazing incidence on a flat plate
they could observe how neutrons climbed and fell back due to gravity with speeds below
a few cm~s.
Obviously, the quantum description is a bit more involved. The lowest energy level for
neutrons due to gravity is 2.3 ë 10−31 J, or 1.4 peV. To get an impression of it smallness, we

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

can compare it to the value of 2.2 ë 10−18 J or 13.6 eV for the lowest state in the hydrogen
atom.

A rephrased large number hypothesis

Despite its weakness, gravitation provides many puzzles. Most famous are a number of
curious coincidences that can be found when quantum mechanics and gravitation are
combined. They are usually called ‘large number hypotheses’ because they usually involve
large dimensionless numbers. A pretty, but less well known version connects the Planck
Ref. 903 length, the cosmic horizon, and the number of baryons:

R0 4 t0 4
(N b )3  ‹  = ‹   10244 (592)
l Pl t Pl

in which N b = 1081 and t 0 = 1.2 ë 1010 a were used. There is no known reason why the
number of baryons and the horizon size R 0 should be related in this way. This coincidence
is equivalent to the one originally stated by Dirac,** namely

ħ2
m p3  . (594)
Gct 0
Copyright © Christoph Schiller November 1997–May 2006

* Due to the influence of gravity on phases of wave functions, some people who do not believe in bath in-
duced decoherence have even studied the influence of gravity on the decoherence process of usual quantum
Ref. 879 systems in flat space-time. Predictably, the calculated results do not reproduce experiments.
Ref. 904 ** The equivalence can be deduced using Gn b m p = 1~t 02 , which, as Weinberg explains, is required by several
Page 1044 cosmological models. Indeed, this can be rewritten simply as

m 02 ~R 02  m Pl
2
~R Pl
2
= c 4 ~G 2 . (593)

Together with the definition of the baryon density n b = N b ~R 03 one gets Dirac’s large number hypothesis,
¼
»
substituting protons for pions. Note that the Planck time and length are defined as ħG~c 5 and ħG~c 3
and are the natural units of length and time. We will study them in detail in the third part of the mountain
ascent.

Dvipsbugw
 870 vii details of quantum theory • 27. including gravitation

where m p is the proton mass. This approximate equality seems to suggest that certain
microscopic properties, namely the mass of the proton, is connected to some general
properties of the universe as a whole. This has lead to numerous speculations, especially
since the time dependence of the two sides differs. Some people even speculate whether
relations (592) or (594) express some long-sought relation between local and global topo-
Ref. 905 logical properties of nature. Up to this day, the only correct statement seems to be that
they are coincidences connected to the time at which we happen to live, and that they
should not be taken too seriously.
Dvipsbugw
Is quantum gravity necessary?
One might think that gravity does not require a quantum description. We remember that
we stumbled onto quantum effects because classical electrodynamics implies, in stark
contrast with reality, that atoms decay in about 0.1 ns. Classically, an orbiting electron
would emit radiation until it falls into the nucleus. Quantum theory is necessary to save

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the situation.
When the same calculation is performed for the emission of gravitational radiation by
Challenge 1360 ny orbiting electrons, one finds a decay time of around 1037 s. (True?) This extremely large
value, trillions of times longer than the age of the universe, is a result of the low emission
Page 407 of gravitational radiation by rotating masses. Therefore, the existence of atoms does not
require a quantum theory of gravity.
Indeed, quantum gravity is unnecessary in every single domain of everyday life. How-
ever, quantum gravity is necessary in domains which are more remote, but also more
fascinating.

Limits to disorder
Die Energie der Welt ist constant.

“ Die Entropie der Welt strebt einem Maximum

zu.*
Rudolph Clausius

Page 471
We have already encountered the famous statement by Clausius, the father of the term
‘entropy’. Strangely, for over hundred years nobody asked whether there actually exists a
theoretical maximum for entropy. This changed in 1973, when Jakob Bekenstein found
”
Ref. 882 the answer while investigating the consequences gravity has for quantum physics. He
found that the entropy of an object of energy E and size L is bound by
Copyright © Christoph Schiller November 1997–May 2006

S D EL
kπ
(595)
ħc
for all physical systems. In particular, he deduced that (nonrotating) black holes saturate
Challenge 1361 n the bound, with an entropy given by

kc 3 A kG
S= = 4πM 2 (596)
Għ 4 ħc

* The energy of the universe is constant. Its entropy tends towards a maximum.

Dvipsbugw
 quantum mechanics with gravitation – the first approach 871

where A is now the area of the horizon of the black hole. It is given by A = 4πR 2 =
4π(2GM~c 2 )2 . In particular, the result implies that every black hole has an entropy. Black
holes are thus disordered systems described by thermostatics. Black holes are the most
disordered systems known.*
As an interesting note, the maximum entropy also gives a memory limit for memory
Challenge 1362 n chips. Can you find out how?
What are the different microstates leading to this macroscopic entropy? It took many
years to convince physicists that the microstates have to do with the various possible
states of the horizon itself, and that they are due to the diffeomorphism invariance at this Dvipsbugw
Ref. 884 boundary. As Gerard ’t Hooft explains, the entropy expression implies that the number
of degrees of freedom of a black hole is about (but not exactly) one per Planck area of the
Challenge 1363 n horizon.
If black holes have entropy, they must have a temperature. What does this temperat-
ure mean? In fact, nobody believed this conclusion until two unrelated developments
confirmed it within a few months. All these results were waiting to be discovered since

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the 1930s, even though, incredibly, nobody had thought about them for over 40 years.

Measuring acceleration with a thermometer –

Fulling–Davies–Unruh radiation
Independently, Stephen Fulling in 1973, Paul Davies in 1975 and William Unruh in 1976
Ref. 885 made the same theoretical discovery while studying quantum theory: if an inertial ob-
server observes that he is surrounded by vacuum, a second observer accelerated with re-
spect to the first does not: he observes black body radiation. The radiation has a spectrum
corresponding to the temperature

T=a
ħ
. (597)
2πkc
The result means that there is no vacuum on Earth, because any observer on its surface
can maintain that he is accelerated with 9.8 m~s2 , thus leading to T = 40 zK! We can thus
measure gravity, at least in principle, using a thermometer. However, even for the largest
practical accelerations the temperature values are so small that it is questionable whether
Ref. 886 the effect will ever be confirmed experimentally. But if it will, it will be a great experiment.
When this effect was predicted, people studied the argument from all sides. For ex-
ample, it was then found that the acceleration of a mirror leads to radiation emission!
Copyright © Christoph Schiller November 1997–May 2006

Mirrors are thus harder to accelerate than other bodies of the same mass.
When the acceleration is high enough, also matter particles can be detected. If a
particle counter is accelerated sufficiently strongly across the vacuum, it will start count-

* The precise discussion that black holes are the most disordered systems in nature is quite subtle. It is
Ref. 883 summarized by Bousso. Bousso claims that the area appearing in the maximum entropy formula cannot be
taken naively as the area at a given time, and gives four arguments why this should be not allowed. However,
all four arguments are wrong in some way, in particular because they assume that lengths smaller than the
Planck length or larger than the universe’s size can be measured. Ironically, he brushes aside some of the
arguments himself later in the paper, and then deduces an improved formula, which is exactly the same as
the one he criticizes first, just with a different interpretation of the area A. In short, the expression of black
hole entropy is the maximum entropy for a physical system with surface A.

Dvipsbugw
 872 vii details of quantum theory • 27. including gravitation

ing particles! We see that the difference between vacuum and matter becomes fuzzy at
large energies.
For completeness, we mention that also an observer in rotational motion detects radi-
ation following expression (597).

Bl ack holes aren’t bl ack

In 1974, the English physicist Stephen Hawking, famous for the courage with which he
fights a disease which forces him into the wheelchair, surprised the world of general Dvipsbugw
relativity with a fundamental theoretical discovery. He found that if a virtual particle–
antiparticle pair appeared in the vacuum near the horizon, there is a finite chance that
one particle escapes as a real particle, while the virtual antiparticle is captured by the
black hole. The virtual antiparticle is thus of negative energy, and reduces the mass of
the black hole. The mechanism applies both to fermions and bosons. From far away this
effect looks like the emission of a particle. Hawking’s detailed investigation showed that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the effect is most pronounced for photon emission. In particular, Hawking showed that
black holes radiate as black bodies.
Black hole radiation confirms both the result on black hole entropy by Bekenstein
and the effect for observers accelerated in vacuum found by Fulling, Davies and Unruh.
When all this became clear, a beautiful Gedanken experiment was published by William
Ref. 887 Unruh and Robert Wald, showing that the whole result could have been deduced already
50 years earlier!
Shameful as this delay of the discovery is
for the community of theoretical physicists, space station
the story itself remains beautiful. It starts in with dynamo
the early 1970s, when Robert Geroch stud-
ied the issue shown in Figure 342. Imagine
a mirror box full of heat radiation, thus full rope
of light. The mass of the box is assumed to
box filled
be negligible, such as a box made of thin alu- with light
minium paper. We lower the box, with all
its contained radiation, from a space station
horizon
towards a black hole. On the space station,
lowering the weight of the heat radiation al-
lows to generate energy. Obviously, when the black
Copyright © Christoph Schiller November 1997–May 2006

box reaches the black hole horizon, the heat support hole
shell
radiation is red-shifted to infinite wavelength.
At that point, the full amount of energy ori-
ginally contained in the heat radiation has
been provided to the space station. We can F I G U R E 342 A Gedanken experiment allowing
now do the following: we can open the box to deduce the existence of black hole radiation
on the horizon, let drop out whatever is still
inside, and wind the empty and massless box back up again. As a result, we have com-
pletely converted heat radiation into mechanical energy. Nothing else has changed: the
black hole has the same mass as beforehand.
But this result contradicts the second principle of thermodynamics! Geroch concluded

Dvipsbugw
 bl ack holes aren’t bl ack 873

that something must be wrong. We must have forgotten an effect which makes this pro-
cess impossible.
In the 1980s, Unruh and Wald showed that black hole radiation is precisely the for-
gotten effect that puts everything right. Because of black hole radiation, the box feels
buoyancy, so that it cannot be lowered down to the horizon. It floats somewhat above
it, so that the heat radiation inside the box has not yet zero energy when it falls out of
the opened box. As a result, the black hole does increase in mass and thus in entropy. In
summary, when the empty box is pulled up again, the final situation is thus the following:
only part of the energy of the heat radiation has been converted into mechanical energy, Dvipsbugw
part of the energy went into the increase of mass and thus of entropy of the black hole.
The second principle of thermodynamics is saved.
Well, it is only saved if the heat radiation has precisely the right energy density at
the horizon and above. Let us have a look. The centre of the box can only be lowered
up to a hovering distance d above the horizon, where the acceleration due to gravity is
д = c 2 ~4GM. The energy E gained by lowering the box is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

dc 2
E = mc 2 − mд = mc 2 (1 − )
d
(598)
2 8GM

The efficiency of the process is η = E~mc 2 . To be consistent with the second law of ther-
modynamics, this efficiency must obey

η= =1−
E TBH
e (599)
mc 2 T

We thus find a black hole temperature TBH given by the hovering distance d. That hov-
ering distance d is roughly given by the size of the box. The box size in turn must be at
least the wavelength of the thermal radiation; in first approximation, Wien’s relation gives
d  ħc~kT. A precise calculation, first performed by Hawking, gives the result

ħc 3 c4
TBH = = = дsurf =
ħc 1 ħ
дsurf with (600)
8πkGM 4πk R 2πkc 4GM
where R and M are the radius and the mass of the black hole. It is either called the black-
hole temperature or Bekenstein-Hawking temperature. As an example, a black hole with
Copyright © Christoph Schiller November 1997–May 2006

the mass of the Sun would have the rather small temperature of 62 nK, whereas a smaller
black hole with the mass of a mountain, say 1012 kg, would have a temperature of 123 GK.
That would make quite a good oven. All known black hole candidates have masses in the
range from a few to a few million solar masses. The radiation is thus extremely weak, the
reason being that the emitted wavelength is of the order of the black hole radius, as you
Challenge 1364 ny might want to check. The radiation emitted by black holes is often also called Bekenstein-
Ref. 888 Hawking radiation.
Black hole radiation is thus so weak that we can speak of an academic effect. It leads
Challenge 1365 ny to a luminosity that increases with decreasing mass or size as

Dvipsbugw
 874 vii details of quantum theory • 27. including gravitation

TA B L E 65 The principles of thermodynamics and those of horizon mechanics

Principle Th e r mody na m i c s Hori z on s

Zeroth principle the temperature T is con- the surface gravity a is con-

stant in a body at equilib- stant on the horizon
rium
First principle energy is conserved: dE = energy is conserved:
TdS − pdV + µdN d(mc ) = 8πG dA + ΩdJ +
2 ac 2
Dvipsbugw
Φdq
Second principle entropy never decreases: surface area never de-
dS E 0 creases: dA E 0
Third principle T = 0 cannot be achieved a = 0 cannot be achieved

c 6 ħ π2
L  2 L = nAσ T 4 = n
1 1

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

or (601)
M 2 R G 2 M 2 15 ë 27
Page 611 where σ is the Stefan–Boltzmann or black body radiation constant, n is the number of
particle degrees of freedom that can be radiated; if only photons are radiated, we have
n = 2. (For example, if neutrinos were massless, they would be emitted more frequently
Ref. 889 than photons.)
Black holes thus shine, and the more the smaller they are. This is a genuine quantum
effect, since classically, black holes, as the name says, cannot emit any light. Even though
the effect is academically weak, it will be of importance later on. In actual systems, many
other effect around black holes increase the luminosity far above this value; indeed, black
holes are usually brighter than normal stars, due to the radiation emitted by the matter
falling into them. But that is another story. Here we are only treating isolated black holes,
surrounded only by vacuum.
Due to the emitted radiation, black holes gradually lose mass. Therefore their theoret-
Challenge 1366 ny ical lifetime is finite. A calculation shows that it is given by

20 480 π G 2
t = M3  M 3 3.4 ë 10−16 s~kg3 (602)
ħc 4
as function of their initial mass M. For example, a black hole with mass of 1 g would have
Copyright © Christoph Schiller November 1997–May 2006

a lifetime of 3.4 ë 10−25 s, whereas a black hole of the mass of the Sun, 2.0 ë 1030 kg, would
have a lifetime of about 1068 years. Obviously, these numbers are purely academic. In any
case, black holes evaporate. However, this extremely slow process for usual black holes
determines their lifetime only if no other, faster process comes into play. We will present
a few such processes shortly. Hawking radiation is the weakest of all known effects. It is
not masked by stronger effects only if the black hole is non-rotating, electrically neutral
and with no matter falling into it from the surroundings.
So far, none of these quantum gravity effects has been confirmed experimentally, as the
values are much too small to be detected. However, the deduction of a Hawking temper-
Ref. 890 ature has been beautifully confirmed by a theoretical discovery of Unruh, who found that
there are configurations of fluids in which sound waves cannot escape, so-called ‘silent

Dvipsbugw
 bl ack holes aren’t bl ack 875

holes’. Consequently, these silent holes radiate sound waves with a temperature satisfying
the same formula as real black holes. A second type of analogue system, namely optical
Ref. 892 black holes, are also being investigated.

Gamma ray bursts

In 1975, a much more dramatic radiation effect than black hole radiation was predicted
Ref. 893 for charged black holes by Damour and Ruffini. Charged black holes have a much shorter
lifetime than just presented, because during their formation a second process takes place. Dvipsbugw
In a region surrounding them the electric field is larger than the so-called vacuum po-
larization value, so that large numbers of electron-positron pairs are produced, which
then almost all annihilate. This process effectively reduces the charge of the black hole to
a value for which the field is below critical everywhere, while emitting large amounts of
high energy light. It turns out that the mass is reduced by up to 30 % in a time of the order
of seconds. That is quite shorter than 1068 years. This process thus produces an extremely

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

intense gamma ray burst.
Such gamma ray bursts had been discovered in the late 1960s by military satellites
which were trying to spot nuclear explosions around the world through their gamma ray
Ref. 894 emission. The satellites found about two such bursts per day, coming from all over the
sky. Another satellite, the Compton satellite, confirmed that they were extragalactic in
origin, and that their duration varied between a sixtieth of a second and about a thousand
seconds. In 1996, the Italian-Dutch BeppoSAX satellite started mapping and measuring
gamma ray bursts systematically. It discovered that they were followed by an afterglow in
the X-ray domain of many hours, sometimes of days. In 1997 afterglow was discovered
also in the optical domain. The satellite also allowed to find the corresponding X-ray,
optical and radio sources for each burst. These measurements in turn allowed to determ-
Ref. 895 ine the distance of the burst sources; red-shifts between 0.0085 and 4.5 were measured.
In 1999 it also became possible to detect optical bursts corresponding to the gamma ray
ones.*
All this data together show that the gamma ray bursts have energies ranging from
10 W to 1045 W. The larger value is about one hundredth of the brightness all stars of
40

Challenge 1367 n the whole visible universe taken together! Put differently, it is the same amount of energy
that is released when converting several solar masses into radiation within a few seconds.
In fact, the measured luminosity is near the theoretical maximum luminosity a body can
have. This limit is given by
Copyright © Christoph Schiller November 1997–May 2006

c5
L < L Pl = = 0.9 ë 1052 W , (603)
4G
Challenge 1368 e as you might want to check yourself. In short, the sources of gamma ray bursts are the
Ref. 895 biggest bombs found in the universe. In fact, more detailed investigations of experimental
data confirm that gamma ray bursts are ‘primal screams’ of black holes in formation.
With all this new data, Ruffini took up his 1975 model again in 1997 and with his col-
laborators showed that the gamma ray bursts generated by the annihilation of electron-
positrons pairs created by vacuum polarization, in the region they called the dyadosphere,
* For more about this fascinating topic, see the http://www.aip.de/~jcg/grb.html website by Jochen Greiner.

Dvipsbugw
 876 vii details of quantum theory • 27. including gravitation

Dvipsbugw

F I G U R E 343 A selection of gamma ray bursters observed in the sky

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

have a luminosity and a duration exactly as measured, if a black hole of about a few up to
30 solar masses is assumed. Charged black holes therefore reduce their charge and mass
through the vacuum polarization and electron positron pair creation process. (The pro-
cess reduces the mass because it is one of the few processes which is reversible; in contrast,
most other attempts to reduce charge on a black hole, e.g. by throwing in a particle with
the opposite charge, increase the mass of the black hole and are thus irreversible.) The left
over remnant then can lose energy in various ways and also turns out to be responsible for
the afterglow discovered by the BeppoSAX satellite. Among others, Ruffini’s team specu-
lates that the remnants are the sources for the high energy cosmic rays, whose origin had
Ref. 896 not been localized so far. All these exciting studies are still ongoing.
Recent studies distinguish two classes of gamma ray bursts. Short gamma ray bursts,
with a duration between a millisecond and two seconds, differ significantly in energy and
spectrum from long gamma ray bursts, with an average length of ten seconds, a higher
Ref. 898 energy content and a softer energy spectrum. It is often speculated that short bursts are
due to merging neutron stars or merging black holes, whereas long bursts are emitted,
as just explained, when a black hole is formed in a supernova or hypernova explosion. It
Copyright © Christoph Schiller November 1997–May 2006

also seems that gamma ray bursts are not of spherical symmetry, but that the emission
takes place in a collimated beam. This puts the energy estimates given above somewhat
into question. The details of the formation process are still subject to intense exploration.
Other processes leading to emission of radiation from black holes are also possible.
Examples are matter falling into the black hole and heating up, matter being ejected from
Page 484 rotating black holes through the Penrose process, or charged particles falling into a black
hole. These mechanisms are at the origin of quasars, the extremely bright quasi-stellar
sources found all over the sky. They are assumed to be black holes surrounded by matter,
in the development stage following gamma ray bursters. The details of what happens in
quasars, the enormous voltages (up to 1020 V) and magnetic fields generated, as well as
their effects on the surrounding matter are still object of intense research in astrophysics.

Dvipsbugw
 bl ack holes aren’t bl ack 877

Material properties of black holes

Once the concept of entropy of a black hole was established, people started to think about
black holes like about any other material object. For example, black holes have a matter
density, which can be defined by relating their mass to a fictitious volume defined by
4πR 3 ~3. This density is given by

3c 6
ρ=
1
(604)
M 2 32πG 3 Dvipsbugw

and can be quite low for large black holes. For the highest black holes known, with 1000
Challenge 1369 e million solar masses or more, the density is of the order of the density of air. Nevertheless,
even in this case, the density is the highest possible in nature for that mass.
By the way, the gravitational acceleration at the horizon is still appreciable, as it is given
by
1 c4 c2
дsurf = =

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(605)
M 4G 2R

Challenge 1370 ny which is still 15 km~s2 for an air density black hole.
Obviously, the black hole temperature is related to the entropy S by its usual definition

= V = W
1 ∂S ∂S
T ∂E ρ ∂(Mc 2 ) ρ
(606)

All other thermal properties can be deduced by the standard relations from thermostatics.
In particular, it looks as if black holes are the matter states with the largest possible
Challenge 1371 ny entropy. Can you confirm this statement?
It also turns out that black holes have a negative heat capacity: when heat is added, they
cool down. In other words, black holes cannot achieve equilibrium with a bath. This is
not a real surprise, since any gravitationally bound material system has negative specific
heat. Indeed, it takes only a bit of thinking to see that any gas or matter system collapsing
Challenge 1372 ny under gravity follows dE~dR A 0 and dS~dR A 0. That means that while collapsing, the
Challenge 1373 n energy and the entropy of the system shrink. (Can you find out where they go?) Since
temperature is defined as 1~T = dS~dE, temperature is always positive; from the temper-
ature increase dT~dR < 0 during collapse one deduces that the specific heat dE~dT is
Copyright © Christoph Schiller November 1997–May 2006

Ref. 899 negative.

Black holes, like any object, oscillate when slightly perturbed. These vibrations have
Ref. 900 also been studied; their frequency is proportional to the mass of the black hole.
Nonrotating black holes have no magnetic field, as was established already in the 1960s
Ref. 889 by Russian physicists. On the other hand, black holes have something akin to a finite elec-
trical conductivity and a finite viscosity. Some of these properties can be understood if the
Ref. 897 horizon is described as a membrane, even though this model is not always applicable. In
any case, one can study and describe macroscopic black holes like any other macroscopic
material body. The topic is not closed.

Dvipsbugw
 878 vii details of quantum theory • 27. including gravitation

How do black holes evaporate?

When a nonrotating and uncharged black hole loses mass by radiating Hawking radi-
ation, eventually its mass reaches values approaching the Planck mass, namely a few mi-
crograms. Expression (602) for the lifetime, applied to a black hole of Planck mass, yields
a value of over sixty thousand Planck times. A surprising large value. What happens in
those last instants of evaporation?
A black hole approaching the Planck mass at some time will get smaller than its own
Compton wavelength; that means that it behaves like an elementary particle, and in par- Dvipsbugw
ticular, that quantum effects have to be taken into account. It is still unknown how these
final evaporation steps take place, whether theºmass continues to diminish smoothly or
in steps (e.g. with mass values decreasing as n when n approaches zero), how its in-
ternal structure changes, whether a stationary black hole starts to rotate (as the author
predicts), how the emitted radiation deviates from black body radiation. There is still
enough to study. However, one important issue has been settled.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The information paradox of black holes
When the thermal radiation of black holes was discovered, one question was hotly de-
bated for many years. The matter forming a black hole can contain lots of information;
e.g., imagine the black hole formed by a large number of books collapsing onto each other.
On the other hand, a black hole radiates thermally until it evaporates. Since thermal radi-
ation carries no information, it seems that information somehow disappears, or equival-
ently, that entropy increases.
An incredible number of papers have been written about this problem, some even
claiming that this example shows that physics as we know it is incorrect and needs to
be changed. As usual, to settle the issue, we need to look at it with precision, laying all
prejudice aside. Three intermediate questions can help us finding the answer.
— What happens when a book is thrown into the Sun? When and how is the information
radiated away?
— How precise is the sentence that black hole radiate thermal radiation? Could there be
a slight deviation?
— Could the deviation be measured? In what way would black holes radiate information?
Challenge 1374 e You might want to make up your own mind before reading on.
Let us walk through a short summary. When a book or any other highly complex – or
low entropy – object is thrown into the Sun, the information contained is radiated away.
Copyright © Christoph Schiller November 1997–May 2006

The information is contained in some slight deviations from black hole radiation, namely
in slight correlations between the emitted radiation emitted over the burning time of the
Sun. A short calculation, comparing the entropy of a room temperature book and the
information contained in it, shows that these effects are extremely small and difficult to
measure.
Ref. 901 A clear exposition of the topic was given by Don Page. He calculated what information
would be measured in the radiation if the system of black hole and radiation together
would be in a pure state, i.e. a state containing specific information. The result is simple.
Even if a system is large – consisting of many degrees of freedom – and in pure state, any
smaller subsystem nevertheless looks almost perfectly thermal. More specifically, if a total
system has a Hilbert space dimension N = nm, where n and m D n are the dimensions

Dvipsbugw
 bl ack holes aren’t bl ack 879

of two subsystems, and if the total system is in a pure state, the subsystem m would have
Challenge 1375 ny an entropy S m given by
1−m mn
Sm =
1
+ Q (607)
2n k=n+1 k

which is approximately given by

S m = ln m − mQ1.
m
for (608) Dvipsbugw
2n
To discuss the result, let us think of n and m as counting degrees of freedom, instead of
Hilbert space dimensions. The first term in equation (608) is the usual entropy of a mixed
state. The second term is a small deviation and describes the amount of specific informa-
tion contained in the original pure state; inserting numbers, one finds that it is extremely
small compared to the first. In other words, the subsystem m is almost indistinguishable

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

from a mixed state; it looks like a thermal system even though it is not.
A calculation shows that the second, small term on the right of equation (608) is in-
deed sufficient to radiate away, during the lifetime of the black hole, any information
contained in it. Page then goes on to show that the second term is so small that not only
it is lost in measurements; it is also lost in the usual, perturbative calculations for physical
systems.
The question whether any radiated information could be measured can now be
answered directly. As Don Page showed, even measuring half of the system only gives
Ref. 902 about 1/2 bit of that information. It is necessary to measure the complete system to meas-
ure all the contained information. In summary, at a given instant, the amount of inform-
ation radiated by a black hole is negligible when compared with the total black hole radi-
ation, and is practically impossible to detect by measurements or even by usual calcula-
tions.

More paradoxes
A black hole is a macroscopic object, similar to a star. Like all objects, it can interact with
its environment. It has the special property to swallow everything that falls into them.
This immediately leads us to ask if we can use this property to cheat around the usual
everyday ‘laws’ of nature. Some attempts have been studied in the section on general
Page 485 relativity and above; here we explore a few additional ones.
Copyright © Christoph Schiller November 1997–May 2006

**
Apart from the questions of entropy, we can look for methods to cheat around conserva-
Challenge 1376 ny tion of energy, angular momentum, or charge. Every Gedanken experiment comes to the
same conclusions. No cheats are possible; in addition, the maximum number of degrees
of freedom in a region is proportional to the surface area of the region, and not to its
volume. This intriguing result will keep us busy for quite some time.
**
A black hole transforms matter into antimatter with a certain efficiency. Thus one might

Dvipsbugw
 880 vii details of quantum theory • 27. including gravitation

Challenge 1377 ny look for departures from particle number conservation. Are you able to find an example?

**
Black holes deflect light. Is the effect polarization dependent? Gravity itself makes no
difference of polarization; however, if virtual particle effects of QED are included, the story
Ref. 906 might change. First calculations seem to show that such a effect exists, so that gravitation
might produce rainbows. Stay tuned.
** Dvipsbugw
If lightweight boxes made of mirrors can float in radiation, one gets a strange con-
sequence: such a box might self-accelerate in free space. In a sense, an accelerated box
could float on the Fulling–Davies–Unruh radiation it creates by its own acceleration.
Are you able to show the following: one reason why this is impossible is a small but
Challenge 1378 ny difference between gravity and acceleration, namely the absence of tidal effects. (Other
reasons, such as the lack of perfect mirrors, also make the effect impossible.)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
In 2003, Michael Kuchiev has made the spectacular prediction that matter and radiation
with a wavelength larger than the diameter of a black hole is partly reflected when it hits
Ref. 907 a black hole. The longer the wavelength, the more efficient the reflection would be. For
stellar or even bigger black holes, only photons or gravitons are predicted to be reflected.
Black holes are thus not complete trash cans. Is the effect real? The discussion is still
ongoing.

Quantum mechanics of gravitation

Let us take a conceptual step at this stage. So far, we looked at quantum theory with
gravitation; now we have a glimpse at quantum theory of gravitation.
If we focus on the similarity between the electromagnetic field and the gravitational
‘field,’ we should try to find the quantum description of the latter. Despite attempts by
many brilliant minds for almost a century, this approach was not successful.* Let us see
why.

The gravitational B ohr atom

A short calculation shows that an electron circling a proton due to gravity alone, without
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1379 ny electrostatic attraction, would do so at a gravitational Bohr radius of

ħ2
r gr.B. = = 1.1 ë 1029 m (609)
G m e2 m p

which is about a thousand times the distance to the cosmic horizon. In fact, even in the
normal hydrogen atom there is not a single way to measure gravitational effects. (Are you
Challenge 1380 ny able to confirm this?) But why is gravity so weak? Or equivalently, why are the universe
and normal atoms so much smaller than a gravitational Bohr atom? At the present point

* Modern approaches take another direction, as explained in the third part of the mountain ascent.

Dvipsbugw
 quantum mechanics of gravitation 881

of our quest these questions cannot be answered. Worse, the weakness of gravity even
means that with high probability, future experiments will provide little additional data
helping to decide among competing answers. The only help is careful thought.

Decoherence of space-time
If the gravitational field evolves like a quantum system, we encounter all issues found in
other quantum systems. General relativity taught us that the gravitational field and space-
time are the same. As a result, we may ask why no superpositions of different macroscopic Dvipsbugw
space-times are observed.
Ref. 908 The discussion is simplified for the simplest case of all, namely the superposition, in a
vacuum region of size l, of a homogeneous gravitational field with value д and one with
Page 796 value д′ . As in the case of a superposition of macroscopic distinct wave functions, such
a superposition decays. In particular, it decays when particles cross the volume. A short
Challenge 1381 ny calculation yields a decay time given by

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2kT 3~2 nl 4
td = ( ) ,
(д − д′ )2
(610)
πm

where n is the particle number density, kT their kinetic energy and m their mass. In-
serting typical numbers, we find that the variations in gravitational field strength are ex-
Challenge 1382 ny tremely small. In fact, the numbers are so small that we can deduce that the gravitational
field is the first variable which behaves classically in the history of the universe. Quantum
gravity effects for space-time will thus be extremely hard to detect.
In short, matter not only tells space-time how to curve, it also tells it to behave with
class. This result calls for the following question.

Do gravitons exist?
Quantum theory says that everything that moves is made of particles. What kind of
particles are gravitational waves made of? If the gravitational field is to be treated
quantum mechanically like the electromagnetic field, its waves should be quantized. Most
properties of these quanta can be derived in a straightforward way.
The 1~r 2 dependence of universal gravity, like that of electricity, implies that the
particles have vanishing mass and move at light speed. The independence of gravity from
electromagnetic effects implies a vanishing electric charge.
Copyright © Christoph Schiller November 1997–May 2006

The observation that gravity is always attractive, never repulsive, means that the field
quanta have integer and even spin. Vanishing spin is ruled out, since it implies no coupling
Ref. 909 to energy. To comply with the property that ‘all energy has gravity’, S = 2 is needed. In fact,
it can be shown that only the exchange of a massless spin 2 particle leads, in the classical
limit, to general relativity.
The coupling strength of gravity, corresponding to the fine structure constant of elec-
tromagnetism, is given either by

m 2 E 2
α G1 = = 2.2 ë 10−15 kg−2 α G2 = =‹  =‹ 
G Gmm
or by (611)
ħc ħc m Pl E Pl

Dvipsbugw
 882 vii details of quantum theory • 27. including gravitation

However, the first expression is not a pure number; the second expression is, but depends
on the mass one inserts. These difficulties reflect the fact that gravity is not properly speak-
ing an interaction, as became clear in the section on general relativity. It is often argued
that m should be taken as the value corresponding to the energy of the system in ques-
tion. For everyday life, typical energies are 1 eV, leading to a value α G2  1~1056 . Gravity
is indeed weak compared to electromagnetism, for which α em = 1~137.04.
If all this is correct, virtual field quanta would also have to exist, to explain static grav-
itational fields.
However, up to this day, the so-called graviton has not yet been detected, and there is Dvipsbugw
in fact little hope that it ever will. On the experimental side, nobody knows yet how to
Challenge 1383 n build a graviton detector. Just try! On the theoretical side, the problems with the coupling
constant probably make it impossible to construct a renormalizable theory of gravity; the
lack of renormalization means the impossibility to define a perturbation expansion, and
thus to define particles, including the graviton. It might thus be that relations such as
E = ħω or p = ħ~2πλ are not applicable to gravitational waves. In short, it may be that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the particle concept has to be changed before applying quantum theory to gravity. The
issue is still open at this point.

Space-time foam
The indeterminacy relation for momentum and position also applies to the gravitational
field. As a result, it leads to an expression for the indeterminacy of the metric tensor д in
a region of size L, which is given by

l Pl 2
∆д  2 , (612)
L2
¼
Challenge 1384 ny where l Pl = ħG~c 3 is the Planck length. Can you deduce the result? Quantum theory
thus shows that like the momentum or the position of a particle, also the metric tensor д
is a fuzzy observable.
But that is not all. Quantum theory is based on the principle that actions below ħ~2
cannot be observed. This implies that the observable values for the metric д in a region
of size L are bound by
дE 3 2 .
2ħG 1
(613)
c L
Copyright © Christoph Schiller November 1997–May 2006

Can you confirm this? The result has far-reaching consequences. A minimum value for
the metric depending inversely on the region size implies that it is impossible to say what
happens to the shape of space-time at extremely small dimensions. In other words, at
extremely high energies, the concept of space-time itself becomes fuzzy. John Wheeler
introduced the term space-time foam to describe this situation. The term makes clear
that space-time is not continuous nor a manifold in those domains. But this was the
basis on which we built our description of nature so far! We are forced to deduce that
our description of nature is built on sand. This issue will form the start of the third part
Page 998 of our mountain ascent.

Dvipsbugw
 quantum mechanics of gravitation 883

No particles
Gravity has another important consequence for quantum theory. To count and define
Page 773 particles, quantum theory needs a defined vacuum state. However, the vacuum state can-
not be defined when the curvature radius of space-time, instead of being larger than
the Compton wavelength, becomes comparable to it. In such highly curved space-times,
particles cannot be defined. The reason is the impossibility to distinguish the environ-
ment from the particle in these situations: in the presence of strong curvatures, the va-
cuum is full of spontaneously generated matter, as black holes show. Now we just saw that Dvipsbugw
at small dimensions, space-time fluctuates wildly; in other words, space-time is highly
curved at small dimensions or high energies. In other words, strictly speaking particles
cannot be defined; the particle concept is only a low energy approximation! We will ex-
plore this strange conclusion in more detail in the third part of our mountain ascent.

No science fiction

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The end of the twentieth century has brought several unexpected but strong results in the
semiclassical quantum gravity.
Ref. 910 In 1995 Ford and Roman found that worm holes, which are imaginable in general re-
lativity, cannot exist if quantum effects are taken into account. They showed that macro-
scopic worm holes require unrealistically large negative energies. (For microscopic worm
holes the issue is still unclear.)
Ref. 912 In 1996 it was found by Kay, Radzikowski and Wald that closed time-like curves do
not exist in semiclassical gravity; there are thus no time machines in nature.
Ref. 911 In 1997 Pfenning and Ford showed that warp drive situations, which are also imagin-
able in general relativity, cannot exist if quantum effects are taken into account. They also
require unrealistically large negative energies.

Not cheating any longer

This short excursion into the theory of quantum gravity showed that a lot of trouble is
waiting. The reason is that up to now, we deluded ourselves. In fact, it was more than that:
we cheated. We carefully hid a simple fact: quantum theory and general relativity contra-
dict each other. That was the real reason that we stepped back to special relativity before
we started exploring quantum theory. In this way we avoided all problems, as quantum
theory does not contradict special relativity. However, it does contradict general relativ-
Copyright © Christoph Schiller November 1997–May 2006

ity. The issues are so dramatic, changing everything from the basis of classical physics to
the results of quantum theory, that we devote the beginning of the third part only to the
exploration of the contradictions. There will be surprising consequences on the nature
of space-time, particles and motion. But before we study these issues, we complete the
theme of the present, second part of the mountain ascent, namely the essence of matter
and interactions.

Dvipsbugw
884 vii details of quantum theory

Biblio graphy

758 Gell-Mann wrote this for the 1976 Nobel Conference. M. Gell-Mann, What are the Build-
ing Blocks of Matter?, in D. Huff & O. Prewitt, editors, The Nature of the Physical Uni-
verse, New York, Wiley, 1979, p. 29. Cited on page 793.
759 See e.g. the reprints of his papers in the standard collection by John A. Wheeler &
Wojciech H. Zurek, Quantum Theory and Measurement, Princeton University Press,
1983. Cited on page 794. Dvipsbugw
760 H.D. Zeh, On the interpretation of measurement in quantum theory, Foundations of Phys-
ics 1, pp. 69–76, 1970. Cited on page 794.
761 Linda R eichl, A Modern Course in Statistical Physics, Wiley, 2nd edition, 1998. An excel-
lent introduction into thermodynamics. Cited on page 795.
762 E. Joos & H.D. Zeh, The emergence of classical properties through interactions with the
environment, Zeitschrift für Physik B 59, pp. 223–243, 1985. See also Erich Joos, Decoher-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ence and the appearance of a classical world in quantum theory, Springer Verlag, 2003. Cited
on page 796.
763 M. Tegmark, Apparent wave function collapse, Foundation of Physics Letters 6, pp. 571–
590, 1993. Cited on page 797.
764 The decoherence time is bound from above by the relaxation time. See A.O. Caldeira &
A.J. Leggett, Influence of damping on quantum interference: an exactly soluble model,
Physical Review A 31, 1985, pp. 1059–1066. This is the main reference about effects of decoher-
ence for a harmonic oscillator. The general approach to relate decoherence to the influence
of the environment is due to Niels Bohr, and has been pursued in detail by Hans Dieter Zeh.
Cited on page 798.
765 G. Lindbl ad, On the generators of quantum dynamical subgroups, Communications in
Mathematical Physics 48, pp. 119–130, 1976. Cited on page 798.
766 Wojciech H. Zurek, Decoherence and the transition from quantum to classical, Physics
Today pp. 36–44, October 1991. An easy but somewhat confusing article. His reply to the
numerous letters of response in Physics Today, April 1993, pp. 13–15, and pp. 81–90, exposes
his ideas in a clearer way and gives a taste of the heated discussions on this topic. Cited on
pages 799 and 803.
767 John Bardeen, explained this regularly in the review talks he gave at the end of his life,
such as the one the author heard in Tokyo in 2000. Cited on page 799.
768 The first decoherence measurement was performed by M. Brune & al., Observing pro-
gressive decoherence of the “meter” in a quantum measurement, Physical Review Letters 77,
Copyright © Christoph Schiller November 1997–May 2006

pp. 4887–4890, 9 December 1996. Cited on page 799.

769 Later experiments confirming the numerical predictions from decoherence were published
by C. Monroe, D.M. Meekhof, B.E. King & D.J. Winel and, A “Schrödinger cat”
superposition state of an atom, Science 272, pp. 1131–1136, 1996, W.P. Schleich, Quantum
physics: engineering decoherence, Nature 403, pp. 256–257, 2000, C.J. Myatt, B.E. King,
Q.A. Turchette, C.A. Sackett, D. Kielpinski, W.M. Itano, C. Monroe &
D.J. Winel and, Decoherence of quantum superpositions through coupling to engineered
reservoirs, Nature 403, pp. 269–273, 2000. See also the summary by W.T. Strunz, G. Al-
ber & F. Haake, Dekohärenz in offenen Quantensystemen, Physik Journal 1, pp. 47–52,
November 2002. Cited on page 799.
770 L. Hackermüller, K. Hornberger, B. Brezger, A. Zeilinger & M. Arndt,

Dvipsbugw
bibliography 885

Decoherence of matter waves by thermal emission of radiation, Nature 427, pp. 711–714,
2004. Cited on page 799.
771 K. Baumann, Quantenmechanik und Objektivierbarkeit, Zeitschrift für Naturforschung
25a, pp. 1954–1956, 1970. Cited on page 800.
772 See for example D. Styer, Physics Today p. 11, September 2000. Cited on page 801.
773 David B ohm, Quantum Theory, Prentice-Hall, 1951, pp. 614–622. Cited on page 802.
774 A. Einstein, B. Pod olsky & N. Rosen, Can quantum-mechanical description of real-
ity be considered complete?, Physical Review 48, pp. 696–702, 1935. Cited on page 802. Dvipsbugw
775 A. Aspect, J. Dalibard & G. Roger, Experimental tests of Bell’s inequalities using
time-varying analyzers, Physical Review Letters 49, pp. 1804–1807, 1982, Cited on page 803.
776 G.C. Hergerfeldt, Causality problems for Fermi’s two-atom system, Physical Review
Letters 72, pp. 596–599, 1994. Cited on page 803.
777 David Z. Albert, Quantum Mechanics and Experience, Harvard University Press, 1994.
No citations.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

778 On the superposition of magnetization in up and down directions there are numerous pa-
pers. Recent experiments on the subject of quantum tunnelling in magnetic systems are
described in D.D. Awschalom, J.F. Smith, G. Grinstein, D.P. DiVicenzo & D.
Loss, Macroscopic quantum tunnelling in magnetic proteins, Physical Review Letters 88,
pp. 3092–3095, 1992, and in C. Paulsen & al., Macroscopic quantum tunnelling effects
of Bloch walls in small ferromagnetic particles, Europhysics Letters 19, pp. 643–648, 1992.
Cited on page 803.
779 The prediction that quantum tunnelling could be observable when the dissipative interac-
tion with the rest of the world is small enough was made by Leggett; the topic is reviewed
in A.J. Leggett, S. Chahravarty, A.T. Dorsey, M.P.A. Fisher, A. Garg & W.
Zwerger, Dynamics of dissipative 2-state systems, Review of Modern Physics 59, pp. 1–85,
1987. Cited on page 807.
780 For example, superpositions were observed in Josephson junctions by R.F. Voss & R.A.
Webb, Macroscopic quantum tunnelling in 1 mm Nb Josephson junctions, Physical Review
Letters 47, pp. 265–268, 1981, Cited on page 804.
781 S. Haroche, Entanglement, decoherence and the quantum-classical transition, Physics
Today 51, pp. 36–42, July 1998. An experiment putting atom at two places at once, distant
about 80 nm, was published by C. Monroe & al., Science 272, p. 1131, 1996. Cited on page
804.
782 M.R. Andres & al., Observations of interference between two Bose condensates, Science
275, pp. 637–641, 31 January 1997. See also the www.aip.org/physnews/special.htm website.
Copyright © Christoph Schiller November 1997–May 2006

Cited on page 804.

783 The most famous reference on the wave function collapse is chapter IV of the book by Kurt
Gottfried, Quantum Mechanics, Benjamin, New York, 1966, It is the favorite reference
by Victor Weisskopf, and cited by him on every occasion he talks about the topic. Cited on
page 805.
784 An experimental measurement of superpositions of left and right flowing currents with 1010
electrons was J.E. Mooij, T.P. Orl and o, L. Levitov, L. Tian, C.H. van der Wal
& S. Lloyd, Josephson persistent-current qubit, Science 285, pp. 1036–1039, 1999. In the
year 2000, superpositions of 1 µA clockwise and anticlockwise have been detected; for more
details, see J.R. Friedman & al., Quantum superposition of distinct macroscopic states,
Nature 406, p. 43, 2000. Cited on page 803.

Dvipsbugw
886 vii details of quantum theory

785 Collapse times have been measured ... Cited on page 809.
786 A clear discussion can be found in S. Haroche & J.-M. R aimond, Quantum computing:
dream or nightmare?, Physics Today 49, pp. 51–52, 1996, as well as the comments in Physics
Today 49, pp. 107–108, 1996. Cited on page 805.
787 S. Kochen & E.P. Specker, The problem of hidden variables in quantum mechanics, 17,
pp. 59–87, 1967. Cited on page 810.
788 J.F. Cl auser, M.A. Horne, A. Shimony & R.A. Holt, Physical Review Letters 23,
p. 880, 1969. The more general and original result is found in J.S. Bell, Physics 1, p. 195, Dvipsbugw
1964, Cited on page 810.
789 Bryce de Witt & Neill Graham, eds., The Many–Worlds Interpretation of Quantum
Mechanics, Princeton University Press, 1973. This interpretation talks about entities which
cannot be observed, namely the many worlds, and often assumes that the wave function of
the universe exists. Both habits are beliefs and in contrast with facts. Cited on page 813.
790 ‘On the other had I think I can safely say that nobody understands quantum mechanics."
From R ichard P. Feynman, The Character of Physical Law, MIT Press, Cambridge, 1965,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

p. 129. He repeatedly made this statement, e.g. in the introduction of his otherwise excellent
QED: the Strange Theory of Light and Matter, Penguin Books, 1990.. Cited on page 813.
791 M. Tegmark, The importance of quantum decoherence in brain processes, Physical Re-
view D 61, pp. 4194–4206, 2000, or also http://www.arxiv.org/abs/quant-ph/9907009. Cited
on page 814.
792 Connections between quantum theory and information theory can be found in Cited on
page 814.
793 J.A. Wheeler, pp. 242–307, in Batelle Recontres: 1967 Lectures in Mathematics and Phys-
ics, C. DeWitt & J.A. Wheeler, editors, W.A. Benjamin, 1968. For a pedagogical ex-
planation, see John W. Norbury, From Newton’s laws to the Wheeler-DeWitt equation,
http://www.arxiv.org/abs/physics/980604 or European Journal of Physics 19, pp. 143–150,
1998. Cited on page 815.
794 E. Fermi, Quantum theory of radiation, Review of Modern Physics 4, p. 87, 1932. No cita-
tions.
795 G. Hegerfeldt, Causality problems for Fermi’s two-atom system, Physical Review Letters
72, pp. 596–599, 1994, No citations.
796 The beautiful analogy between optics and two-dimensional electron systems, in GaAs at he-
lium temperature, has been studied in detail in C.W.J. Beenakker & H. van Hou ten,
Solid State Physics, 44, pp. 1–228, 1991, No citations.
Copyright © Christoph Schiller November 1997–May 2006

797 The use of radioactivity for breeding is described by ... Cited on page 818.
798 S.M. Block, Real engines of creation, Nature 386, pp. 217–219, 1997. Cited on page 820.
799 Early results and ideas on molecular motors are summarised by Barbara Goss Levi,
Measured steps advance the understanding of molecular motors, Physics Today pp. 17–19,
April 1995. Newer results are described in R.D. Astumian, Making molecules into motors,
Scientific American pp. 57–64, July 2001. Cited on page 820.
800 The motorized screw used by viruses was described by A.A. Simpson & al., Structure of
the bacteriophage phi29 DNA packaging motor, Nature 408, pp. 745–750, 2000. Cited on
page 820.
801 R. Bartussek & P. Hänggi, Brownsche Motoren, Physikalische Blätter 51, pp. 506–507,
1995. Cited on page 821.

Dvipsbugw
bibliography 887

802 The table and the evolutionary tree are taken from J.O. McInerney, M. Mull arkey,
M.E. Wernecke & R. Powell, Bacteria and Archaea: Molecular techniques reveal as-
tonishing diversity, Biodiversity 3, pp. 3–10, 2002. The tree might still change a little in the
coming years. Cited on page 823.
803 This is taken from the delightful children text Hans J. Press, Spiel das Wissen schafft,
Ravensburger Buchverlag 1964, 2004. Cited on page 825.
804 The discovery of a specific taste for fat was published by F. L augerette, P. Passilly-
Degrace, B. Patris, I. Niot, M. Febbraio, J.P. Montmayeur & P. Besnard,
CD36 involvement in orosensory detection of dietary lipids, spontaneous fat preference, Dvipsbugw
and digestive secretions, Journal of Clinical Investigation 115, pp. 3177–3184, 2005. Cited on
page 826.
805 Linda Buck and Richard Axel received the 2004 Nobel Prize for medicine and physiology
for their unravelling of the working of the sense of smell. Cited on page 826.
806 To learn to enjoy life to the maximum there is no standard way. A good foundation can be
found in those books which teach the ability to those which have lost it. Cited on page 827.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The best experts are those who help others to overcome traumas. Peter A. Levine &
Ann Frederick, Waking the Tiger – Healing Trauma – The Innate Capacity to Transform
Overwhelming Experiences, North Atlantic Books, 1997. Geoff Graham, How to Become
the Parent You Never Had - a Treatment for Extremes of Fear, Anger and Guilt, Real Options
Press, 1986.
A good complement to these texts is the systemic approach, as presented by Bert
Hellinger, Zweierlei Glück, Carl Auer Verlag, 1997. Some of his books are also available
in English. The author explains how to achieve a life of contentness, living with the highest
possible responsibility of one’s actions, by reducing entaglements with one’s past and with
other people. He presents a simple and efficient technique for realising this disentanglement.
The next step, namely full mastery in the enjoyment of life, can be found in any book
written by somebody who has achieved mastery in any one topic. The topic itself is not
important, only the passion is.
A. de l a Garanderie, Le dialogue pédagogique avec l’élève, Centurion, 1984, A. de la
Garanderie/ Pour une pédagogie de l’intelligence, Centurion, 1990, A. de la Garanderie/
Réussir ça s’apprend, Bayard, 1994. De la Garanderie explains how the results of teaching
and learning depend in particular on the importance of evocation, imagination and motiv-
ation.
Pl ato, Phaedrus, Athens, 380 bce.
Françoise Dolto, La cause des enfants, Laffont, 1985, and her other books. Dolto
(1908–1988), a child psychiatrist, is one of the world experts on the growth of the child; her
main theme was that growth is only possible by giving the highest possible responsibility to
Copyright © Christoph Schiller November 1997–May 2006

every child during its development.

In the domain of art, many had the passion to achieve full pleasure. A good piece of
music, a beautiful painting, an expressive statue or a good movie can show it. On a smaller
scale, the art to typeset beautiful books, so different from what many computer programs
do by default, the best introduction is by Jan Tschichold (1902–1974), the undisputed master
of the field. Among the many books he designed are the beautiful Penguin books of the late
1940s; he also was a type designer, e.g. of the Sabon typeface. A beautiful summary of his
views is the short but condensed text Jan Tschichold, Ausgewählte Aufsätze über Fragen
der Gestalt des Buches und der Typographie, Birkhäuser Verlag, Basel, 1993. An extensive and
beautiful textbook on the topic is Hans Peter Willberg & Friedrich Forssman,
Lesetypographie, Verlag Hermann Schmidt, Mainz, 1997. See also Robert Bringhurst,
The Elements of Typographic Style, Hartley & Marks, 2004.

Dvipsbugw
888 vii details of quantum theory

Many scientists passionately enjoyed their occupation. Any biography of Charles Darwin
will purvey his fascination for biology, of Friedrich Bessel for astronomy, of Albert Einstein
for physics and of Linus Pauling for chemistry.
807 The group of John Wearden in Manchester has shown by experiments with persons that the
accuracy of a few per cent is possible for any action with a duration between a tenth of a
second and a minute. See J. McCrone, When a second lasts forever, New Scientist pp. 53–
56, 1 November 1997. Cited on page 828.
808 E.J. Zimmerman, The macroscopic nature of space-time, American Journal of Physics 30,
pp. 97–105, 1962. Cited on page 829. Dvipsbugw
809 The potential of single-ion clocks is explained in ... Cited on page 830.
810 P.D. Peş ić, The smallest clock, European Journal of Physics 14, pp. 90–92, 1993. Cited on
page 829.
811 D.J Morré & al., Biochemical basis for the biological clock, Biochemistry 41, pp. 11941–
11945, 2002. Cited on page 830.
The chemical clocks in our body are described in John D. Palmer, The Living Clock,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

812
Oxford University Press, 2002, or in A. Ahlgren & F. Halberg, Cycles of Nature: An
Introduction to Biological Rhythms, National Science Teachers Association, 1990. See also
the http://www.msi.umn.edu/~halberg/introd/ website. Cited on page 830.
813 An introduction to the sense of time as result of clocks in the brain is found in R.B. Ivry
& R. Spencer, The neural representation of time, Current Opinion in Neurobiology 14,
pp. 225–232, 2004. The interval timer is explain in simple words in K. Wright, Times in
our lives, Scientific American pp. 40–47, September 2002. The MRI research used is S.M.
R ao, A.R. Mayer & D.L. Harrington, The evolution of brain activation during tem-
poral processing, Nature Neuroscience 4, pp. 317–323, 2001. Cited on page 831.
814 See for example ... Cited on page 832.
815 A pretty example of a quantum mechanical system showing exponential behaviour at all
times is given by H. Nakazato, M. Namiki & S. Pascazio, Exponential behaviour
of a quantum system in a macroscopic medium, Physical Review Letters 73, pp. 1063–1066,
1994. Cited on page 833.
816 See the delightful book about the topic by Paolo Facchi & Saverio Pascazio, La
regola d’oro di Fermi, Bibliopolis, 1999. An experiment observing deviations at short times is
S.R. Wilkinson, C.F. Bharucha, M.C. Fischer, K.W. Madison, P.R. Morrow,
Q. Niu, B. Sundaram & M.G. R aizen, Nature 387, p. 575, 1997. Cited on page 833.
817 See ... Cited on page 833.
818 H. Kobayashi & S. Kohshima, Unique morphology of the human eye, Nature 387,
Copyright © Christoph Schiller November 1997–May 2006

pp. 767–768, 1997. Cited on page 836.

819 James W. Prescott, Body pleasure and the origins of violence, The Futurist Bethseda,
1975, also available at http://www.violence.de/prescott/bullettin/article.html. Cited on page
836.
820 See for example P. Py ykkö, Relativity, gold, closed-shell interactions, and CsAu.NH3, An-
gewandte Chemie, International Edition 41, pp. 3573–3578, 2002, or L.J. Norrby, Why is
mercury liquid? Or, why do relativistic effects not get into chemistry textbooks?, Journal of
Chemical Education 68, pp. 110–113, 1991. Cited on page 837.
821 The switchable mirror effect was discovered by J.N. Huiberts, R. Griessen, J.H.
R ector, R.J. Wijngarden, J.P. Dekker, D.G. de Groot & N.J. Koeman, Yt-
trium and lanthanum hydride films with switchable optical properties, Nature 380, pp. 231–

Dvipsbugw
bibliography 889

234, 1996. A good introduction is R. Griessen, Schaltbare Spiegel aus Metallhydriden,

Physikalische Blätter 53, pp. 1207–1209, 1997. Cited on page 842.
822 The present attempts to build cheap THz wave sensors – which allow to see through clothes
– is described by D. Clery, Brainstorming their way to an imaging revolution, Science 297,
pp. 761–763, 2002. Cited on page 843.
823 V.F. Weisskopf, Of atoms, mountains and stars: a study in qualitative physics, Science
187, p. 605, 1975. Cited on page 844.
824 K. Au tumn, Y. Liang, T. Hsich, W. Zwaxh, W.P. Chan, T. Kenny, R. Fearing &
Dvipsbugw
R.J. Full, Adhesive force of a single gecko foot-hair, Nature 405, pp. 681–685, 2000. Cited
on page 845.
825 A.B. Kesel, A. Martin & T. Seidl, Getting a grip on spider attachment: an AFM
approach to microstructure adhesion in arthropods, Smart Materials and Structures 13,
pp. 512–518, 2004. Cited on page 845.
826 H. Inaba, T. Saitou, K.-I. Tozaki & H. Hayashi, Effect of the magnetic field on
the melting transition of H2O and D2O measured by a high resolution and supersensitive

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

differential scanning calorimeter, Journal of Applied Physics 96, pp. 6127–6132, 2004. Cited
on page 846.
827 R. Hemley & al., presented these results at the International Conference on New Dia-
mond Science in Tsukuba (Japan) 2005. See figures at http://www.carnegieinstitution.org/
diamond-13may2005/. Cited on page 846.
828 On the internet, the ‘spherical’ periodic table is credited to Timothy Stowe; but there is no ref-
erence for that claim, except an obscure calendar from a small chemical company. The table
(containing a few errors) is from the http://chemlab.pc.maricopa.edu/periodic/stowetable.
html website. Cited on page 837.
829 For good figures of atomic orbitals, take any modern chemistry text. See for example the
text by ... Cited on page 837.
830 For experimentally determined pictures of the orbitals of dangling bonds, see for example
F. Giessibl & al., Subatomic features on the silicon (111)-(7x7) surface observed by atomic
force microscopy, Science 289, pp. 422–425, 2000. Cited on page 837.
831 A.P. Finne, T. Araki, R. Bl aauwgeers, V.B. Eltsov, N.B. Kopnin, M. Krusius,
L. Skrbek, M. Tsubota & G.E. Volovik, An intrinsic velocity-independent criterion
for superfluid turbulence, Nature 424, pp. 1022–1025, 2003. Cited on page 848.
832 J.L. Costa–Krämer, N. Garcia, P. Garcia–Molchales & P.A. Serena, Con-
ductance in disordered nanowires: foward and backscattering, Physical Review B 53,
p. 102658, 1996. Cited on page 848.
Copyright © Christoph Schiller November 1997–May 2006

833 The first experimental evidence is by V.J. Goldman & B. Su, Resonance tunnelling in the
fractional quantum Hall regime: measurement of fractional charge, Science 267, pp. 1010–
1012, 1995. The first unambiguous measurements are by R. de Picciotto & al., Direct
observation of a fractional charge, Nature 389, pp. 162–164, 1997, and L. Saminadayar,
D.C. Gl attli, Y. Jin & B. Etienne, Observation of the e,3 fractionally charged Laugh-
lin quasiparticle/ Physical Review Letters 79, pp. 2526–2529, 1997. or http://www.arxiv.org/
abs/cond-mat/9706307. Cited on page 849.
834 The original prediction, which earned him a Nobel Prize for physics, is in R.B. L aughlin,
Anomalous quantum Hall effect: an incompressible quantum fluid with fractionally charged
excitations, Physical Review Letters 50, pp. 1395–1398, 1983. Cited on page 849.
835 The fractional quantum Hall effect was discovered by D.C. Tsui, H.L. Stormer & A.C.

Dvipsbugw
890 vii details of quantum theory

Gossard, Two-dimensional magnetotransport in the extreme quantum limit, Physical Re-

view Letters 48, pp. 1559–1562, 1982. Cited on page 849.
836 The discovery of the original quantum Hall effect, which earned von Klitzing a Nobel Prize,
was published as K. von Klitzing, G. Dorda & M. Pepper, New method for high-
accuracy determination of the fine-structure constant based on quantised Hall resistance,
Physical Review Letters 45, pp. 494–497, 1980. Cited on page 849.
837 [3] The first paper was J. Bardeen, L.N. Cooper & J.R. Schrieffer, Microscopic
Theory of Superconductivity, Physical Review 106, pp. 162–164, 1957. The full, so-called BCS
theory, is then given in the masterly paper J. Bardeen, L.N. Cooper & J.R. Schrief- Dvipsbugw
fer, Theory of Superconductivity, Physical Review 108, pp. 1175–1204, 1957. Cited on page
847.
838 The story is found on many places in the internet. Cited on page 854.
839 H.B.G. Casimir, On the attraction between two perfectly conducting bodies, Proc. Kon.
Ned. Acad. Wetensch. 51, pp. 793–795, 1948. No citations.
840 H.B.G. Casimir & D. Polder, The influence of retardation on the London-Van der

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Waals forces, Physical Review 73, pp. 360–372, 1948. No citations.
841 S.L. B oersma, Aantrekkende kracht tussen schepen, Nederlands tidschrift voor
natuuurkunde 67, pp. 244–246, Augustus 2001. See also his paper S.L. B oersma,A
maritime analogy of the Casimir effect, American Journal of Physics 64, pp. 539–541, 1996.
Cited on page 854.
842 A. L arraza & B. Denard o, An acoustic Casimir effect, Physics Letters A 248, pp. 151–
155, 1998. See also A. L arraza, A demonstration apparatus for an acoustic analog of the
Casimir effect, American Journal of Physics 67, pp. 1028–1030, 1999. Cited on page 855.
843 B.W. Derjaguin, I.I. Abrikosova & E.M. Lif shitz, Direct measurement of molecu-
lar attraction between solids separated by a narrow gap, Quaterly Review of the Chemical
Society (London) 10, pp. 295–329, 1956. Cited on page 856.
844 M.J. Sparnaay, Nature 180, p. 334, 1957. M.J. Sparnaay, Measurements of attractive
forces between flat plates, Physica (Utrecht) 24, pp. 751–764, 1958. Cited on page 856.
845 S.K. L amoreaux, Demonstration of the Casimir force in the 0.6 to 6 µm range, Physical
Review Letters 78, pp. 5–8, 1997. U. Mohideen & A. Roy, Precision measurement of the
Casimir force from 0.1 to 0.9 µm, Physical Review Letters 81, pp. 4549–4552, 1998. Cited on
page 856.
846 A. L ambrecht, Das Vakuum kommt zu Kräften, Physik in unserer Zeit 36, pp. 85–91, 2005.
Cited on page 856.
847 T.H. B oyer, Van der Waals forces and zero-point energy for dielectric and permeable ma-
Copyright © Christoph Schiller November 1997–May 2006

terials, Physical Review A 9, pp. 2078–2084, 1974. This was taken up again in O. Kenneth,
I. Klich, A. Mann & M. R evzen, Repulsive Casimir forces, Physical Review Letters 89,
p. 033001, 2002. However, none of these effects has been verified yet. Cited on page 856.
848 The effect was discovered by K. Scharnhorst, On propagation of light in the vacuum
between plates, Physics Letters B 236, pp. 354–359, 1990. See also G. Barton, Faster-than-
c light between parallel mirrors: the Scharnhorst effect rederived, Physics Letters B 237,
pp. 559–562, 1990, and P.W. Milonni & K. Svozil, Impossibility of measuring faster-
than-c signalling by the Scharnhorst effect, Physics Letters B 248, pp. 437–438, 1990. The
latter corrects an error in the first paper. Cited on page 856.
849 D.L. Burke & al., Positron production in multiphoton light-by-light scattering, Phys-
ical Review Letters 79, pp. 1626–1629, 1997. A simple summary is given in Bertram

Dvipsbugw
bibliography 891

Schwarzschild, Gamma rays create matter just by plowing into laser light, Physics
Today 51, pp. 17–18, February 1998. Cited on page 859.
850 M. Kardar & R. Golestanian, The ‘friction’ of the vacuum, and other fluctuation-
induced forces, Reviews of Modern Physics 71, pp. 1233–1245, 1999. Cited on page 859.
851 G. Gour & L. Sriramkumar, Will small particles exhibit Brownian motion in the
quantum vacuum?, http://www.arxiv.org/abs/quant/phys/9808032 Cited on page 859.
852 P.A.M. Dirac, ..., 1985. This article transcribes a speech by Paul Dirac just before his death.
Disturbed by the infinities of quantum electrodynamics, he rejects as wrong the theory
Dvipsbugw
which he himself found, even though it correctly describes all experiments. The humility
of a great man who is so dismissive of his main and great achievement was impressive when
hearing him, and still is impressive when reading the speech. Cited on page 860.
853 See for example the overview by M. B orn & T. Jüstel, Umweltfreundliche Lichtquellen,
Physik Journal 2, pp. 43–49, 2003. Cited on page 849.
854 J. Kasparian, Les lasers femtosecondes : applications atmosphériques, La Recherche
pp. RE14 1–7, February 2004/ See also the http://www.teramobile.org website. Cited on

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

page 851.
855 Misleading statements are given in the introduction and in the conclusion of the review by
H. Paul, Interference between independent photons, Review of Modern Physics, volume
58, pages 209-231; however, in the bulk of the article the author in practice retracts the state-
ment, e.g. on page 221. Cited on pages 851 and 852.
856 L. Cser, Gy. Tőrők, G. Krexner, I. Sharkov & B. Faragó, Holographic imaging
of atoms using thermal neutrons, Physical Review Letters 89, p. 75504, 2002. Cited on page
852.
857 H. Euler & B. Kockel, Naturwissenschaften 23, p. 246, 1935, H. Euler, Annalen der
Physik 26, p. 398, 1936, W. Heisenberg & H. Euler, Folgerung aus der Diracschen The-
orie des Electrons, Zeitschrift für Physik 98, pp. 714–722, 1936. Cited on page 861.
858 Thomas Stoehlker & al., Physical Review Letters 9 October 2000. Cited on page 861.
859 M. Roberts & al., Observation of an electric octupole transition in a single ion, Physical
Review Letters 78, pp. 1876–1879, 1997. A lifetime of 3700 days was determined. Cited on
page 862.
860 K. An, J.J. Childs, R.R. Desari & M.S. Field, Microlaser: a laser with one atom in
an optical resonator, Physical Review Letters 73, pp. 3375–3378, 19 December 1994. Cited on
page 851.
861 A. Takita, H. Yamamoto, Y. Hayasaki, N. Nishida & H. Misawa, Three-
dimensional optical memory using a human fingernail, Optics Express 13, pp. 4560–4567,
Copyright © Christoph Schiller November 1997–May 2006

2005. They used a Ti:sapphire oscillator and a Ti:sapphire amplifier. Cited on page 851.
862 See the website by Tom Hales on http://www.math.lsa.umich.edu/~hales/countdown. An
earlier book on the issue is the delightful text by Max Leppmaier, Kugelpackungen von
Kepler bis heute, Vieweg Verlag, 1997. Cited on page 862.
863 For a short overview, also covering binary systems, see C. Bechinger, H.-H. von
Grünberg & P. Leiderer, Entropische Kräfte - warum sich repulsiv wechselwirkende
Teilchen anziehen können, Physikalische Blätter 55, pp. 53–56, 1999. Cited on page 862.
864 Hans–Werner Fink, Direct observation of three-body interactions in adsorbed layers:
Re on W(111), Physical Review Letters 52, pp. 1532–1534, 1984. Cited on page 862.
865 See for example the textbook by A. Pimpinelli & J. Vill ain, Physics of Crystal Growth,
Cambridge University Press, 1998. Cited on page 863.

Dvipsbugw
892 vii details of quantum theory

866 Y. Furukawa, Faszination der Schneekristalle - wie ihre bezaubernden Formen entstehen,
Chemie in unserer Zeit 31, pp. 58–65, 1997. His http://www.lowtem.hokudai.ac.jp/~frkw/
index_e.html website gives more information. Cited on page 863.
867 P.-G. de Gennes, Soft matter, Reviews of Modern Physics 63, p. 645, 1992. Cited on page
863.
868 The fluid helium based quantum gyroscope was first built by the Czech group ... and the
Berkeley group, Nature 10. April 1997. Cited on page 863.
869 K. Schwab & al., Nature 404, p. 974, 2000. For the optical analog, see Ref. 519. Cited on
page 864. Dvipsbugw
870 J.P. Briand & al., Production of hollow atoms by the excitation of highly charged ions in
interaction with a metallic surface, Physical Review Letters 65, pp. 159–162, 1990. See also C.
R hodes, Hollow atoms, ..., and ...Physikalische Blätter Oktober 1996. Cited on page 864.
871 The present knowledge on how humans move is summarised by ... Cited on page 867.
872 See e.g. D.A. Bryant, Electron acceleration in the aurora, Contemporary Physics 35,
pp. 165–179, 1994. Cited on page 867.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

873 The discharges above clouds are described e.g. by ... Cited on page 867.
874 See e.g. L.O’C. Drury, Acceleration of cosmic rays, Contemporary Physics 35, pp. 231–242,
1994. Cited on page 867.
875 About Coulomb explosions, see ... Cited on page 867.
876 See Dirk Kreimer, New mathematical structures in renormalizable quantum field the-
ories, http://www.arxiv.org/abs/hep-th/0211136 or Annals of Physics 303, pp. 179–202, 2003,
and the erratum ibid., 305, p. 79, 2003. Cited on page 867.
877 S. Fray, C. Alvarez Diez, T.W. Hänsch & Martin Weitz, Atomic interferometer
with amplitude gratings of light and its applications to atom based tests of the equivalence
principle, Physical Review Letters 93, p. 240404, 2004. Cited on page 868.
878 R. Colell a, A.W. Overhauser & S.A. Werner, Observation of gravitationally in-
duced quantum mechanics, Physical Review Letters 34, pp. 1472–1475, 1975. This experiment
is usually called the COW experiment. Cited on page 869.
879 G.C. Ghirardi, A. R imini & T. Weber, Unified dynamics for microscopic and macro-
scopic systems, Physical Review D 34, pp. 470–491, 1986. Cited on page 869.
880 Ch.J. B ordé & C. L ämmerzahl, Atominterferometrie und Gravitation, Physikalische
Blätter 52, pp. 238–240, 1996. See also A. Peters, K.Y. Chung & S. Chu, Observation
of gravitational acceleration by dropping atoms, Nature 400, pp. 849–852, 26 August 1999.
Cited on page 869.
881 V.V. Nesvizhevsky & al., Quantum states of neutrons in the Earth’s gravitational field,
Copyright © Christoph Schiller November 1997–May 2006

Nature 415, p. 297, 17 January 2002. Cited on page 869.

882 J.D. Bekenstein, Black holes and entropy, Physical Review D 7, pp. 2333–2346, 1973.
Cited on page 870.
883 R. B ousso, The holographic principle, Review of Modern Physics 74, pp. 825–874, 2002,
also available as http://www.arxiv.org/abs/hep-th/0203101. The paper is an excellent review;
however, it has some argumentation errors, as explained on page 871. Cited on page 871.
884 This is clearly argued by S. Carlip, Black hole entropy from horizon conformal field the-
ory, Nuclear Physics B Proceedings Supplement 88, pp. 10–16, 2000. Cited on page 871.
885 S.A. Fulling, Nonuniqueness of canonical field quantization in Riemannian space-time,
Physical Review D 7, pp. 2850–2862, 1973, P. Davies/ Scalar particle production in Schwar-
zchild and Rindler metrics, Journal of Physics A: General Physics 8, pp. 609–616, 1975, W.G.

Dvipsbugw
bibliography 893

Unruh, Notes on black hole evaporation, Physical Review D 14, pp. 870–892, 1976. Cited
on page 871.
886 About the possibility to measure Fulling–Davies–Unruh radiation, see for example the pa-
per by H. Rosu, On the estimates to measure Hawking effect und Unruh effect in the labor-
atory, International Journal of Modern Physics D3 p. 545, 1994. http://www.arxiv.org/abs/
gr-qc/9605032 or the paper P. Chen & T. Tajima, Testing Unruh radiation with ultra-
intense lasers, Physical Review Letters 83, pp. 256–259, 1999. Cited on page 871.
887 W.G. Unruh & R.M. Wald, Acceleration radiation and the generalised second law of
thermodynamics, Physical Review D 25, pp. 942–958 , 1982. Cited on page 872. Dvipsbugw
888 R.M. Wald, The thermodynamics of black holes, Living Reviews of Relativity 2001,
www-livingreviews.org/lrr-2001-6. Cited on page 873.
889 For a delightful popular account from a black hole expert, see Igor Novikov, Black Holes
and the Universe, Cambridge University Press 1990. Cited on pages 874 and 877.
890 The original paper is W.G. Unruh, Experimental black hole evaporation?, Physical Review
Letters 46, pp. 1351–1353, 1981. A good explanation with good literature overview is the one

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

by Matt Visser, Acoustical black holes: horizons, ergospheres and Hawking radiation,
http://www.arxiv.org/abs/gr-qc/9712010. Cited on page 874.
891 G.W. Gibbons & S.W. Hawking, Action integrals and partition functions in quantum
gravity, Physical Review D 15, pp. 2752–2756, 1977. No citations.
892 Optical black holes ... Cited on page 875.
893 T. Damour & R. Ruffini, Quantum electrodynamical effects in Kerr–Newman geomet-
ries, Physical Review Letters 35, pp. 463–466, 1975. Cited on page 875.
894 These were the Vela satellites; their existence and results were announced officially only in
1974, even though they were working already for many years. Cited on page 875.
895 The best general introduction into the topic of gamma ray bursts is S. Klose, J. Greiner
& D. Hartmann, Kosmische Gammastrahlenausbrüche – Beobachtungen und Modelle,
Teil I und II, Sterne und Weltraum March and April 2001. Cited on page 875.
896 When the gamma ray burst encounters the matter around the black hole, it is broadened.
The larger the amount of matter, the broader the pulse is. See G. Preparata, R. Ruffini
& S.-S. Xue, The dyadosphere of black holes and gamma-ray bursts, Astronomy and As-
trophysics 338, pp. L87–L90, 1998, R. Ruffini, J.D. Salmonson, J.R. Wilson & S.-S.
Xue, On the pair electromagnetic pulse of a black hole with electromagnetic structure, As-
tronomy and Astrophysics 350, pp. 334–343, 1999, R. Ruffini, J.D. Salmonson, J.R.
Wilson & S.-S. Xue, On the pair electromagnetic pulse from an electromagnetic black
hole surrounded by a baryonic remnant, Astronomy and Astrophysics 359, pp. 855–864,
2000, and C.L. Bianco, R. Ruffini & S.-S. Xue, The elementary spike produced by
Copyright © Christoph Schiller November 1997–May 2006

a pure e+ e− pair-electromagnetic pulse from a black hole: the PEM pulse, Astronomy and
Astrophysics 368, pp. 377–390, 2001. For a very personal account by Ruffini on his involve-
ment in gamma ray bursts, see his paper Black hole formation and gamma ray bursts, http://
www.arxiv.org/abs/astro-ph/0001425. Cited on page 876.
897 About the membrane despription of black holes, see ... Cited on page 877.
898 http://www.arxiv.org/abs/astro-ph/0301262, http://www.arxiv.org/abs/astro-ph/0303539,
D.W. Fox & al., Early optical emission from the γ-ray burst of 4 October 2002, Nature
422, pp. 284–286, 2003. Cited on page 876.
899 Negative heat capacity has also been found in atom clusters and in nuclei. See e.g. M.
Schmidt & al., Negative heat capacity for a cluster of 147 sodium atoms, Physical Review
Letters 86, pp. 1191–1194, 2001. Cited on page 877.

Dvipsbugw
894 vii details of quantum theory

900 H.-P. Nollert, Quasinormal modes: the characteristic ‘sound’ of black holes and neutron
stars, Classical and Quantum Gravity 16, pp. R159–R216, 1999. Cited on page 877.
901 He wrote a series of papers on the topic; a beautiful summary is Don N. Page, How fast
does a black hole radiate information?, International Journal of Modern Physics 3, pp. 93–
106, 1994, based on his earlier papers, such as Information in black hole radiation, Physical
Review Letters 71, pp. 3743–3746, 1993. See also his electronic preprint at http://www.arxiv.
org/abs/hep-th/9305040. Cited on page 878.
902 See Don N. Page, Average entropy of a subsystem, Physical Review Letters 71, pp. 1291–
1294, 1993. The entropy formula of this paper, used above, was proven by S.K. Foong & S. Dvipsbugw
Kanno, Proof of Page’s conjecture on the average entropy of a subsystem, Physical Review
Letters 72, pp. 1148–1151, 1994. Cited on page 879.
903 The author speculates that this version of the coincidences could be original; he has not
found it in the literature. Cited on page 869.
904 Steven Weinberg, Gravitation and Cosmology, Wiley, 1972. See equation 16.4.3 on page
619 and also page 620. Cited on page 869.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

905 It could be that knot theory provides a relation between local knot invariant and a global
one. Cited on page 870.
906 R. L afrance & R.C. Myers, Gravity’s rainbow: limits for the applicability of the equi-
valence principle, Physical Review D 51, pp. 2584–2590, 1995, http://www.arxiv.org/abs/
hep-th/9411018. Cited on page 880.
907 M.Yu. Kuchiev & V.V. Fl ambaum, Scattering of scalar particles by a black hole, http://
www.arxiv.org/abs/gr-qc/0312065. See also M.Yu. Kuchiev & V.V. Fl ambaum, Reflec-
tion on event horizon and escape of particles from confinement inside black holes, http://
www.arxiv.org/abs/gr-qc/0407077. Cited on page 880.
908 E. Joos, Why do we observe a classical space-time?, Physics Letters A 116, pp. 6–8, 1986.
Cited on page 881.
909 This point was made repeatedly by Steven Weinberg, namely in Derivation of gauge
invariance and the equivalence principle from Lorentz invariance of the S-matrix , Phys-
ics Letters 9, pp. 357–359, 1964, in Photons and gravitons in S-matrix theory: derivation of
charge conservation and equality of gravitational and inertial mass, Physical Review B 135,
pp. 1049–1056, 1964, and in Photons and gravitons in perturbation theory: derivation of
Maxwell’s and Einstein’s equations, Physical Review B 138, pp. 988–1002, 1965. Cited on
page 881.
910 L.H. Ford & T.A. Roman, Quantum field theory constrains traversable wormhole geo-
metries, http://www.arxiv.org/abs/gr-qc/9510071 or Physical Review D 53, pp. 5496–5507,
Copyright © Christoph Schiller November 1997–May 2006

1996. Cited on page 883.

911 M.J. Pfenning & L.H. Ford, The unphysical nature of ‘warp drive’, Classical and
Quantum Gravity 14, pp. 1743–1751, 1997. Cited on page 883.
912 B.S. Kay, M. R adzikowski & R.M. Wald, Quantum field theory on spacetimes with a
compactly generated Cauchy horizon, http://www.arxiv.org/abs/gr-qc/9603012 or Commu-
nications in Mathematical Physics 183 pp. 533–556, 1997. Cited on page 883.
913 An overview on squeezed light is given in ... No citations.
914 An introduction into the history of the concept of precision is given by N.W. Wise, The
values of precision, Princeton University Press, 1994. No citations.
915 E.H. Kennard, Zur Quantenmechanik einfacher Bewegungstypen, Zeitschrift für Physik
44, pp. 326–352, 1927. The priority of the result is often wrongly attributed to the much

Dvipsbugw
bibliography 895

later paper by H.P. Robertson, The uncertainty principle, Physical Review 34, pp. 163–
164, 1929. The complete result is by Erwin Schrödinger, Zum Heisenbergschem Un-
schärfeprinzip, Sitzungsberichte der Preussischen Akademie der Wissenschaften, Physikalisch-
Mathematische Klasse Berlin, pp. 296–303, 1930. No citations.
916 W. Heisenberg, Über den anschaulichen Inhalt der quantentheoretischen Kinematik
und Mechanik, Zeitschrift für Physik 43, pp. 172–198, 1927. No citations.
917 M.G. R aymer, Uncertainty principle for joint measurement of noncommuting variables,
American Journal of Physics 62, pp. 986–993, 1994. No citations.
Dvipsbugw
918 The question about the origin of blue colour in flowers has been discussed already in the
beginning of the twentieth century. A short summary is given by K. Yoshida & al., Cause
of blue petal colour, Nature 373, p. 291, 26 January 1995. No citations.
919 The main reference on nondemolition measurements is the text by C.M. Caves & P.D.
Drumond, Quantum limits on bosonic communication rates, Reviews of Modern Physics
66, pp. 481–537, 1994. No citations.
920 Mark P. Silverman, And Yet it Moves: Strange Systems and Subtle Questions in Physics,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Cambridge University Press, 1993. It is a treasure chest for anybody interested in the details
of modern physics.
The first of the two other books by Mark P. Silverman, More than one Mystery,
Springer 1995, talks about the Aharonov-Bohm effect, about the EPR paradox and about the
Hanbury–Brown–Twiss experiments; his book Waves and Grains: Reflections on Light and
Learning, Princeton University Press, 1998, completes his exposition of interesting thoughts
and insights about physics. No citations.
921 N. Ashcroft & D. Mermin, Solid State Physics, Saunders, 1976, is probably the best
as well as the most original and entertaining introduction to the topic available in English
language. No citations.
922 The famous quantum theory book is P.A.M. Dirac, Quantum Mechanics, 4th revised
edition, Oxford University Press, 1967. No citations.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 C h a p t e r VIII

I NSI DE THE NUC LEUS

Dvipsbugw

28. the structure of the nucleus – the densest

clouds

N
uclear physics was born in 1896 in France, but is now a small activity.
ot many researchers are working on the topic. It produced

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ot more than one daughter, experimental high energy physics, which was born
around 1930. But since 1985, also these activities are in strong decline. Despite the short-
Ref. 923 ness, the family history is impressive; the two fields uncovered why stars shine, how
powerful bombs work, how cosmic evolution produced the atoms we are made of and
how medical doctors can dramatically improve their healing rate.

Nuclear physics is just low-density astrophysics.

“
A physical wonder – magnetic resonance imaging
”
Arguably, the most spectacular tool that physical research has produced in the twenti-
eth century was magnetic resonance imaging, or MRI for short. This technique allows to
image human bodies with a high resolution and with (almost) no damage, in strong con-
trast to X-ray imaging. Though the machines are still expensive – costing 400 000 euro
and more – there is hope that they will become cheaper in the future. Such a machine
consists essentially of a large magnetic coil, a radio transmitter and a computer. Some
results of putting part of a person into the coil are shown in Figure 344.
In these machines, a radio transmitter emits radio waves that are absorbed because
hydrogen nuclei are small spinning magnets. The magnets can be parallel or antiparallel
to the magnetic field produced by the coil. The transition energy E can be absorbed from
Copyright © Christoph Schiller November 1997–May 2006

a radio wave whose frequency ω is tuned to the magnetic field B. The energy absorbed
by a single hydrogen nucleus is given by

E = ħω = ħγB (614)

The material constant γ~2π has a value of 42.6 MHz~T for hydrogen nuclei; it results from
the non-vanishing spin of the proton. This is a quantum effect, as stressed by the appear-
ance of the quantum of action ħ. Using some cleverly applied magnetic fields, typically
with a strength between 0.3 and 1.5 T, the machines are able to measure the absorption for
each volume element separately. Interestingly, the precise absorption level depends on the
chemical compound the nucleus is built into. Thus the absorption value will depend on

Dvipsbugw
 898 viii inside the nucleus • the structure of the nucleus

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 344 Sagittal images of the head and the spine – used with permission from Joseph P. Hornak,
The Basics of MRI, http://www.cis.rit.edu/htbooks/mri, Copyright 2003

the chemical environment. When the intensity of the absorption is plotted as grey scale,
an image is formed that retraces the different chemical composition. Two examples are
shown in Figure 344. Using additional tricks, modern machines can picture blood flow
in the heart or air flow in lungs; they can even make films of the heart beat. Other tech-
niques show how the location of sugar metabolism in the brain depends on what you are
thinking about.* In fact, also what you are thinking about all the time has been imaged:
Ref. 924 the first image of people making love has been taken by Willibrord Weijmar Schultz and
his group in 1999. It is shown in Figure 345.
Each magnetic resonance image thus proves that atoms have spinning nuclei. Like for
any other object, nuclei have size, colour, composition and interactions that ask to be
explored.

The size of nuclei

The magnetic resonance signal shows that hydrogen nuclei are quite sensitive to magnetic
Copyright © Christoph Schiller November 1997–May 2006

fields. The д-factor of protons, defined using the magnetic moment µ, their mass and
Page 761 charge as д = µ4m~eħ, is about 5.6. Using expression (537) that relates the д-factor and
the radius of a composite object, we deduce that the radius of the proton is about 0.9 fm;
this value is confirmed by experiment. Protons are thus much smaller than hydrogen
atoms, the smallest of atoms, whose radius is about 30 pm. In turn, the proton is the
smallest of all nuclei; the largest nuclei have radii 7 times the proton value.

* The website http://www.cis.rit.edu/htbooks/mri by Joseph P. Hornak gives an excellent introduction to

magnetic resonance imaging, both in English and Russian, including the physical basis, the working of the
machines, and numerous beautiful pictures. The method of studying nuclei by putting them at the same
time into magnetic and radio fields is also called nuclear magnetic resonance.

Dvipsbugw
the structure of the nucleus – the densest clouds 899

The small size of nuclei is no news. It is known since the beginning of the twenti-
eth century. The story starts on the first of March in 1896, when Henri Becquerel* dis-
covered a puzzling phenomenon: minerals of uranium potassium sulphate blacken pho-
tographic plates. Becquerel had heard that the material is strongly fluorescent; he conjec-
tured that fluorescence might have some connection to the X-rays discovered by Conrad
Röngten the year before. His conjecture was wrong; nevertheless it led him to an import-
ant new discovery. Investigating the reason for the effect of uranium on photographic
plates, Becquerel found that these minerals emit an undiscovered type of radiation, dif-
ferent from anything known at that time; in addition, the radiation is emitted by any Dvipsbugw
substance containing uranium. In 1898, Bémont named the property of these minerals
radioactivity.
Radioactive rays are also emitted from many elements other than
uranium. The radiation can be ‘seen’: it can be detected by the tiny
flashes of light that are emitted when the rays hit a scintillation
screen. The light flashes are tiny even at a distance of several metre

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

from the source; thus the rays must be emitted from point-like
sources. Radioactivity has to be emitted from single atoms. Thus
radioactivity confirmed unambiguously that atoms do exist. In fact,
radioactivity even allows to count them, as we will find out shortly.
The intensity of radioactivity cannot be influenced by magnetic
or electric fields; it does not depend on temperature or light irradi-
ation. In short, radioactivity does not depend on electromagnetism
Henri Becquerel
and is not related to it. Also the high energy of the emitted radi-
ation cannot be explained by electromagnetic effects. Radioactivity must thus be due to
another, new type of force. In fact, it took 30 years and a dozen of Nobel Prizes to fully un-
derstand the details. It turns out that several types of radioactivity exist; the types behave
differently when they fly through a magnetic field or when they encounter matter. They
are listed in Table 66. All have been studied in great detail, with the aim to understand
the nature of the emitted entity and its interaction with matter.
In 1909, radioactivity inspired the 37 year old physicist Ernest Rutherford,**who had
won the Nobel Prize just the year before, to another of his brilliant experiments. He asked
his collaborator Hans Geiger to take an emitter of alpha radiation – a type of radioactivity
which Rutherford had identified and named 10 years earlier – and to point the radiation
at a thin metal foil. The quest was to find out where the alpha rays would end up. The Copyright © Christoph Schiller November 1997–May 2006

* Henri Becquerel (b. 1852 Paris, d. 1908 Le Croisic), important French physicist; his primary topic was the
study of radioactivity. He was the thesis adviser of Marie Curie, the wife of Pierre Curie, and was central to
bringing her to fame. The SI unit for radioactivity is named after him. For his discovery of radioactivity he
received the 1903 Nobel Prize for physics; he shared it with the Curies.
** Ernest Rutherford (1871–1937), important New Zealand physicist. He emigrated to Britain and became
professor at the University of Manchester. He coined the terms alpha particle, beta particle, proton and
neutron. A gifted experimentalist, he discovered that radioactivity transmutes the elements, explained the
nature of alpha rays, discovered the nucleus, measured its size and performed the first nuclear reactions.
Ironically, in 1908 he received the Nobel price for chemistry, much to the amusement of himself and of
the world-wide physics community; this was necessary as it was impossible to give enough physics prizes
to the numerous discoverers of the time. He founded a successful research school of nuclear physics and
many famous physicists spent some time at his institute. Ever an experimentalist, Rutherford deeply disliked
quantum theory, even though it was and is the only possible explanation for his discoveries.

Dvipsbugw
 900 viii inside the nucleus • the structure of the nucleus

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 345 The origin of human life (© Willibrord
Weijmar Schultz)

research group followed the path of the particles by using scintillation screens; later on
they used an invention by Charles Wilson: the cloud chamber. A cloud chamber, like its
successor, the bubble chamber, produces white traces along the path of charged particles;
the mechanism is the same as the one than leads to the white lines in the sky when an
aeroplane flies by.
The radiation detectors gave a strange result: most alpha
particles pass through the metal foil undisturbed, whereas
a few are reflected. In addition, those few which are reflec-
ted are not reflected by the surface, but in the inside of
Challenge 1385 n the foil. (Can you imagine how they showed this?) Ruther-
ford deduced from this scattering experiment that first of all,
atoms are mainly transparent. Only transparency explains
why most alpha particles pass the foil without disturbance,
even though it was over 2000 atoms thick. But some particles
Copyright © Christoph Schiller November 1997–May 2006

were scattered by large angles or even reflected. Rutherford

showed that the reflections must be due to a single scattering
point. By counting the particles that were reflected (about 1
in 20000 for his 0.4 µm gold foil), Rutherford was also able Marie Curie
to deduce the size of the reflecting entity and to estimate its
mass. He found that it contains almost all of the mass of the atom in a diameter of around
1 fm. He thus named it the nucleus. Using the knowledge that atoms contain electrons,
Rutherford then deduced from this experiment that atoms consist of an electron cloud
that determines the size of atoms – of the order of 0.1 nm – and of a tiny but heavy nucleus
at the centre. If an atom had the size of a basketball, its nucleus would have the size of a
dust particle, yet contain 99.9 % of the basketball’s mass. Atoms resemble thus candy floss

Dvipsbugw
 the structure of the nucleus – the densest clouds 901

TA B L E 66 The main types of radioactivity and rays emitted by matter

Ty pe Pa r t - Example Range Dan- Use

icle ger Shield
235
α rays helium U, 238 U, 238 Pu, a few cm in when any thickness
238
3 to 10 MeV nuclei Pu, 241 Am air eaten, material, measurement
inhaled, e.g. paper
touched
Dvipsbugw
β rays electrons 14
C, 40 K, 3 H, < 1 mm in serious metals cancer
101
0 to 5 MeV and Tc metal treatment
antineu- light years none none research
trinos
40
β + rays positrons K, 11 C, 11 C, 13 N, less than β medium any tomography
15
and O material

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

neutrinos light years none none research
110
γ rays high Ag several m in high thick lead preservation
energy air of herbs,
photons disinfection
252
n reactions neutrons Cf, Po-Li ... high 0.3 m of ...
c. 1 MeV (α,n), 38 Cl-Be paraffin
(γ,n)
9
n emission neutrons He, 24 N ... high 0.3 m of ...
typ. 40 MeV paraffin
5
p emission protons Be, 161 Re like α rays small solids
typ. 20 MeV
232
spontaneous nuclei Cm, 263 Rf like α rays small solids detection
fission of new
typ. 100 MeV elements

around a heavy dust particle. Even though the candy floss – the electron cloud – around
Copyright © Christoph Schiller November 1997–May 2006

the nucleus is extremely thin and light, it is strong enough to avoid that two atoms in-
terpenetrate; thus it keeps the neighbouring nuclei at constant distance. For the tiny and
massive alpha however, particles the candy floss is essentially empty space, so that they
simply fly through the electron clouds until they exit on the other side or hit a nucleus.
The density of the nucleus is impressive: about 5.5 ë 1017 kg~m3 . At that density, the
mass of the Earth would fit in a sphere of 137 m radius and a grain of sand would have a
Challenge 1386 e mass larger than the largest existing oil tanker. (True?) Now we know that oil tankers are
complex structures. What then is the structure of a nucleus?

I now know how an atom looks like!

“ Ernest Rutherford

”
Dvipsbugw
 902 viii inside the nucleus • the structure of the nucleus

Nuclei are composed

The magnetic resonance images also show that nuclei are composed. Images can be taken
also using heavier nuclei instead of hydrogen, such as certain fluorine or oxygen nuclei.
The д-factors of these nuclei also depart from the value 2 characteristic of point particles;
Page 760 the more massive they are, the bigger the departure. Such objects have a finite size; indeed,
the size of nuclei can be measured directly and confirm the values predicted by the д-
factor. Both the values of the д-factor and the non-vanishing sizes show that nuclei are
composed. Dvipsbugw
Interestingly, the idea that nuclei are composed is older than the concept of nucleus
itself. Already in 1815, after the first mass measurements of atoms by John Dalton and oth-
ers, researchers noted that the mass of the various chemical elements seem to be almost
perfect multiples of the weight of the hydrogen atom. William Prout then formulated the
hypothesis that all elements are composed of hydrogen. When the nucleus was discovered,
knowing that it contains almost all mass of the atom, it was therefore first thought that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

all nuclei are made of hydrogen nuclei. Being at the origin of the list of constituents, the
hydrogen nucleus was named proton, from the greek term for ‘first’ and reminding the
name of Prout at the same time. Protons carry a positive unit of electric charge, just the
opposite of that of electrons, but are almost 2000 times as heavy.
However, the charge and the mass numbers of the other nuclei do not match. On aver-
age, a nucleus that has n times the charge of a proton, has a mass that is about 2.6 n times
than of the proton. Additional experiments then confirmed an idea formulated by Werner
Heisenberg: all nuclei heavier than hydrogen nuclei are made of positively charged pro-
tons and neutral neutrons. Neutrons are particles a tiny bit more massive than protons
(the difference is less than a part in 700), but without any electrical charge. Since the
mass is almost the same, the mass of nuclei – and thus that of atoms – is still an (almost
perfect) integer multiple of the proton mass. But since neutrons are neutral, the mass and
the charge number of nuclei differ. Being neutral, neutrons do not leave tracks in clouds
chambers and are more difficult to detect. For this reason, they were discovered much
later than other subatomic particles.
Today it is possible to keep single neutrons suspended between suitably shaped coils,
with the aid of teflon ‘windows’. Such traps were proposed in 1951 by Wolfgang Paul. They
work because neutrons, though they have no charge, do have a small magnetic moment.
(By the way, this implies that neutrons are composed of charged particles.) With a suitable
arrangement of magnetic fields, neutrons can be kept in place, in other words, they can
Copyright © Christoph Schiller November 1997–May 2006

be levitated. Obviously, a trap only makes sense if the trapped particle can be observed.
In case of neutrons, this is achieved by the radio waves absorbed when the magnetic
moment switches direction with respect to an applied magnetic field. The result of these
experiments is simple: the lifetime of free neutrons is around 888(1) s. Nevertheless, in-
side most nuclei we are made of, neutrons do not decay, as the result does not lead to a
Challenge 1387 n state of lower energy. (Why not?)
Magnetic resonance images also show that some elements have different types of
atoms. These elements have atoms that with the same number of protons, but with dif-
ferent numbers of neutrons. One says that these elements have several isotopes.* This
* The name is derived from the Greek words for ‘same’ and ‘spot’, as the atoms are on the same spot in the
periodic table of the elements.

Dvipsbugw
 the structure of the nucleus – the densest clouds 903

Half-life

> 10+15 s 10-1 s

10+10 s 10-2 s
10+7 s 10-3 s
10+5 s 10-4 s
10+4 s 10-5 s
Dvipsbugw
10+3 s 10-6 s
10+2 s 10-7 s
10+1 s 10-15 s
10+0 s < 10-15 s
unknown

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Decay type

stable
el. capture (beta+)
beta - emission
alpha emission
proton emission
neutron emission
spontaneous fission
unknown

F I G U R E 346 All known nuclides with their lifetimes and main decay modes (data from
http://www.nndc.bnl.gov/nudat2)

also explains why some elements radiate with a mixture of different decay times. Though
chemically they are (almost) indistinguishable, isotopes can differ strongly in their nuc-
Copyright © Christoph Schiller November 1997–May 2006

lear properties. Some elements, such as tin, caesium, or polonium, have over thirty iso-
topes each. Together, the about 100 known elements have over 2000 nuclides.*
The motion of protons and neutrons inside nuclei allows to understand the spin and
the magnetic moment of nuclei. Since nuclei are so extremely dense despite containing nu-
merous positively charged protons, there must be a force that keeps everything together
against the electrostatic repulsion. We saw that the force is not influenced by electromag-
netic or gravitational fields; it must be something different. The force must be short range;
otherwise nuclei would not decay by emitting high energy alpha rays. The new force is
called the strong nuclear interaction. We shall study it in detail shortly.

* Nuclides is the standard expression for a nucleus with a given number of neutrons and protons.

Dvipsbugw
 904 viii inside the nucleus • the structure of the nucleus

metal wire
(e.g. paper clip)

thin
aluminium
foils
Dvipsbugw

F I G U R E 347 An electroscope (or electrometer) (© Harald Chmela) and its charged (left) and
uncharged state (right)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Nuclei can move alone – cosmic rays
In everyday life, nuclei are mostly found inside atoms. But in some situations, they move
all by themselves. The first to discover an example was Rutherford, who had shown that
alpha particles are helium nuclei. Like all nuclei, alpha particles are small, so that they are
quite useful as projectiles.
Then, in 1912, Viktor Heß* made a completely unexpected dis-
covery. Heß was intrigued by electroscopes (also called electromet-
ers). These are the simplest possible detectors of electric charge.
They mainly consist of two hanging, thin metal foils, such as two
strips of aluminium foil taken from a chocolate bar. When the elec-
troscope is charged, the strips repel each other and move apart, as
shown in Figure 347. (You can build one easily yourself by covering
an empty glass with some transparent cellophane foil and suspend-
Challenge 1388 e ing a paper clip and the aluminium strips from the foil.) An electro-
scope thus measures electrical charge. Like many before him, Heß
noted that even for a completely isolated electroscope, the charge
disappears after a while. He asked: why? By careful study he elim- Viktor Heß
inated one explanation after the other, he and others were left with
only one possibility: that the discharge could be due to charged rays, such as those of the
Copyright © Christoph Schiller November 1997–May 2006

recently discovered radioactivity. He thus prepared a sensitive electrometer and took it

with him on a balloon flight.
As expected, the balloon flight showed that the discharge effect diminished with
height, due to the larger distance from the radioactive substances on the Earth’s surface.

* Viktor Franz Heß, (1883–1964), Austrian nuclear physicist, received the Nobel Prize for physics in 1936 for
his discovery of cosmic radiation. Heß was one of the pioneers of research into radioactivity. Heß’ discovery
also explained why the atmosphere is always somewhat charged, a result important for the formation and
behaviour of clouds. Twenty years after the discovery of cosmic radiation, in 1932 Carl Anderson discovered
the first antiparticle, the positron, in cosmic radiation; in 1937 Seth Neddermeyer and Carl Anderson dis-
covered the muon; in 1947 a team led by Cecil Powell discovered the pion; in 1951, the Λ 0 and the kaon K 0
are discovered. All discoveries used cosmic rays and most of these discoveries led to Nobel Prizes.

Dvipsbugw
the structure of the nucleus – the densest clouds 905

TA B L E 67 The main types of cosmic radiation

Pa r t i c l e Energy Origin Detector Shield

At high altitude, the primary particles:

Protons (90 %) 109 to 1022 eV stars, supernovae, ex- scintillator in mines
tragalactic, unknown
Alpha rays (9 %) ... ... ... ...
Dvipsbugw
Other nuclei, such 109 to 1019 eV stars, novae ... ...
as iron (1 %)
Neutrinos MeV, GeV Sun, stars chlorine, none
gallium, water
Electrons (0.1 %) 106 to A 1012 eV supernova remnants

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Gammas (10−6 ) 1 eV to 50 TeV stars, pulsars, galactic, semiconductor in mines
extragalactic detectors
At sea level, secondary particles are produced in the atmosphere:
Muons 3 GeV, protons hit atmosphere, drift chamber 15 m of water
150~ m2 s produce pions which or 2.5 m of
decay into muons soil
Oxygen and other
nuclei
Positrons
Neutrons ...
Pions ...
In addition, there are slowed down primary beam particles.

But above about 1000 m of height, the discharge effect increased again, and the higher he
flew, the stronger it became. Risking his health and life, he continued upwards to more
than 5000 m; there the discharge effect was several times stronger than on the surface
of the Earth. This result is exactly what is expected from a radiation coming from outer
space and absorbed by the atmosphere. In one of his most important flights, performed
Copyright © Christoph Schiller November 1997–May 2006

during an (almost total) solar eclipse, Heß showed that most of the ‘height radiation’ did
not come from the Sun, but from further away. He – and Millikan – thus called the ra-
diation cosmic rays. During the last few centuries, many people have drunk from a glass
and eaten chocolate; but only Heß combined these activities with such careful observa-
tion and deduction that he earned a Nobel Prize.*
Today, the most impressive detectors for cosmic rays are Geiger–Müller counters and
spark chambers. Both share the same idea; a high voltage is applied between two metal
parts kept in a thin and suitably chosen gas (a wire and a cylindrical mesh for the Geiger-
Müller counter, two plates or wire meshes in the spark chambers). When a high energy
* In fact, Hess gold foils in his electrometer.

Dvipsbugw
 906 viii inside the nucleus • the structure of the nucleus

ionizing particle crosses the counter, a spark is generated, which can either be observed
through the generated spark (as you can do yourself in the entrance hall of the CERN
main building), or detected by the sudden current flow. Historically, the current was first
amplified and sent to a loudspeaker, so that the particles can be heard by a ‘click’ noise.
With a Geiger counter, one cannot see atoms or particles, but one can hear them. Finally,
ionized atoms could be counted. Finding the right gas mixture is tricky; it is the reason
that the counter has a double name. One needs a gas that extinguishes the spark after a
while, to make the detector ready for the next particle. Müller was Geiger’s assistant; he
made the best counters by adding the right mixture of alcohol to the gas in the chamber. Dvipsbugw
Nasty rumours maintained that this was discovered when another assistant tried, without
success, to build counters while Müller was absent. When Müller, supposedly a heavy
drinker, came back, everything worked again. However, the story is apocryphal. Today,
Geiger–Müller counters are used around the world to detect radioactivity; the smallest
fit in mobile phones and inside wrist watches.
The particle energy in cosmic rays spans a large range between 103 eV and at least

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1020 eV; the latter is the same energy as a tennis ball after serve. Understanding the origin
of cosmic rays is a science by its own. Some are galactic in origin, some are extragalactic.
For most energies, supernova remnants – pulsars and the like – seem the best candidates.
However, the source of the highest energy particles is still unknown.
In other words, cosmic rays are probably the only
type of radiation discovered without the help of shad-
ows. But shadows have been found later on. In a
beautiful experiment performed in 1994, the shadow
thrown by the Moon on high energy cosmic rays
(about 10 TeV) was studied. When the position of the
shadow is compared with the actual position of the
Moon, a shift is found. Due to the magnetic field of
the Earth, the cosmic ray Moon shadow would be shif-
ted westwards for protons and eastwards for antipro- Figure to be included

tons. The data are consistent with a ratio of antiprotons

Ref. 944 between 0 % and 30 %. By studying the shadow, the
experiment thus showed that high energy cosmic rays F I G U R E 348 A Geiger–Müller
are mainly positively charged and thus consist mainly counter
of matter, and only in small part, if at all, of antimatter.
Detailed observations showed that cosmic rays arrive on the surface of the Earth as a
Copyright © Christoph Schiller November 1997–May 2006

mixture of many types of particles, as shown in Table 67. They arrive from outside the
atmosphere as a mixture of which the largest fraction are protons, alpha particles, iron
and other nuclei. Nuclei can thus travel alone over large distances. The number of charged
cosmic rays depends on their energy. At the lowest energies, charged cosmic rays hit the
human body many times a second. The measurements also show that the rays arrive in
Page 938 irregular groups, called showers. The neutrino flux is many orders of magnitude higher,
but does not have any effect on human bodies.
The distribution of the incoming direction of cosmic rays shows that many rays must
be extragalactic in origin. The typical nuclei of cosmic radiation are ejected from stars
and accelerated by supernova explosions. When they arrive on Earth, they interact with
the atmosphere before they reach the surface of the Earth. The detailed acceleration mech-

Dvipsbugw
 the structure of the nucleus – the densest clouds 907

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 349 An aurora borealis produced by charged particles in the night
sky

anisms are still a topic of research.

Cosmic rays have several effects on everyday life. Through the charges they produce in
the atmosphere, they are probably responsible for the non-straight propagation of light-
ning. Cosmic rays are also important in the creation of rain drops and ice particles inside
clouds, and thus indirectly in the charging of the clouds. Cosmic rays, together with am-
Page 520 bient radioactivity, also start the Kelvin generator.
If the Moon would not exist, we would die from cosmic rays. The Moon helps to give
the Earth a high magnetic field via a dynamo effect, which then diverts most rays towards
the magnetic poles. Also the upper atmosphere helps animal life to survive, by shielding
life from the harmful effects of cosmic rays. Indeed, aeroplane pilots and airline employ-
ees have a strong radiation exposure that is not favourable to their health. Cosmic rays
are one of several reasons that long space travel, such as a trip to mars, is not an option
for humans. When cosmonauts get too much radiation exposure, the body weakens and
eventually they die. Space heroes, including those of science fiction, would not survive
much longer than two or three years.
Copyright © Christoph Schiller November 1997–May 2006

Cosmic rays also produce beautifully coloured flashes inside the eyes of cosmonauts;
they regularly enjoy these events in their trips. But cosmic rays are not only dangerous
and beautiful. They are also useful. If cosmic rays would not exist at all, we would not
exist either. Cosmic rays are responsible for mutations of life forms and thus are one of
the causes of biological evolution. Today, this effect is even used artificially; putting cells
into a radioactive environment yields new strains. Breeders regularly derive new mutants
in this way.
Cosmic rays cannot be seen directly, but their cousins, the ‘solar’ rays, can. This is
most spectacular when they arrive in high numbers. In such cases, the particles are inev-
itably deviated to the poles by the magnetic field of the Earth and form a so-called aurora
borealis (at the North Pole) or an aurora australis (at the South pole). These slowly mov-

Dvipsbugw
 908 viii inside the nucleus • the structure of the nucleus

Dvipsbugw
F I G U R E 350 An aurora australis on Earth seen from space (in the X-ray domain) and
one on Saturn

ing and variously coloured curtains of light belong to the most spectacular effects in the
night sky. Visible light and X-rays are emitted at altitudes between 60 and 1000 km. Seen
from space, the aurora curtains typically form a circle with a few thousand kilometres

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

diameter around the magnetic poles.*
Cosmic rays are mainly free nuclei. With time, researchers found that nuclei appear
without electron clouds also in other situations. In fact, the vast majority of nuclei in the
universe have no electron clouds at all: in the inside of stars no nucleus is surrounded
by bound electrons; similarly, a large part of intergalactic matter is made of protons. It is
known today that most of the matter in the universe is found as protons or alpha particles
inside stars and as thin gas between the galaxies. In other words, in contrast to what the
Greeks said, matter is not usually made of atoms; it is mostly made of nuclei. Our everyday
environment is an exception when seen on cosmic scales. In nature, atoms are rare.
By the way, nuclei are in no way forced to move; nuclei can also be stored with almost
no motion. There are methods – now commonly used in research groups – to superpose
electric and magnetic fields in such a way that a single nucleus can be kept floating in
Page 602 mid-air; we discussed this possibility in the section on levitation earlier on.

Nuclei decay
Not all nuclei are stable over time. The first measurement that provided a hint was the
way radioactivity changes with time. The number N of atoms decreases with time. More
precisely, radioactivity follows an exponential decay:

N(t) = N(0)e−t~τ
Copyright © Christoph Schiller November 1997–May 2006

(615)

The parameter τ, the so-called life time, depends on the type of nucleus emitting the rays.
It can vary from much less than a microsecond to millions of millions of years. The ex-
pression has been checked for as long as 34 multiples of the duration τ; its validity and
precision is well-established by experiments. Radioactivity is the decay of unstable nuclei.
Formula (615) is an approximation for large numbers of atoms, as it assumes that N(t) is a
continuous variable. Despite this approximation, deriving this expression from quantum
Page 832 theory is not a simple exercise, as we saw in the section on atomic physics. Though the

* In the solar system, aurorae due to core magnetic fields have been observed on Jupiter, Saturn, Uranus,

Dvipsbugw
 the structure of the nucleus – the densest clouds 909

quantum Zeno effect can appear for small times t, for the case of radioactivity it has not
been observed so far.
Most of all, the expression (615) allows to count the number of atoms in a given mass
of material. Imagine to have measured the mass of radioactive material at the beginning
of your experiment; you have chosen an element that has a lifetime of about a day. Then
you put the material inside a scintillation box. After a few weeks the number of flashes
has become so low that you can count them; using the formula you can then determine
how many atoms have been in the mass to begin with. Radioactivity thus allows us to
determine the number of atoms, and thus their size, in addition to the size of nuclei. Dvipsbugw
The decay (615) and the release of energy is typical of metastable systems. In 1903,
Rutherford and Soddy discovered what the state of lower energy is for alpha and beta
emitters. In these cases, radioactivity changes the emitting atom; it is a spontaneous trans-
mutation of the atom. An atom emitting alpha or beta rays changes its chemical nature.
Page 249 Radioactivity confirms what statistical mechanics of gases had concluded long time be-
fore: atoms have a structure that can change. In alpha decay, the radiating nucleus emits a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(doubly charged) helium nucleus. The kinetic energy is typically a handful of MeV. After
the emission, the nucleus has changed to one situated two places earlier in the periodic
system of the elements.
In beta decay, a neutron transforms itself into a proton, emitting an electron and an
antineutrino. Also beta decay changes the chemical nature of the atom, but to the place
following the original atom in the periodic table of the elements. A variation is the beta+
decay, in which a proton changes into a neutron and emits a neutrino and a positron. We
Page 936 will study these important decay processes below.
In gamma decay, the nucleus changes from an excited to a lower energy state by emit-
ting a high energy photon. In this case, the chemical nature is not changed. Typical en-
ergies are in the MeV range. Due to the high energy, such rays ionize the material they
encounter; since they are not charged, they are not well absorbed by matter and penetrate
deep into materials. Gamma radiation is thus by far the most dangerous type of (outside)
radioactivity.
By the way, in every human body about nine thousand radioactive decays take place
every second, mainly 4.5 kBq (0.2 mSv~a) from 40 K and 4 kBq from 14 C (0.01 mSv~a).
Challenge 1389 n Why is this not dangerous?
All radioactivity is accompanied by emission of energy. The energy emitted by an atom
trough radioactive decay or reactions is regularly a million time large than that emitted
by a chemical process. That is the reason for the danger of nuclear weapons. More than a
Copyright © Christoph Schiller November 1997–May 2006

decay, a radioactive process is thus an explosion.

What distinguishes those atoms that decay from those which do not? An exponential
decay law implies that the probability of decay is independent of the age of the atom. Age
Page 772 or time plays no role. We also know from thermodynamics, that all atoms are exactly
identical. So how is the decaying atom singled out? It took around 40 years to discover
that decays are triggered by the statistical fluctuations of the vacuum, as described by
quantum theory. Indeed, radioactivity is one of the clearest observations that classical
physics is not sufficient to describe nature. Radioactivity, like all decays, is a pure quantum
effect. Only a finite quantum of action makes it possible that a system remains unchanged

Neptune, Earth, Io and Ganymede. Aurorae due to other mechanisms have been seen on Venus and Mars.

Dvipsbugw
 910 viii inside the nucleus • the structure of the nucleus

until it suddenly decays. Indeed, in 1928 George Gamow explained alpha decay with the
tunnelling effect. The tunnelling effect explains the relation between the lifetime and the
range of the rays, as well as the measured variation of lifetimes – between 10 ns and 1017
years – as the consequence of the varying potentials to be overcome.
By the way, massless particles cannot decay. There is a simple reason for it: massless
particles do not experience time, as their paths are null. A particle that does not experi-
Challenge 1390 ny ence time cannot have a half-life. (Can you find another argument?)
As a result of the chemical effects of radioactivity, the composition ratio of certain ele-
Page 919 ments in minerals allows to determine the age of the mineral. Using radioactive decay to Dvipsbugw
deduce the age of a sample is called radiometric dating. With this technique, geologists
determined the age of mountains, the age of sediments and the age of the continents.
They determined the time that continents moved apart, the time that mountains formed
when the continents collided and the time when igneous rocks were formed. The times
found in this way are consistent with the relative time scale that geologists had defined
independently for centuries before the technique appeared. With the appearance of ra-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

diometric dating, all fell into place. Equally successful was the radiocarbon method; with
it, historians determined the age of civilizations and the age of human artefacts.* Many
false beliefs were shattered. In some communities the shock is still not over, even though
Ref. 926 over hundred years have passed since these results became known.
With the advent of radiometric dating, for the first time it became possible to reliably
date the age of rocks, to compare it with the age of meteorites and, when space travel
became fashionable, with the age of the Moon. The result of the field of radiometric dating
was beyond all estimates and expectations: the oldest rocks and the oldest meteorites
studied independently using different dating methods, are 4570(10) million years old. But
if the Earth is that old, why did the Earth not cool down in its core in the meantime?

Why is hell hot?

The lava seas and streams found in and around volcanoes are the origin of the images
that many cultures ascribe to hell: fire and suffering. Because of the high temperature of
lava, hell is inevitably depicted as a hot place. A striking example is the volcano Erta Ale,
shown in Figure 351. But why is lava still hot, after 4570 million years?
A straightforward calculation shows that if the Earth had been a hot sphere in the
Challenge 1391 ny beginning, it should have cooled down and solidified already long time ago. The Earth
should be a solid object, like the moon: the Earth should not contain any lava and hell
Copyright © Christoph Schiller November 1997–May 2006

would not be hot.

The solution to the riddle is provided by radioactivity: the centre of the Earth contains
an oven fuelled by radioactive potassium 40 K, radioactive uranium 235 U and 238 U and
Ref. 925 radioactive thorium 232 Th. The radioactivity of these elements, and to minor degree a
few others, keeps the centre of the Earth glowing. More precise investigations, taking into
Page 920 account the decay times and material concentrations, show that this mechanism indeed
explains the internal heat of the Earth. (By the way, the decay of potassium is the origin
for the 1 % of argon found in the Earth’s atmosphere.)

* In 1960, the developer of the radiocarbon dating technique, Willard Libby, received the Nobel Prize for
chemistry.

Dvipsbugw
 the structure of the nucleus – the densest clouds 911

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 351 The lava sea in the volcano Erta Ale in Ethiopia (© Marco Fulle)

This brings up a challenge: why is the radioactivity of lava and of the Earth in general
Challenge 1392 n not dangerous to humans?

Nuclei can form composites

Nuclei are highly unstable when they contain more than about 280 nucleons. Higher mass
values inevitably decay into smaller fragments. But when the mass is above 1057 nucleons,
nuclear composites are stable again: such systems are then called neutron stars. This is
the most extreme example of pure nuclear matter found in nature. Neutron stars are left
overs of (type II) supernova explosions. They do not run any fusion reactions any more,
as other stars do; in first approximation they are simply a large nucleus.
Neutron stars are made of degenerate matter. Their density of 1018 kg~m3 is a few times
that of a nucleus, as gravity compresses the star. This density value means that tea spoon
of such a star has a mass of several 100 million tons. Neutron stars are about 10 km in
Copyright © Christoph Schiller November 1997–May 2006

diameter. They are never much smaller, as such stars are unstable. They are never much
larger, because more massive neutron stars turn into black holes.

Nuclei have colours and shapes

In everyday life, the colour of objects is determined by the wavelength of light that is
least absorbed, or if they shine, by the wavelength that is emitted. Also nuclei can absorb
photons of suitably tuned energies and get into an excited state. In this case, the photon
energy is converted into a higher energy of one or several of the nucleons whirling around
inside the nucleus. Many radioactive nuclei also emit high energy photons, which then
are called gamma rays, in the range of 1 keV (or 0.2 fJ) to about 20 MeV (or 3.3 pJ). The
process is similar to the emission of light by electrons in atoms. From the energy, the

Dvipsbugw
 912 viii inside the nucleus • the structure of the nucleus

Dvipsbugw

F I G U R E 352 Various nuclear shapes – fixed (left) and oscillating (right), shown
realistically as clouds (above) and simplified as geometric shapes (below)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

number and the lifetime of the excited states – they range from 1 ps to 300 d – researchers
can deduce how the nucleons move inside the nucleus.
The photon energies define the ‘colour’ of the nucleus. It can be used, like all colours, to
distinguish nuclei from each other and to study their motion. in particular, the colour of
the γ-rays emitted by excited nuclei can be used to determine the chemical composition
of a piece of matter. Some of these transition lines are so narrow that they can been used
to study the change due to the chemical environment of the nucleus, to measure their
motion or to detect the gravitational Doppler effect.
The study of γ-rays also allows to determine the shape of nuclei. Many nuclei are spher-
ical; but many are prolate or oblate ellipsoids. Ellipsoids are favoured if the reduction in
average electrostatic repulsion is larger than the increase in surface energy. All nuclei –
except the lightest ones such as helium, lithium and beryllium – have a constant mass
density at their centre, given by about 0.17 fermions per fm3 , and a skin thickness of
about 2.4 fm, where their density decreases. Nuclei are thus small clouds, as shown in
Figure 352.
We know that molecules can be of extremely involved shape. In contrast, nuclei are
mostly spheres, ellipsoids or small variations of these. The reason is the short range, or
better, the fast spatial decay of nuclear interactions. To get interesting shapes like in mo-
lecules, one needs, apart from nearest neighbour interactions, also next neighbour inter-
Copyright © Christoph Schiller November 1997–May 2006

actions and next next neighbour interactions. The strong nuclear interaction is too short
ranged to make this possible. Or does it? It might be that future studies will discover that
some nuclei are of more unusual shape, such as smoothed pyramids. Some predictions
Ref. 939 have been made in this direction; however, the experiments have not been performed yet.
The shape of nuclei does not have to be fixed; nuclei can also oscillate in shape. Such
oscillations have been studied in great detail. The two simplest cases, the quadrupole and
octupole oscillations, are shown in Figure 352. Obviously, nuclei can also rotate. Rapidly
spinning nuclei, with a spin of up to 60ħ and more, exist. They usually slow down step by
step, emitting a photon and reducing their angular momentum at each step. Recently it
Ref. 940 was discovered that nuclei can also have bulges that rotate around a fixed core, a bit like
tides rotate around the Earth.

Dvipsbugw
 the structure of the nucleus – the densest clouds 913

Motion in the nuclear domain – four types of motion

Nuclei are small because the nuclear interactions are short-ranged. Due to this short
range, nuclear interactions play a role only in types of motion: scattering, bound motion,
decay and a combination of these three called nuclear reactions. The history of nuclear
physics showed that the whole range of observed phenomena can be reduced to these
four fundamental processes. In each motion type, the main interest is what happens at
the start and at the end; the intermediate situations are less interesting. Nuclear interac-
tions thus lack the complex types of motion which characterize everyday life. That is the Dvipsbugw
reason for the shortness of this chapter.
Scattering is performed in all accelerator experiments. Such experiments repeat for
nuclei what we do when we look at an object. Seeing is a scattering process, as seeing
is the detection of scattered light. Scattering of X-rays was used to see atoms for the first
time; scattering of high energy alpha particles was used to discover and study the nucleus,
and later the scattering of electrons with even higher energy was used to discover and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

study the components of the proton.
Bound motion is the motion of protons and neutrons inside nuclei or the motion
of quarks inside hadrons. Bound motion determines shape and shape changes of com-
pounds.
Decay is obviously the basis of radioactivity. Decay can be due to the electromagnetic,
the strong or the weak nuclear interaction. Decay allows to study the conserved quantities
of nuclear interactions.
Nuclear reactions are combinations of scattering, decay and possibly bound motion.
Nuclear reactions are for nuclei what the touching of objects is in everyday life. Touching
an object we can take it apart, break it, solder two objects together, throw it away, and
much more. The same can be done with nuclei. In particular, nuclear reactions are re-
sponsible for the burning of the Sun and the other stars; they also tell the history of the
nuclei inside our bodies.
Quantum theory showed that all four types of motion can be described in the same
way. Each type of motion is due to the exchange of virtual particles. For example, scat-
tering due to charge repulsion is due to exchange of virtual photons, the bound motion
inside nuclei due to the strong nuclear interaction is due to exchange of virtual gluons,
beta decay is due to the exchange of virtual W bosons, and neutrino reactions are due to
the exchange of virtual Z bosons. The rest of this chapter explains these mechanisms in
more details.
Copyright © Christoph Schiller November 1997–May 2006

Nuclei react
The first man who thought to have made transuranic elements, the Italian genius Enrico
Page 775 Fermi, received the Nobel Prize for the discovery. Shortly afterwards, Otto Hahn and his
collaborators Lise Meitner and Fritz Strassman showed that Fermi was wrong, and that
his prize was based on a mistake. Fermi was allowed to keep his prize, the Nobel com-
mittee gave Hahn the Nobel Prize as well, and to make the matter unclear to everybody
and to women physicists in particular, the prize was not given to Lise Meitner. (After her
death, a new element was named after her.)
When protons or neutrons were shot into nuclei, they usually remained stuck inside
them, and usually lead to the transformation of an element into a heavier one. After hav-

Dvipsbugw
914 viii inside the nucleus • the structure of the nucleus

ing done this with all elements, Fermi used uranium; he found that bombarding it with
neutrons, a new element appeared, and concluded that he had created a transuranic ele-
ment. Alas, Hahn and his collaborators found that the element formed was well-known:
it was barium, a nucleus with less than half the mass of uranium. Instead of remaining
stuck as in the previous 91 elements, the neutrons had split the uranium nucleus. Hahn,
Meitner and Strassmann had observed reactions such as:
235 143
U+n Ba + 90 Kr + 3n + 170 MeV . (616)
Dvipsbugw
Meitner called the splitting process nuclear fission. A large amount of energy is liberated
in fission. In addition, several neutrons are emitted; they can thus start a chain reaction.
Later, and (of course) against the will of the team, the discovery would be used to make
nuclear bombs.
Reactions and decays are transformations. In each transformation, already the Greek
taught us to search, first of all, for conserved quantities. Besides the well-known cases

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of energy, momentum, electric charge and angular momentum conservation, the results
of nuclear physics lead to several new conserved quantities. The behaviour is quite con-
strained. Quantum field theory implies that particles and antiparticles (commonly de-
noted by a bar) must behave in compatible ways. Both experiment and quantum field
theory show for example that every reaction of the type A + B C + D implies that the
reactions A + C B + D or C + D A + B or, if energy is sufficient, A C + D + B, are
also possible. Particles thus behave like conserved mathematical entities.
Experiments show that antineutrinos differ from neutrinos. In fact, all reactions con-
firm that the so-called lepton number is conserved in nature. The lepton number L is zero
for nucleons or quarks, is 1 for the electron and the neutrino, and is −1 for the positron
and the antineutrino.
In addition, all reactions conserve the so-called baryon number. The baryon number B
is 1 for protons and neutrons (and 1/3 for quarks), and −1 for antiprotons and antineutrons
(and thus −1~3 for antiquarks). So far, no process with baryon number violation has ever
been observed. Baryon conservation is one reason for the danger of radioactivity, fission
and fusion.

B ombs and nuclear reactors

Uranium fission is triggered by a neutron, liberates energy and produces several addi-
Copyright © Christoph Schiller November 1997–May 2006

tional neutrons. It can trigger a chain reaction which can lead to an explosion or a con-
trolled generation of heat. Once upon a time, in the middle of the twentieth century, these
processes were studied by quite a number of researchers. Most of them were interested
in making weapons or in using nuclear energy, despite the high toll these activities place
on the economy, on human health and on the environment.
Most stories around this topic are absurd. The first nuclear weapons were built during
the second world war with the smartest physicists that could be found. Everything was
ready, including the most complex physical models, factories and an organization of in-
credible size. There was just one little problem: there was no uranium of sufficient quality.
The mighty United States thus had to go around the world to shop for good uranium.
They found it in the Belgian colony of Congo, in central Africa. In short, without the sup-

Dvipsbugw
 the structure of the nucleus – the densest clouds 915

port of Belgium, which sold the Congolese uranium to the USA, there would have been
no nuclear bomb, no early war end and no superpower status.
Congo paid a high price for this important status. It was ruled by a long chain of milit-
ary dictators up to this day. But the highest price was paid by the countries that actually
built nuclear weapons. Some went bankrupt, others remained underdeveloped, still other
countries have amassed huge debts and have a large underprivileged population. There is
no exception. The price of nuclear weapons has also been that some regions of our planet
became uninhabitable, such as numerous islands, deserts and marine environments. But
it could have been worse. When the most violent physicist ever, Edward Teller, made his Dvipsbugw
first calculations about the hydrogen bomb, he predicted that the bomb would set the
atmosphere into fire. Nobel Prize winner Hans Bethe corrected the mistake and showed
that nothing of this sort would happen. Nevertheless, the military preferred to explode
the hydrogen bomb in the Bikini atoll, the most distant place from their homeland they
Ref. 943 could find. Today it is even dangerous simply to fly over that island. It was them noticed
that nuclear test explosions increased ambient radioactivity in the atmosphere all over

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the world. Of the produced radioactive elements, 3 H is absorbed by humans in drinking
water, 14 C and 90 Sr through food, and 137 Cs in both ways. In the meantime, all countries
have agreed to perform their nuclear tests underground.
But even peaceful nuclear reactors are dangerous. The reason was discovered in 1934
by Frédéric Joliot and his wife Irène, the daughter of Pierre and Marie Curie: artificial
radioactivity. The Joliot–Curies discovered that materials irradiated by alpha rays become
radioactive in turn. They found that alpha rays transformed aluminium into radioactive
phosphorous:
27 4 4
13 Al + 2 α 15 P . (617)

In fact, almost all materials become radioactive when irradiated with alpha particles, neut-
rons or gamma rays. As a result, radioactivity itself can only be contained with difficulty.
After a time which depends on the material and the radiation, the box that contains ra-
dioactive material has itself become radioactive.
The dangers of natural and artificial radioactivity are the reason for the high costs of
nuclear reactors. After about thirty years of operation, reactors have to be dismantled.
The radioactive pieces have to be stored in specially chosen, inaccessible places, and at
the same time the workers’ health must not be put in danger. The world over, many dis-
mantlings are now imminent. The companies performing the job sell the service at high
price. All operate in a region not far from the border to criminal activity, and since radio-
Copyright © Christoph Schiller November 1997–May 2006

activity cannot be detected by the human senses, many crossed it. In fact, an important
nuclear reactor is (usually) not dangerous to humans: the Sun.

The Sun
Nuclear physics is the most violent part of physics. But despite this bad image, nuclear
physics has something to offer which is deeply fascinating: the understanding of the Sun,
the stars and the early universe.
The Sun emits 385 YW of light. Where does this energy come from? If it came by burn-
ing coal, the Sun would stop burning after a few thousands of years. When radioactivity
was discovered, researchers tested the possibility that this process was at the heart of the

Dvipsbugw
 916 viii inside the nucleus • the structure of the nucleus

TA B L E 68 Some radioactivity measurements

M at e r i a l Activity in
B q/kg

air c. 10−2
sea water 101
human body c. 102
cow milk max. 103
pure 238 U metal c. 107 Dvipsbugw
highly radioactive α emitters A 107
radiocarbon: 14 C (β emitter) 108
highly radioactive β and γ emitters A9
main nuclear fallout: 137 Cs, 90 Sr (α emitter) 2 ë 109
polonium, one of the most radioactive materials (α) 1024

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Sun’s shining. However, even though radioactivity can produce more energy than chem-
ical burning, the composition of the Sun – mostly hydrogen and helium – makes this
impossible. In fact, the study of nuclei showed that the Sun burns by hydrogen fusion.
Fusion is the composition of a large nucleus from smaller ones. In the Sun, the fusion
reaction
4 1 H 4 He + 2 e + + 2 ν + 4.4 pJ (618)

is the result of a continuous cycle of three separate nuclear reactions:

1
H + 1H 2
H + e + + ν(a weak nuclear reaction)
2
H + 1H 3
He + γ(a strong nuclear reaction)
3 3 4
He + He He + 2 1 H + γ . (619)

In total, four protons are thus fused to one helium nucleus; if we include the electrons,
four hydrogen atoms are fused to one helium atom with the emission of neutrinos and
Copyright © Christoph Schiller November 1997–May 2006

light with a total energy of 4.4 pJ (26.7 MeV). Most of the energy is emitted as light;
around 10 % is carried away by neutrinos. The first of the three reaction of equation 619
is due to the weak nuclear interaction; this avoids that it happens too rapidly and ensures
that the Sun will shine still for some time. Indeed, in the Sun, with a luminosity of 385 YW,
Ref. 941 there are thus about 1038 fusions per second. This allows to deduce that the Sun will last
another handful of Ga (Gigayears) before it runs out of fuel.
The fusion reaction (619) takes place in the centre of the Sun. The energy carried away
by the photons arrives at the Sun’s surface about two hundred thousand years later; this
delay is due to the repeated scattering of the photon by the constituents inside the Sun.
After two-hundred thousand years, the photons take another 8.3 minutes to reach the
Earth and to sustain the life of all plants and animals.

Dvipsbugw
 the structure of the nucleus – the densest clouds 917

TA B L E 69 Human exposure to radioactivity and the corresponding doses

Exposure Dose

Daily human exposure:

Average exposure to cosmic radiation in Europe, at sea levelc. 0.3 mSv~a (1.2 mSv~a)
(3 km)
Average (and maximum) exposure to soil radiation, without0.4 mSv~a (2 mSv~a)
radon Dvipsbugw
Average (and maximum) inhalation of radon 1 mSv~a (100 mSv~a)
Average exposure due to internal radionuclides 0.3 mSv~a
natural content of 40 K in human muscles 10−4 Gy and 4500 Bq
natural content of Ra in human bones 2 ë 10−5 Gy and 4000 Bq
14
natural content of C in humans 10−5 Gy
Total average (and maximum) human exposure 2 mSv~a (100 mSv~a)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Common situations:
Dental X-ray c. 10 mSv equivalent dose
Lung X-ray c. 0.5 mSv equivalent dose
Short one hour flight (see http://www.gsf.de/epcard) c. 1 µSv
Transatlantic flight c. 0.04 mSv
Maximum allowed dose at work 30 mSv~a
Deadly exposures:
Ionization 0.05 C~kg can be deadly
Dose 100 Gy=100 J~kg is deadly in 1 to 3
days
Equivalent dose more than 3 Sv~a leads to death

Curiosities and fun challenges on radioactivity

It is still not clear whether the radiation of the Sun is constant over long time scales. There
is an 11 year periodicity, the famous solar cycle, but the long term trend is still unknown.
Precise measurements cover only the years from 1978 onwards, which makes only about
Copyright © Christoph Schiller November 1997–May 2006

3 cycles. A possible variation of the solar constant might have important consequences
for climate research; however, the issue is still open.

**
Not all γ-rays are due to radioactivity. In the year 2000, an Italian group discovered that
Ref. 937 thunderstorms also emit γ-rays, of energies up to 10 MeV. The mechanisms are still being
investigated.
**
Chain reactions are quite common in nature. Fire is a chemical chain reaction, as are
exploding fireworks. In both cases, material needs heat to burn; this heat is supplied by a

Dvipsbugw
 918 viii inside the nucleus • the structure of the nucleus

neighbouring region that is already burning.

**
Radioactivity can be extremely dangerous to humans. The best example is plutonium.
Only 1 µg of this alpha emitter inside the human body are sufficient to cause lung cancer.
**
Lead is slightly radioactive, because it contains the 210 Pb isotope, a beta emitter. This
lead isotope is produced by the uranium and thorium contained in the rock from where Dvipsbugw
the lead is extracted. For sensitive experiments, such as for neutrino experiments, one
needs radioactivity shields. The best shield material is lead, but obviously it has to be low
radioactivity lead. Since the isotope 210 Pb has a half-life of 22 years, one way to do it is to
use old lead. In a precision neutrino experiment in the Gran Sasso in Italy, the research
team uses lead mined during Roman times in order to reduce spurious signals.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Not all reactors are human made. Natural reactors have been predicted in 1956 by Paul
Kuroda. In 1972 the first example was found. In Oklo, in the African country of Gabon,
there is a now famous geological formation where uranium is so common that two thou-
sand million years ago a natural nuclear reactor has formed spontaneously – albeit a small
one, with an estimated power generation of 100 kW. It has been burning for over 150 000
years, during the time when the uranium 235 percentage was 3% or more, as required for
chain reaction. (Nowadays, the uranium 235 content on Earth is 0.7%.) The water of a
nearby river was periodically heated to steam during an estimated 30 minutes; then the
reactor cooled down again for an estimated 2.5 hours, since water is necessary to mod-
erate the neutrons and sustain the chain reaction. The system has been studied in great
detail, from its geological history up to the statements it makes about the constancy of
the ‘laws’ of nature. The studies showed that 2000 million years ago the mechanisms were
the same as those used today.

**
High energy radiation is dangerous to humans. In the 1950s, when nuclear tests were still
made above ground by the large armies in the world, the generals overruled the orders
of the medical doctors. They positioned many soldiers nearby to watch the explosion,
and worse, even ordered them to walk to the explosion site as soon as possible after the
Copyright © Christoph Schiller November 1997–May 2006

explosion. One does not need to comment on the orders of these generals. Several of these
unlucky soldiers made a strange observation: during the flash of the explosion, they were
Challenge 1393 ny able to see the bones in their own hand and arms. How can this be?
**
The SI units for radioactivity are now common; in the old days, 1 Sv was called 100 rem
or ‘Röntgen equivalent man’; The SI unit for dose, 1 Gy = 1 J~kg, replaces what used to be
called 100 rd or Rad. The SI unit for exposition, 1 C~kg, replaces the older unit ‘Röntgen’,
for which the relation is 1 R = 2.58 ë 10−4 C~kg.

**

Dvipsbugw
 the structure of the nucleus – the densest clouds 919

Nuclear bombs are terrible weapons. To experience their violence but also the criminal
Ref. 938 actions of many military people during the tests, have a look at the pictures of explosions.
In the 1950 and 60s, nuclear tests were performed by generals who refused to listen to
doctors and scientists. Generals ordered to explode these weapons in the air, making the
complete atmosphere of the world radioactive, hurting all mankind in doing so; worse,
they even obliged soldiers to visit the radioactive explosion site a few minutes after the
explosion, thus doing their best to let their own soldiers die from cancer and leukaemia.
** Dvipsbugw
The technique of radiometric dating has deeply impacted astronomy, geology, evolu-
tionary biology, archaeology and history. (And it has reduced the number of violent
Ref. 926 believers.) Half-lives can usually be measured to within one or two percent of accuracy,
and they are known both experimentally and theoretically not to change over geological
time scales. As a result, radiometric dating methods can be be surprisingly precise. But
Challenge 1394 n how does one measure half-lives of thousands of millions of years to high precision?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
The beta decay of the radioactive carbon isotope 14 C has a decay time of 5568 a. This
isotope is continually created in the atmosphere through the influence of cosmic rays.
This happens through the reaction 14 N + n p + 14 C. As a result, the concentration of
radiocarbon in air is relatively constant over time. Inside living plants, the metabolism
thus (unknowingly) maintains the same concentration. In dead plants, the decay sets in.
The decay time of a few thousand years is particularly useful to date historic material. The
method, called radiocarbon dating, has been used to determine the age of mummies, the
age of prehistoric tools and the age of religious relics. The original version of the technique
measured the radiocarbon content through its radioactive decay and the scintillations
it produced. A quality jump was achieved when accelerator mass spectroscopy became
commonplace. It was not necessary any more to wait for decays: it is now possible to
determine the 14 C content directly. As a result, only a tiny amount of carbon, as low as
0.2 mg, is necessary for a precise dating. This technique showed that numerous religious
relics are forgeries, such as a cloth in Turin, and several of their wardens turned out to be
crooks.
**
Researchers have developed an additional method to date stones using radioactivity.
Copyright © Christoph Schiller November 1997–May 2006

Whenever an alpha ray is emitted, the emitting atom gets a recoil. If the atom is part
of a crystal, the crystal is damaged by the recoil. The damage can be seen under the mi-
croscope. By counting the damaged regions it is possible to date the time at which rocks
have been crystallized. In this way it has been possible to determine when material from
volcanic eruptions has become rock.
**
Several methods to date wine are used, and more are in development. A few are given in
Table 70.

**

Dvipsbugw
920 viii inside the nucleus • the structure of the nucleus

TA B L E 70 Natural isotopes used in radiometric dating

Isotope D e c ay Half-life Method using it Examples

product
147 143
Sm Nd 106 Ga samarium–neodynium rocks, lunar soil,
method meteorites
87 87
Rb Sr 48.8 Ga rubidium–strontium rocks, lunar soil,
method meteorites
187 187 Dvipsbugw
Rh Os 42 Ga rhenium–osmium rocks, lunar soil,
method meteorites
176 176
Lu Hf 38 Ga lutetium–hafnium rocks, lunar soil,
method meteorites
232 208
Th Pb 14 Ga thorium–lead method, rocks, lunar soil,
lead–lead method meteorites
238 206
U Pb 4.5 Ga uranium–lead method, rocks, lunar soil,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

lead–lead method meteorites
40 40
K Ar 1.26 Ga potassium–argon rocks, lunar soil,
method, argon–argon meteorites
method
235 207
U Pb 0.7 Ga uranium–lead method, rocks, lunar soil,
lead–lead method meteorites
10 10
Be B 1.52 Ma cosmogenic radiometric ice cores
dating
60
Fe ... 1.5 Ma supernova debris dating deep sea crust
36 36
Cl Ar 0.3 Ma cosmogenic radiometric ice cores
dating
234 230
U Th 248 ka uranium–thorium corals, stalactites,
method bones, teeth
230 226
Th Ra 75, 4 ka
14 14
C N 5715 a radiocarbon method wood, clothing, bones,
organic material, wine
137
Cs 30 a gamma-ray counting dating food and wine
after Chernobyl
nuclear accident
210
Pb 22 a gamma-ray counting dating wine
Copyright © Christoph Schiller November 1997–May 2006

3
H 12.3 a gamma-ray counting dating wine

Selected radioactive decay times can be changed by external influence. Electron capture,
as observed in beryllium-7, is one of the rare examples were the decay time can change, by
up to 1.5%, depending on the chemical environment. The decay time for the same isotope
has also been found to change by a fraction of a percent under pressures of 27 GPa. On
the other hand, these effects are predicted (and measured) to be negligible for nuclei of
larger mass.

Dvipsbugw
 the strong nuclear interaction and the birth of matter 921

29. the strong nuclear interaction and the birth

of mat ter
Lernen ist Vorfreude auf sich selbst.*

“ Peter Sloterdijk

Since protons are positively charged, inside nuclei they must be bound by a force strong ”
enough to keep them together against their electromagnetic repulsion. This is the strong
nuclear interaction. Most of all, the strong interaction tells a good story about the stuff Dvipsbugw
we are made of.

Why do the stars shine?

Don’t the stars shine beautifully? I am the only

“ person in the world who knows why they do.

Frits Houtermans

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

All stars shine because of fusion. When two light nuclei are fused to a heavier one, some
energy is set free, as the average nucleon is bound more strongly. This energy gain is
possible until the nuclei of iron 56 Fe are made. For nuclei beyond this nucleus, the binding
energies per nucleon then decrease again; thus fusion is not energetically possible.**
The different stars observed in the sky*** can be distinguished by the type of fusion
nuclear reaction that dominates. Most stars, in particular young or light stars run hy-
drogen fusion. In fact, there are at least two main types of hydrogen fusion: the direct
hydrogen-hydrogen (p-p) cycle and the CNO cycle(s).
Page 916 The hydrogen cycle described above is the main energy source of the Sun. The simple
description does not fully purvey the fascination of the process. On average, protons in
the Sun’s centre move with 600 km~s. Only if they hit each other precisely head-on can
a nuclear reaction occur; in all other cases, the electrostatic repulsion between the pro-
tons keeps them apart. For an average proton, a head-on collision happens once every 7
thousand million years. Nevertheless, there are so many proton collisions in the Sun that
every second four million tons of hydrogen are burned to helium.
Fortunately for us, the photons generated in the Sun’s centre are ‘slowed’ down by the
outer parts of the Sun. In this process, gamma photons are progressively converted to
visible photons. As a result, the sunlight of today was in fact generated at the time of the
Ref. 927 Neandertalers: a typical estimate is about 200 000 years ago. In other words, the effective
Copyright © Christoph Schiller November 1997–May 2006

speed of light right at the centre of the Sun is estimated to be around 10 km~year.
If a star has heavier elements inside it, the hydrogen fusion uses these elements as

* ‘Learning is anticipated joy about yourself.’

** Thus fission becomes interesting as energy source for heavy nuclei.
*** For the stars above you, see the http://me.in-berlin-de/~jd/himmel/himmel.00.11.html website.

Dvipsbugw
 922 viii inside the nucleus • the strong nuclear interaction

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 353 Photographs of the Sun at wavelengths of 30.4 nm (in the extreme ultraviolet, left) and
around 677 nm (visible light, right, at a different date), by the SOHO mission (ESA and NASA)

catalysts. This happens through the so-called CNO cycle, which runs as
12
C + 1H 13
N+γ
13 13
N C + e+ + ν
13
C + 1H 14
N+γ
14
N + 1H 15
O+γ
15 15
O N + e+ + ν
15
N + 1H 12
C + 4 He (620)

The end result of the cycle is the same as that of the hydrogen cycle, both in nuclei and in
energy. The CNO cycle is faster than hydrogen fusion, but requires higher temperatures,
as the protons must overcome a higher energy barrier before reacting with carbon or
Challenge 1395 n nitrogen than when they react with another proton. (Why?) Due to the comparatively low
temperature of a few tens of million kelvin inside the Sun, the CNO cycle is less important
than the hydrogen cycle. (This is also the case for the other CNO cycles that exist.) These
Copyright © Christoph Schiller November 1997–May 2006

studies also explain why the Sun does not collapse. The Sun is a ball of hot gas, and the
high temperature of its constituents prevents their concentration into a small volume. For
some stars, the radiation pressure of the emitted photons prevents collapse; for others it is
the Pauli pressure; for the Sun, like for the majority of stars, it is the usual thermal motion
of the gas.
The nuclear reaction rates at the interior of a star are extremely sensitive to temperat-
ure. The carbon cycle reaction rate is proportional to between T 13 for hot massive O stars
and T 20 for stars like the Sun. In red giants and supergiants, the triple alpha reaction rate
is proportional to T 40 ; these strong dependencies imply that stars shine with constancy
over medium times, since any change in temperature would be damped by a very effi-
cient feedback mechanism. (Of course, there are exceptions: variable stars get brighter

Dvipsbugw
 the strong nuclear interaction and the birth of matter 923

and darker with periods of a few days; and the Sun shows small oscillations in the minute
range.)
How can the Sun’s surface have a temperature of 6000 K, whereas the corona around
it, the thin gas emanating from the Sun, reaches two million Kelvin? In the latter part of
the twentieth century it was shown, using satellites, that the magnetic field of the Sun is
the cause; through the violent flows in the Sun’s matter, magnetic energy is transferred to
the corona in those places were flux tubes form knots, above the bright spots in the left
of Figure 353 or above the dark spots in the right photograph. As a result, the particles of
the corona are accelerated and heat the whole corona. Dvipsbugw
When the Sun erupts, as shown in the lower left corner in Figure 353, matter is ejected
far into space. When this matter reaches the Earth,* after being diluted by the journey,
it affects the environment. Solar storms can deplete the higher atmosphere and can thus
possibly trigger usual Earth storms. Other effects of the Sun are the formation of auroras
and the loss of orientation of birds during their migration; this happens during exception-
ally strong solar storms, as the magnetic field of the Earth is disturbed in these situations.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The most famous effect of a solar storm was the loss of electricity in large parts of Canada
in March of 1989. The flow of charged solar particles triggered large induced currents in
the power lines, blew fuses and destroyed parts of the network, shutting down the power
system. Millions of Canadians had no electricity, and in the most remote places it took
two weeks to restore the electricity supply. Due to the coldness of the winter and a train
accident resulting from the power loss, over 80 people died. In the meantime the network
has been redesigned to withstand such events.
The proton cycle and the CNO cycles are not the only options. Heavier and older stars
than the Sun can also shine through other fusion reactions. In particular, when hydrogen
is consumed, such stars run helium burning:

3 4 He 12
C. (621)

This fusion reaction is of low probability, since it depends on three particles being at the
same point in space at the same time. In addition, small amounts of carbon disappear
rapidly via the reaction α + 12 C 16 O. Nevertheless, since 8 Be is unstable, the reaction
with 3 alpha particles is the only way for the universe to produce carbon. All these negative
odds are countered only by one feature carbon has an excited state at 7.65 MeV, which is
0.3 MeV above the sum of the alpha particle masses; the excited state resonantly enhances
the low probability of the three particle reaction. Only in this way the universe is able to
Copyright © Christoph Schiller November 1997–May 2006

produce the atoms necessary for pigs, apes and people. The prediction of this resonance
by Fred Hoyle is one of the few predictions in physics that used the simple experimental
observation that humans exist. The story has lead to an huge outflow of metaphysical
Page 689 speculations, most of which are unworthy of being even mentioned.

Why are fusion reactors not common yet?

Across the world, for over 50 years, a large number of physicists and engineers have tried
to build fusion reactors. Fusion reactors try to copy the mechanism of energy release used

* It might even be that the planets affect the solar wind; the issue is not settled and is still under study.

Dvipsbugw
 924 viii inside the nucleus • the strong nuclear interaction

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 354 A simplified drawing of the Joint European Torus
in operation at Culham, showing the large toroidal chamber
and the magnets for the plasma confinement (© EFDA-JET)

by the Sun. The first machine that realized macroscopic energy production was the Joint
European Torus* (JET for short) located in Culham in the United Kingdom.
Ref. 928 The idea of JET is to produce an extremely hot plasma that is as dense as possible.
At high enough temperature and density, fusion takes place; the energy is released as a
particle flux that is transformed (like in a fission reactor) into heat and then into electricity.
To achieve ignition, JET used the fusion between deuterium and tritium, because this
reaction has the largest cross section and energy gain:

D+T He4 + n + 17.6 MeV . (622)

Because tritium is radioactive, most research experiments are performed with the much
less efficient deuterium–deuterium reactions, which have a lower cross section and a
Copyright © Christoph Schiller November 1997–May 2006

lower energy gain:

D+D T + H + 4 MeV
3
D+D He + n + 3.3 MeV . (623)

Fusion takes place when deuterium and tritium (or deuterium) collide at high energy.
The high energy is necessary to overcome the electrostatic repulsion of the nuclei. In
other words, the material has to be hot. To release energy from deuterium and tritium,

* See www.jet.edfa.org.

Dvipsbugw
 the strong nuclear interaction and the birth of matter 925

one therefore first needs energy to heat it up. This is akin to the ignition of wood: in order
to use wood as a fuel, one first has to heat it with a match.
Following the so-called Lawson criterium, published in 1957 by the English engineer
Ref. 929 John Lawson, (but already known to Russian researchers) a fusion reaction releases en-
ergy only if the triple product of density n, reaction (or containment) time τ and temper-
ature T exceeds a certain value. Nowadays this criterium is written as

nτT A 3 ë 1028 s K~m3 . (624)

Dvipsbugw
In order to realize the Lawson criterium, JET uses temperatures of 100 to 200 MK, particle
densities of 2 to 3 ë 1020 m−3 , and confinement times of 1 s. The temperature is much
higher than the 20 MK at the centre of the Sun, because the densities and the confine-
ment times are lower for JET.
Matter at these temperatures is in form of plasma: nuclei and electrons are completely
separated. Obviously, it is impossible to pour a plasma at 100 MK into a container: the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

walls would instantaneously evaporate. The only option is to make the plasma float in
a vacuum, and to avoid that the plasma touches the container wall. The main challenge
of fusion research in the past has been to find a way to keep a hot gas mixture of deu-
terium and tritium suspended in a chamber so that the gas never touches the chamber
walls. The best way is to suspend the gas using a magnetic field. This works because in
the fusion plasma, charges are separated, so that they react to magnetic fields. The most
successful geometric arrangement was invented by the famous Russian physicists Igor
Tamm and Andrei Sakharov: the tokamak. Of the numerous tokamaks around the world,
JET is the largest and most successful. Its concrete realization is shown in Figure 354. JET
manages to keep the plasma from touching the walls for about a second; then the situ-
ation becomes unstable: the plasma touches the wall and is absorbed there. After such
a disruption, the cycle consisting of gas injection, plasma heating and fusion has to be
restarted. As mentioned, JET has already achieved ignition, that is the state were more
energy is released than is added for plasma heating. However, so far, no sustained com-
mercial energy production is planned or possible, because JET has no attached electrical
power generator.
The successor project, ITER, an international tokamak built with European, Japanese,
US-American and Russian funding, aims to pave the way for commercial energy gen-
eration. Its linear reactor size will be twice that of JET; more importantly, ITER plans
to achieve 30 s containment time. ITER will use superconducting magnets, so that it will
Copyright © Christoph Schiller November 1997–May 2006

have extremely cold matter at 4 K only a few metres from extremely hot matter at 100 MK.
In other words, ITER will be a high point of engineering. The facility will be located in
Cadarache in France and is planned to start operation in the year 2016.
Like many large projects, fusion started with a dream: scientists spread the idea that
fusion energy is safe, clean and inexhaustible. These three statements are still found on
every fusion website across the world. In particular, it is stated that fusion reactors are not
dangerous, produce much lower radioactive contamination than fission reactors, and use
water as basic fuel. ‘Solar fusion energy would be as clean, safe and limitless as the Sun.’
In reality, the only reason that we do not feel the radioactivity of the Sun is that we are far
away from it. Fusion reactors, like the Sun, are highly radioactive. The management of
radioactive fusion reactors is much more complex than the management of radioactive

Dvipsbugw
 926 viii inside the nucleus • the strong nuclear interaction

fission reactors.
Fusion fuels are almost inexhaustible: deuterium is extracted from water and the tri-
tium – a short-lived radioactive element not found in nature in large quantities – is pro-
duced from lithium. The lithium must be enriched, but since material is not radioactive,
this is not problematic. However, the production of tritium from lithium is a dirty process
that produces large amounts of radioactivity. Fusion energy is thus inexhaustible, but not
safe and clean.
In short, of all technical projects ever started by mankind, fusion is by far the most
challenging and ambitious. Whether fusion will ever be successful – or whether it ever Dvipsbugw
should be successful – is another issue.

Where do our atoms come from?

People consist of electrons and various nuclei. Electrons, hydrogen and helium nuclei
Page 462 are formed during the big bang. All other nuclei are formed in stars. Young stars run

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

hydrogen burning or helium burning; heavier and older stars run neon-burning or even
silicon-burning. These latter processes require high temperatures and pressures, which
are found only in stars with a mass at least eight times that of the Sun. However, all fusion
processes are limited by photodissociation and will not lead to nuclei heavier than 56 Fe.
Heavier nuclei can only be made by neutron capture. There are two main processes;
the s-process (for ‘slow’) runs inside stars, and gradually builds up heavy elements until
the most heavy, lead, from neutron flying around. The rapid r-process occurs in stellar
explosions. Many stars die this violent death. Such an explosion has two main effects: on
one hand it distributes most of the matter of the star, such carbon, nitrogen or oxygen,
into space in the form of neutral atoms. On the other hand, new elements are synthesized
during the explosion. The abundances of the elements in the solar system can be precisely
measured. These several hundred data points correspond exactly with what is expected
from the material ejected by a (type II) supernova explosion. In other words, the solar
system formed from the remnants of a supernova, as did, somewhat later, life on Earth.*
We all are recycled stardust.

The weak side of the strong interaction

Both radioactivity and medical images show that nuclei are composed. But quantum the-
ory makes an additional prediction: protons and neutrons themselves must be composed.
There are two reasons: nucleons have a finite size and their magnetic moments do not
Copyright © Christoph Schiller November 1997–May 2006

match the value predicted for point particles. The prediction of components inside the
protons was confirmed in the late 1960s when Kendall, Friedman and Taylor shot high
Ref. 930 energy electrons into hydrogen atoms. They found what that a proton contains three con-
stituents with spin 1/2, which they called called partons. The experiment was able to ‘see’
the constituents through large angle scattering of electrons, in the same way that we see
objects through large angle scattering of photons. These constituents correspond in num-
ber and properties to the so-called quarks predicted in 1964 by Murray Gell-Mann** and,
* By chance, the composition ratios between carbon, nitrogen and oxygen inside the Sun are the same as
inside the human body.
** Murray Gell-Mann (b. 1929 New York, d. ) received the Nobel Prize for physics in 1969. He is the origin-
ator of the term ‘quark’. (The term has two origins: officially, it is said to be taken from Finnegans Wake, a

Dvipsbugw
 the strong nuclear interaction and the birth of matter 927

figure to be inserted
Dvipsbugw

F I G U R E 355 A selection of mesons and baryons and their classification as bound states of
quarks

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 931 independently, by George Zweig.
It turns out that the interaction keeping the protons together in a nucleus, which was
first described by Yukawa Hideki,* is only a shadow of the interaction that keeps quarks
together in a proton. Both are called by the same name. The two cases correspond some-
what to the two cases of electromagnetism found in atomic matter. Neon atoms show
the cases most clearly: the strongest aspect of electromagnetism is responsible for the
attraction of the electrons to the neon nuclei and its feebler ‘shadow’ is responsible for
the attraction of neon atoms in liquid neon and for processes like evaporation. Both at-
tractions are electromagnetic, but the strengths differ markedly. Similarly, the strongest
aspect of the strong interaction leads to the formation of the proton and the neutron; the
feeble aspect leads to the formation of nuclei and to alpha decay. Obviously, most can be
learned by studying the strongest aspect.

B ound motion, the particle zoo and the quark model

Physicists are simple people. To understand the constituents of matter, and of nuclei in
particular, they had no better idea than to take all particles they could get hold of and to
smash them into each other. Many played this game for several decades.**
Imagine that you want to study how cars are built just by crashing them into each other.
Before you get a list of all components, you must perform and study a non-negligible num-
ber of crashes. Most give the same result, and if you are looking for a particular part, you
Copyright © Christoph Schiller November 1997–May 2006

novel by James Joyce; in reality, he took it from a German and Yiddish term meaning ‘lean soft cheese’ and
used figuratively in those langauges to mean ‘silly idea’.)
Gell-Mann is the central figure of particle physics; he introduced the concept of strangeness, the renor-
malization group, the V-A interaction, the conserved vector current, the partially conserved axial current,
the eightfold way, the quark model and quantum chromodynamics.
Gell-Mann is also known for his constant battle with Richard Feynman about who deserves to be called
the most arrogant physicist of their university.
* Yukawa Hideki (1907–1981), important Japanese physicist specialized in nuclear and particle physics. He
founded the journal Progress of Theoretical Physics and together with his class mate Tomonaga he was an
example to many scientists in Japan. He received the 1949 Nobel Prize for physics for this theory of mesons.
Challenge 1396 n ** In fact, quantum theory forbids any other method. Can you explain why?

Dvipsbugw
 928 viii inside the nucleus • the strong nuclear interaction

might have to wait for a long time. If the part is tightly attached to others, the crashes have
to be especially energetic. Since quantum theory adds the possibility of transformations,
reactions and excited states, the required diligence and patience is even greater than for
car crashes. Therefore, for many decades, researchers collected an ever increasing num-
ber of debris. The list was overwhelming. Then came the quark model, which explained
the whole mess as a consequence of only a few types of bound constituents.
Other physicists then added a few details and as a result, the whole list of debris could
be ordered in tables such as the ones given in Figure 355 These tables were the beginning
of the end of high energy physics. When the proton scattering experiments found that Dvipsbugw
protons are made of three constituents, the quark model became accepted all over the
world.
The proton and the neutron are seen as combinations of two quarks, called up (u) and
down (d). Later, other particles lead to the addition of four additional types of quarks.
Their names are somewhat confusing: they are called strange (s), charm (c), bottom (b) –
also called ‘beauty’ in the old days – and top (t) – called ‘truth’ in the past.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

All quarks have spin one half; their electric charges are multiples of 1~3 of the electron
charge. In addition, quarks carry a strong charge, which in modern terminology is called
colour. In contrast to electromagnetism, which has only positive, negative, and neutral
charges, the strong interaction has red, blue, green charges on one side, and anti-red,
anti-blue and anti-green on the other. The neutral state is called ‘white’. All baryons and
mesons are white, in the same way that all atoms are neutral.

– CS – details to be added – CS –

The mass, shape and colour of protons

Frank Wilczek mentions that one of the main results of QCD, the theory of strong inter-
Ref. 932 actions, is to explain mass relations such as

m proton  e−k~α m Planck and k = 11~2π , α unif = 1~25 . (625)

Here, the value of the coupling constant α unif is taken at the unifying energy, a factor of
Page 941 1000 below the Planck energy. (See the section of unification below.) In other words, a
general understanding of masses of bound states of the strong interaction, such as the
Copyright © Christoph Schiller November 1997–May 2006

proton, requires almost purely a knowledge of the unification energy and the coupling
constant at that energy. The approximate value α unif = 1~25 is an extrapolation from the
low energy value, using experimental data.
The proportionality factor in expression (625) is still missing. Indeed, it is not easy to
calculate. Many calculations are now done on computers. The most promising calcula-
tion simplify space-time to a lattice and then reduce QCD to lattice QCD. Using the most
powerful computers available, these calculations have given predictions of the mass of
Ref. 933 the proton and other baryons within a few per cent.
But the mass is not the only property of the proton. Being a cloud of quarks and gluons,
it also has a shape. Surprisingly, it took a long time before people started to become inter-
ested in this aspect. The proton is made of two u quarks and one d quark. It thus resembles

Dvipsbugw
 the strong nuclear interaction and the birth of matter 929

Dvipsbugw
figure to be inserted

F I G U R E 356 The spectrum of the excited states of proton and neutron

a ionized H 2+ molecule, where one electron forms a cloud around two protons. Obviously,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the H 2+ molecule is elongated. Is that also the case for the proton? First results from 2003
seem to point into this direction.
The shape of a molecule will depend on whether other molecules surround it. Recent
research showed that both the size and the shape of the proton in nuclei is slightly variable;
Ref. 934 both seem to depend on the nucleus in which the proton is built-in.
Apart from shapes, molecules also have a colour. The colour of a molecule, like that
of any object, is due to the energy absorbed when it is irradiated. For example, the H 2+
molecule can absorb certain light frequencies by changing to an excited state. Protons
Ref. 935 and neutrons can also be excited; in fact, their excited states have been studied in detail; a
summary is shown in Figure 356. It turns out that all these excitations can be explained as
excited quarks states. For several excitations, the masses (or colours) have been calculated
by lattice QCD to within 10%. The quark model and QCD thus structure and explain a large
part of the baryon spectrum.
Obviously, in our everyday environment the energies necessary to excite nucleons do
not appear – in fact, they do not even appear inside the Sun – and these excited states can
be neglected. They only appear in particle accelerators. In a way, we can say that in our
corner of the universe energies are to low to show the colour of protons.

Experimental consequences of the quark model

How can we pretend that quarks exist, even though they are never found alone? There
Copyright © Christoph Schiller November 1997–May 2006

are a number of arguments in favour.

— The quark model explains the non-vanishing magnetic moment of the neutron and
explains the magnetic moments µ of the baryons. By describing the proton as a uud
state and the neutron a udd state with no orbital angular momentum, we get

µ u = 51 (4µ p + µ n )µ 0 µ d = 51 (4µ n + µ p )µ 0 where µ 0 = ħ 2 ~Mc (626)

This means that mu = m d = 330 MeV,a bit more than a third of the nucleon, whose
mass is c. 940 MeV. If we assume that the quark magnetic moment is proportional to
their charge, we predict a ratio of the magnetic moments of the proton and the neut-

Dvipsbugw
 930 viii inside the nucleus • the strong nuclear interaction

q
g g g
g g

g g
g

Dvipsbugw
F I G U R E 357 The essence of the QCD Lagrangian

ron of µ p ~µ n = −1.5; this prediction differs from measurements only by 3 %. Using the
values for the magnetic moment of the quarks, magnetic moment values of over half
a dozen of other baryons can be predicted. The results typically deviate from measure-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ments by around 10 %; the sign is always correctly calculated.
— The quark model describes the quantum numbers of mesons and baryons. All texts
on the quark model are full of diagrams such as those shown in Figure 355. These
generalizations of the periodic table of the elements were filled in during the twentieth
century; they allow a complete classification of all mesons and baryons as bound states
of quarks.
— The quark model also explains the mass spectrum of baryons and mesons. The best
predictions are made by lattice calculations. After one year of computer time, research-
ers were able to reproduce the masses of proton and neutron to within a few per cent.
Even if one sets the u and d quark masses to zero, the resulting proton and neutron
Ref. 945 mass differ from experimental values only by 10 %.

The L agrangian of quantum chromodynamics

All motion due to the strong interaction can be described by the three fundamental pro-
cesses shown in Figure 357. Two gluons can scatter, a gluon can emit another, and a quark
can emit or absorb a gluon. In electrodynamics, only the last diagram is possible, in the
strong interaction, the first two appear as well. Among others, the first two diagrams are
responsible for the confinement of quarks.
Let us have now a look at the official Lagrangian of the strong interaction, and show
that it is just a complicated rewriting of Figure 357. The Lagrangian density of quantum
Copyright © Christoph Schiller November 1997–May 2006

chromodynamics, abbreviated QCD, is

LQ C D = − 41 F µν − c 2 Q m q ψ q ψ qk + i ħc c Q ψ q γ µ (D µ ) k l ψ ql
(a) (a)µν k k
F (627)
q q

where
(a)
F µν = ∂ µ Aaν − ∂ ν Aaµ + дs f abc Abµ Acν
(D µ ) k l = δ k l ∂ µ − i Q λ ak , l Aaµ .
дs
2 a

We remember from the section on the principle of least action that Lagrangians are always

Dvipsbugw
 the strong nuclear interaction and the birth of matter 931

Page 197 sums of scalar products; this is clearly seen in the formula. Furthermore, the index a =
1 . . . 8 numbers the eight types of gluons, k = 1, 2, 3 numbers the three colours and q =
1, . . . 6 numbers the six quark flavours. The fields Aaµ (x) are the eight gluon fields: they
are the coiled lines in Figure 357. The field ψ qk (x) is the field of the quark of flavour q
and colour k: it is the straight line in the figure. The quark fields are 4-component Dirac
spinors.
The first term of the Lagrangian (627) represents the kinetic energy of the radiation
(gluons), the second term the kinetic energy of the matter particles (the quarks) and the Dvipsbugw
third term the interaction between the two. The third term in the Lagrangian, the inter-
action term, thus corresponds to the third diagram in Figure 357.
Gluons are massless; therefore no gluon mass term appears in the Lagrangian. In op-
position to electromagnetism, where the gauge group U(1) is abelian, the gauge group
SU(3) of the strong interactions is non-abelian. As a consequence, the colour field itself
is charged, i.e. carries colour: the index a appears on A and F. As a result, gluons can
interact with each other, in contrast to photons, which pass each other undisturbed. The

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

first two diagrams of Figure 357 are thus reflected in the somewhat complicated defini-
(a)
tion of the function F µν . In contrast to electrodynamics, the definition has an extra term
that is quadratic in the fields A; it is described by the interaction strength дs and by the
so-called structure constants f abc . These are the structure constants of the SU(3) algebra.
The interaction between the quarks and the gluons, the third term of the Lagrangian,
is described by the matrices λ ak , l ; they are a fundamental, 3-dimensional representation
of the generators of the SU(3) algebra.*

* In their simplest form, the matrices γ µ can be written as

σi
γ0 = Œ ‘ γn = Œ ‘
I 0 0
and for n = 1, 2, 3 (628)
0 −I −σ i 0

Page 1201 where the σ i are the Pauli spin matrices.

The matrices λ a , a = 1..8, and the structure constants f abc obey the relations

[λ a , λ b ] = 2i f abc λ c
˜λ a , λ b  = 4~3δ ab I + 2d abc λ c (629)

where I is the unit matrix. The structure constants f abc , which are odd under permutation of any pair of
indices, and d abc , which are even, are
Copyright © Christoph Schiller November 1997–May 2006

abc f abc abc d abc abc d abc

º
123 1 118 1~ 3 355 1~2
147 1~2 146 1~2 366 −1~2
156 −1~2 157 1~2º 377 −1~2 º
246 1~2 228 1~ 3 448 −1~(2º3 ) (630)
257 1~2 247 −1~2 558 −1~(2º3 )
345 1~2 256 1~2º 668 −1~(2º3 )
367 º−1~2 338 1~ 3 778 º 3
−1~(2 )
458 º3 ~2 344 1~2 888 −1~ 3
678 3 ~2

All other elements vanish. A fundamental 3-dimensional representation of the generators λ a is given for

Dvipsbugw
932 viii inside the nucleus • the strong nuclear interaction

We see that only quarks and gluons appear in the Lagrangian of QCD, because only
quarks and gluons interact via the strong force. This can be also expressed by saying that
only quarks and gluons carry colour; colour is the source of the strong force in the same
way that electric charge is the source of the electromagnetic field. In the same way as
electric charge, colour charge is conserved in all interactions. Electric charge comes in
two types, positive and negative; in contrast, colour comes in three types, called red, green
and blue. The neutral state, with no colour charge, is called white. Protons and neutrons,
but also electrons or neutrinos, are thus white in this sense.
Like in all of quantum field theory, also in the case of QCD the mathematical form Dvipsbugw
of the Lagrangian is uniquely defined by requiring renormalizability, Lorentz invariance
and gauge invariance (SU(3) in this case) and by specifying the different particles types
(6 quarks in this case) with their masses and the coupling constant.
The Lagrangian is thus almost fixed by construction. We say ‘almost’, because it con-
tains a few parameters that remain unexplained:

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

— The number and the masses of the quarks are not explained by QCD.
— The coupling constant дs of the strong interaction is unexplained. Often also the equi-
valent quantity α s = дs2 ~4π is used to describe the coupling. Like for the case of the
electroweak interactions, α s and thus дs depend on the energy Q of the experiment.
This energy dependence is indeed observed in experiments; it is described by the renor-
malization procedure:
α s (Q 2 ) =
12π 1
+ ... (632)
33 − 2n f ln Q 22
Λ

where n f is the number of quarks with mass less than the energy scale Q and lies
between 3 and 6. The strong coupling is thus completely described by the energy para-
meter Λ  0.25 GeV~c2 . If α s is known for one energy, it is known for all of them.
Presently, the experimental value is α s (Q 2 = 34 GeV) = 0.14  0.02. Expression (632)
also illustrates asymptotic freedom: α s vanishes for high energies. In other words, at
high energies quarks are freed from the strong interaction.*
At low energies, the coupling increases, and leads to quark confinement.** This be-

example by the set of the Gell-Mann matrices

’ 0 1 0 “ ’ 0 −i 0 “ ’ 1 0 0 “
λ1 = – 1 0 0 — λ2 = – i 0 0 — λ3 = – 0 −1 0 —
” 0 0 • ” 0 0 • ” 0 0 •
Copyright © Christoph Schiller November 1997–May 2006

0 0 0
’ 0 0 1 “ ’ 0 0 −i “ ’ 0 0 0 “
λ4 = – 0 0 0 — λ5 = – 0 0 0 — λ6 = – 0 0 1 —
” 1 0 0 • ” i 0 0 • ” 0 1 0 •
’ 0 0 0 “
1 ’
1 0 0 “
λ7 = – 0 0 −i — λ 8 = º – 0 1 0 — . (631)
” 0 i 0 • 3 ” 0 0 −2 •

There are eight matrices, one for each gluon type, with 3  3 elements, corresponding to the three colours of
the strong interactions.
* Asymptotic freedom was first discovered by Gerard ‘t Hooft; since he had the Nobel Prize already, it the
2004 Prize was then given to the next people who found it: David Gross, David Politzer and Frank Wilczek.
** Only at energies much larger than Λ can a perturbation expansion be applied.

Dvipsbugw
 the strong nuclear interaction and the birth of matter 933

haviour is in contrast to the electroweak interactions, where the coupling increases

with energy. We thus find that one parameter describing the strong coupling, Λ, re-
mains unexplained and must be introduced into the Lagrangian from the beginning.
— The properties of space-time, its Lorentz invariance, its continuity and the number of
its dimensions are obviously all unexplained and assumed from the outset.

The sizes and masses of quarks

The size of quarks, like that of all elementary particles, is predicted to be zero by quantum Dvipsbugw
field theory. So far, no experiment has found an effect due to a finite quark size. Meas-
Ref. 946 urements show that quarks are surely smaller than 10−19 m. No definite size predictions
have been made; quarks might have a size of the order of the grand unification scale,
Page 941 i.e. 10−32 m; however, so far this is pure speculation.
We noted in several places that a compound is always less massive than its components.
But when the mass values for quarks are looked up in most tables, the masses of u and d
quarks are only of the order of a few MeV~c 2 , whereas the proton’s mass is 938 MeV~c2 .

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What is the story here?
The definition of the masses for quarks is more involved than for other particles.
Quarks are never found as free particles, but only in bound states. Quarks behave almost
like free particles at high energies; this property is called asymptotic freedom. The mass of
such a free quark is called current quark mass; for the light quarks it is only a few MeV~c2 .
At low energy, for example inside a proton, quarks are not free, but must carry along a
large amount of energy due to the confinement process. As a result, bound quarks have a
much larger constituent quark mass, which takes into account this confinement energy. To
give an idea of the values, take a proton; the indeterminacy relation for a particle inside a
sphere of radius 0.9 fm gives a momentumº indeterminacy of around 190 MeV~c. In three
dimensions this gives an energy of 3 times that value, or a mass of about 330 MeV~c2 .
Three confined quarks are thus heavier than a proton, whose mass is 938 MeV~c2 ; we can
thus still say that a compound proton is less massive than its constituents. In short, the
mass of the proton and the neutron is (almost exclusively) the kinetic energy of the quarks
Ref. 947 inside them, as their own rest mass is negligible. As Frank Wilczek says, some people put
on weight even though they never eat anything heavy.
To complicate the picture, the distinction of the two mass types makes no sense for
the top quark; this quark decays so rapidly that the confinement process has no time to
set in. As a result, the top mass is again a mass of the type we are used to.
Copyright © Christoph Schiller November 1997–May 2006

Confinement and the future of the strong interaction

The description of the proton mass using confined quarks should not hide the fact that
the complete explanation of quark confinement, the lack of single quarks in nature, is the
biggest challenge of theoretical high energy physics.
The Lagrangian of QCD differs from that of electromagnetism in a central aspect. So
far, bound states cannot be deduced with a simple approximation method. In particu-
lar, the force dependence between two coloured particles, which does not decrease with
increasing distance, but levels off at a constant value, does not follow directly from the
Lagrangian. The constant value, which then leads to confinement, has been reproduced
only in involved computer calculations.

Dvipsbugw
 934 viii inside the nucleus • the strong nuclear interaction

In fact, the challenge is so tough that the brightest minds have been unable to solve it,
so far. In a sense, it can be seen as the biggest challenge of all of physics, as its solution
probably requires the unification of all interactions and most probably the unification
with gravity. We have to leave this issue for later in our adventure.

Curiosities about the strong interactions

The computer calculations necessary to extract particle data from the Lagrangian of
quantum chromodynamics are among the most complex calculations ever performed. Dvipsbugw
They beat weather forecasts, fluid simulations and the like by orders of magnitude.
Nobody knows whether this is necessary: the race for a simple approximation method
for finding solutions is still open.
**
Even though gluons are massless, like photons and gravitons, there is no colour radiation.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Gluons carry colour and couple to themselves; as a result, free gluons were predicted
to directly decay into quark–antiquark pairs. This decay has indeed been observed in
experiments at particle accelerators.
Something similar to colour radiation, but still stranger might have been found
in 1997. First results seem to confirm the prediction of glueballs from numerical
Ref. 948 calculations.
**
The latest fashion in high energy physics is the search for hybrid mesons, particles made
Ref. 948 of gluons and quarks. This fashion is not over yet; the coming years should settle whether
the candidates known so far really are hybrids.
**
Do particles made of five quarks, so-called pentaquarks, exist? So far, they seem to exist
only in a few laboratories in Japan, whereas other laboratories across the world fail to see
them. The issue is still open.
**
Whenever we look at a periodic table of the elements, we look at a manifestation of the
strong interaction. The Lagrangian of the strong interaction describes the origin and prop-
Copyright © Christoph Schiller November 1997–May 2006

erties of the presently known 115 elements.

Nevertheless a central aspect of nuclei is determined together with the electromag-
netic interaction. Why are there around one hundred different elements? Because the
electromagnetic coupling constant α is 1~137.036(1). Indeed, if the charge of a nucleus
was much higher than around 130, the electric field around nuclei would lead to spon-
taneous electron–positron pair generation; the electron would fall into the nucleus and
transform one proton into a neutron, thus inhibiting a larger proton number.
**
To know more about radioactivity, its effects, its dangers and what a government can do
about it, see the English and German language site of the Federal Office for Radiation

Dvipsbugw
 the weak nuclear interaction and the handedness of nature 935

Protection at http://www.bfs.de.

**
From the years 1990 onwards, it has often been claimed that extremely poor countries
Challenge 1397 n are building nuclear weapons. Why is this not possible?
**
In the 1960s and 70s, it was discovered that the Sun pulsates with a frequency of 5 minutes.
The effect is small, only 3 kilometres out of 1.4 million; still it is measurable. In the mean- Dvipsbugw
time, helioseismologists have discovered numerous additional oscillations of the Sun, and
in 1993, even on other stars. Such oscillations allow to study what is happening inside
stars, even separately in each of the layers they consist of.
**
Historically, nuclear reactions also provided the first test of the relation E = γmc 2 . This

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

was achieved in 1932 by Cockcroft and Walton. They showed that by shooting protons
into lithium one gets the reaction
7 1 8 4 4
3 Li + 1 H 4 Be 2 He + 2 He + 17 MeV . (633)

The measured energy on the right is exactly the value that is derived from the differences
in total mass of the nuclei on both sides.

**
Some stars shine like a police siren: their luminosity increases and decreases regularly.
Such stars, called Cepheids, are important because their period depends on their average
(absolute) brightness. Measuring their period and their brightness on Earth thus allows
astronomers to determine their distance.

30. the weak nuclear interaction and the

handedness of nature
No interaction is as weird as the weak interaction. First of all, the corresponding ‘weak
Copyright © Christoph Schiller November 1997–May 2006

radiation’ consists of massive particles; there are two types, the neutral Z boson with a
mass of 91.2 GeV – that is the mass of a silver atom – and the electrically charged W
boson with a mass of 80.8 GeV. The masses are so large that free radiation exists only
for an extremely short time, about 0.1 ys; then the particles decay. The large mass is the
reason that the interaction is extremely short range and weak; any exchange of virtual
particles scales with the negative exponential of the intermediate particle’s mass.
The existence of a massive intermediate vector boson was already deduced in the 1940s;
but theoretical physicists did not accept the idea until the Dutch physicist Gerard ’t Hooft
proved that it was possible to have such a mass without having problems in the rest of
the theory. For this proof he later received the Nobel price of physics. Experimentally,

Dvipsbugw
 936 viii inside the nucleus • the weak nuclear interaction

the Z boson was found found as a virtual particle in 1973 and as a real particle in 1983,
both times at CERN in Geneva. The last experiment was a year-long effort by thousands
of people working together.
A central effect of the weak interaction is its ability to transform quarks. It is this prop-
erty that is responsible for beta decay, where a d quark in a neutron is changed into a u
quark, or for a crucial step in the Sun, where the opposite happens.
The next weird characteristic of the weak interaction is the nonconservation of parity
under spatial inversion. The weak interaction distinguishes between mirror systems, in
contrast to everyday life, gravitation, electromagnetism, and the strong interactions. Par- Dvipsbugw
ity non-conservation had been predicted by 1956 by Lee Tsung-Dao and Yang Chen Ning,
and was confirmed a few months later, earning them a Nobel Prize. (They had predicted
the effect in order to explain the ability of K mesons to decay either into 2 or into 3 pions.)
The most beautiful consequence of parity non-conservation property is its influence on
Ref. 952 the colour of certain atoms. This prediction was made in 1974 by Bouchiat and Bouchiat.
The weak interaction is triggered by the weak charge of electrons and nuclei. Therefore,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

electrons in atoms do not exchange only virtual photons with the nucleus, but also virtual
Z particles. The chance for this latter process is extremely small, around 10−11 times smal-
ler than virtual photon exchange. But since the weak interaction is not parity conserving,
this process allows electron transitions which are impossible by purely electromagnetic ef-
fects. In 1984, measurements confirmed that certain optical transitions of caesium atoms
that are impossible via the electromagnetic interaction, are allowed when the weak inter-
Ref. 953 action is taken into account. Several groups have improved these results and have been
able to confirm the prediction of the weak interaction, including the charge of the nucleus,
Ref. 958 to within a few per cent.

– CS – The section on weak interactions will be inserted here – CS –

Curiosities about the weak interactions

The weak interaction is responsible for the burning of hydrogen to helium. Without he-
lium, there would be no path to make still heavier elements. Thus we owe our own exist-
ence to the weak interaction.
**
Copyright © Christoph Schiller November 1997–May 2006

The weak interaction is required to have an excess of matter over antimatter. Without the
parity breaking of the weak interactions, there would be no matter at all in the universe.

**
Through the emitted neutrinos, the weak interaction helps to get the energy out of a
supernova. If that were not the case, black holes would form, heavier elements – of which
we are made – would not have been spread out into space, and we would not exist.
**
Ref. 951 The paper by Peter Higgs on the boson named after him is only 79 lines long, and has
only five equations.

Dvipsbugw
the weak nuclear interaction and the handedness of nature 937

**
The weak interaction is not parity invariant. In other words, when two electrons collide,
the fraction of the collisions that happens through the weak interaction should behave
differently than a mirror experiment. In 2004, polarized beams of electrons – either left-
handed or right-handed – were shot at a matter target and the reflected electrons were
counted. The difference was 175 parts per billion – small, but measurable. The experiment
also confirmed the predicted weak charge of -0.046 of the electron.
** Dvipsbugw
The weak interaction is also responsible for the heat produced inside the Earth. This heat
keeps the magma liquid. As a result, the weak interaction, despite its weakness, is respons-
ible for all earthquakes, tsunamis and volcanic eruptions.
**
Beta decay, due to the weak interaction, separates electrons and protons. Only in 2005

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

people have managed to propose practical ways to use this effect to build long-life batter-
ies that could be used in satellites. Future will tell whether the method will be successful.

Mass, the Higgs boson and a ten thousand million dollar lie
Difficile est satiram non scribere.

“
In the years 1993 and 1994 an intense marketing campaign was carried out across the
Juvenal*

”
United States of America by numerous particle physicists. They sought funding for the
‘superconducting supercollider’, a particle accelerator with a circumference of 80 km. This
should have been the largest human machine ever built, with a planned cost of more than
ten thousand million dollars, aiming at finding the Higgs boson before the Europeans
would do so, at a fraction of the cost. The central argument brought forward was the
following: since the Higgs boson was the basis of mass, it was central to science to know
about it. Apart from the discussion on the relevance of the argument, the worst is that it
is wrong.
We have even seen that 95 % of the mass of protons, and thus of the universe, is due
to confinement; it appears even if the quarks are approximated as massless. The Higgs
boson is not responsible for the origin of mass itself; it just might shed some light on the
Copyright © Christoph Schiller November 1997–May 2006

issue. The whole campaign was a classic case of disinformation and many people involved
have shown their lack of honesty. In the end, the project was stopped, mainly for financial
reasons. But the disinformation campaign had deep consequences. US physicist lost their
credibility. Even in Europe the budget cuts became so severe that the competing project
in Geneva, though over ten times cheaper and financed by thirty countries instead of
only one, was almost stopped as well. (Despite this hick-up, the project is now under way,
scheduled for completion in 2007/8.)

* ‘It is hard not to be satirical.’ 1, 30

Dvipsbugw
 938 viii inside the nucleus • the weak nuclear interaction

Neutrinium and other curiosities on the electroweak

interaction
The weak interaction, with its breaking of parity and the elusive neutrino, exerts a deep
Ref. 949 fascination on all those who have explored it.
**
16
Ref. 950 Every second around 10 neutrinos fly through our body. They have five sources:
— Solar neutrinos arrive on Earth at 6 ë 1014 ~m2 s, with an energy from 0 to 0.42 MeV; Dvipsbugw
they are due to the p-p reaction in the sun; a tiny is due to the 8 B reaction and has
energies up to 15 MeV.
— Atmospheric neutrinos are products of cosmic rays hitting the atmosphere, consist of
2/3 of muon neutrinos and one third of electron neutrinos, and have energies mainly
between 100 MeV and 5 GeV.
Page 910 — Earth neutrinos from the radioactivity that keeps the Earth warm form a flux of
6 ë 1010 ~m2 s.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

— Fossil neutrinos from the big bang, with a temperature of 1.95 K are found in the uni-
verse with a density of 300 cm−3 , corresponding to a flux of 1015 ~m2 s.
— Man-made neutrinos are produced in nuclear reactors (at 4 MeV) and in as neutrino
beams in accelerators, using pion and kaon decay. A standard nuclear plant produces
5ë1020 neutrinos per second. Neutrino beams are produced, for example, at the CERN
in Geneva. They are routinely sent 700 km across the Earth to central Italy, where they
are detected.
They are mainly created in the atmosphere by cosmic radiation, but also coming
directly from the background radiation and from the centre of the Sun. Nevertheless,
during our whole life – around 3 thousand million seconds – we have only a 10 %
chance that one of them interacts with one of the 3ë1027 atoms of our body. The reason
is that the weak interaction is felt only over distances less than 10−17 m, about 1/100th
of the diameter of a proton. The weak interaction is indeed weak.
**
The weak interaction is so weak that a neutrino–antineutrino annihilation – which is only
possible by producing a massive intermediate Z boson – has never been observed up to
this day. Copyright © Christoph Schiller November 1997–May 2006

**
Only one type of particles interacts (almost) only weakly: neutrinos. Neutrinos carry no
electric charge, no colour charge and almost no gravitational charge (mass). To get an
impression of the weakness of the weak interaction, it is usually said that the probability
of a neutrino to be absorbed by a lead screen of the thickness of one light-year is less than
50 %. The universe is thus essentially empty for neutrinos. Is there room for bound states
of neutrinos circling masses? How large would such a bound state be? Can we imagine
bound states, which would be called neutrinium, of neutrinos and antineutrinos circling
each other? The answer depends on the mass of the neutrino. Bound states of massless

Dvipsbugw
 the weak nuclear interaction and the handedness of nature 939

particles do not exist. They could and would decay into two free massless particles.*
Since neutrinos are massive, a neutrino–antineutrino bound state is possible in prin-
Challenge 1398 ny ciple. How large would it be? Does it have excited states? Can they ever be detected? These
issues are still open.
**
Do ruminating cows move their jaws equally often in clockwise and anticlockwise dir-
ection? In 1927, the theoretical physicists Pascual Jordan and Ralph de Laer Kronig pub-
Ref. 954 lished a study showing that in Denmark the two directions are almost equally distributed. Dvipsbugw
The rumination direction of cows is thus not related to the weak interaction.
**
The weak interaction plays an important part in daily life. First of all, the Sun is shining.
The fusion of two protons to deuterium, the first reaction of the hydrogen cycle, implies
Page 916 that one proton changes into a neutron. This transmutation and the normal beta decay

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1399 e have the same first-order Feynman diagram. The weak interaction is thus essential for
the burning of the Sun. The weakness of the process is one of the guarantees that the Sun
will continue burning for quite some time.
**
Of course, the weak interaction is responsible for radioactive beta decay, and thus for part
of the radiation background that leads to mutations and thus to biological evolution.

**
What would happen if the Sun suddenly stopped shining? Obviously, temperatures would
fall by several tens of degrees within a few hours. It would rain, and then all water would
freeze. After four or five days, all animal life would stop. After a few weeks, the oceans
would freeze; after a few months, air would liquefy.
**
Not everything about the Sun is known. For example, the neutrino flux from the Sun
oscillates with a period of 28.4 days. That is the same period with which the magnetic
field of the Sun oscillates. The connections are still being studied.
**
Copyright © Christoph Schiller November 1997–May 2006

The energy carried away by neutrinos is important in supernovas; if neutrinos would not
carry it away, supernovas would collapse instead of explode. That would have prevented
the distribution of heavier elements into space, and thus our own existence.
**
Even earlier on in the history of the universe, the weak interaction is important, as it pre-
vents the symmetry between matter and antimatter, which is required to have an excess
of one over the other in the universe.

* In particular, this is valid for photons bound by gravitation; this state is not possible.

Dvipsbugw
 940 viii inside the nucleus • the standard model

**
Due to the large toll it placed on society, research in nuclear physics, like poliomyelitis, has
almost disappeared from the planet. Like poliomyelitis, nuclear research is kept alive only
in a few highly guarded laboratories around the world, mostly by questionable figures, in
order to build dangerous weapons. Only a small number of experiments carried on by a
few researchers are able to avoid this involvement and continue to advance the topic.
**
Interesting aspects of nuclear physics appear when powerful lasers are used. In 1999, a Dvipsbugw
British team led by Ken Ledingham observed laser induced uranium fission in 238 U nuc-
lei. In the meantime, this has even be achieved with table-top lasers. The latest feat, in
2003, was the transmutation of 129 I to 128 I with a laser. This was achieved by focussing
Ref. 955 a 360 J laser pulse onto a gold foil; the ensuing plasma accelerates electrons to relativ-
istic speed, which hit the gold and produce high energy γ rays that can be used for the
transmutation.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

31. the standard model of elementary particle
physics – as seen on television

– CS – the section will appear soon – CS –

Conclusion and open questions about the standard model

The standard model clearly distinguishes elementary from composed particles. It
provides the full list of properties that characterizes a particle and thus any object
in nature: charge, spin, isospin, parity, charge parity, strangeness, charm, topness, beauty,
lepton number, baryon number and mass. The standard model also describes interac-
tions as exchange of virtual radiation particles. It describes the types of radiation that are
found in nature at experimentally accessible energy. In short, the standard model realizes
the dream of the ancient Greeks, plus a bit more: we have the bricks that compose all of
matter and radiation, and in addition we know precisely how they move and interact.
Copyright © Christoph Schiller November 1997–May 2006

But we also know what we still do not know:

— we do not know the origin of the coupling constants;
— we do not know the origin of the symmetry groups;
— we do not know the details of confinement;
— we do not know whether the particle concept survives at high energy;
— we do not know what happens in curved space-time.
To study these issues, the simplest way is to explore nature at particle energies that are
as high as possible. There are two methods: building large experiments or making some
calculations. Both are important.

Dvipsbugw
 grand unification – a simple dream 941

32. grand unification – a simple dream

Materie ist geronnenes Licht.*

“ Albertus Magnus

Is there a common origin of the three particle interactions? We have seen in the preced- ”
ing sections that the Lagrangians of the electromagnetic, the weak and the strong nuclear
interactions are determined almost uniquely by two types of requirements: to possess a
certain symmetry and to possess mathematical consistency. The search for unification of Dvipsbugw
the interactions thus requires the identification of th unified symmetry of nature. In re-
cent decades, several candidate symmetries have fuelled the hope to achieve this program:
grand unification, supersymmetry, conformal invariance and coupling constant duality.
The first of them is conceptually the simplest.
At energies below 1000 GeV there are no contradictions between the Lagrangian of
the standard model and observation. The Lagrangian looks like a low energy approxima-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

tion. It should thus be possible (attention, this a belief) to find a unifying symmetry that
contains the symmetries of the electroweak and strong interactions as subgroups and
thus as different aspects of a single, unified interaction; we can then examine the physical
properties that follow and compare them with observation. This approach, called grand
unification, attempts the unified description of all types of matter. All known elementary
particles are seen as fields which appear in a Lagrangian determined by a single symmetry
group.
Like for each gauge theory described so far, also the grand unified Lagrangian is mainly
determined by the symmetry group, the representation assignments for each particle, and
the corresponding coupling constant. A general search for the symmetry group starts
Ref. 956 with all those (semisimple) Lie groups which contain U(1)SU(2)SU(3). The smallest
groups with these properties are SU(5), SO(10) and E(8); they are defined in Appendix D.
For each of these candidate groups, the experimental consequences of the model must be
studied and compared with experiment.

Experimental consequences
Grand unification makes several clear experimental predictions.
Any grand unified model predicts relations between the quantum numbers of all ele-
mentary particles – quarks and leptons. As a result, grand unification explains why the
electron charge is exactly the opposite of the proton charge.
Copyright © Christoph Schiller November 1997–May 2006

Grand unification predicts a value for the weak mixing angle θ W that is not determined
by the standard model. The predicted value,

sin2 θ W,th = 0.2 (634)

is close to the measured value of

sin2 θ W,ex = 0.231(1) . (635)

* ‘Matter is coagulated light.’ Albertus Magnus (b. c. 1192 Lauingen, d. 1280 Cologne), the most important
thinker of his time.

Dvipsbugw
 942 viii inside the nucleus • 32. grand unification – a simple dream

All grand unified models predict the existence of magnetic monopoles, as was shown
Ref. 959 by Gerard ’t Hooft. However, despite extensive searches, no such particles have been
found yet. Monopoles are important even if there is only one of them in the whole uni-
verse: the existence of a single monopole implies that electric charge is quantized. Grand
unification thus explains why electric charge appears in multiples of a smallest unit.
Grand unification predicts the existence of heavy intermediate vector bosons, called
X bosons. Interactions involving these bosons do not conserve baryon or lepton number,
but only the difference B − L between baryon and lepton number. To be consistent with
experiment, the X bosons must have a mass of the order of 1016 GeV. Dvipsbugw
Most spectacularly, the X bosons grand unification implies that the proton decays.
This prediction was first made by Pati and Salam in 1974. If protons decay, means that
neither coal nor diamond* – nor any other material – is for ever. Depending on the pre-
cise symmetry group, grand unification predicts that protons decay into pions, electrons,
kaons or other particles. Obviously, we know ‘in our bones’ that the proton lifetime is
rather high, otherwise we would die of leukaemia; in other words, the low level of cancer

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

already implies that the lifetime of the proton is larger than 1016 a.
Detailed calculations for the proton lifetime τ p using SU(5) yield the expression

M X4
τp   10311 a
1
αG2 (M X ) M p5
(636)

where the uncertainty is due to the uncertainty of the mass M X of the gauge bosons in-
volved and to the exact decay mechanism. Several large experiments aim to measure this
lifetime. So far, the result is simple but clear. Not a single proton decay has ever been
Ref. 960 observed. The experiments can be summed up by

τ(p e + π 0 ) A 5 ë 1033 a
τ(p K + ν̄) A 1.6 ë 1033 a
τ(n e + π− ) A 5 ë 1033 a
τ(n K 0 ν̄) A 1.7 ë 1032 a (637)

These values are higher than the prediction by SU(5). To settle the issue definitively, one
last prediction of grand unification remains to be checked: the unification of the coupling
constants.
Copyright © Christoph Schiller November 1997–May 2006

The state of grand unification

The estimates of the grand unification energy are near the Planck energy, the energy at
which gravitation starts to play a role even between elementary particles. As grand unific-
ation does not take gravity into account, for a long time there was a doubt whether some-
thing was lacking in the approach. This doubt changed into certainty when the precision
Ref. 961 measurements of the coupling constants became available. This happened in 1991, when

* As is well known, diamond is not stable, but metastable; thus diamonds are not for ever, but coal might be,
if protons do not decay.

Dvipsbugw
grand unification – a simple dream 943

Dvipsbugw

F I G U R E 358 The behaviour of the three coupling constants with

energy for the standard model (left) and for the minimal
supersymmetric model (right) (© Dmitri Kazakov)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

these measurements were shown as Figure 358. It turned out that the SU(5) prediction of
the way the constants evolve with energy imply that the three constants do not meet at
the grand unification energy. Simple grand unification by SU(5) is thus definitively ruled
out.
This state of affairs is changed if supersymmetry is taken into account. Supersymmetry
is the low-energy effect of gravitation in the particle world. Supersymmetry predicts new
particles that change the curves at intermediate energies, so that they all meet at a grand
unification energy of about 1016 GeV. The inclusion of supersymmetry also puts the pro-
ton lifetime prediction back to a value higher (but not by much) than the present exper-
imental bound and predicts the correct value of the mixing angle. With supersymmetry,
we can thus retain all advantages of grand unification (charge quantization, fewer para-
meters) without being in contradiction with experiments. The predicted particles, not
yet found, are in a region accessible to the LHC collider presently being built at CERN in
Geneva. We will explore supersymmetry later on.
Eventually, some decay and particle data will become available. Even though these
experimental results will require time and effort, a little bit of thinking shows that they
probably will be only partially useful. Grand unification started out with the idea to unify
the description of matter. But this ambitious goal cannot been achieved in this way. Grand
Copyright © Christoph Schiller November 1997–May 2006

unification does eliminate a certain number of parameters from the Lagrangians of QCD
and QFD; on the other hand, some parameters remain, even if supersymmetry is added.
Most of all, the symmetry group must be put in from the beginning, as grand unification
cannot deduce it from a general principle.
If we look at the open points of the standard model, grand unification reduces their
number. However, grand unification only shifts the open questions of high energy physics
to the next level, while keeping them unanswered. Grand unification remains a low en-
ergy effective theory. Grand unification does not tell us what elementary particles are; the
name ‘grand unification’ is ridiculous. In fact, the story of grand unification is a first hint
that looking at higher energies using only low-energy concepts is not the way to solve the
mystery of motion. We definitively need to continue our adventure.

Dvipsbugw
 Dvipsbug

Dvipsbug
Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net Copyright © Christoph Schiller November 1997–May 2006

viii inside the nucleus • 32. grand unification – a simple dream

944
bibliography 945

Biblio graphy

923 A excellent technical introduction to nuclear physics is B ogdan Povh, Kl aus R ith,
Christoph Scholz & Frank Zetsche, Teilchen und Kerne, Springer, 5th edition,
1999. It is also available in English translation.
One of the best introductions into particle physics is Kurt Gottfried & Victor F.
Weisskopf, Concepts of Particle Physics, Clarendon Press, Oxford, 1984. Victor Weisskopf
was one of the heroes of the field, both in theoretical research and in the management of
CERN, the particle research institution. Cited on page 897.
Dvipsbugw
924 W.C.M. Weijmar Schultz & al., Magnetic resonance imaging of male and female
genitals during coitus and female sexual arousal, British Medical Journal 319, pp. 1596–1600,
December 18, 1999, available online as http://www.bmj.com/cgi/content/full/319/7225/1596.
Cited on page 898.
925 A good overview is given by A.N. Halliday, Radioactivity, the discovery of time and the
earliest history of the Earth, Contemporary Physics 38, pp. 103–114, 1997. Cited on page 910.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

926 An excellent summary on radiometric dating is by R. Wiens, Radiometric dating – a chris-
tian perpective, http://www.asa3.org/ASA/resources/Wiens.html. The absurd title is due to
the habit in many religious circles to put into question radiometric dating results. Apart
from the extremely few religious statements in the review, the content is well explained.
Cited on pages 910 and 919.
927 The slowness of the speed of light inside stars is due to the frequent scattering of photons
by the star matter. The most common estimate for the Sun is an escape time of 40 000 to 1
million years, but estimates between 17 000 years and 50 million years can be found in the
literature. Cited on page 921.
928 See the freely downloadable book by John Wesson, The Science of JET - The Achievements
of the Scientists and Engineers Who Worked on the Joint European Torus 1973-1999, JET
Joint Undertaking, 2000, available at http://www.jet.edfa.org/documents/wesson/wesson.
html. Cited on page 924.
929 J.D. L awson, Some criteria for a power producing thermonuclear reactor, Proceedings of
the Physical Society, London B 70, pp. 6–10, 1957. The paper had been kept secret for two
years. Cited on page 925.
930 Kendall, Friedman and Taylor received the 1990 Nobel Prize for Physics for a series of ex-
periments they conducted in the years 1967 to 1973. The story is told in the three Nobel
lectures R.E. Taylor, Deep inelastic scattering: the early years, Review of Modern Phys-
ics 63, pp. 573–596, 1991, H.W. Kendall, Deep inelastic scattering: Experiments on the
Copyright © Christoph Schiller November 1997–May 2006

proton and the observation of scaling, Review of Modern Physics 63, pp. 597–614, 1991, and
J.I. Friedman, Deep inelastic scattering: Comparisons with the quark model, Review of
Modern Physics 63, pp. 615–620, 1991. Cited on page 926.
931 G. Zweig, An SU3 model for strong interaction symmetry and its breaking II, CERN Re-
port No. 8419TH. 412, February 21, 1964. Cited on page 927.
932 F. Wilczek, Getting its from bits, Nature 397, pp. 303–306, 1999. Cited on page 928.
933 For an overview of lattice QCD calculations, see ... Cited on page 928.
934 S. Strauch & al., Polarization transfer in the 4 He (e,e’p) 3 H reaction up to Q 2 = 2.6(GeV/
c)2 , Physical Review Letters 91, p. 052301, 2003. Cited on page 929.
935 The excited states of the proton and the neutron can be found in on the particle data group
website at http://pdg.web.cern.ch. Cited on page 929.

Dvipsbugw
946 viii inside the nucleus

936 An older but fascinating summary of solar physics is R. Kippenhahn, Hundert Milliarden
Sonnen, Piper, 1980. which is available also in English translation. No citations.
937 M. Brunetti, S. Cecchini, M. Galli, G. Giovannini & A. Pagliarin, Gamma-
ray bursts of atmospheric origin in the MeV energy range, Geophysical Research Letters 27,
p. 1599, 1 June 2000. Cited on page 917.
938 A book with nuclear explosion photographs is Michael Light, 100 Suns, Jonathan Cape,
2003. Cited on page 919.
939 J. Dudek, A. God, N. Schunck & M. Mikiewicz, Nuclear tetrahedral symmetry:
Dvipsbugw
possibly present throughout the periodic table, Physical Review Letters 88, p. 252502, 24
June 2002. Cited on page 912.
940 A good introduction is R. Cl ark & B. Wodsworth, A new spin on nuclei, Physics World
pp. 25–28, July 1998. Cited on page 912.
941 M. Nauenberg & V.F. Weisskopf, Why does the sun shine?, American Journal of Phys-
ics 46, pp. 23–31, 1978. Cited on page 916.
942 R.A. Alpher, H. Bethe & G. Gamow, The Origin of Chemical Elements, Physical Re-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

view 73, p. 803-804, 1948. No citations.
943 John Horgan, The End of Science – Facing the Limits of Knowledge in the Twilight of the
Scientific Age, Broadway Books, 1997, chapter 3, note 1. Cited on page 915.
944 M. Chantell, T.C. Weekes, X. Sarazin & M. Urban, Antimatter and the moon,
Nature 367, p. 25, 1994. M. Amenomori & al., Cosmic ray shadow by the moon observed
with the Tibet air shower array, Proceedings of the 23rd International Cosmic Ray Confer-
ence, Calgary 4, pp. 351–354, 1993. M. Urban & al., Nuclear Physics, Proceedings Supple-
ment 14B, pp. 223–236, 1990. Cited on page 906.
945 See also C. Bernard & al., Light hadron spectrum with Kogut–Susskind quarks, Nuclear
Physics, Proceedings Supplement 73, p. 198, 1999, and references therein. Cited on page 930.
946 F. Abe & al., Measurement of dijet angular distributions by the collider detector at Ferm-
ilab, Physical Review Letters 77, pp. 5336–5341, 1996. Cited on page 933.
947 The approximation of QCD with zero mass quarks is described by F. Wilczek, Getting its
from bits, Nature 397, pp. 303–306, 1999. Cited on page 933.
948 F. Close, Glueballs and hybrids: new states of matter, Contemporary Physics 38, pp. 1–12,
1997. Cited on page 934.
949 K. Grotz & H.V. Kl apd or, Die schwache Wechselwirkung in Kern-, Teilchen- und As-
trophysik, Teubner Verlag, Stuttgart, 1989. Also available in English and in several other
languages. Cited on page 938.
D. Treille, Particle physics from the Earth and from teh sky: Part II, Europhysics News
Copyright © Christoph Schiller November 1997–May 2006

950
35, no. 4, 2004. Cited on page 938.
951 P.W. Higgs, Broken symmetries, massless particles and gauge fields, Physics Letters 12,
pp. 132–133, 1964. He then expanded the story in P.W. Higgs, Spontaneous symmetry
breakdown without massless bosons, Physical Review 145, p. 1156-1163, 1966. Higgs gives
most credit to Anderson, instead of to himself; he also mentions Brout and Englert, Gural-
nik, Hagen, Kibble and ‘t Hooft. Cited on page 936.
952 M.A. B ouchiat & C.C. B ouchiat, Weak neutral currents in atomic physics, Physics
Letters B 48, pp. 111–114, 1974. U. Amaldi, A. B öhm, L.S. Durkin, P. L angacker,
A.K. Mann, W.J. Marciano, A. Sirlin & H.H. Williams, Comprehensive analysis
of data pertaining to the weak neutral current and the intermediate-vector-boson masses,
Physical Review D 36, pp. 1385–1407, 1987. Cited on page 936.

Dvipsbugw
bibliography 947

953 M.C. Noecker, B.P. Masterson & C.E. Wiemann, Precision measurement of parity
nonconservation in atomic cesium: a low-energy test of electroweak theory, Physical Review
Letters 61, pp. 310–313, 1988. See also D.M. Meekhof & al., High-precision measurement
of parity nonconserving optical rotation in atomic lead, Physical Review Letters 71, pp. 3442–
3445, 1993. Cited on page 936.
954 Rumination is studied in P. Jordan & R. de L aer Kronig, in Nature 120, p. 807, 1927.
Cited on page 939.
955 K.W.D. Ledingham & al., Photonuclear physics when a multiterawatt laser pulse inter-
acts with solid targets, Physical Review Letters 84, pp. 899–902, 2000. K.W.D. Ledingham Dvipsbugw
& al., Laser-driven photo-transmutation of Iodine-129 – a long lived nuclear waste product,
Journal of Physics D: Applied Physics 36, pp. L79–L82, 2003. R.P. Singhal, K.W.D. Led-
ingham & P. McKenna, Nuclear physics with ultra-intense lasers – present status and fu-
ture prospects, Recent Research Developments in Nuclear Physics 1, pp. 147–169, 2004. Cited
on page 940.
956 For the bibliographic details of the latest print version of the Review of Particle Physics, see

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Appendix C. The online version can be found at http://pdg.web.cern.ch/pdg. The present
status on grand unification can also be found in the respective section of the overview. Cited
on page 941.
957 J. Tran Thanh Van, ed., CP violation in Particle Physics and Astrophysics, Proc. Conf.
Chateau de Bois, France, May 1989, Editions Frontières, 1990. No citations.
958 S.C. Bennet & C.E. Wiemann, Measurement of the 6S – 7S transition polarizability
in atomic cesium and an improved test of the standard model, Physical Review Letters 82,
pp. 2484–2487, 1999. The group has also measured the spatial distribution of the weak
charge, the so-called the anapole moment; see C.S. Wood & al., Measurement of parity
nonconservation and an anapole moment in cesium, Science 275, pp. 1759–1763, 1997. Cited
on page 936.
959 H. Jeon & M. Longo, Search for magnetic monopoles trapped in matter, Physical Review
Letters 75, pp. 1443–1447, 1995. Cited on page 942.
960 On proton decay rates, see the data of the particle data group, at http://pdg.web.cern.ch.
Cited on page 942.
961 U. Amaldi, W. de B oer & H. Fürstenau, Comparison of grand unified theories with
elektroweak and strong coupling constants measured at LEP, Physics Letters 260, pp. 447–
455, 1991. This widely cited paper is the standard reference for this issue. Cited on page 942.
Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 Dvipsbugw

Dvipsbugw
Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net Copyright © Christoph Schiller November 1997–May 2006
ADVANC ED QUANTUM THEORY (NOT

– CS – this chapter will be made available in the future – CS –

YET AVAI L ABLE)
C h a p t e r IX
C hap t e r X

QUANTUM PHYSIC S
IN A NUTSHELL
Dvipsbugw

Quantum theory’s essence – the l ack of the infinitely

small

C
ompared to classical physics, quantum theory is remarkably more
omplex. The basic idea however, is simple: in nature there is a minimum
hange, or a minimum action, or again, a minimum angular momentum ħ~2.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The minimum action leads to all the strange observations made in the microscopic
domain, such as wave behaviour of matter, tunnelling, indeterminacy relations, ran-
domness in measurements, quantization of angular momentum, pair creation, decay,
indistinguishability and particle reactions. The mathematics is often disturbingly in-
volved. Was this part of the walk worth the effort? It was. The accuracy is excellent and
the results profound. We give an overview of both and then turn to the list of questions
that are still left open.

Achievements in precision
Quantum theory improved the accuracy of predictions from the few – if any – digits
common in classical mechanics to the full number of digits – sometimes fourteen – that
can be measured today. The limited precision is usually not given by the inaccuracy of
theory, it is given by the measurement accuracy. In other words, the agreement is only
limited by the amount of money the experimenter is willing to spend. Table 71 shows this
in more detail.
TA B L E 71 Some comparisons between classical physics, quantum theory and experiment

O b se r va b l e C l as- Prediction M e a su r e - C o st
si c a l of q ua n t um ment e st i -
Copyright © Christoph Schiller November 1997–May 2006

pre- theorya m at e b
dic-
tion
Simple motion of bodies
Indeterminacy 0 ∆x∆p E ħ~2 (1  10−2 ) ħ~2 10 k€
Wavelength of matter none λp = 2πħ (1  10−2 ) ħ 10 k€
beams
Tunnelling rate in alpha 0 τ = ... (1  10−2 ) τ 0.5 M€
decay
Compton wavelength none λ c = h~m e c (1  10−3 ) λ 20 k€

Dvipsbugw
 952 x quantum physics in a nu tshell

O b se r va b l e C l as- Prediction M e a su r e - C o st b
si c a l of q ua n t um ment e st i -
pre- theorya m at e
dic-
tion
Pair creation rate 0 ... ... 20 M€
Radiative decay time in hy- none τ  1~n 3 ... 5 k€
drogen Dvipsbugw
Smallest action and angu- 0 ħ~2 (1  10−6 ) ħ~2 10 k€
lar momentum
Casimir effect 0 ma~A = (π 2 ħc)~(240r 4 ) (1  10−3 ) ma 30 k€
Colours of objects
Lamb shift none ∆λ = 1057.86(1) MHz (1  10−6 ) ∆λ 50 k€
Rydberg constant none Rª = m e cα 2 ~2h (1  10−9 ) Rª 50 k€

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Stefan–Boltzmann con- none σ = π 2 k 4 ~60ħ 3 c 2 (1  3 ë 10−8 ) σ 20 k€
stant
Wien displacement con- none b = λ max T (1  10−5 ) b 20 k€
stant
Refractive index of ... none ... ... ...
Photon-photon scattering 0 ... ... 50 M€
Particle and interaction properties
Electron gyromagnetic ra- 1 or 2 2.002 319 304 3(1) 2.002 319 304 30 M€
tio 3737(82)
Z boson mass none m 2Z = m 2W (1 + sin θ 2W ) (1  10−3 ) m Z 100 M€
proton mass none (1  5 %) m p m p =1.67 yg 1 M€
reaction rate 0 ... ...
Composite matter properties
Atom lifetime  1 µs ª A 1020 a 10 k€
Molecular size none from QED within 10−3 20 k€
Von Klitzing constant ª h~e 2 = µ 0 c~2α (1  10−7 ) h~e 2 1 M€
AC Josephson constant 0 2e~h (1  10−6 ) 2e~h 5 M€
Heat capacity of metals at 0 25 J~K < 10−3 J~K 10 k€
Copyright © Christoph Schiller November 1997–May 2006

0K
Water density none ... 1000 kg~m3 10 k€
Minimum electr. conduct- 0 G = 2e 2 ~ħ G(1  10−3 ) 3 k€
ivity
Proton lifetime  1 µs ª A 1035 a 100 M€

Page 959 a. All these predictions are calculated from the quantities of Table 72, and no other input. Their
Page 1154 most precise experimental values are given in Appendix B.
b. Sometimes the cost for the calculation of the prediction is higher than that of its measurement.
Challenge 1400 n (Can you spot the examples?) The sum of the two is given.

Dvipsbugw
 quantum theory’s essence – the l ack of the infinitely small 953

We notice that the predicted values are not noticeably different from the measured ones.
If we remember that classical physics does not allow to calculate any of the predicted
values we get an idea of the progress quantum physics has allowed. But despite this im-
pressive agreement, there still are unexplained observations. In fact, these unexplained
Page 959 observations provide the input for the calculations just cited; we list them in detail below,
in Table 72.
In summary, in the microscopic domain we are left with the impression that quantum
theory is in perfect correspondence with nature; despite prospects of fame and riches,
despite the largest number of researchers ever, no contradiction with observation has Dvipsbugw
been found yet.

Physical results of quantum theory

Deorum offensae diis curae.

“ Voltaire, Traité sur la tolérance.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

All of quantum theory can be resumed in two sentences.

U In nature, actions smaller than ħ~2 = 0.53 ë 10−34 Js are not observed.

U All intrinsic properties in nature – with the exception of mass – such as elec-
tric charge, spin, parities, etc., appear as integer numbers; in composed systems
they either add or multiply.

The second statement in fact results from the first. The existence of a smallest action in
nature directly leads to the main lesson we learned about motion in the second part of
our adventure:

U If it moves, it is made of particles.

This statement applies to everything, thus to all objects and to all images, i.e. to matter
and to radiation. Moving stuff is made of quanta. Stones, water waves, light, sound waves,
earthquakes, gelatine and everything else we can interact with is made of particles. We
started the second part of our mountain ascent with the title question: what is matter and
what are interactions? Now we know: they are composites of elementary particles.
Copyright © Christoph Schiller November 1997–May 2006

To be clear, an elementary particle is a countable entity, smaller than its own Compton
Page 1174 wavelength, described by energy, momentum, and the following complete list of intrinsic
properties: mass, spin, electric charge, parity, charge parity, colour, isospin, strangeness,
charm, topness, beauty, lepton number, baryon number and R-parity. Experiments so far
failed to detect a non-vanishing size for any elementary particle.
Moving entities are made of particles. To see how deep this result is, you can apply
it to all those moving entities for which it is usually forgotten, such as ghosts, spirits,
angels, nymphs, daemons, devils, gods, goddesses and souls. You can check yourself what
Challenge 1401 e happens when their particle nature is taken into account.
From the existence of a minimum action, quantum theory deduces all its statements
about particle motion. We go through the main ones.

Dvipsbugw
954 x quantum physics in a nu tshell

There is no rest for microscopic particles. All objects obey the indeterminacy principle,
which states that the indeterminacies in position x and momentum p follow

∆x∆p E ħ~2 with ħ = 1.1 ë 10−34 Js (638)

and making rest an impossibility. The state of particles is defined by the same observables
as in classical physics, with the difference that observables do not commute. Classical
physics appears in the limit that the Planck constant ħ can effectively be set to zero.
Quantum theory introduces a probabilistic element into motion. It results from the Dvipsbugw
minimum action value through the interactions with the baths in the environment of
any system.
Large number of identical particles with the same momentum behave like waves. The
so-called de Broglie wavelength λ is given by the momentum p of a single particle through

λ= =
h 2πħ

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(639)
p p

both in the case of matter and of radiation. This relation is the origin of the wave be-
haviour of light. The light particles are called photons; their observation is now standard
practice. All waves interfere, refract and diffract. This applies to electrons, atoms, photons
and molecules. All waves being made of particles, all waves can be seen, touched and
moved. Light for example, can be ‘seen’ in photon-photon scattering, can be ‘touched’ us-
ing the Compton effect and it can be ‘moved’ by gravitational bending. Matter particles,
such as molecules or atoms, can be seen, e.g. in electron microscopes, as well as touched
and moved, e.g. with atomic force microscopes. The interference and diffraction of wave
particles is observed daily in the electron microscope.
Particles cannot be enclosed. Even though matter is impenetrable, quantum theory
shows that tight boxes or insurmountable obstacles do not exist. Waiting long enough
always allows to overcome boundaries, since there is a finite probability to overcome any
obstacle. This process is called tunnelling when seen from the spatial point of view and is
called decay when seen from the temporal point of view. Tunnelling explains the working
of television tubes as well as radioactive decay.
Particles are described by an angular momentum called spin, specifying their beha-
viour under rotations. Bosons have integer spin, fermions have half integer spin. An even
number of bound fermions or any number of bound bosons yield a composite boson; an
Copyright © Christoph Schiller November 1997–May 2006

odd number of bound fermions or an infinite number of interacting bosons yield a low-
energy fermion. Solids are impenetrable because of the fermion character of its electrons
in the atoms.
Identical particles are indistinguishable. Radiation is made of bosons, matter of fermi-
ons. Under exchange, fermions commute at space-like separations, whereas bosons anti-
commute. All other properties of quantum particles are the same as for classical particles,
namely countability, interaction, mass, charge, angular momentum, energy, momentum,
position, as well as impenetrability for matter and penetrability for radiation.
In collisions, particles interact locally, through the exchange of other particles. When
matter particles collide, they interact through the exchange of virtual bosons, i.e. off-shell
bosons. Motion change is thus due to particle exchange. Exchange bosons of even spin

Dvipsbugw
 quantum theory’s essence – the l ack of the infinitely small 955

mediate only attractive interactions. Exchange bosons of odd spin mediate repulsive in-
teractions as well.
Quantum theory defines elementary particles as particles smaller than
The properties of collisions imply the existence of antiparticles, as regularly observed
in experiments. Elementary fermions, in contrast to many elementary bosons, differ from
their antiparticles; they can be created and annihilated only in pairs. Apart from neutri-
nos, elementary fermions have non-vanishing mass and move slower than light.
Images, made of radiation, are described by the same properties as matter. Images can
only be localized with a precision of the wavelength λ of the radiation producing it. Dvipsbugw
The appearance of Planck’s constant ħ implies that length scales exist in nature.
Quantum theory introduces a fundamental jitter in every example of motion. Thus the
infinitely small is eliminated. In this way, lower limits to structural dimensions and to
many other measurable quantities appear. In particular, quantum theory shows that it
is impossible that on the electrons in an atom small creatures live in the same way that
humans live on the Earth circling the Sun. Quantum theory shows the impossibility of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Lilliput.
Clocks and metre bars have finite precision, due to the existence of a smallest action
and due to their interactions with baths. On the other hand, all measurement apparatuses
must contain baths, since otherwise they would not be able to record results.
Ref. 962 Quantum physics leaves no room for cold fusion, astrology, teleportation, telekinesis,
supernatural phenomena, multiple universes, or faster than light phenomena – the EPR
paradox notwithstanding.

Results of quantum field theory

Quantum field theory is that part of quantum theory that includes the process of trans-
formation of particles into each other. The possibility of transformation results from the
Page 713 existence of a minimum action. Transformations have several important consequences.
Quantum electrodynamics is the quantum field description of electromagnetism. Like
all the other interactions, its Lagrangian is determined by the gauge group, the require-
ments of space-time (Poincaré) symmetry, permutation symmetry and renormalizability.
The latter requirement follows from the continuity of space-time. Through the effects of
virtual particles, QED describes decay, pair creation, vacuum energy, Unruh radiation for
accelerating observers, the Casimir effect, i.e. the attraction of neutral conducting bodies,
and the limit for the localization of particles. In fact, an object of mass m can be localized
Copyright © Christoph Schiller November 1997–May 2006

only within intervals of the Compton wavelength

λC = =
h 2πħ
, (640)
mc mc
where c is the speed of light. At the latest at these distances we must abandon the classical
description and use quantum field theory. Quantum field theory introduces corrections
to classical electrodynamics; among others, the nonlinearities thus appearing produce
small departures from the superposition principle for electromagnetic fields, resulting in
photon-photon scattering.
Composite matter is separable because of the finite interaction energies of the con-

Dvipsbugw
 956 x quantum physics in a nu tshell

stituents. Atoms are made of a nucleus made of quarks, and of electrons. They provide an
effective minimal length scale to all everyday matter.
Elementary particles have the same properties as either objects or images, except di-
visibility. The elementary fermions (objects) are: the six leptons electron, muon, tau, each
with its corresponding neutrino, and the six quarks. The elementary bosons (images) are
Page 1174 the photon, the eight gluons and the two weak interaction bosons.
Quantum chromodynamics, the field theory of the strong interactions, explains the
masses of mesons and baryons through its descriptions as bound quark states. At funda-
mental scales, the strong interaction is mediated by the elementary gluons. At femtometer Dvipsbugw
scales, the strong interaction effectively acts through the exchange of spin 0 pions, and is
thus strongly attractive.
The theory of electroweak interactions describes the unification of electromagnetism
and weak interactions through the Higgs mechanism and the mixing matrix.
Objects are composed of particles. Quantum field theory provides a complete list of
Page 1174 the intrinsic properties which make up what is called an ‘object’ in everyday life, namely

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the same which characterize particles. All other properties of objects, such as shape, tem-
perature, (everyday) colour, elasticity, density, magnetism, etc., are merely combinations
of the properties from the particle properties. In particular, quantum theory specifies an
object, like every system, as a part of nature interacting weakly and incoherently with its
environment.
Since quantum theory explains the origin of material properties, it also explains the
origin of the properties of life. Quantum theory, especially the study of the electroweak
and the strong forces, has allowed to give a common basis of concepts and descriptions
to materials science, nuclear physics, chemistry, biology, medicine and to most of astro-
nomy.
For example, the same concepts allow to answer questions such as why water is liquid
at room temperature, why copper is red, why the rainbow is coloured, why the Sun and the
stars continue to shine, why there are about 110 elements, where a tree takes the material
to make its wood and why we are able to move our right hand at our own will.
Matter objects are permanent because, in contrast to radiation, matter particles can
only disappear when their antiparticles are present. It turns out that in our environment
antimatter is almost completely absent, except for the cases of radioactivity and cosmic
rays, where it appears in tiny amounts.
The particle description of nature, e.g. particle number conservation, follows from the
possibility to describe interactions perturbatively. This is possible only at low and medium
Copyright © Christoph Schiller November 1997–May 2006

energies. At extremely high energies the situation changes and non-perturbative effects
come into play.

Is quantum theory magic?

Studying nature is like experiencing magic. Nature often looks different from what it is.
During magic we are fooled – but only if we forget our own limitations. Once we start to
see ourselves as part of the game, we start to understand the tricks. That is the fun of it.
The same happens in physics.

**

Dvipsbugw
 quantum theory’s essence – the l ack of the infinitely small 957

The world looks irreversible, even though it isn’t. We never remember the future. We are
fooled because we are macroscopic.

**
The world looks decoherent, even though it isn’t. We are fooled again because we are
macroscopic.
**
There are no clocks possible in nature. We are fooled because we are surrounded by a Dvipsbugw
huge number of particles.
**
Motion seems to disappear, even though it is eternal. We are fooled again, because our
senses cannot experience the microscopic domain.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
The world seems dependent on the choice of the frame of reference, even though it is not.
We are fooled because we are used to live on the surface of the Earth.
**
Objects seem distinguishable, even though the statistical properties of their components
show that they are not. We are fooled because we live at low energies.

**
Matter looks continuous, even though it isn’t. We are fooled because of the limitations of
our senses.

In short, our human condition permanently fools us. The answer to the title question is
affirmative: quantum theory is magic. That is its main attraction.

The dangers of buying a can of beans

Another summary of our walk so far is given by the ultimate product warning, which
according to certain well-informed lawyers should be printed on every cans of beans and
Ref. 963 on every product package. It shows in detail how deeply our human condition fools us.
Copyright © Christoph Schiller November 1997–May 2006

Warning: care should be taken when looking at this product:

It emits heat radiation.
Bright light has the effect to compress this product.
Warning: care should be taken when touching this product:
Part of it could heat up while another part cools down, causing severe burns.
Warning: care should be taken when handling this product:
This product consists of at least 99.999 999 999 999 % empty space.
This product contains particles moving with speeds higher than one million kilometres per
hour.

Dvipsbugw
 958 x quantum physics in a nu tshell

Every kilogram of this product contains the same amount of energy as liberated by about
one hundred nuclear bombs.*
In case this product is brought in contact with antimatter, a catastrophic explosion will
occur.
In case this product is rotated, it will emit gravitational radiation.
Warning: care should be taken when transporting this product:
The force needed depends on its velocity, as does its weight. Dvipsbugw
This product will emit additional radiation when accelerated.
This product attracts, with a force that increases with decreasing distance, every other object
around, including its purchaser’s kids.
Warning: care should be taken when storing this product:
It is impossible to keep this product in a specific place and at rest at the same time.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Except when stored underground at a depth of several kilometres, over time cosmic radi-
ation will render this product radioactive.
This product may disintegrate in the next 1035 years.
It could cool down and lift itself into the air.
Parts of this product are hidden in other dimensions.
This product warps space and time in its vicinity, including the storage container.
Even if stored in a closed container, this product is influenced and influences all other ob-
jects in the universe, including your parents in law.
This product can disappear from its present location and reappear at any random place in
the universe, including your neighbour’s garage.
Warning: care should be taken when travelling away from this product:
It will arrive at the expiration date before the purchaser does so.
Warning: care should be taken when using this product:
Any use whatsoever will increase the entropy of the universe.
The constituents of this product are exactly the same as those of any other object in the
Copyright © Christoph Schiller November 1997–May 2006

universe, including those of rotten fish.

The use could be disturbed by the (possibly) forthcoming collapse of the universe.

The impression of a certain paranoid side to physics is purely coincidental.

* A standard nuclear warhead has an explosive yield of about 0.2 megatons (implied is the standard explosive
Ref. 964 trinitrotoluene or TNT), about thirteen times the yield of the Hiroshima bomb, which was 15 kilotonne. A
megatonne is defined as 1 Pcal=4.2 PJ, even though TNT delivers about 5 % slightly less energy that this value.
In other words, a megaton is the energy content of about 47 g of matter. That is less than a handful for most
solids or liquids.

Dvipsbugw
the essence and the limits of quantum theory 959

The essence and the limits of quantum theory

We can summarize quantum physics with a simple statement: quantum physics is the de-
scription of matter and radiation without the concept of infinitely small. Matter and radi-
ation are described by finite quantities. We had already eliminated the infinitely large in
our exploration of relativity. On the other hand, some types of infinities remain. We had
to retain the infinitely small in the description of space or time, and in topics related to
them, such as renormalization. We did not manage to eliminate all infinities yet. We are
thus not yet at the end of our quest. Surprisingly, we shall soon find out that a completely Dvipsbugw
finite description of all of nature is equally impossible. To find out more, we focus on the
path that remains to be followed.

What is unexplained by quantum theory and general relativity?

The material gathered in this second part of our mountain ascent, together with the earlier
summary of general relativity, allows us to describe all observed phenomena connected

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

to motion. Therefore, we are also able to provide a complete list of the unexplained prop-
erties of nature. Whenever we ask ‘why?’ about an observation and continue doing so
after each answer, we arrive at one of the points listed in Table 72.

TA B L E 72 Everything quantum field theory and general relativity do not explain; in other words, a list of
the only experimental data and criteria available for tests of the unified description of motion
O b se r va b l e P r o p e r t y u n e x p l a i n e d so f a r

Local quantities, from quantum theory

α em the low energy value of the electromagnetic coupling constant
αw the low energy value of the weak coupling constant
αs the low energy value of the strong coupling constant
mq the values of the 6 quark masses
ml the values of 3 lepton masses (or 6, if neutrinos have masses)
mW the values of the independent mass of the W vector boson
θW the value of the Weinberg angle
β1 , β2 , β3 three mixing angles (or 7, if neutrinos have masses)
θ CP the value of the CP parameter
θ st the value of the strong topological angle
Copyright © Christoph Schiller November 1997–May 2006

3 the number of particle generations

3+1 the number of space and time dimensions
0.5 nJ~m3 the value of the observed vacuum energy density or cosmological constant
Global quantities, from general relativity
1.2(1) ë 1026 m (?) the distance of the horizon, i.e. the ‘size’ of the universe (if it makes sense)
1082 (?) the number of baryons in the universe, i.e. the average matter density in the
universe (if it makes sense)
1092 (?) the initial conditions for more than 1092 particle fields in the universe, includ-
ing those at the origin of galaxies or stars (if or as long as they make sense)

Local structures, from quantum theory

Dvipsbugw
960 x quantum physics in a nu tshell

O b se r va b l e P r o p e r t y u n e x p l a i n e d so f a r

S(n) the origin of particle identity, i.e. of permutation symmetry

Ren. group the renormalization properties, i.e. the existence of point particles
SO(3,1) the origin of Lorentz (or Poincaré) symmetry
(i.e. of spin, position, energy, momentum)
C‡ the origin of the algebra of observables
Gauge group the origin of gauge symmetry
Dvipsbugw
(and thus of charge, strangeness, beauty, etc.)
in particular, for the standard model:
U(1) the origin of the electromagnetic gauge group (i.e. of the quantization of elec-
tric charge, as well as the vanishing of magnetic charge)
SU(2) the origin of weak interaction gauge group
SU(3) the origin of strong interaction gauge group

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Global structures, from general relativity
maybe R  S3 (?) the unknown topology of the universe (if it makes sense)

The table has several notable aspects.* First of all, neither quantum mechanics nor general
relativity explain any property unexplained in the other field. The two theories do not
help each other; the unexplained parts of both fields simply add up. Secondly, both in
quantum theory and in general relativity, motion still remains the change of position
with time. In short, in the first two parts of this walk we did not achieve our goal: we still
do not understand motion. Our basic questions remain: What is time and space? What is
mass? What is charge and what are the other properties of objects? What are fields? Why
are all the electrons the same?
We also note that Table 72 lists extremely different concepts. That means that at this
point of our walk there is a lot we do not understand. Finding the answers will not be
easy, but will require effort.
On the other hand, the list of unexplained properties of nature is also short. The de-
scription of nature our adventure has produced so far is concise and precise. No discrep-
ancies from experiments are known. In other words, we have a good description of mo-
tion in practice. Going further is unnecessary if we only want to improve measurement
precision. Simplifying the above list is mainly important from the conceptual point of
view. For this reason, the study of physics at university often stops at this point. However,
Copyright © Christoph Schiller November 1997–May 2006

even though we have no known discrepancies with experiments, we are not at the top of
Motion Mountain, as Table 72 shows.
An even more suggestive summary of the progress and open issues of physics is shown
in Figure 359. From one corner of a cube, representing Galilean physics, three edges – la-
* Every now and then, researchers provide other lists of open questions. However, they all fall into the list
above. The elucidation of dark matter and of dark energy, the details of the big bang, the modifications of
general relativity by quantum theory, the mass of neutrinos, the quest for unknown elementary particles
such as the inflaton field, magnetic monopoles or others, the functioning of cosmic high-energy particle
accelerators, the stability or decay of protons, the origins of the heavy chemical elements, other interactions
between matter and radiation or the possibility of higher spatial dimensions are questions that all fall into
the table above.

Dvipsbugw
 the essence and the limits of quantum theory 961

Classical Quantum
gravitation gravity
1685 ca 1990

Classical ħ, e, k Theory Dvipsbugw

mechanics of motion
1650 Quantum General 2020?
theory relativity
1925 1915

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Special Quantum field
relativity theory
1905 1950

F I G U R E 359 A simplified history of the description of motion in physics, by giving the limits
to motion included in each description

belled G, c and ħ, e, k – lead to classical gravity, special relativity and quantum theory.
Each constant implies a limit to motion; in the corresponding theory, this limit is taken
into account. From these first level theories, corresponding parallel edges lead to general
relativity, quantum field theory and quantum gravity, which take into account two of the
limits.* From the second level theories, all edges lead to the last missing corner; that is the
theory of motion. It takes onto account all limits found so far. Only this theory is a full
or unified description of motion. The important point is that we already know all limits
to motion. To arrive at the last point, no new experiments are necessary. No new know-
ledge is required. We only have to advance in the right direction, with careful thinking.
Reaching the final theory of motion is the topic of the third part of our adventure.
Finally, we note from Table 72 that all progress we can expect about the foundations
of motion will take place in two specific fields: cosmology and high energy physics.
Copyright © Christoph Schiller November 1997–May 2006

How to delude oneself that one has reached the top of Motion
Mountain
Nowadays it is deemed chic to pretend that the adventure is over at the stage we have just
reached.** The reasoning is as follows. If we change the values of the unexplained con-

* Of course, Figure 359 gives a simplified view of the history of physics. A more precise diagram would
use three different arrows for ħ, e and k, making the figure a five-dimensional cube. However, not all of its
Challenge 1402 ny corners would have dedicated theories (can you confirm this?), and moreover, the diagram would be much
less appealing.
** Actually this attitude is not new. Only the arguments have changed. Maybe the greatest physicist ever,
Ref. 965 James Clerk Maxwell, already fought against this attitude over a hundred years ago: ‘The opinion seems to

Dvipsbugw
 962 x quantum physics in a nu tshell

stants from Table 72 only ever so slightly, nature would look completely different from
Ref. 966 what it does. The consequences have been studied in great detail; Table 73 gives an over-
view of the results.
TA B L E 73 A selection of the consequences of changing the properties of nature

O b se r va b l e C h a n g e R e su lt

Local quantities, from quantum theory

α em smaller: only short lived, smaller and hotter stars; no Sun Dvipsbugw
larger: darker Sun, animals die of electromagnetic radiation,
too much proton decay, no planets, no stellar
explosions, no star formation, no galaxy formation
+60%: quarks decay into leptons
+200%: proton-proton repulsion makes nuclei impossible
αw −50%: carbon nucleus unstable

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

very weak: no hydrogen, no p-p cycle in stars, no C-N-O cycle
+2%: no protons from quarks
»F m e ~
2
G either no or only helium in the universe
Gm 2e :
much larger: no stellar explosions, faster stellar burning
αs −9%: no deuteron, stars much less bright
−1%: no C resonance, no life
+3.4%: diproton stable, faster star burning
much larger: carbon unstable, heavy nuclei unstable, widespread
leukaemia
n-p mass differ- larger: neutron decays in proton inside nuclei; no elements
ence
smaller: free neutron not unstable, all protons into neutrons
during big bang; no elements
smaller than protons would capture electrons, no hydrogen atoms,
me : star life much shorter
m l changes:
e-p mass ratio much differ- no molecules
ent:
Copyright © Christoph Schiller November 1997–May 2006

much smaller: no solids

3 generations 6-8: only helium in nature
A8: no asymptotic freedom and confinement
Global quantities, from general relativity

have got abroad that, in a few years, all great physical constants will have been approximately estimated, and
that the only occupation which will be left to men of science will be to carry these measurements to another
place of decimals. ... The history of science shows that even during that phase of her progress in which she
devotes herself to improving the accuracy of the numerical measurement of quantities with which she has
long been familiar, she is preparing the materials for the subjugation of new regions, which would have
remained unknown if she had been contented with the rough methods of her early pioneers.’

Dvipsbugw
 the essence and the limits of quantum theory 963

O b se r va b l e C h a n g e R e su lt

horizon size much smaller: no people

baryon number very different: no smoothness
much higher: no solar system
Initial condition changes:
Moon mass smaller: small Earth magnetic field; too much cosmic radiation;
widespread child skin cancer
Moon mass larger: large Earth magnetic field; too little cosmic radiation;
Dvipsbugw
no evolution into humans
Sun’s mass smaller: too cold for the evolution of life
Sun’s mass larger: Sun too short lived for the evolution of life
Jupiter mass smaller: too many comet impacts on Earth; extinction of animal
life
Jupiter mass larger: too little comet impacts on Earth; no Moon; no

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

dinosaur extinction
Oort cloud ob- smaller: no comets; no irregular asteroids; no Moon; still
ject number dinosaurs
galaxy centre dis- smaller: irregular planet motion; supernova dangers
tance
initial cosmicspeed +0.1%: 1000 times faster universe expansion
−0.0001%: universe recollapses after 10 000 years
vacuum energy change by no flatness
density 10−55 :
3 + 1 dimensions different: no atoms, no planetary systems
Local structures, from quantum theory
permutation none: no matter
symmetry
Lorentz symmetry none: no communication possible
U(1) different: no Huygens principle, no way to see anything
SU(2) different: no radioactivity, no Sun, no life
SU(3) different: no stable quarks and nuclei
Global structures, from general relativity
Copyright © Christoph Schiller November 1997–May 2006

topology other: unknown; possibly correlated gamma ray bursts or star

images at the antipodes

Some even speculate that the table can be condensed into a single sentence: if any para-
Ref. 967 meter in nature is changed, the universe would either have too many or too few black
Challenge 1403 r holes. However, the proof of this condensed summary is not complete yet.
Table 73, on the effects of changing nature, is overwhelming. Obviously, even the tini-
est changes in the properties of nature are incompatible with our existence. What does
this mean? Answering this question too rapidly is dangerous. Many fall into a common
trap, namely to refuse admitting that the unexplained numbers and other properties need

Dvipsbugw
 964 x quantum physics in a nu tshell

to be explained, i.e. deduced from more general principles. It is easier to throw in some
irrational belief. The three most fashionable beliefs are that the universe is created or
designed, that the universe is designed for people, or that the values are random, as our
universe happens to be one of many others.
All these beliefs have in common that they have no factual basis, that they discourage
further search and that they sell many books. Physicists call the issue of the first belief fine
tuning, and usually, but not always, steer clear from the logical errors contained in the so
Page 683 common belief in ‘creation’ discussed earlier on. However, many physicists subscribe to
the second belief, namely that the universe is designed for people, calling it the anthropic Dvipsbugw
principle, even though we saw that it is indistinguishable both from the simian principle
Page 689 or from the simple request that statements be based on observations. In 2004, this be-
lief has even become fashionable among older string theorists. The third belief, namely
multiple universes, is a minority view, but also sells well.
Stopping our mountain ascent with a belief at the present point is not different from
doing so directly at the beginning. Doing so used to be the case in societies which lacked

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the passion for rational investigation, and still is the case in circles which discourage the
use of reason among their members. Looking for beliefs instead of looking for answers
means to give up the ascent of Motion Mountain while pretending to have reached the
top.
That is a pity. In our adventure, accepting the powerful message of Table 73 is one
of the most awe-inspiring, touching and motivating moments. There is only one possible
implication based on facts: the evidence implies that we are only a tiny part of the universe,
but linked with all other aspects of it. Due to our small size and to all the connections with
our environment, any imagined tiny change would make us disappear, like a water droplet
is swept away by large wave. Our walk has repeatedly reminded us of this smallness and
dependence, and overwhelmingly does so again at this point.
Having faced this powerful experience, everybody has to make up his own mind on
Challenge 1404 n whether to proceed with the adventure or not. Of course, there is no obligation to do so.

What awaits us?

The shortness of the list of unexplained aspects of nature means that no additional experi-
mental data are available as check of the final description of nature. Everything we need to
arrive at the final description of motion will probably be deduced from the experimental
data given in this list, and from nothing else. In other words, future experiments will not
Copyright © Christoph Schiller November 1997–May 2006

help us – except if they change something in the list, as supersymmetry might do with
the gauge groups or astronomical experiments with the topology issue.
This lack of new experimental data means that to continue the walk is a conceptual
adventure only. We have to walk into storms raging near the top of Motion Mountain,
keeping our eyes open, without any other guidance except our reason: this is not an ad-
venture of action, but an adventure of the mind. And it is an incredible one, as we shall
soon find out. To provide a feeling of what awaits us, we rephrase the remaining issues in
six simple challenges.
What determines colours? In other words, what relations of nature fix the famous fine
structure constant? Like the hero of Douglas Adams’ books, physicists know the answer
to the greatest of questions: it is 137.036. But they do not know the question.

Dvipsbugw
 the essence and the limits of quantum theory 965

What fixes the contents of a teapot? It is given by its size to the third power. But why
are there only three dimensions? Why is the tea content limited in this way?
Was Democritus right? Our adventure has confirmed his statement up to this point;
nature is indeed well described by the concepts of particles and of vacuum. At large scales,
relativity has added a horizon, and at small scales, quantum theory added vacuum energy
and pair creation. Nevertheless, both theories assume the existence of particles and the
existence of space-time, and neither predicts them. Even worse, both theories completely
fail to predict the existence of any of the properties either of space-time – such as its
dimensionality – or of particles – such as their masses and other quantum numbers. A Dvipsbugw
lot is missing.
Was Democritus wrong? It is often said that the standard model has only about twenty
unknown parameters; this common mistake negates about 1093 initial conditions! To get
an idea of the problem, we simply estimate the number N of possible states of all particles
in the universe by
N=nvd p f (641)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

where n is the number of particles, v is the number of variables (position, momentum,
spin), d is the number of different values each of them can take (limited by the maximum
of 61 decimal digits), p is the number of visible space-time points (about 10183 ) and f is
a factor expressing how many of all these initial conditions are actually independent of
each other. We thus have the following number of possibilities

N = 1092 ë 8 ë 1061 ë 10183 ë f = 10336 ë f (642)

from which the 1093 actual initial conditions have to be explained. There is a small prob-
lem that we know nothing whatsoever about f . Its value could be 0, if all data were in-
Page 815 terdependent, or 1, if none were. Worse, above we noted that initial conditions cannot
be defined for the universe at all; thus f should be undefined and not be a number at all!
Whatever the case, we need to understand how all the visible particles get their 1093 states
assigned from this range of options.
Page 157 Were our efforts up to this point in vain? Quite at the beginning of our walk we
noted that in classical physics, space and time are defined using matter, whereas matter
is defined using space-time. Hundred years of general relativity and of quantum theory,
including dozens of geniuses, have not solved this oldest paradox of all. The issue is still
Challenge 1405 e open at this point of our walk, as you might want to check by yourself.
Copyright © Christoph Schiller November 1997–May 2006

The answers to these six questions define the top of Motion Mountain. Answering
them means to know everything about motion. In summary, our quest for the unravelling
of the essence of motion gets really interesting only from this point onwards!

Dvipsbugw
 966 x quantum physics in a nu tshell

That is why Leucippus and Democritus, who say

“
that the atoms move always in the void and the
unlimited, must say what movement is, and in
what their natural motion consists.
Aristotle, Treaty of the Heaven

”
Ref. 968

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
bibliography 967

Biblio graphy

962 An informative account of the world of psychokinesis and the paranormal is given by the
famous professional magician James R andi, Flim-flam!, Prometheus Books, Buffalo 1987,
as well as in several of his other books. See also the http://www.randi.org website. Cited on
page 955.
963 This way to look at things goes back to the text by Susan Hewitt & Edward Subitzky,
A call for more scientific truth in product warning labels, Journal of Irreproducible Results Dvipsbugw
36, nr. 1, 1991. Cited on page 957.
964 J. Malik, The yields of the Hiroshima and Nagasaki nuclear explosions, Technical Report
LA-8819, Los Alamos National Laboratory, September 1985. Cited on page 958.
965 James Clerk Max well, Scientific Papers, 2, p. 244, October 1871. Cited on page 961.
966 A good introduction is C.J. Hogan, Why the universe is just so, Reviews of Modern Physics
72, pp. 1149–1161, 2000. Most of the material of Table 73 is from the mighty book by John D.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Barrow & Frank J. Tipler, The Anthropic Cosmological Principle, Oxford University
Press, 1986. Discarding unrealistic options is also an interesting pastime. See for example
the reasons why life can only be carbon-based, as explained in the essay by I. Asimov, The
one and only, in his book The Tragedy of the Moon, Doubleday, Garden City, New York, 1973.
Cited on page 962.
967 L. Smolin, The fate of black hole singularities and the parameters of the standard models
of particle physics and cosmology, http://www.arxiv.org/abs/gr-qc/9404011. Cited on page
963.
968 Aristotle, Treaty of the heaven, III, II, 300 b 8. See Jean-Paul Dumont, Les écoles
présocratiques, Folio Essais, Gallimard, p. 392, 1991. Cited on page 966.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
Intermezzo

BAC TER IA, FLI ES AND K NOT S

Dvipsbugw

La première et la plus belle qualité de la nature

“ est le mouvement qui l’agite sans cesse ; mais ce

mouvement n’est qu’une suite perpétuelle de
crimes ; ce n’est que par des crimes qu’elle le
conserve.
Donatien de Sade, Justine, ou les malheurs de la

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

vertu.*

W
obbly entities, in particular jellyfish or amoebas, open up a fresh vision of the
orld of motion, if we allow to be led by the curiosity to study them in detail.
e have missed many delightful insights by leaving them aside. In particular,
”
wobbly entities yield surprising connections between shape change and motion which
will be of great use in the last part of our mountain ascent. Instead of continuing to look
at the smaller and smaller, we now take a look back, towards everyday motion and its
mathematical description.
To enjoy this intermezzo, we change a dear habit. So far, we always described any gen-
eral example of motion as composed of the motion of point particles. This worked well
in classical physics, in general relativity and in quantum theory; we based the approach
on the silent assumption that during motion, each point of a complex system can be fol-
lowed separately. We will soon discover that this assumption is not realized at smallest
scales. Therefore the most useful description of motion of extended bodies uses methods
that do not require that body parts be followed one by one. We explore this issue in this
intermezzo; doing so is a lot of fun in its own right.
If we imagine particles as extended entities – as we soon will have to – a particle mov-
ing through space is similar to a dolphin swimming through water or to a bee flying
through air. Let us explore how these animals do this.
Copyright © Christoph Schiller November 1997–May 2006

Bumblebees and other miniature flying systems

When a butterfly passes by, as can happen to anybody ascending a mountain as long as
flowers are present, we can stop a moment to appreciate a simple fact: a butterfly flies, and
it is rather small. If you leave some cut fruit in the kitchen until it is rotten, we find the
even smaller fruit flies. If you have ever tried to build small model aeroplanes, or if you
even only compare them to paper aeroplanes (probably the smallest man-made flying

* ‘The primary and most beautiful of nature’s qualities is motion, which agitates her at all times; but this
motion is simply a perpetual consequence of crimes; she conserves it by means of crimes only.’ Donatien Al-
phonse François de Sade (1740–1814) is the French writer from whom the term ‘sadism’ was deduced.

Dvipsbugw
 970 intermezzo: bacteria, flies and knots

Dvipsbugw
F I G U R E 360 A flying F I G U R E 361 Vortices around a butterfly wing (© Robert
fruit fly, tethered to a Srygley/Adrian Thomas)
string

thing you ever saw) you start to get a feeling for how well evolution optimized insects.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Compared to paper planes, insects also have engines, flapping wings, sensors, navigation
systems, gyroscopic stabilizers, landing gear and of course all the features due to life, re-
production and metabolism, built into an incredibly small volume. Evolution really is an
excellent engineering team. The most incredible flyers, such as the common house fly
(Musca domestica), can change flying direction in only 30 ms, using the stabilizers that
nature has built by reshaping the original second pair of wings. Human engineers are
getting more and more interested in the technical solutions evolution has chosen and are
Ref. 970 trying to achieve the same miniaturization. The topic is extremely vast, so that we will
pick out only a few examples.
How does an insect such as a fruit fly (Drosophila melanogaster) fly? The lift mд gen-
erated by a fixed wing follows the relation

mд = f A v 2 ρ (643)

where A is the surface of the wing, v is the speed of the wing in the fluid of density ρ.
The factor f is a pure number, usually with a value between 0.2 and 0.4, that depends on
Ref. 969 the angle of the wing and its shape; here we use the average value 0.3. For a Boeing 747,
the surface is 511 m2 , the top speed is 250 m~s; at an altitude of 12 km the density of air
is only a quarter of that on the ground, thus only 0.31 kg~m3 . We deduce (correctly) that
Challenge 1406 e a Boeing 747 has a mass of about 300 ton. For bumblebees with a speed of 3 m~s and a
Copyright © Christoph Schiller November 1997–May 2006

wing surface of 1 cm2 , we get a lifted mass of about 35 mg, much less than the weight of
the bee, namely about 1 g. In other words, a bee, like any other insect, cannot fly if it keeps
its wings fixed. It could not fly with fixed wings even if it had propellers! Therefore, all
insects must move their wings, in contrast to aeroplanes, not only to advance or to gain
height, but also to simply remain airborne. Aeroplanes generate enough lift with fixed
wings. Indeed, if you look at flying animals, you note that the larger they are, the less they
need to move their wings.
Challenge 1407 n Can you deduce from equation (643) that birds or insects can fly but people cannot?
The formula also (partly) explains why human powered aeroplanes must be so large.*

* The rest of the explanation requires some aerodynamics, which we will not study here. Aerodynamics

Dvipsbugw
 intermezzo: bacteria, flies and knots 971

But how do insects, small birds, flying fish or bats have to move their wings? This is a
tricky question. In fact, the answer is just being uncovered by modern research. The main
Ref. 971 point is that insect wings move in a way to produce eddies at the front edge which in turn
Ref. 972 thrust the insect upwards. The aerodynamic studies of butterflies – shown in Figure 361 –
and the studies of enlarged insect models moving in oil instead of in air are exploring the
way insects make use of vortices. Researchers try to understand how vortices allow con-
trolled flight at small dimensions. At the same time, more and more mechanical birds and
model ‘aeroplanes’ that use flapping wings for their propulsion are being built around the
world. The field is literally in full swing.* The aim is to reduce the size of flying machines. Dvipsbugw
However, none of the human-built systems is yet small enough that it actually requires
wing motion to fly, as is the case for insects.
Formula (643) also shows what is necessary for take-off and landing. The lift of wings
decreases for smaller speeds. Thus both animals and aeroplanes increase their wing sur-
face in these occasions. But even strongly flapping enlarged wings often are not sufficient
at take-off. Many flying animals, such as swallows, therefore avoid landing completely.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

For flying animals which do take off from the ground, nature most commonly makes
them hit the wings against each other, over their back, so that when the wings separate
again, the vacuum between them provides the first lift. This method is used by insects
and many birds, such as pheasants. As every hunter knows, pheasants make a loud ‘clap’
when they take off.
Both wing use and wing construction thus depend
on size. There are four types of wings in nature. First
of all, all large flying objects, such aeroplanes and
large birds, fly using fixed wings, except during take-
off and landing. Second, common size birds use flap-
ping wings. These first two types of wings have a thick-
ness of about 10 to 15 % of the wing depth. At smal- Figure to be inserted
ler dimensions, a third wing type appears, as seen in
dragonflies and other insects. At these scales, at Reyn-
olds numbers of around 1000 and below, thin mem- F I G U R E 362 Two large wing types
brane wings are the most efficient. The Reynolds num-
ber measures the ratio between inertial and viscous effects in a fluid. It is defined as

R=
lvρ
(644)
η
Copyright © Christoph Schiller November 1997–May 2006

where l is a typical length of the system, v the speed, ρ the density and η the dynamic
viscosity of the fluid.** A Reynolds number much larger than one is typical for rapid air

shows that the power consumption, and thus the resistance of a wing with given mass and given cruise
speed, is inversely proportional to the square of the wingspan. Large wingspans with long slender wings are
thus of advantage in (subsonic) flying, especially when energy needs to be conserved.
* The website http://www.aniprop.de presents a typical research approach and the sites http://ovirc.free.fr
and http://www.ornithopter.org give introductions into the way to build such systems for hobbyists.
** The viscosity is the resistance to flow a fluid poses. It is defined by the force F necessary to move a layer of
surface A with respect to a second, parallel one at distance d; in short, the (coefficient of) dynamic viscosity
is defined as η = d F~A v. The unit is 1 kg~s m or 1 Pa s or 1 N s~m2 , once also called 10 P or 10 poise. In other

Dvipsbugw
 972 intermezzo: bacteria, flies and knots

flow and fast moving water. In fact, the Reynolds numbers specifies what is meant by a
‘rapid’ or ‘fluid’ flow on one hand, and a ‘slow’ or ‘viscous’ flow on the other. The first three
wing types are all for rapid flows.
The fourth type of wings is found at the smallest possible dimensions, for insects smal-
ler than one millimetre; their wings are not membranes at all. Typical are the cases of
thrips and of parasitic wasps, which can be as small as 0.3 mm. All these small insects
have wings which consist of a central stalk surrounded by hair. In fact, Figure 363 shows
that some species of thrips have wings which look like miniature toilet brushes.
At even smaller dimensions, corresponding to Dvipsbugw
Reynolds number below 10, nature does not use
wings any more, though it still makes use of air trans-
port. In principle, at the smallest Reynolds numbers
gravity plays no role any more, and the process of fly-
ing merges with that of swimming. However, air cur-
rents are too strong compared with the speeds that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

such a tiny system could realize. No active naviga-
tion is then possible any more. At these small dimen-
sions, which are important for the transport through F I G U R E 363 The wings of a few types
air of spores and pollen, nature uses the air currents of insects smaller than 1 mm (thrips,
for passive transport, making use of special, but fixed Encarsia, Anagrus, Dicomorpha)
shapes. (HortNET)
We can summarize that active flying is only pos-
sible through shape change. Only two types of shape changes are possible: that of pro-
pellers (or turbines) and that of wings.* Engineers are studying with intensity how these
Ref. 973 shape changes have to take place in order to make flying most effective. Interestingly, the
same challenge is posed by swimming.

Swimming
Swimming is a fascinating phenomenon. The Greeks argued that the ability of fish to
swim is a proof that water is made of atoms. If atoms would not exist, a fish could not
advance through it. Indeed, swimming is an activity that shows that matter cannot be
continuous. Studying swimming can thus be quite enlightening. But how exactly do fish
swim?
Whenever dolphins, jellyfish, submarines or humans swim, they take water with their
Copyright © Christoph Schiller November 1997–May 2006

fins, body, propellers, hands or feet and push it backwards. Due to momentum conserva-

words, given a horizontal tube, the viscosity determines how strong the pump needs to be to pump the fluid
through the tube at a given speed. The viscosity of air 20°C is 1.8  10−5 kg~s m or 18 µPa s and increases
Challenge 1408 ny with temperature. In contrast, the viscosity of liquids decreases with temperature. (Why?) The viscosity of
water at 0°C is 1.8 mPa s, at 20°C it is 1.0 mPa s (or 1 cP), and at 40°C is 0.66 mPa s. Hydrogen has a viscosity
smaller than 10 µPa s, whereas honey has 25 Pa s and pitch 30 MPa s.
Physicists also use a quantity ν called the kinematic viscosity. It is defined with the help of the mass density
of the fluid as ν = η~ρ and is measured in m2 ~s, once called 104 stokes. The kinematic viscosity of water at
20°C is 1 mm2 ~s (or 1 cSt). One of the smallest values is that of acetone, with 0.3 mm2 ~s; a larger one is
glycerine, with 2000 mm2 ~s. Gases range between 3 mm2 ~s and 100 mm2 ~s.
* The book by John Brackenbury, Insects in Flight, 1992. is a wonderful introduction into the biomech-
anics of insects, combining interesting science and beautiful photographs.

Dvipsbugw
 intermezzo: bacteria, flies and knots 973

tion they then move forward.* In short, people swim in the same way that fireworks or
rockets fly: by throwing matter behind them. Does all swimming work in this way? In
particular, do small organisms advancing through the molecules of a liquid use the same
method? No.
It turns out that small organisms such as bacteria do not have the capacity to propel or
accelerate water against their surroundings. From far away, the swimming of microorgan-
isms thus resembles the motion of particles through vacuum. Like microorganisms, also
particles have nothing to throw behind them. Indeed, the water remains attached around
a microorganism without ever moving away from it. Physically speaking, in these cases Dvipsbugw
of swimming the kinetic energy of the water is negligible. In order to swim, unicellular
beings thus need to use other effects. In fact, their only possibility is to change their body
shape in controlled ways.
Let us go back to everyday scale for a moment.
Swimming scallops, molluscs up to a few cm in size,
can be used to clarify the difference between macro-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

scopic and microscopic swimming. Scallops have
a double shell connected by a hinge that they can
open and close. If they close it rapidly, water is ex-
pelled and the mollusc is accelerated; the scallop
then can glide for a while through the water. Then
the scallop opens the shell again, this time slowly,
and repeats the feat. When swimming, the larger F I G U R E 364 A swimming scallop (here
scallops look like clockwork false teeth. If we re- from the genus Chlamys) (© Dave
duce the size of the scallop by a thousand times to Colwell)
the size of single cells we get a simple result: such a
tiny scallop cannot swim.
The origin of the lack of scalability of swimming methods is the changing ratio
between inertial and dissipative effects at different scales. This ratio is measured by the
Reynolds number. For the scallop the Reynolds number is about 100, which shows that
when it swims, inertial effects are much more important than dissipative, viscous effects.
For a bacterium the Reynolds number is much smaller than 1, so that inertial effects effect-
ively play no role. There is no way to accelerate water away from a bacterial-sized scallop,
and thus no way to glide. In fact one can even show the stronger result that no cell-sized
being can move if the shape change is the same in the two halves of the motion (opening
Ref. 976 and closing). Such a shape change would simply make it move back and forward. Thus
Copyright © Christoph Schiller November 1997–May 2006

there is no way to move at cell dimensions with a method the scallop uses on centimetre
scale; in fact the so-called scallop theorem states that no microscopic system can swim if
it uses movable parts with only one degree of freedom.
Microorganisms thus need to use a more evolved, two-dimensional motion of their
shape to be able to swim. Indeed, biologists found that all microorganisms use one of the
following three swimming styles:
— Microorganisms of compact shape of diameter between 20 µm and about 20 mm, use

Page 73 * Fish could use propellers, as the arguments against wheels we collected at the beginning of our walk do
not apply for swimming. But propellers with blood supply would be a weak point in the construction, and
thus in the defence of a fish.

Dvipsbugw
 974 intermezzo: bacteria, flies and knots

figure to be added

F I G U R E 365 Ciliated and flagellate motion

Dvipsbugw
cilia. Cilia are hundreds of little hairs on the surface of the organism. The organisms
move the cilia in waves wandering around their surface, and these surface waves make
the body advance through the fluid. All children watch with wonder Paramecium, the
unicellular animal they find under the microscope when they explore the water in
which some grass has been left for a few hours. Paramecium, which is between 100 µm
and 300 µm in size, as well as many plankton species* use cilia for its motion. The cilia

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

and their motion are clearly visible in the microscope. A similar swimming method is
even used by some large animals; you might have seen similar waves on the borders of
certain ink fish; even the motion of the manta (partially) belongs into this class. Ciliate
Ref. 977 motion is an efficient way to change the shape of a body making use of two dimensions
and thus avoiding the scallop theorem.
— Sperm and eukaryote microorganisms whose sizes are in the range between 1 µm and
50 µm swim using an (eukaryote) flagellum.** Flagella, Latin for ‘small whips’, work
like flexible oars. Even though their motion sometimes appears to be just an oscilla-
tion, flagella get a kick only during one half of their motion, e.g. at every swing to
the left. Flagella are indeed used by the cells like miniature oars. Some cells even twist
their flagellum in a similar way that people rotate an arm. Some microorganisms, such
as Chlamydomonas, even have two flagella which move in the same way as people
Ref. 980 move their legs when they perform the breast stroke. Most cells can also change the
sense in which the flagellum is kicked, thus allowing them to move either forward
Ref. 984 or backward. Through their twisted oar motion, bacterial flagella avoid retracing the
same path when going back and forward. As a result, the bacteria avoid the scallop
theorem and manage to swim despite their small dimensions. The flexible oar motion
they use is an example of a non-adiabatic mechanism; an important fraction of the
energy is dissipated.
— The smallest swimming organisms, bacteria with sizes between 0.2 µm and 5 µm, swim
Copyright © Christoph Schiller November 1997–May 2006

Ref. 981 using bacterial flagella. These flagella, also called prokaryote flagella, are different from
the ones just mentioned. Bacterial flagella move like turning corkscrews. They are used
by the famous Escherichia coli bacterium and by all bacteria of the genus Salmonella.
This type of motion is one of the prominent exceptions to the non-existence of wheels
Page 73 in nature; we mentioned it in the beginning of our walk. Corkscrew motion is an ex-
ample of an adiabatic mechanism.

* See the http://www.liv.ac.uk/ciliate/ website for an overview.

Ref. 978 ** The largest sperm, of 6 cm length, are produced by the 1.5 mm sized Drosophila bifurca fly, a relative of
the famous Drosophila melanogaster. Even when thinking about the theory of motion, it is impossible to
avoid thinking about sex.

Dvipsbugw
 intermezzo: bacteria, flies and knots 975

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 366 A
well-known ability of
cats

A Coli bacterium typically has a handful of flagella, each about 30 nm thick and of
corkscrew shape, with up to six turns; the turns have a ‘wavelength’ of 2.3 µm. Each
flagellum is turned by a sophisticated rotation motor built into the cell, which the cell
can control both in rotation direction and in angular velocity. For Coli bacteria, the
Ref. 979 range is between 0 and about 300 Hz.
Copyright © Christoph Schiller November 1997–May 2006

A turning flagellum does not propel a bacterium like a propeller; as mentioned, the
velocities involved are much too small, the Reynolds number being only about 10−4 . At
these dimensions and velocities, the effect is better described by a corkscrew turning in
honey or in cork: a turning corkscrew produces a motion against the material around
it, in the direction of the corkscrew axis. The flagellum moves the bacterium in the
same way that a corkscrew moves the turning hand with respect to the cork.
Note that still smaller bacteria do not swim at all. Each bacterium faces a minimum
Ref. 982 swimming speed requirement: is must outpace diffusion in the liquid it lives in. Slow
swimming capability makes no sense; numerous microorganisms therefore do not man-
age or do not try to swim at all. Some microorganisms are specialized to move along
liquid–air interfaces. Others attach themselves to solid bodies they find in the liquid.

Dvipsbugw
 976 intermezzo: bacteria, flies and knots

Some of them are able to move along these solids. The amoeba is an example for a mi-
croorganism moving in this way. Also the smallest active motion mechanisms known,
Page 820 namely the motion of molecules in muscles and in cell membranes, work this way.
Let us summarize these observations in a different way. All known active motion, or
self-propulsion, takes place in fluids – be it air or liquids. All active motion requires shape
change. In order that shape change leads to motion, the environment, e.g. the water, must
itself consist of moving components always pushing onto the swimming entity. The mo-
tion of the swimming entity can then be deduced from the particular shape change it
performs. To test your intuition, you may try the following puzzle: is microscopic swim- Dvipsbugw
Challenge 1409 ny ming possible in two spatial dimensions? In four?

Falling cats and the theory of shape change

In the last decades, the theory of shape change has changed from a fashionable piece of
research to a topic whose results are both appealing and useful. We have seen that shape

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

change of a body in a fluid can lead to translation. But shape change can also lead to
a rotation of the body. In particular, the theory of shape change is useful in explaining
how falling cats manage to fall on their feet. Cats are not born with this ability; they have
to learn it. But the feat remains fascinating. The great British physicist Michael Berry
understood that this ability of cats can be described by an angular phase in a suitably
defined shape space.

– CS – to be inserted – CS –

Page 91 In fact, cats confirm in three dimensions what we already knew for two dimensions: a
deformable body can change its own orientation in space without outside help.
But shape change bears more surprises.

Turning a sphere inside out

A text should be like a lady’s dress; long enough

“ ” to cover the subject, yet short enough to keep it

interesting.

Continuing the theme of motion of wobbly entities, a famous example cannot be avoided.
In 1957, the mathematician Stephen Smale proved that a sphere can be turned inside out.
The discovery brought him the Fields medal in 1966, the highest prize for discoveries in
Copyright © Christoph Schiller November 1997–May 2006

mathematics. Mathematicians call his discovery the eversion of the sphere.

To understand the result, we need to describe more clearly the rules of mathematical
eversion. First of all, it is assumed that the sphere is made of a thin membrane which
has the ability to stretch and bend without limits. Secondly, the membrane is assumed
to be able to intersect itself. Of course, such a ghostly material does not exist in everyday
life; but in mathematics, it can be imagined. A third rule requires that the moves must
be performed in such a way that the membrane is not punctured, ripped nor creased; in
short, everything must happen smoothly (or differentiably, as mathematicians like to say).
Even though Smale proved that eversion is possible, the first way to actually perform it
was discovered by the blind topologist Bernard Morin in 1961, based on ideas of Arnold

Dvipsbugw
intermezzo: bacteria, flies and knots 977

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 367 A way to turn a sphere inside out, with intermediate steps ordered
clockwise (© John Sullivan)

Shapiro. After him, several additional methods have been discovered.

Several computer videos of sphere eversions are now available.* The most famous ones
are Outside in, which shows an eversion due to William P. Thurston, and The Optiverse,
which shows the most efficient method known so far, discovered by a team led by John
Sullivan and shown in Figure 367.
Why is sphere eversion of interest to physicists? If elementary particles were extended
and at the same time were of spherical shape, eversion might be a symmetry of particles.
To make you think, we mention the effects of eversion on the whole surrounding space,
Copyright © Christoph Schiller November 1997–May 2006

not only on the sphere itself. The final effect of eversion is the transformation

(x, y, −z) R 2
(x, y, z) (645)
r2

where R is the radius of the sphere and r is the length of the coordinate vector (x, y, z),

* Summaries of the videos can be seen at the http://www.geom.umn.edu/docs/outreach/oi website, which

also has a good pedagogical introduction. Another simple eversion and explanation is given by Erik de Neve
on the http://www.xs4all.nl/~alife/sphere1.htm website. It is even possible to run the film software at home;
see the http://www.cslub.uwaterloo.ca/~mjmcguff/eversion website. Figure 367 is from the http://new.math.
uiuc.edu/optiverse website.

Dvipsbugw
 978 intermezzo: bacteria, flies and knots

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 368 The knot diagrams for the simplest prime knots (© Robert Scharein)

»
thus r = x 2 + y 2 + z 2 . Due to the minus sign in the z-coordinate, eversion is thus dif-
ferent from inversion, but not by too much. As we will find out shortly, a transformation
Page 1051 similar to eversion, space-time duality, is a fundamental symmetry of nature.

Knots, links and braids

Don’t touch this, or I shall tie your fingers into

“
Copyright © Christoph Schiller November 1997–May 2006

knots!
(Surprisingly efficient child education
technique.)

Knots and their generalization are central to the study of wobbly entity motion. A (math-
ematical) knot is a closed piece of rubber string, i.e. a string whose ends have been glued
together, which cannot be deformed into a circle or a simple loop. The simple loop is
”
also called the trivial knot. If knots are ordered by their crossing numbers, as shown in
Figure 368, the trivial knot (01 ) is followed by the trefoil knot (31 ) and by the figure-eight
knot (41 ). The figure only shows prime knots, i.e., knots that cannot be decomposed into
two knots that are connected by two parallel strands. In addition, the figure only shows
one of two possible mirror images.

Dvipsbugw
 intermezzo: bacteria, flies and knots 979

right-hand left-hand
crossing +1 crossing -1
Redemeister Redemeister Redemeister
move I move II move III

Dvipsbugw
the flype

a nugatory crossing
F I G U R E 369 Crossing types in F I G U R E 370 The Reidemeister moves and the flype
knots

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 371 The diagrams for the simplest links with two and three components (© Robert
Scharein)

Knots are of importance in the context of this intermezzo as they visualize the limit-
ations of the motion of wobbly entities. In addition, we will find other reasons to study
knots later on. In this section, we just have a bit of fun.*
How do we describe such a knot through the telephone? Mathematicians have spent
Copyright © Christoph Schiller November 1997–May 2006

a lot of time to figure out smart ways to achieve it. The simplest way is to flatten the knot
onto a plane and to list the position and the type (below or above) of the crossings.
Mathematicians are studying the simplest way to describe knots by the telephone. The
task is not completely finished, but the end is in sight. Of course, the flat diagrams can be
characterized by the minimal number of crossings. The knots in Figure 368 are ordered
in this way. There is 1 knot with zero, 1 with three and 1 with four crossings (not counting
Ref. 990 mirror knots); there are 2 knots with five and 3 with six crossings, 7 knots with seven,
21 knots with eight, 41 with nine, 165 with ten, 552 with eleven, 2176 with twelve, 9988

* Pretty pictures and other information about knots can be found on the KnotPlot site, i.e. at the http://www.
cs.ubc.ca/nest/imager/contributions/scharein/KnotPlot.html site.

Dvipsbugw
 980 intermezzo: bacteria, flies and knots

with thirteen, 46 972 with fourteen, 253 293 with fifteen and 1 388 705 knots with sixteen
crossings.
Mathematicians do not talk about ‘telephone messages’, they talk about invariants,
i.e. about quantities that do not depend on the precise shape of the knot. At present,
the best description of knots is a polynomial invariant based on a discovery by Vaughan
Jones in 1984. However, though the polynomial allows to uniquely describe most simple
knots, it fails to do so for more complex ones. But the Jones polynomial finally allowed
to prove that a diagram which is alternating and eliminates nugatory crossings (i.e. if it
is ‘reduced’) is indeed one which has minimal number of crossings. The polynomial also Dvipsbugw
allows to show that any two reduced alternating diagrams are related by a sequence of
flypes.
Together with the search for invariants, the tabulation of knots is a modern mathemat-
ical sport. In 1949, Schubert proved that every knot can be decomposed in a unique way
Ref. 990 as sum of prime knots. Knots thus behave similarly to integers.
The mirror image of a knot usually, but not always, is different from the original. If

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

you want a challenge, try to show that the trefoil knot, the knot with three crossings, is
different from its mirror image. The first proof was by Max Dehn in 1914.
Antiknots do not exist. An antiknot would be a knot on a rope that cancels out the
corresponding knot when the two are made to meet along the rope. It is easy to prove
that this is impossible. We take an infinite sequence of knots and antiknots on a string,
Ref. 992 K − K + K − K + K − K.... On one hand, we could make them disappear in this way
K − K + K − K + K − K... = (K − K)+(K − K)+(K − K)... = 0. On the other hand, we could
do the same thing using K −K +K −K +K −K... = K(−K +K)+(−K +K)+(−K +K)... = K.
The only knot K with an antiknot is thus the unknot K = 0.*

– CS – Several topics to be included – CS –

Since knots are stable in time, a knotted line in three dimensions is equivalent to a
knotted surface in space-time. When thinking in higher dimensions, we need to be care-
ful. Every knot (or knotted line) can be untied in four or more dimensions; however, there
is no surface embedded in four dimensions which has as t = 0 slice a knot, and as t = 1
slice the circle. Such a surface embedding needs at least five dimensions.
In higher dimensions, knots are possible only n-spheres are tied instead of circles; for
example, as just said, 2-spheres can be tied into knots in 4 dimensions, 3-spheres in 5
dimensions and so forth.
Copyright © Christoph Schiller November 1997–May 2006

Mathematicians also study more elaborate structures. Links are the generalization of
knots to several closed strands. Braids are the generalization of links to open strands.
Braids are especially interesting, as they form a group; can you state what the group op-
Challenge 1411 e eration is?

Knots in nature and on paper

Knots do not play a role only in shoe laces and in sailing boats.

* This proof does not work when performed with numbers; we would be able to deduce 1 = 0 by setting K=1.
Challenge 1410 n Why is this proof valid with knots but not with numbers?

Dvipsbugw
 intermezzo: bacteria, flies and knots 981

Figure to be included
Dvipsbugw
F I G U R E 372 A hagfish tied into F I G U R E 373 How to simulate order for long ropes
a knot

**
Proteins, the molecules that make up many cell structures, are chains of aminoacids. It

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 994 seems that very few proteins are knotted, and that most of these form trefoil knots. How-
ever, a figure-eight knotted protein has been discovered in 2000 by William Taylor.
**
Knots form also in other polymers. They seem to play a role in the formation of radicals in
carbohydrates. Research on knots in polymers is presently in full swing.
**
This is the simplest unsolved knot problem: Imagine an ideally wobbly rope, that is, a
rope that has the same radius everywhere, but whose curvature can be changed as one
Challenge 1412 r prefers. Tie a trefoil knot into the rope. By how much do the ends of the rope get nearer?
In 2006, there are only numerical estimates for the answer: about 10.1 radiuses. There is
no formula yielding the number 10.1. Alternatively, solve the following problem: what
is the rope length of a closed trefoil knot? Also in this case, only numerical values are
known – about 16.33 radiuses – but no exact formula. The same is valid for any other
knot, of course.
**
A famous type of eel, the knot fish Myxine glutinosa, also called hagfish or slime eel, is
Ref. 991 able to make a knot in his body and move this knot from head to tail. It uses this motion
Copyright © Christoph Schiller November 1997–May 2006

to cover its body with a slime that prevents predators from grabbing it; it also uses this
motion to escape the grip of predators, to get rid of the slime after the danger is over, and
to push against a prey it is biting in order to extract a piece of meat. All studied knot fish
form only left handed trefoil knots, by the way; this is another example of chirality in
nature.
**
One of the most incredible discoveries of recent years is related to knots in DNA molecules.
The DNA molecules inside cell nuclei can be hundreds of millions of base pairs long; they
regularly need to be packed and unpacked. When this is done, often the same happens
as when a long piece of rope or a long cable is taken out of a closet.

Dvipsbugw
 982 intermezzo: bacteria, flies and knots

It is well known that you can roll up a rope and put it into a closet in such a way that
it looks orderly stored, but when it is pulled out at one end, a large number of knots is
suddenly found. Figure 373 shows how to achieve this.
To make a long story short, this also happens to nature when it unpacks DNA in cell
nuclei. Life requires that DNA molecules move inside the cell nucleus without hindrance.
So what does nature do? Nature takes a simpler approach: when there are unwanted cross-
ings, it cuts the DNA, moves it over and puts the ends together again. In cell nuclei, there
are special enzymes, the so-called topoisomerases, which perform this process. The de-
tails of this fascinating process are still object of modern research. Dvipsbugw

**
The great mathematician Carl-Friedrich Gauß was the first person to ask what would hap-
pen when an electrical current I flows along a wire A linked with a wire B. He discovered
Ref. 995 a beautiful result by calculating the effect of the magnetic field of one wire onto the other.
Gauss found the expression

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(xA − xB )
ë BB = =n,
1 1
4πI ∫ dx
A
A
4π ∫ dx ë ∫ dx
A
A
B
B 
SxA − xB S3
(646)

where the integrals are performed along the wires. Gauss found that the number n does
not depend on the precise shape of the wires, but only on the way they are linked. De-
forming the wires does not change it. Mathematicians call such a number a topological
invariant. In short, Gauss discovered a physical method to calculate a mathematical in-
variant for links; the research race to do the same for other invariants, also for knots and
braids, is still going on today.
In the 1980s, Edward Witten was able to generalize this approach to include the nuclear
interactions, and to define more elaborate knot invariants, a discovery that brought him
the Fields medal.
**
Knots are also of importance at Planck scales, the smallest dimensions possible in nature.
We will soon explore how knots and the structure of elementary particles are related.

**
Knots appear rarely in nature. For example, tree roots do not seem to grow many knots
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1413 r during the lifetime of a plant. How do plants avoid this? In other words, why are there no
knotted bananas in nature?
**
If we move along the knot and count the crossings where we stay above and subtract the
number of crossings where we pass below, we get a number called the writhe of the knot.
It is not an invariant, but usually a tool in building them. The writhe is not necessarily
Challenge 1414 ny invariant under one of the three Reidemeister moves. Can you see which one? However,
the writhe is invariant under flypes.

Dvipsbugw
 intermezzo: bacteria, flies and knots 983

Clouds
Clouds are another important class of extended entities. The lack of a definite boundary
makes them even more fascinating than amoebas, bacteria or falling cats. We can observe
the varieties of clouds from an aeroplane. We also have encountered clouds as the basic
structure determining the size of atoms. Comparing these two and other types of clouds
teaches us several interesting things about nature.
Galaxies are clouds of stars; stars are clouds of plasma; the atmosphere is a gas cloud.
Page 597 Obviously, the common cumulus or cumulonimbus in the sky are vapour and water Dvipsbugw
droplet clouds. Clouds of all types can be described by a shape and a size, even though in
theory they have no bound. An effective shape can be defined by that region in which the
cloud density is only, say, 1 % of the maximum density; slightly different procedures can
also be used. All clouds are described by probability densities of the components making
up the cloud. All clouds show conservation of their number of constituents.
Whenever we see a cloud, we can ask why it does not collapse. Every cloud is an aggreg-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 183 ate. All aggregates are kept from collapse in only three ways: through rotation, through
pressure or through the Pauli principle, i.e. the quantum of action. Galaxies are kept from
collapsing by rotation. Most stars and the atmosphere are kept from collapsing by gas pres-
sure. Neutron stars, the Earth, atomic nuclei, protons or the electron clouds of atoms are
kept apart by the quantum of action.
A rain cloud can contain several thousand tons of water; can you explain what keeps
it afloat, and what else keeps it from continuously diffusing into a thinner and thinner
Challenge 1415 ny structure?
Two rain clouds can merge. So can two atomic electron clouds. But only atomic clouds
are able to cross each other. We remember that a normal atom can be inside a Rydberg
atom and leave it again without change. Rain clouds, stars, galaxies or other macroscopic
clouds cannot cross each other. When their paths cross, they can only merge or be ripped
into pieces. Due to this lack of crossing ability, it is in fact easier to count atomic clouds
than macroscopic clouds. In the macroscopic case, there is no real way to define a ‘single’
cloud in an accurate way. If we aim for full precision, we are unable to claim that there
is more than one rain cloud, as there is no clear-cut boundary between them. Electronic
clouds are different. True, in a piece of solid matter we can argue that there is only a single
electronic cloud throughout the object; however, when the object is divided, the cloud is
divided in a way that makes the original atomic clouds reappear. We thus can speak of
‘single’ electronic clouds.
Copyright © Christoph Schiller November 1997–May 2006

Let us explore the limits of the topic. In our definition of the term ‘cloud’ we assumed
that space and time are continuous. We also assumed that the cloud constituents were
localized entities. This does not have to be the case.

A one-dimensional classical analogue of the Schrödinger

equation
Fluid dynamics is a topic with many interesting aspects. A beautiful result from the 1960s
Ref. 974 is that a linear, deformable vortex in a rotating liquid is (almost) described by the one-
dimensional Schrödinger equation. A simple physical system of this type is the vortex that
can be observed in any emptying bath tub: it is extended in one dimension, and it wriggles
around. As we will see shortly, this wriggling motion is described by a Schrödinger-like

Dvipsbugw
 984 intermezzo: bacteria, flies and knots

w v

Dvipsbugw
F I G U R E 374 The mutually perpendicular
tangent e, normal n, torsion w and velocity v of
a vortex in a rotating fluid

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

equation.
Any deformable linear vortex, as illustrated in Figure 374, is described by a continuous
set of position vectors r(t, s) that depend on time t and on a single parameter s. The
parameter s specifies the relative position along the vortex. At each point on the vortex,
there is a unit tangent vector e(t, s), a unit normal curvature vector n(t, s) and a unit
torsion vector w(t, s). The three vectors, shown in Figure 374, are defined as usual as

e=
∂r
,
∂s
κn =
∂e
,
∂s
∂(e  n)
τw = − , (647)
∂s
where κ specifies the value of the curvature and τ specifies the value of the torsion. In
general, both numbers depend on time and on the position along the line.
We assume that the rotating environment induces a local velocity v for the vortex that
is proportional to the curvature κ, perpendicular to the tangent vector e and perpendic-
ular to the normal curvature vector n:
Copyright © Christoph Schiller November 1997–May 2006

v = ηκ(e  n) , (648)

Ref. 974 where η is the so-called coefficient of local self-induction that describes the coupling
between the liquid and the vortex motion.
Ref. 975 Any vortex described by the evolution equation (648) obeys the one-dimensional
Schrödinger equation. To repeat his argument, we assume that the filament is deformed
only slightly from the straight configuration. (Technically, we are thus in the linear re-
gime.) For such a filament, directed along the x-axis, we can write

r = (x, y(x, t), z(x, t)) . (649)

Dvipsbugw
intermezzo: bacteria, flies and knots 985

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 375 Motion of a vortex: the fundamental helical solution and a moving
helical ‘wave packet’

Slight deformations imply ∂s  ∂x and therefore

e = (1, , )  (1, 0, 0) ,
∂y ∂z
∂x ∂x
∂2 y ∂2 z
κn  (0, 2 , 2 ) , and
∂x ∂x
v = (0, , ) .
∂y ∂z
(650)
∂t ∂t
We can thus rewrite equation (648) as

∂2 z ∂2 y
(0, , ) = η (0, − 2 , 2 ) .
∂y ∂z
(651)
∂t ∂t ∂x ∂x
Copyright © Christoph Schiller November 1997–May 2006

This equation is well known; if we drop the first coordinate and introduce complex num-
bers by setting Φ = y + iz, we can rewrite it as

∂2 Φ
= iη 2 .
∂Φ
(652)
∂t ∂x
This is the one-dimensional Schrödinger equation for the evolution of a free wave func-
tion! The complex function Φ specifies the transverse deformation of the vortex. In other
words, we can say that the Schrödinger equation in one dimension describes the evolu-
tion of the deformation for an almost linear vortex surrounded by a rotating liquid. We
note that there is no constant ħ in the equation, as we are exploring a classical system.

Dvipsbugw
 986 intermezzo: bacteria, flies and knots

Schrödinger’s equation is linear in Φ. Therefore the fundamental solution is

Φ(x, y, z, t) = a e i(τx−ωt) with ω = ητ 2 and κ = aτ 2 ., (653)

The amplitude a and the wavelength or pitch b = 1~τ can be freely chosen, as long as
the approximation of small deviation is fulfilled; this condition translates as a  b.* In
the present interpretation, the fundamental solution corresponds to a vortex line that is
deformed into a helix, as shown in Figure 375. The angular speed ω is the rotation speed
Dvipsbugw
around the axis of the helix.
Challenge 1416 ny A helix moves along the axis with a speed given by

v helix along axis = 2ητ . (654)

In other words, for extended entities following evolution equation (648), rotation and
translation are coupled.** The momentum p can be defined using ∂Φ~∂x, leading to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

p=τ=
1
. (655)
b
Momentum is thus inversely proportional to the helix wavelength or pitch, as expected.
The energy E is defined using ∂Φ~∂t, leading to

E = ητ 2 =
η
. (656)
b2
Energy and momentum are connected by

p2
E= µ=
1
where . (657)
2µ 2η

In other words, a vortex with a coefficient η – describing the coupling between environ-
Page 984 ment and vortex – is thus described by a number µ that behaves like an effective mass. We
can also define the (real) quantity SΦS = a; it describes the amplitude of the deformation.
In the Schrödinger equation (652), the second derivative implies that the deformation
‘wave packet’ has tendency to spread out over space. Can you confirm that the wavelength–
frequency relation for a vortex wave group leads to something like the indeterminacy
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1418 ny relation (however, without a ħ appearing explicitly)?

In summary, the complex amplitude Φ for a linear vortex in a rotating liquid behaves
like the one-dimensional wave function of a non-relativistic free particle. In addition,
we found a suggestion for the reason why complex numbers appear in the Schrödinger
equation of quantum theory : they could be due to the intrinsic rotation of an underlying
substrate. We will see later on whether this is correct.

* The curvature is given by κ = a~b 2 , the torsion by τ = 1~b. Instead of a  b one can thus also write κ  τ.
Challenge 1417 ny ** A wave packet moves along the axis with a speed given by v packet = 2ητ 0 , where τ 0 is the torsion of the
helix of central wavelength.

Dvipsbugw
 intermezzo: bacteria, flies and knots 987

Fluid space-time
So far, we have looked at the motion of wobbly entities in continuous space-time. But
that is an unnecessary restriction. Looking at space-time itself in this way is also interest-
Ref. 996 ing. The most intriguing approach was published in 1995 by Ted Jacobson. He explored
what happens if space-time, instead of assumed to be continuous, is assumed to be the
statistical average of numerous components moving in a disordered fashion.
The standard description of general relativity describes space-time as an entity similar
Page 383 to a flexible mattress. Jacobson studied what happens if the mattress is assumed to be Dvipsbugw
made of a liquid. A liquid is a collection of (undefined) components moving randomly
and described by a temperature varying from place to place. He thus explored what hap-
pens if space-time is made of fluctuating entities.
Page 858 Jacobson started from the Fulling–Davies–Unruh effect and assumed that the local
temperature is given by the same multiple of the local gravitational acceleration. He also
used the proportionality – correct on horizons – between area and entropy. Since the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

energy flowing through a horizon can be called heat, one can thus translate the expression
δQ = TδS into the expression δE = aδA(c 2 ~4G), which describes the behaviour of space-
Page 1072 time at horizons. As we have seen, this expression is fully equivalent to general relativity.
In other words, imagining space-time as a liquid is a powerful analogy that allows to
deduce general relativity. Does this mean that space-time actually is similar to a liquid?
So far, the analogy is not sufficient to answer the question. In fact, just to confuse the
reader a bit more, there is an old argument for the opposite statement.

Solid space-time
The main reason to try to model empty space as a solid is a famous property of the mo-
tion of dislocations. To understand it, a few concepts need to be introduced. Dislocations
are one-dimensional construction faults in crystals, as shown in Figure 376. A general
dislocation is a mixture of the two pure dislocation types: edge dislocations and screw
dislocations. Both are shown in Figure 376. If one studies how the involved atoms can
rearrange themselves, one finds that edge dislocations can only move perpendicularly to
the added plane. In contrast, screw dislocations can move in all directions.* An import-
ant case of general, mixed dislocations, i.e. of mixtures of edge and screw dislocations,
are closed dislocation rings. On such a dislocation ring, the degree of mixture changes
continuously from place to place.
A dislocation is described by its strength and by its effective size; they are shown, re-
Copyright © Christoph Schiller November 1997–May 2006

spectively, in red and blue in Figure 376. The strength of a dislocation is measured by
the so-called Burgers vector; it measures the misfits of the crystal around the disloca-
tion. More precisely, the Burgers vector specifies by how much a section of perfect crystal
needs to be displaced, after it has been cut open, to produce the dislocation. Obviously,
the strength of a dislocation is quantized in multiples of a minimal Burgers vector. In
fact, dislocations with large Burgers vectors can be seen as composed of dislocations of
minimal Burgers vector.
The size or width of a dislocation is measured by an effective width w. Also the width

* See the http://uet.edu.pk/dmems/edge_dislocation.htm, http://uet.edu.pk/dmems/screw_dislocation.htm

and http://uet.edu.pk/dmems/mixed_dislocation.htm web pages for seeing a moving dislocation.

Dvipsbugw
988 intermezzo: bacteria, flies and knots

Dvipsbugw

effective size
Burgers vector

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 376 The two pure dislocation types: edge and screw dislocations

Copyright © Christoph Schiller November 1997–May 2006

is a multiple of the lattice vector. The width measures the size of the deformed region of
the crystal around the dislocation. Obviously, the size of the dislocation depends on the
elastic properties of the crystal, can take continuous values and is direction-dependent.

Dvipsbugw
 intermezzo: bacteria, flies and knots 989

The width is thus related to the energy content of a dislocation.

A general dislocation can move, though only in directions which are both perpendic-
ular to its own orientation and to its Burgers vector. Let us study this motion in more
detail. We call c the speed of sound in a pure (cubic) crystal. As Frenkel and Kontorowa
Ref. 997 found in 1938 it turns out that when a screw dislocation moves with velocity v, its width
w changes as
w=»
w0
. (658)
1 − v 2 ~c 2
Dvipsbugw
In addition, the energy of the moving dislocation obeys

E=»
E0
. (659)
1 − v 2 ~c 2

A screw dislocation thus cannot move faster than the speed of sound in a crystal and its

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

width shows a speed-dependent contraction. (Edge dislocations have similar, but more
complex behaviour.) The motion of screw dislocations in solids is described by the same
effects and formulae that describe the motion of bodies in special relativity; the speed of
sound is the limit speed for dislocations in the same way that the speed of light is the limit
speed for objects.
Does this mean that elementary particles are dislocations of space or even of space-
time, maybe even dislocation rings? The speculation is appealing, even though it supposes
that space-time is a solid, and thus contradicts the model of space or space-time as a fluid.
Page 1010 Worse, we will soon encounter good reasons to reject modelling space-time as a lattice;
Challenge 1419 n maybe you can find a few ones already by yourself. Still, expressions (658) and (659) for
dislocations continue to fascinate. For the time being, we do not study them further.

Swimming in curved space

There is an additional reason to see space as a liquid. It is possible to swim through
Ref. 998 empty space. This discovery was published in 2003 by Jack Wisdom. He found that cyclic
changes in the shape of a body can lead to net translation, a rotation of the body, or both.
Swimming in space-time does not happen at high Reynolds numbers. That would im-
ply that a system would be able to throw empty space behind it, and to propel itself for-
ward as a result. No such effects have ever been found. However, Jack Wisdom found
Copyright © Christoph Schiller November 1997–May 2006

a way to swim that corresponds to low Reynolds numbers, where swimming results of
simple shape change.
There is a simple system that shows the main idea. We know from Galilean physics
that on a frictionless surface it is impossible to move, but that it is possible to turn oneself.
This is true only for a flat surface. On a curved surface, one can use the ability to turn and
translate it into motion.
Take to massive discs that lie on the surface of a frictionless, spherical planet, as shown
in Figure 377. Consider the following four steps: The disc separation φ is increased by
the angle ∆φ, then the discs are rotated oppositely about their centres by the angle ∆θ,
their separation is decreased by −∆φ, and they are rotated back by −∆θ. Due to to the
Challenge 1420 ny conservation of angular momentum, the two-disc system changes its longitude ∆ψ as

Dvipsbugw
990 intermezzo: bacteria, flies and knots

y
φ
ψ θ
φ
Equator
Dvipsbugw
θ
x

F I G U R E 377 Swimming on a curved

surface using two discs

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 378 A large F I G U R E 379 Is this possible?
raindrop falling downwards

∆ψ = γ 2 ∆θ∆φ ,
1
(660)
2
where γ is the angular radius of the discs. This cycle can be repeated over and over. The
cycle it allows a body on the surface of the Earth, to swim along the surface. However, for
a body of metre size, the motion for each swimming cycle is only around 10−27 m.
Wisdom showed that the mechanism also works in curved space-time. The mechan-
ism thus allows a falling body to swim away from the path of free fall. Unfortunately, the
achievable distances for everyday objects are negligible. Nevertheless, the effect exists.
At this point, we are thoroughly confused. Space-time seems to be solid and liquid
at the same time. Despite this contrast, the situation gives the impression that extended,
wobbly and fluctuating entities might lead us towards a better understanding of the struc-
Copyright © Christoph Schiller November 1997–May 2006

ture of space and time. That exploration is left for the third and last part of our adventure.

Curiosities and fun challenges on wobbly entities

Any pair of shoes proves that we live on the

“ inside of a sphere. Their soles are worn out at

the ends, and hardly at all in between.
Anonymous

The topic of wobbly entities is full of fascinating details. Here are a few.

**
”

Dvipsbugw
 intermezzo: bacteria, flies and knots 991

Challenge 1421 n What is the shape of raindrops? Try to picture it. However, use your reason, not your
prejudice! By the way, it turns out that there is a maximum size for raindrops, with a
Ref. 999 value of about 4 mm. The shape of such a large raindrop is shown in Figure 378. Can you
imagine where the limit comes from?
For comparison, the drops in clouds, fog or mist are in the range of 1 to 100 µm, with
a peak at 10 to 15 µm. In those cases when all droplets are of similar size one and when
light is scattered only once by the droplets, one can observe coronae, glories or fogbows.
** Dvipsbugw
Challenge 1422 n What is the entity shown in Figure 379 – a knot, a braid or a link?
**
Challenge 1423 d Can you find a way to classify tie knots?
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1424 n Are you able to find a way to classify the way shoe laces can be threaded?

Outlook
We have studied one example of motion of extended bodies already earlier on: solitons.
We can thus sum up the possible motions of extended entities in four key themes. We first
studied solitons and interpenetration, then knots and their rearrangement, continued
with duality and eversion and finally explored clouds and extension. The sum of it all
seems to be half liquid and half solid.
The motion of wobbly bodies probably is the most neglected topic in all textbooks on
motion. Research is progressing at full speed; it is expected that many beautiful analo-
gies will be discovered in the near future. For example, in this intermezzo we have not
described any good analogy for the motion of light; similarly, including quantum theory
into the description of wobbly bodies’ motion remains a fascinating issue for anybody
Challenge 1425 d aiming to publish in a new field.
The ideas introduced in this intermezzo were sufficient to prepare us for the third part
of our ascent of Motion Mountain. We can now tackle the final part of our adventure.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
992 intermezzo: bacteria, flies and knots

Biblio graphy

969 Henk Tennekes, De wetten van de vliegkunst – over stijgen, dalen, vliegen en zweven, Ara-
mith Uitgevers, 1992. This clear and interesting text is also available in English. Cited on
page 970.
970 In 1997, the smallest human-made flying object was the helicopter built by a group of the
Institut für Mikrotechnik in Mainz, in Germany. A picture is available at their web page,
to be found at http://www.imm-mainz.de/English/billboard/f_hubi.html. The helicopter is
Dvipsbugw
24 mm long, weighs 400 mg and flies (though not freely) using two built-in electric mo-
tors driving two rotors, running at between 40 000 and 100 000 revolutions per minute. See
also the helicopter from Stanford University at http://www-rpl.stanford.edu/RPL/htmls/
mesoscopic/mesicopter/mesicopter.html, with an explanation of its battery problems. Cited
on page 970.
971 The most recent computational models of lift still describe only two-dimensional wing mo-
tion, e.g. Z.J. Wang, Two dimensional mechanism for insect hovering, Physical Review Let-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ters 85 pp. 2216–2219, 2000. A first example of a mechanical bird has been constructed by
Wolfgang Send; it can be studied on the http://www.aniprop.de website. See also W. Send,
Physik des Fliegens, Physikalische Blätter 57, pp. 51–58, June 2001. Cited on page 971.
972 R.B. Srygley & A.L.R. Thomas, Unconventional lift-generating mechanisms in free-
flying butterflies, Nature 420, pp. 660–664, 2002. Cited on page 971.
973 The simulation of insect flight using enlarged wing models flapping in oil instead of air
is described for example in http://www.dickinson.caltech.edu/research_robofly.html. The
higher viscosity of oil allows to achieve the same Reynolds number with larger sizes and
lower frequencies than in air. Cited on page 972.
974 George K. Batchelor, An Introduction to Fluid Mechanics, Cambridge University
Press, 1967, and H. Hashimoto, A soliton on a vortex filament, Journal of Fluid Mech-
anics 51, pp. 477–485, 1972. A summary is found in H. Zhou, On the motion of slender
vortex filaments, Physics of Fluids 9, p. 970-981, 1997. Cited on pages 983 and 984.
975 V.P. Dmitriyev, Helical waves on a vortex filament, American Journal of Physics 73,
pp. 563–565, 2005, and V.P. Dmitriyev, Mechanical analogy for the wave-particle: helix
on a vortex filament, http://www.arxiv.org/abs/quant-ph/0012008. Cited on page 984.
976 E. Purcell, Life at low Reynolds number, American Journal of Physics 45, p. 3, 1977. Cited
on page 973.
977 Most bacteria are flattened, ellipsoidal sacks kept in shape by the membrane enclosing the
cytoplasma. But there are exceptions; in salt water, quadratic and triangular bacteria have
been found. More is told in the corresponding section in the interesting book by Bernard
Copyright © Christoph Schiller November 1997–May 2006

Dixon, Power Unseen – How Microbes Rule the World, W.H. Freeman, New York, 1994.
Cited on page 974.
978 S. Pitnick, G. Spicer & T.A. Markow, How long is a giant sperm?, Nature 375, p. 109,
1995. Cited on page 974.
979 For an overview of the construction and the motion of coli bacteria, see H.C. Berg, Motile
behavior of bacteria, Physics Today 53, pp. 24–29, January 2000. Cited on page 975.
980 M. Kawamura, A. Sugamoto & S. Nojiri, Swimming of microorganisms viewed
from string and membrane theories, Modern Journal of Physics Letters A 9, pp. 1159–1174,
1994. Also available as http://www.arxiv.org/abs/hep-th/9312200. Cited on page 974.
981 They are also called prokaryote flagella. See for example S.C. Schuster & S. Khan,
The bacterial flagellar motor, Annual Review of Biophysics and Biomolecular Structure 23,

Dvipsbugw
bibliography 993

pp. 509–539, 1994, or S.R. Capl an & M. Kara-Ivanov, The bacterial flagellar motor,
International Review of Cytology 147, pp. 97–164, 1993. See also the information on the
topic that can be found on the website http://www.id.ucsb.edu:16080/fscf/library/origins/
graphics-captions/flagellum.html. Cited on page 974.
982 This is from the book by David Dusenbery, Life at a Small Scale, Scientific American
Library, 1996. Cited on page 975.
983 A. Shapere & F. Wilczek, Gauge kinematics of deformable bodies, American Journal
of Physics 57, pp. 514–518, 1989, A. Shapere & F. Wilczek, Geometry of self-propulsion
at low Reynolds number, Journal of Fluid Mechanics 198, pp. 557–585, 1989, A. Shapere Dvipsbugw
& F. Wilczek, Efficiencies of self-propulsion at low Reynolds number, Journal of Fluid
Mechanics 198, pp. 587–599, 1989. See also R. Montgomery, Gauge theory of the falling
cat, Field Institute Communications 1, pp. 75–111, 1993. No citations.
984 W. Nu tsch & U. Rüffer, Die Orientierung freibeweglicher Organismen zum Licht,
dargestellt am Beispiel des Flagellaten Chlamydomonas reinhardtii, Naturwissenschaften 81,
pp. 164–174, 1994. Cited on page 974.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

985 A lot here is from Louis H. Kauffman, Knots and Physics, World Scientific, second
edition, 1994, which gives a good, simple, and visual introduction to the mathematics of
knots and its main applications to physics. No citations.
986 S. Smale, A classification of immersions of the two-sphere, Transactions of the American
Mathematical Society 90, pp. 281–290, 1958. No citations.
987 G.K. Francis & B. Morin, Arnold Shapiro’s Eversion of the Sphere, Mathematical Intel-
ligencer pp. 200–203, 1979. See also Ref. 1000. No citations.
988 Silvio Lev y, Delle Max well & Tamara Munzner, Making Waves – a Guide to
the Ideas Behind Outside In, Peters, 1995. No citations.
989 B. Morin & J.-P. Petit, Le retournement de la sphere, Pour la Science 15, pp. 34–41, 1979.
See also the clear article by A. Phillips, Turning a surface inside out, Scientific American
pp. 112–120, May 1966. No citations.
990 A good introduction to knots is the paper by J. Hoste, M. Thistlethwaite & J.
Weeks, The first 1,701,936 knots, The Mathematical Intelligencer 20, pp. 33–47, 1998. Cited
on pages 979 and 980.
991 Alexei Sossinsky, Nœuds – histoire d’une théorie mathématique, Editions du Seuil, 1999.
D. Jensen, Le poisson noué, Pour la science, dossier hors série, pp. 14–15, April 1997. Cited
on page 981.
992 I. Stewart, Game, Set and Math, Penguin Books, 1989, pp. 58–67. Cited on page 980.
993 For some modern knot research, see P. Holdin, R.B. Kusner & A. Stasiak, Quantiz-
ation of energy and writhe in self-repelling knots, New Journal of Physics 4, pp. 20.1–20.11,
Copyright © Christoph Schiller November 1997–May 2006

2002. No citations.
994 W.R. Taylor, A deeply knotted protein structure and how it might fold, Nature 406,
pp. 916–919, 2000. Cited on page 981.
995 A.C. Hirshfeld, Knots and physics: Old wine in new bottles, American Journal of Physics
66, pp. 1060–1066, 1998. Cited on page 982.
996 T. Jacobson, Thermodynamics of spacetime: the Einstein equation of state, Physical Re-
view Letters 75, pp. 1260–1263, 1995, or http://www.arxiv.org/abs/gr-qc/9504004. Cited on
page 987.
997 J. Frenkel & T. Kontorowa, Über die Theorie der plastischen Verformung, Physikalis-
che Zeitschrift der Sowietunion 13, pp. 1–10, 1938. F.C. Frank, On the equations of mo-
tion of crystal dislocations, Proceedings of the Physical Society A 62, pp. 131–134, 1949, J.

Dvipsbugw
994 x intermezzo: bacteria, flies and knots

Eshelby, Uniformly moving dislocations, Proceedings of the Physical Society A 62, pp. 307–
314, 1949. See also G. Leibfried & H. Dietze, Zeitschrift für Physik 126, p. 790, 1949. A
general introduction can be found in A. Seeger & P. Schiller, Kinks in dislocations
lines and their effects in internal friction in crystals, Physical Acoustics 3A, W.P. Mason,
ed., Academic Press, 1966. See also the textbooks by Frank R.N. Nabarro, Theory of
Crystal Dislocations, Oxford University Press, 1967, or J.P. Hirth & J. Lothe, Theory of
Dislocations, McGraw Hills Book Company, 1968. Cited on page 989.
998 J. Wisd om, Swimming in spacetime: motion by cyclic changes in body shape, Science 299,
pp. 1865–1869, 21st of March, 2003. Cited on page 989. Dvipsbugw
999 H.R. Pruppacher & J.D. Klett, Microphysics of Clouds and Precipitation, Reidel, 1978,
pp. 316–319. Falling drops are flattened and look like a pill, due to the interplay between
surface tension and air flow. Cited on page 991.
1000 A manual for the ways to draw manifolds is the unique text by George Francis, The
Topological Picturebook, Springer Verlag, 1987. It also contains a chapter on sphere eversion.
Cited on page 993.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1001 M.Yu. Konstantinov, Some problems of topology change description in the theory of
space-time, http://www.arxiv.org/abs/gr-qc/9508050 preprint. No citations.
1002 A. Unzicker, Teleparallel space-time with torsion yields geometrization of electro-
dynamcis with quantised charges, http://www.arxiv.org/abs/gr-qc/9612061 preprint. No
citations.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
Third Part

Motion Without Motion:

What Are Space, Time
and Particles?

Where through the combination of

quantum mechanics and general relativity,
the top of Motion Mountain is reached
and it is discovered
that vacuum is indistinguishable from matter,
that space, time and mass are easily confused,
that there is no difference between the very large and the very small,
and that a complete description of motion is possible.
(Well, wait a few more years for the last line.)
 C h a p t e r XI

GENER AL R EL ATI V I T Y V ER SUS

QUANTUM MEC HANIC S
Dvipsbugw

Man muß die Denkgewohnheiten durch

“ Denknotwendigkeiten ersetzen.*
Albert Einstein

”
T
he two stories told in the two parts of the path we have followed up to now, namely
hat on general relativity and that on quantum field theory, are both beautiful and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

horoughly successful. Both are confirmed by experiments. We have reached a con-
siderable height in our mountain ascent. The precision we achieved in the description of
nature is impressive, and we are now able to describe all known examples of motion. So
far we have encountered no exceptions.
However, the most important aspects of any type of motion, the masses of the particles
involved and the strength of their coupling, are still unexplained. Furthermore, the origin
of the number of particles in the universe, their initial conditions and the dimensionality
of space-time remain hidden from us. Obviously, our adventure is not yet complete.
This last part of our hike will be the most demanding. In the ascent of any high moun-
tain, the head gets dizzy because of the lack of oxygen. The finite energy at our disposal
requires that we leave behind all unnecessary baggage and everything that slows us down.
In order to determine what is unnecessary, we need to focus on what we want to achieve.
Our aim is the precise description of motion. But even though the general relativity and
quantum theory are extremely precise, we carry are a burden: the two theories and their
concepts contradict each other. To pinpoint this useless baggage, we first list these contra-
dictions.

The contradictions
In classical physics and in general relativity, the vacuum, or empty space-time, is a region
Copyright © Christoph Schiller November 1997–May 2006

with no mass, no energy and no momentum. If matter or gravitational fields are present,
space-time is curved. The best way to measure the mass or energy content of space-time
Ref. 1003 is to measure the average curvature of the universe. Cosmology tells us how we can do
Page 460 this; measurements yield an average energy density of the ‘vacuum’ of

E~V  1 nJ~m3 . (661)

Ref. 1004 However, quantum field theory tells a different story. Vacuum is a region with zero-point
fluctuations. The energy content of vacuum is the sum of the zero-point energies of all

* ‘One needs to replace habits of thought by necessities of thought.’

Dvipsbugw
 the contradictions 997

the fields it contains. Indeed, the Casimir effect ‘proves’ the reality of these zero-point
Page 858 energies. Their energy density is given, within one order of magnitude, by

= 3 ν 3 dν =
E 4πh ν max πh 4
V c ∫0
ν
c 3 max
. (662)

The approximation is valid for the case in which the cut-off frequency ν max is much lar-
ger than the rest mass m of the particles corresponding to the field under consideration.
Particle physicists argue that the cut-off energy has to be at least the energy of grand Dvipsbugw
unification, about 1016 GeV= 1.6 MJ. That would give a vacuum energy density of

 1099 J~m3 ,
E
(663)
V

which is about 10108 times higher than the experimental limit deduced from spatial

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

curvature using general relativity estimates. In other words, something is slightly wrong
here.
General relativity and quantum theory contradict each other in other ways. Gravity is
Ref. 1005 curved space-time. Extensive research has shown that quantum field theory, the descrip-
tion of electrodynamics and of nuclear forces, fails for situations with strongly curved
space-times. In these cases the concept of ‘particle’ is not uniquely defined; quantum field
theory cannot be extended to include gravity consistently and thus to include general re-
lativity. Without the concept of the particle as a countable entity, the ability to perform per-
turbation calculations is also lost; and these are the only calculations possible in quantum
field theory. In short, quantum theory only works because it assumes that gravity does
not exist! Indeed, the gravitational constant does not appear in any consistent quantum
field theory.
On the other hand, general relativity neglects the commutation rules between physical
quantities discovered in experiments on a microscopic scale. General relativity assumes
that the position and the momentum of material objects can be given the meaning that
they have in classical physics. It thus ignores Planck’s constant ħ and only works by neg-
lecting quantum theory.
Measurements also lead to problems. In general relativity, as in classical physics, it is
assumed that infinite precision of measurement is possible, e.g. by using finer and finer
Page 655 ruler marks. In contrast, in quantum mechanics the precision of measurement is limited.
Copyright © Christoph Schiller November 1997–May 2006

Page 828 The indeterminacy principle gives the limits that result from the mass M of the apparatus.
Time shows the contradictions most clearly. Relativity explains that time is what is
read from clocks. Quantum theory says that precise clocks do not exist, especially if the
coupling with gravitation is included. What does waiting 10 minutes mean, if the clock
goes into a quantum mechanical superposition as a result of its coupling to space-time
geometry?
In addition, quantum theory associates mass with an inverse length via the Compton
wavelength; general relativity associates mass with length via the Schwarzschild radius.
Similarly, general relativity shows that space and time cannot be distinguished,
whereas quantum theory says that matter does make a distinction. Quantum theory is a
theory of – admittedly weirdly constructed – local observables. General relativity doesn’t

Dvipsbugw
 998 xi general rel ativity versus quantum mechanics

Page 494 have any local observables, as Einstein’s hole argument shows.
Most dramatically, the contradiction is shown by the failure of general relativity to de-
Ref. 1006, Ref. 1007 scribe the pair creation of particles with spin 1/2, a typical and essential quantum process.
John Wheeler and others have shown that, in such a case, the topology of space necessar-
Ref. 1008, Ref. 1009 ily has to change; in general relativity, however, the topology of space is fixed. In short,
quantum theory says that matter is made of fermions, while general relativity cannot in-
corporate fermions.
To sum up, general relativity and quantum theory clash. As long as an existing descrip-
tion of nature contains contradictions, it cannot lead to a unified description, to useful Dvipsbugw
explanations, or even to a correct description. In order to proceed, let us take the shortest
and fastest path: let us investigate the contradictions in more detail.

33. d oes mat ter differ from vacuum?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 1010 There is a simple way to state the origin of all contradictions between general relativ-
ity and quantum mechanics.* Both theories describe motion with objects made up of
particles and with space-time made up of events. Let us see how these two concepts are
defined.
A particle – and in general any object – is defined as a conserved entity to which a
position can be ascribed and which can move. (The etymology of the term ‘object’ is
connected to the latter fact.) In other words, a particle is a small entity with conserved
mass, charge etc., which can vary its position with time.
Ref. 1011 In every physics text time is defined with the help of moving objects, usually called
‘clocks’, or with the help of moving particles, such as those emitted by light sources. Sim-
ilarly, the length is defined in terms of objects, either with an old-fashioned ruler or with
the help of the motion of light, which in turn is motion of particles.
Modern physics has further sharpened the definitions of particle and space-time.
Quantum mechanics assumes that space-time is given (it is included as a symmetry of the
Hamiltonian), and studies the properties and the motion of particles, both for matter and
for radiation. General relativity, and especially cosmology, takes the opposite approach:
it assumes that the properties of matter and radiation are given, e.g. via their equations
of state, and describes in detail the space-time that follows from them, in particular its
curvature.
However, one fact remains unchanged throughout all these advances in physics: the
Copyright © Christoph Schiller November 1997–May 2006

two concepts of particles and of space-time are each defined with the help of the other. To
avoid the contradiction between quantum mechanics and general relativity and to elim-
inate their incompleteness requires the elimination of this circular definition. As argued
in the following, this necessitates a radical change in our description of nature, and in
particular of the continuity of space-time.
For a long time, the contradictions between the two descriptions of nature were
avoided by keeping them separate. One often hears the statement that quantum mech-

* The main results of this section are standard knowledge among specialists of unification; there are given
here in simple arguments. For another way to derive the results, see the summary section on limit statements
in nature, on page 1068.

Dvipsbugw
 d oes matter differ from vacuum? 999

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

F I G U R E 380 ‘Tekenen’ by Maurits Escher, 1948 – a metaphor
for the way in which ‘particles’ and ‘space-time’ are usually
defined: each with the help of the other (© M.C. Escher Heirs)

anics is valid at small dimensions and general relativity is valid at large dimensions, but
this artificial separation is not justified; worse, it prevents the solution of the problem. The
situation resembles the well-known drawing (Figure 380) by Maurits Escher (1898–1972)
where two hands, each holding a pencil, seem to be drawing each other. If one hand is
taken as a symbol of space-time and the other as a symbol of particles, with the act of
drawing taken as a symbol of the act of defining, the picture gives a description of stand-
ard twentieth century physics. The apparent contradiction is solved by recognizing that
the two concepts (the two hands) result from a third, hidden concept from which the
other two originate. In the picture, this third entity is the hand of the painter.
In the case of space-time and matter, the search for the underlying common concept
Ref. 1012, Ref. 1013 is presently making renewed progress. The required conceptual changes are so dramatic
Ref. 1014, Ref. 1015 that they should be of interest to anybody who has an interest in physics. The most ef-
Copyright © Christoph Schiller November 1997–May 2006

fective way to deduce the new concepts is to focus in detail on that domain where the
contradiction between the two standard theories becomes most dramatic and where both
theories are necessary at the same time. That domain is given by a well-known argument.

Planck scales
Both general relativity and quantum mechanics are successful theories for the description
of nature. Each provides a criterion for determining when classical Galilean physics is no
longer applicable. (In the following, we use the terms ‘vacuum’ and ‘empty space-time’
interchangeably.)
General relativity shows that it is necessary to take into account the curvature of space-
time whenever we approach an object of mass m to within a distance of the order of the

Dvipsbugw
 1000 xi general rel ativity versus quantum mechanics

TA B L E 74 The size, Schwarzschild radius and Compton wavelength of some objects appearing in
nature. The lengths between quotes make no physical sense, as explained in the text.

Object Size: Mass m S c h wa r z - R a t i o C o m p t o n R at i o

diam. schild d~r S wav e - d~λ C
d r a d i u s rS length
λC
galaxy  1 Zm  5 ë 1040 kg  70 Tm  107  10−83 m  10104
neutron star 10 km 2.8 ë 1030 kg 4.2 km 2.4 ‘1.3 ë 10−73 m’ 8.0 ë 1076 Dvipsbugw
Sun 1.4 Gm 2.0 ë 1030 kg 3.0 km 4.8 ë 105 ‘1.0 ë 10−73 m’ 8.0 ë 1081
Earth 13 Mm 6.0 ë 1024 kg 8.9 mm 1.4 ë 109 ‘5.8 ë 10−68 m’ 2.2 ë 1074
human 1.8 m 75 kg 0.11 ym 1.6 ë 1025 ‘4.7 ë 10−45 m’ 3.8 ë 1044
molecule 10 nm 0.57 zg ‘8.5 ë 10−52 m’ 1.2 ë 1043 6.2 ë 10−19 m 1.6 ë 1010
atom (12 C) 0.6 nm 20 yg ‘3.0 ë 10−53 m’ 2.0 ë 1043 1.8 ë 10−17 m 3.2 ë 107
proton p 2 fm 1.7 yg ‘2.5 ë 10−54 m’ 8.0 ë 1038 2.0 ë 10−16 m 9.6

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pion π 2 fm 0.24 yg ‘3.6 ë 10−55 m’ 5.6 ë 1039 1.5 ë 10−15 m 1.4
up-quark u < 0.1 fm 0.6 yg ‘9.0 ë 10−55 m’ < 1.1 ë 1038 5.5 ë 10−16 m < 0.18
electron e < 4 am 9.1 ë 10−31 kg ‘1.4 ë 10−57 m’ 3.0 ë 1039 3.8 ë 10−13 m < 1.0 ë 10−5
neutrino ν e < 4 am < 3.0 ë 10−35 kg ‘< 4.5 ë 10−62 m’ n.a. A 1.1 ë 10−8 m < 3.4 ë 10−10

Schwarzschild radius r S , given by

r S = 2Gm~c 2 . (664)

The gravitational constant G and the speed of light c act as conversion constants. Indeed,
as the Schwarzschild radius of an object is approached, the difference between general re-
lativity and the classical 1~r 2 description of gravity becomes larger and larger. For example,
the barely measurable gravitational deflection of light by the Sun is due to approaching
Ref. 1006, Ref. 1016 it to within 2.4 ë 105 times its Schwarzschild radius. Usually however, we are forced to
stay away from objects at a distance that is an even larger multiple of the Schwarzschild
radius, as shown in Table 74. For this reason, general relativity is unnecessary in everyday
life. (An object smaller than its own Schwarzschild radius is called a black hole. Accord-
ing to general relativity, no signals from inside the Schwarzschild radius can reach the
outside world; hence the name ‘black hole’.)
Copyright © Christoph Schiller November 1997–May 2006

Ref. 1017

Similarly, quantum mechanics shows that Galilean physics must be abandoned and
quantum effects must be taken into account whenever an object is approached to within
distances of the order of the (reduced) Compton wavelength λ C , given by

λC =
ħ
. (665)
mc

In this case, Planck’s constant h and the speed of light c act as conversion factors to trans-
form the mass m into a length scale. Of course, this length only plays a role if the object
itself is smaller than its own Compton wavelength. At these dimensions we get relativistic

Dvipsbugw
 d oes matter differ from vacuum? 1001

quantum effects, such as particle–antiparticle creation or annihilation. Table 74 shows

that the approach distance is near or smaller than the Compton wavelength only in the
microscopic world, so that such effects are not observed in everyday life. We do not there-
fore need quantum field theory to describe common observations.
The combined concepts of quantum field theory and general relativity are required in
situations in which both conditions are satisfied simultaneously. The necessary approach
distance for such situations is calculated by setting r S = 2λ C (the factor 2 is introduced
for simplicity). We find that this is the case when lengths or times are (of the order of)
Dvipsbugw
¼
l Pl = = 1.6 ë 10−35 m, the Planck length,
¼
ħG~c 3
(666)
t Pl = ħG~c 5 = 5.4 ë 10−44 s , the Planck time.

Whenever we approach objects at these scales, both general relativity and quantum mech-
anics play a role; at these scales effects of quantum gravity appear. Because the values of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the Planck dimensions are extremely small, this level of sophistication is unnecessary in
everyday life, in astronomy and even in particle physics.
However, to answer the questions posted at the beginning of the book – why do we live
in three dimensions and why is the proton 1836.15 times heavier than the electron? – we re-
quire a precise and complete description of nature. The contradictions between quantum
mechanics and general relativity appear to make the search for answers impossible. How-
ever, while the unified theory describing quantum gravity is not yet complete, we can
already get a few glimpses at its implications from its present stage of development.
Note that the Planck scales specify one of only two domains of nature where quantum
Challenge 1426 n mechanics and general relativity apply at the same time. (What is the other?) As Planck
scales are the easier of the two to study, they provide the best starting point for the fol-
Ref. 1018 lowing discussion. When Max Planck discovered the existence of Planck scales or Planck
units, he was interested in them mainly as natural units of measurement, and that is what
he called them. However, their importance in nature is much more widespread, as we
will shall see in the new section. We will discover that they determine what is commonly
called quantum geometry.

Farewell to instants of time

Time is composed of time atoms ... which in fact

“ are indivisible.
Copyright © Christoph Schiller November 1997–May 2006

Moses Maimonides, 12th century.

The appearance of the quantum of action in the description of motion leads to quantum
limits to all measurements. These limits have important consequences at Planck dimen-
sions. Measurement limits appear most clearly when we investigate the properties of
”
clocks and metre rules. Is it possible to construct a clock that is able to measure time
Ref. 1019, Ref. 1020 intervals shorter than the Planck time? Surprisingly, the answer is no, even though the
time–energy indeterminacy relation ∆E∆t E ħ seems to indicate that by making ∆E ar-
bitrary large, we can make ∆t arbitrary small.
Every clock is a device with some moving parts. Parts can be mechanical wheels,
particles of matter in motion, changing electrodynamic fields, i.e. photons, or decaying

Dvipsbugw
 1002 xi general rel ativity versus quantum mechanics

Ref. 1021, Ref. 1022 radioactive particles. For each moving component of a clock, such as the two hands, the
Ref. 1023 indeterminacy principle applies. As discussed most clearly by Michael Raymer, the inde-
terminacy relation for two non-commuting variables describes two different, but related
situations: it makes a statement about standard deviations of separate measurements on
many identical systems; and it describes the measurement precision for a joint measure-
ment on a single system. Throughout this article, only the second situation is considered.
For any clock to work, we need to know both the time and the energy of each hand.
Otherwise it would not be a recording device. Put more generally, a clock must be a clas-
sical system. We need the combined knowledge of the non-commuting variables for each Dvipsbugw
moving component of the clock. Let us focus on the component with the largest time in-
determinacy ∆t. It is evident that the smallest time interval δt that can be measured by a
clock is always larger than the quantum limit, i.e. larger than the time indeterminacy ∆t
for the most ‘uncertain’ component. Thus we have

δt E ∆t E
ħ
,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(667)
∆E
where ∆E is the energy indeterminacy of the moving component, and this energy inde-
terminacy ∆E must be smaller than the total energy E = mc 2 of the component itself.*
Furthermore, a clock provides information and thus signals have to be able to leave it.
To make this possible, the clock must not be a black hole and its mass m must therefore
be smaller than the Schwarzschild mass for its size, i.e. m D c 2 l~G, where l is the size of
the clock (neglecting factors of order unity). Finally, for a sensible measurement of the
time interval δt, the size l of the clock must be smaller than c δt itself, because otherwise
different parts of the clock could not work together to produce the same time display.**
If we combine all these conditions, we get

δt E
ħG
(668)
c 5 δt
or ¾
δt E = t Pl .
ħG
(669)
c5

In summary, from three simple properties of any clock, namely that there is only a single
clock, that we can read its dial and that it gives sensible read-outs, we get the general
Copyright © Christoph Schiller November 1997–May 2006

conclusion that clocks cannot measure time intervals shorter than the Planck time. Note
that this argument is independent of the nature of the clock mechanism. Whether the
clock is powered by gravitational, electrical, plain mechanical or even nuclear means, the
limit still applies.***
Ref. 1027 The same result can also be found in other ways. For example, any clock small enough
to measure small time intervals necessarily has a certain energy indeterminacy due to

* Physically, this condition means being sure that there is only one clock; the case ∆E A E would mean that
it is impossible to distinguish between a single clock and a clock–anticlock pair created from the vacuum,
or a component plus two such pairs, etc.
** It is amusing to explore how a clock larger than c δt would stop working, as a result of the loss of rigidity

Dvipsbugw
 d oes matter differ from vacuum? 1003

the indeterminacy relation. At the same time, on the basis of general relativity, any energy
density induces a deformation of space-time and signals from the deformed region arrive
Page 410 with a certain delay due to that deformation. The energy indeterminacy of the source
leads to an indeterminacy in the deformation and thus in the delay. The expression from
Ref. 1016 general relativity for the deformation of the time part of the line element due to a mass
m is δt = mG~l c 3 . From the mass–energy relation, an energy spread ∆E produces an
indeterminacy ∆t in the delay
∆t =
∆E G
. (670) Dvipsbugw
l c5
This indeterminacy determines the precision of the clock. Furthermore, the energy in-
determinacy of the clock is fixed by the indeterminacy relation for time and energy
∆E E ħ~∆t, in turn fixed by the precision of the clock. Combining all this, we again
find the relation δt E t Pl for the minimum measurable time. We are forced to conclude
that in nature there is a minimum time interval. In other words, at Planck scales the term

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

‘instant of time’ has no theoretical or experimental basis. It therefore makes no sense to use
the term.

Farewell to points in space

In a similar way, we can deduce that it is impossible to make a metre rule or any other
Ref. 1028 length measuring device that is able to measure lengths shorter than the Planck length.
Obviously, we can already deduce this from l Pl = c t Pl , but a separate proof is also possible.
The straightforward way to measure the distance between two points is to put an object
at rest at each position. In other words, joint measurements of position and momentum
are necessary for every length measurement. Now, the minimal length δl that can be
measured must be larger than the position indeterminacy of the two objects. From the
indeterminacy principle it is known that each object’s position cannot be determined with
a precision ∆l better than that given by the indeterminacy relation ∆l ∆p = ħ, where ∆p
is the momentum indeterminacy. The requirement that there is only one object at each
end, i.e. avoiding pair production from the vacuum, means that ∆p < mc; together, these
requirements give
δl E ∆l E
ħ
. (671)
mc
Furthermore, the measurement cannot be performed if signals cannot leave the objects;
Copyright © Christoph Schiller November 1997–May 2006

thus they may not be black holes. Therefore their masses must be small enough for their
Schwarzschild radius r S = 2Gm~c 2 to be smaller than the distance δl separating them.

Challenge 1427 n in its components.

*** Note that gravitation is essential here. The present argument differs from the well-known study on the
limitations of clocks due to their mass and their measuring time which was published by Salecker and
Ref. 1025
1024 Wigner and summarized in pedagogical form by Zimmerman. Here, both quantum mechanics and grav-
ity are included, and therefore a different, lower and much more fundamental limit is found. Note also that
Ref. 1026 the discovery of black hole radiation does not change the argument; black hole radiation notwithstanding,
measurement devices cannot exist inside black holes.

Dvipsbugw
 1004 xi general rel ativity versus quantum mechanics

Again omitting the factor of 2, we get

¾
δl E = l Pl .
ħG
(672)
c3

Another way to deduce this limit reverses the roles of general relativity and quantum the-
ory. To measure the distance between two objects, we have to localize the first object with
respect to the other within a certain interval ∆x. The corresponding energy indetermin-
acy obeys ∆E = c(c 2 m 2 + (∆p)2 )1~2 E cħ~∆x. However, general relativity shows that a
Dvipsbugw
Ref. 1006, Ref. 1016 small volume filled with energy changes the curvature of space-time, and thus changes
the metric of the surrounding space. For the resulting distance change ∆l, compared to
Ref. 1028, Ref. 1029, empty space, we find the expression ∆l  G∆E~c 4 . In short, if we localize the first particle
Ref. 1030 in space with a precision ∆x, the distance to a second particle is known only with preci-
Ref. 1031, Ref. 1032, sion ∆l. The minimum length δl that can be measured is obviously larger than either of
Ref. 1033 these quantities; inserting the expression for ∆E, we find again that the minimum meas-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

urable length δl is given by the Planck length.
We note that, as the Planck length is the shortest possible length, it follows that there
can be no observations of quantum mechanical effects for situations in which the corres-
ponding de Broglie or Compton wavelength is smaller than the Planck length. In proton–
proton collisions we observe both pair production and interference effects. In contrast,
the Planck limit implies that in everyday, macroscopic situations, such as car–car colli-
sions, we cannot observe embryo–antiembryo pair production and quantum interference
effects.
In summary, from two simple properties common to all length measuring devices,
namely that they can be counted and that they can be read out, we arrive at the conclusion
that lengths smaller than the Planck length cannot be found in measurements. Whatever
method is used, be it a metre rule or time-of-flight measurement, we cannot overcome
this fundamental limit. It follows that the concept of a ‘point in space’ has no experimental
basis. In the same way, the term ‘event’, being a combination of a ‘point in space’ and an
‘instant of time’, also loses its meaning for the description of nature.
A simple way to deduce the minimum length using the limit statements which struc-
ture this ascent is the following. General relativity is based on a maximum force in nature,
Page 349 or alternatively, on a minimum mass change per time; its value is given by dm~dt = c 3 ~4G.
Quantum theory is based on a minimum action in nature, given by L = ħ~2. Since a dis-
tance d can be expressed like
Copyright © Christoph Schiller November 1997–May 2006

d2 =
L
, (673)
dm~dt

one sees directly that a minimum action and a maximum mass rate imply a minimum
distance. In other words, quantum theory and general relativity, when put together, imply
a minimum distance.
Ref. 1034 These results are often expressed by the so-called generalized indeterminacy principle

∆p∆x E ħ~2 + f (∆p)2

G
(674)
c3

Dvipsbugw
 d oes matter differ from vacuum? 1005

or
2
∆p∆x E ħ~2 + f (∆p)2 ,
l Pl
(675)
ħ

where f is a numerical factor of order unity. A similar expression holds for the time–
energy indeterminacy relation. The first term on the right hand side is the usual quantum
mechanical indeterminacy. The second term, negligible for everyday life energies, plays a
role only near Planck energies and is due to the changes in space-time induced by gravity
Challenge 1428 e at these high energies. You should be able to show that the generalized principle (674) Dvipsbugw
automatically implies that ∆x can never be smaller than f 1~2 l Pl .
The generalized indeterminacy principle is derived in exactly the same way in which
Heisenberg derived the original indeterminacy principle ∆p∆x E ħ~2, namely by study-
Ref. 1034 ing the deflection of light by an object under a microscope. A careful re-evaluation of the
Ref. 1035 process, this time including gravity, yields equation (674). For this reason, all approaches
Ref. 1036 that try to unify quantum mechanics and gravity must yield this relation; indeed, it

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

appears in the theory of canonical quantum gravity, in superstring theory and in the
Ref. 1037, Ref. 1038, quantum group approach.
Ref. 1039 We remember that quantum mechanics starts when we realize that the classical
concept of action makes no sense below the value of ħ~2; similarly, unified theories start
when we realize that the classical concepts of time and length make no sense below Planck
values. However, the usual description of space-time does contain such small values; the
usual description involves the existence of intervals smaller than the smallest measurable
one. Therefore, the continuum description of space-time has to be abandoned in favour of
a more appropriate description.
Ref. 1040 The new indeterminacy relation appearing at Planck scales shows that continuity can-
not be a good description of space-time. Inserting c∆p E ∆E E ħ~∆t into equation (674),
we get
∆x∆t E ħG~c 4 = t Pl l Pl , (676)

which of course has no counterpart in standard quantum mechanics. It shows that space-
time events do not exist. A final way to convince oneself that points have no meaning is
that a point is an entity with vanishing volume; however, the minimum volume possible
in nature is the Planck volume VPl = l Pl3
.
While space-time points are idealizations of events, this idealization is incorrect. The
use of the concept of ‘point’ is similar to the use of the concept of ‘aether’ a century ago:
Copyright © Christoph Schiller November 1997–May 2006

it is impossible to detect and it is only useful for describing observations until a way to
describe nature without it has been found. Like ‘aether´, also ‘point´ leads reason astray.
In other words, the Planck units do not only provide natural units, they also provide
— within a factor of order one — the limit values of space and time intervals.

Farewell to the space-time manifold

The consequences of the Planck limits for measurements of time and space can be taken
much further. It is commonplace to say that given any two points in space or any two
instants of time, there is always a third in between. Physicists sloppily call this property
continuity, while mathematicians call it denseness. However, at Planck dimensions this

Dvipsbugw
 1006 xi general rel ativity versus quantum mechanics

property cannot exist, since intervals smaller than the Planck time can never be found.
Thus points and instants are not dense, and between two points there is not always a third.
This means that space and time are not continuous. Of course, at large scales they are –
approximately – continuous, in the same way that a piece of rubber or a liquid seems
continuous at everyday dimensions, even though it is not at a small scale.
All paradoxes resulting from the infinite divisibility of space and time, such as Zeno’s
argument on the impossibility to distinguish motion from rest, or the Banach–Tarski
paradox, are now avoided. We can dismiss the paradoxes straight away because of their
incorrect premises concerning the nature of space and time. Dvipsbugw
But let us go on. Special relativity, quantum mechanics and general relativity all rely
on the idea that time can be defined for all points of a given reference frame. However,
two clocks a distance l apart cannot be synchronized with arbitrary precision. Since the
distance between two clocks cannot be measured with an error smaller than the Planck
length l Pl , and transmission of signals is necessary for synchronization, it is not possible
to synchronize two clocks with a better precision than the time l Pl ~c = t Pl , the Planck

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

time. Because it is impossible to synchronize clocks precisely, a single time coordinate for
a whole reference frame is only an approximation, and this idea cannot be maintained in
a precise description of nature.
Moreover, since the time difference between events can only be measured within a
Planck time, for two events distant in time by this order of magnitude, it is not possible
Ref. 1041 to say with complete certainty which of the two precedes the other! This is an important
result. If events cannot be ordered, the concept of time, which was introduced into phys-
ics to describe sequences, cannot be defined at all at Planck scales. In other words, after
dropping the idea of a common time coordinate for a complete frame of reference, we are
forced to drop the idea of time at a single ‘point’ as well. Therefore, the concept of ‘proper
time’ loses its meaning at Planck scales.
It is straightforward to use the same arguments to show that length measurements do
not allow us to speak of continuous space, but only of approximately continuous space.
As a result of the lack of measurement precision at Planck scales, the concepts of spatial
order, translation invariance, isotropy of vacuum and global coordinate systems have no
experimental basis.
But there is more to come. The very existence of a minimum length contradicts special
relativity theory, in which it is shown that lengths undergo Lorentz contraction when the
frame of reference is changed. A minimum length thus cannot exist in special relativity.
But we just deduced that there must be such a minimum distance in nature. There is only
Copyright © Christoph Schiller November 1997–May 2006

one conclusion: special relativity cannot be correct at smallest distances. Thus, space-time
is neither Lorentz invariant nor diffeomorphism invariant nor dilatation invariant at Planck
dimensions. All symmetries that are at the basis of special and general relativity are thus
only approximately valid at Planck scales.
As a result of the imprecision of measurement, most familiar concepts used to describe
spatial relations become useless. For example, the concept of metric loses its usefulness
at Planck scales. Since distances cannot be measured with precision, the metric cannot
be determined. We deduce that it is impossible to say precisely whether space is flat or
curved. In other words, the impossibility of measuring lengths exactly is equivalent to fluc-
Ref. 1028, Ref. 1042 tuations of the curvature, and thus equivalent to fluctuations of gravity.
In addition, even the number of spatial dimensions makes no sense at Planck scales.

Dvipsbugw
 d oes matter differ from vacuum? 1007

Let us remind ourselves how to determine this number experimentally. One possible
way is to determine how many points we can choose in space such that all the distances
between them are equal. If we can find at most n such points, the space has n − 1 dimen-
sions. We can see that if reliable length measurement at Planck scale is not possible, there
is no way to determine reliably the number of dimensions of a space with this method.
Another way to check for three spatial dimensions is to make a knot in a shoe string
and glue the ends together: since it stays knotted we know that space has three dimensions,
because there is a mathematical theorem that in spaces with greater or fewer than three
dimensions, knots do not exist. Again, at Planck dimensions the errors in measurement Dvipsbugw
do not allow to say whether a string is knotted or not, because measurement limits at
crossings make it impossible to say which strand lies above the other; in short, at Planck
scales we cannot check whether space has three dimensions or not.
There are many other methods for determining the dimensionality of space.* All these
methods start from the definition of the concept of dimensionality, which is based on
a precise definition of the concept of neighbourhood. However, at Planck scales, as just

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

mentioned, length measurements do not allow us to say whether a given point is inside or
outside a given volume. In short, whatever method we use, the lack of reliable length meas-
urements means that at Planck scales, the dimensionality of physical space is not defined. It
should therefore not come as a surprise that when we approach these scales, we may get a
scale-dependent answer for the number of dimensions, that may be different from three.
The reason for the problems with space-time become most evident when we remem-
Ref. 1043 ber Euclid’s well-known definition: ‘A point is that which has no part.’ As Euclid clearly
understood, a physical point, and here the stress is on physical, cannot be defined without
some measurement method. A physical point is an idealization of position, and as such
includes measurement right from the start. In mathematics, however, Euclid’s definition
is rejected; mathematical points do not need a metric for their definition. Mathematical
points are elements of a set, usually called a space. In mathematics, a measurable or met-
ric space is a set of points equipped afterwards with a measure or a metric. Mathematical
points do not need a metric for their definition; they are basic entities. In contrast to the
mathematical situation, the case of physical space-time, the concepts of measure and of
metric are more fundamental than that of a point. The difficulties distinguishing physical
and mathematical space and points arise from the failure to distinguish a mathematical
metric from a physical length measurement.**

* For example, we can determine the dimension using only the topological properties of space. If we draw a
Copyright © Christoph Schiller November 1997–May 2006

so-called covering of a topological space with open sets, there are always points that are elements of several
sets of the covering. Let us call p the maximal number of sets of which a point can be an element in a given
covering. This number can be determined for all possible coverings. The minimum value of p, minus one,
gives the dimension of the space.
In fact, if physical space is not a manifold, the various methods may give different answers for the dimen-
sionality. Indeed, for linear spaces without norm, a unique number of dimensions cannot be defined. The
value then depends on the specific definition used and is called e.g. fractal dimension, Lyapunov dimension,
etc.
** Where does the incorrect idea of continuous space-time have its roots? In everyday life, as well as in
physics, space-time is introduced to describe observations. Space-time is a book-keeping device. Its prop-
erties are extracted from the properties of observables. Since observables can be added and multiplied, we
extrapolate that they can take continuous values. This extrapolation implies that length and time intervals
can take continuous values, and, in particular, arbitrary small values. From this result we get the possibil-

Dvipsbugw
 1008 xi general rel ativity versus quantum mechanics

Perhaps the most beautiful way to make this point is the Banach–Tarski theorem,
Ref. 1044 which clearly shows the limits of the concept of volume. The theorem states that a sphere
made up of mathematical points can be cut into five pieces in such a way that the pieces
can be put together to form two spheres, each of the same volume as the original one.
However, the necessary cuts are ‘infinitely’ curved and detailed: they are wildly discon-
nected. For physical matter such as gold, unfortunately – or fortunately – the existence of
a minimum length, namely the atomic distance, makes it impossible to perform such a
cut. For vacuum, the puzzle reappears: for example, the energy of zero-point fluctuations
is given by the density times the volume; following the Banach–Tarski theorem, the zero Dvipsbugw
point energy content of a single sphere should be equal to the zero point energy of two
similar spheres each of the same volume as the original one. The paradox is solved by
the Planck length, because it also provides a fundamental length scale for vacuum, thus
making infinitely complex cuts impossible. Therefore, the concept of volume is only well
defined at Planck scales if a minimum length is introduced.
To sum up, physical space-time cannot be a set of mathematical points. But the surprises

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

are not finished. At Planck dimensions, since both temporal and spatial order break down,
there is no way to say if the distance between two space-time regions that are close enough
together is space-like or time-like. Measurement limits make it impossible to distinguish
the two cases. At Planck scales, time and space cannot be distinguished from each other.
In addition, it is impossible to state that the topology of space-time is fixed, as gen-
eral relativity implies. The topology changes – mentioned above – required for particle
reactions do become possible. In this way another of the contradictions between general
relativity and quantum theory is resolved.
In summary, space-time at Planck scales is not continuous, not ordered, not endowed
with a metric, not four-dimensional and not made up of points. If we compare this with
the definition of the term manifold,* not one of its defining properties is fulfilled. We
arrive at the conclusion that the concept of a space-time manifold has no backing at Planck
scales. This is a strong result. Even though both general relativity and quantum mechanics
use continuous space-time, the combination of both theories does not.
There is nothing in the world but matter in

“ motion, and matter in motion cannot move

otherwise than in space and time.
Lenin

Farewell to observables and measurements

To complete this review of the situation, if space and time are not continuous, no quant-
” Copyright © Christoph Schiller November 1997–May 2006

ities defined as derivatives with respect to space or time are precisely defined. Velocity,
acceleration, momentum, energy, etc., are only well-defined under the assumption of con-
tinuous space and time. That important tool, the evolution equation, is based on deriv-
atives and thus can no longer be used. Therefore the Schrödinger or the Dirac equation
ity of defining points and sets of points. A special field of mathematics, topology, shows how to start from
a set of points and construct, with the help of neighbourhood relations and separation properties, first a
topological space. Then, with the help of a metric, a metric space can be built. With the appropriate compact-
ness and connectedness relations, a manifold, characterized by its dimension, metric and topology, can be
constructed.
* A manifold is what locally looks like an Euclidean space. The exact definition can be found in Appendix D.

Dvipsbugw
 d oes matter differ from vacuum? 1009

lose their basis. Concepts such as ‘derivative’, ‘divergence-free’, ‘source free’ etc., lose their
meaning at Planck scales.
In fact, all physical observables are defined using length and time measurements. A
list of physical units shows that each is a product of powers of length, time (and mass)
units. (Even though in the SI system electrical quantities have a separate base quantity, the
ampere, the argument still holds; the ampere is itself defined in terms of a force, which
is measured using the three base units length, time and mass.) Since time and length are
not continuous, observables themselves are not defined, because their value is not fixed.
This means that at Planck scales, observables cannot be described by real numbers. Dvipsbugw
In addition, if time and space are not continuous, the usual expression for an observ-
able field A, namely A(t, x), does not make sense: we have to find a more appropriate
description. Physical fields cannot exist at Planck scales.
The consequences for quantum mechanics are severe. It makes no sense to define mul-
tiplication of observables by real numbers, thus by a continuous range of values, but only
by a discrete set of numbers. Among other implications, this means that observables do

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

not form a linear algebra. We recognize that, because of measurement errors, we cannot
prove that observables do form such an algebra. This means that observables are not de-
scribed by operators at Planck scales. And, because quantum mechanics is based on the
superposition principle, without it, everything comes crumbling down. In particular, the
most important observables are the gauge potentials. Since they do not now form an al-
gebra, gauge symmetry is not valid at Planck scales. Even innocuous looking expressions
such as [x i , x j ] = 0 for x i x x j , which are at the root of quantum field theory, become
meaningless at Planck scales. Since at those scales also the superposition principle cannot
be backed up by experiment, even the famous Wheeler–DeWitt equation, often assumed
to describe quantum gravity, cannot be valid.
Similarly, permutation symmetry is based on the premise that we can distinguish two
points by their coordinates, and then exchange particles between those two locations. As
we have just seen, this is not possible if the distance between the two particles is small;
we conclude that permutation symmetry has no experimental basis at Planck scales.
Even discrete symmetries, like charge conjugation, space inversion and time reversal
cannot be correct in this domain, because there is no way to verify them exactly by meas-
urement. CPT symmetry is not valid at Planck scales.
Finally we note that all types of scaling relations do not work at smallest scales. As a
result, renormalization symmetry is also destroyed at Planck scales.
All these results are consistent: if there are no symmetries at Planck scales, there are
Copyright © Christoph Schiller November 1997–May 2006

also no observables, since physical observables are representations of symmetry groups.

In fact, the limits on time and length measurements imply that the concept of measurement
has no significance at Planck scales.

Can space-time be a lattice? – A glimpse of quantum geometry

Ref. 1045 Discretization of space-time has been studied already since 1940s. Recently, the idea that
Ref. 1046 space-time could be described as a lattice has also been explored most notably by David
Ref. 1047 Finkelstein and by Gerard ’t Hooft. The idea of space-time as a lattice is based on the idea
Ref. 1049 that, if a minimum distance exists, then all distances are a multiple of this minimum. It
is generally agreed that, in order to get an isotropic and homogeneous situation for large,

Dvipsbugw
 1010 xi general rel ativity versus quantum mechanics

Ref. 1048 everyday scales, the structure of space-time cannot be periodic, but must be random. In
addition, any fixed structure of space-time violates the result that there are no lengths
smaller than the Planck length: as a result of the Lorentz contraction, any moving ob-
server would find lattice distances smaller than the Planck value. Worse still, the lattice
idea conflicts with general relativity, in particular with the diffeomorphism invariance of
vacuum. Finally, where would a particle be during the jump from one lattice point to the
next? Thus, in summary, space-time cannot be a lattice. A minimum distance does exist
in nature; however, the hope that all other distances are simple multiples of the smallest
distance is not correct. We will discover more evidence for this later on. Dvipsbugw
If space-time is not a set of points or events, it must be something else. Three hints
already appear at this stage. The first step necessary to improve the description of mo-
tion is the recognition that abandoning ‘points’ means abandoning the local description
of nature. Both quantum mechanics and general relativity assume that the phrase ‘ob-
servable at a point’ has a precise meaning. Because it is impossible to describe space as
a manifold, this expression is no longer useful. The unification of general relativity and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

quantum physics forces the adoption of a non-local description of nature at Planck scales.
The existence of a minimum length implies that there is no way to physically distin-
guish locations that are even closer together. We are tempted to conclude therefore that
no pair of locations can be distinguished, even if they are one metre apart, since on any
path joining two points, no two locations that are close together can be distinguished.
This situation is similar to the question about the size of a cloud or of an atom. If we
measure water density or electron density, we find non-vanishing values at any distance
from the centre of the cloud or the atom; however, an effective size can still be defined,
because it is very unlikely that the effects of the presence of a cloud or of an atom can
be seen at distances much larger than this effective size. Similarly, we can guess that two
points in space-time at a macroscopic distance from each other can be distinguished be-
cause the probability that they will be confused drops rapidly with increasing distance. In
short, we are thus led to a probabilistic description of space-time. Space-time becomes a
macroscopic observable, a statistical or thermodynamic limit of some microscopic entit-
ies.
We note that a fluctuating structure for space-time would also avoid the problems
of fixed structures with Lorentz invariance. This property is of course compatible with
a statistical description. In summary, the experimental observations of special relativity,
i.e. Lorentz invariance, isotropy and homogeneity, together with that of a minimum dis-
tance, point towards a fluctuating description of space-time. Research efforts in quantum
Copyright © Christoph Schiller November 1997–May 2006

Ref. 1050 gravity, superstring theory and quantum groups have confirmed independently of each
other that a probabilistic and non-local description of space-time at Planck dimensions,
resolves the contradictions between general relativity and quantum theory. This is our
first result on quantum geometry. To clarify the issue, we have to turn to the concept of
the particle.

Farewell to particles
In every example of motion, some object is involved. One of the important discoveries
of the natural sciences was that all objects are composed of small constituents, called
elementary particles. Quantum theory shows that all composite, non-elementary objects

Dvipsbugw
 d oes matter differ from vacuum? 1011

have a finite, non-vanishing size. This property allows us to determine whether a particle
is elementary or not. If it behaves like a point particle, it is elementary. At present, only
the leptons (electron, muon, tau and the neutrinos), the quarks and the radiation quanta
of the electromagnetic, weak and strong nuclear interactions (the photon, the W and Z
Page 1174 bosons, the gluons) have been found to be elementary. A few more elementary particles
are predicted by various refinements of the standard model. Protons, atoms, molecules,
Page 1000 cheese, people, galaxies etc., are all composite, as shown in Table 74. Elementary particles
are characterized by their vanishing size, their spin and their mass.
Even though the definition of ‘elementary particle’ as point particle is all we need in Dvipsbugw
the following argument, it is not complete, because it seems to leave open the possibility
that future experiments could show that electrons or quarks are not elementary. This is
not so! In fact, any particle smaller than its own Compton wavelength is elementary. If
it were composite, there would be a lighter component inside it and this lighter particle
would have a larger Compton wavelength than the composite particle. This is impossible,
since the size of a composite particle must be larger than the Compton wavelength of its

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

components. (The possibility that all components are heavier than the composite, which
would avoid this argument, does not lead to satisfying physical properties; for example,
Ref. 1051 it leads to intrinsically unstable components.)
The size of an object, such as those given in Table 74, is defined as the length at which
differences from point-like behaviour are observed. This is the way in which, using alpha
particle scattering, the radius of the atomic nucleus was determined for the first time in
Rutherford’s experiment. In other words, the size d of an object is determined by measur-
ing how it scatters a beam of probe particles. In daily life as well, when we look at objects,
we make use of scattered photons. In general, in order to make use of scattering, the ef-
fective wavelength λ = ħ~mv of the probe must be smaller than the object size d to be
determined. We thus need d A λ = ħ~(mv) E ħ~(mc). In addition, in order to make
a scattering experiment possible, the object must not be a black hole, since, if it were, it
would simply swallow the approaching particle. This means that its mass m must be smal-
ler than that of a black hole of the same size; in other words, from equation (664) we must
have m < dc 2 ~G. Combining this with the previous condition we get
¾
dA = l Pl .
ħG
(677)
c3

In other words, there is no way to observe that an object is smaller than the Planck length.
Copyright © Christoph Schiller November 1997–May 2006

There is thus no way in principle to deduce from observations that a particle is point-like. In
fact, it makes no sense to use the term ‘point particle’ at all! Of course, there is a relation
between the existence of a minimum length for empty space and a minimum length for
objects. If the term ‘point of space’ is meaningless, then the term ‘point particle’ is also
meaningless. As in the case of time, the lower limit on length results from the combina-
tion of quantum mechanics and general relativity.*
The size d of any elementary particle must by definition be smaller than its own
Compton wavelength ħ~(mc). Moreover, the size of a particle is always larger than the

* Obviously, the minimum size of a particle has nothing to do with the impossibility, in quantum theory, of
localizing a particle to within less than its Compton wavelength.

Dvipsbugw
 1012 xi general rel ativity versus quantum mechanics

Planck length: d A l Pl . Combining these two requirements and eliminating the size d we
get the condition for the mass m of any elementary particle, namely
¾
m< = = m Pl = 2.2 ë 10−8 kg = 1.2 ë 1019 GeV~c2 .
ħ ħc
(678)
c l Pl G

The limit m Pl , the so-called Planck mass, corresponds roughly to the mass of a human
embryo that is ten days old, or equivalently, to that of a small flea. In short, the mass Dvipsbugw
of any elementary particle must be smaller than the Planck mass. This fact was already
Ref. 1052 noted as ‘well-known’ by Andrei Sakharov* in 1968; he explains that these hypothetical
particles are sometimes called ‘maximons’. And indeed, the known elementary particles
all have masses well below the Planck mass. (In fact, the question why their masses are
so incredibly much smaller than the Planck mass is one of the most important questions
of high-energy physics. We will come back to it.)
There are many other ways to arrive at the mass limit for particles.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

For example, in order to measure mass by scattering – and that is
the only way for very small objects – the Compton wavelength of the
scatterer must be larger than the Schwarzschild radius; otherwise the
probe will be swallowed. Inserting the definition of the two quantities
and neglecting the factor 2, we get again the limit m < m Pl . (In fact it
is a general property of descriptions of nature that a minimum space-
Ref. 1053 time interval leads to an upper limit for elementary particle masses.)
The importance of the Planck mass will become clear shortly.
Another property connected with the size of a particle is its elec-
tric dipole moment. It describes the deviation of its charge distribu- Andrei Sakharov
tion from spherical. Some predictions from the standard model of ele-
Ref. 1054 mentary particles give as upper limit for the electron dipole moment d e a value of

Sd e S
< 10−39 m , (679)
e
where e is the charge of the electron. This value is ten thousand times smaller than the
Planck length l Pl e. Since the Planck length is the smallest possible length, we seem to
Ref. 1055 have a potential contradiction here. However, a more recent prediction from the standard
model is more careful and only states
Copyright © Christoph Schiller November 1997–May 2006

Sd e S
< 3 ë 10−21 m , (680)
e
which is not in contradiction with a minimal length in nature. The issue is still not settled.
We will see below that the experimental limit is expected to allow to test these predictions

* Andrei Dmitrievich Sakharov, famous Soviet nuclear physicist (1921–1989). One of the keenest thinkers
in physics, Sakharov, among others, invented the Tokamak, directed the construction of nuclear bombs,
and explained the matter-antimatter asymmetry of nature. Like many others, he later campaigned against
nuclear weapons, a cause for which he was put into jail and exile, together with his wife, Yelena Bonner. He
received the Nobel Peace Prize in 1975.

Dvipsbugw
 d oes matter differ from vacuum? 1013

in the foreseeable future.

Planck scales have other strange consequences. In quantum field theory, the difference
between a virtual particle and a real particle is that a real particle is ‘on shell’, obeying
E 2 = m 2 c 4 + p2 c 2 , whereas a virtual particle is ‘off shell’, obeying E 2 x m 2 c 4 + p2 c 2 .
Because of the fundamental limits of measurement precision, at Planck scales we cannot
determine whether a particle is real or virtual.
However, that is not all. Antimatter can be described as matter moving backwards
in time. Since the difference between backwards and forwards cannot be determined at
Planck scales, matter and antimatter cannot be distinguished at Planck scales. Dvipsbugw
Particles are also characterized by their spin. Spin describes two properties of a particle:
its behaviour under rotations (and if the particle is charged, its behaviour in magnetic
fields) and its behaviour under particle exchange. The wave function of particles with
spin 1 remains invariant under a rotation of 2π, whereas that of particles with spin 1/2
changes sign. Similarly, the combined wave function of two particles with spin 1 does not
change sign under exchange of particles, whereas for two particles with spin 1/2 it does.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We see directly that both transformations are impossible to study at Planck scales.
Given the limit on position measurements, the position of a rotation axis cannot be well
defined, and rotations become impossible to distinguish from translations. Similarly, pos-
ition imprecision makes impossible the determination of precise separate positions for ex-
change experiments. In short, spin cannot be defined at Planck scales, and fermions cannot
be distinguished from bosons, or, phrased differently, matter cannot be distinguished from ra-
diation at Planck scales. We can thus easily see that supersymmetry, a unifying symmetry
between bosons and fermions, somehow becomes natural at Planck dimensions.
But let us now move to the main property of elementary particles.

Farewell to mass
The Planck mass divided by the Planck volume, i.e. the Planck density, is given by

c5
ρ Pl = = 5.2 ë 1096 kg~m3 (681)
G2ħ
and is a useful concept in the following. If we want to measure the (gravitational) mass
M enclosed in a sphere of size R and thus (roughly) of volume R 3 , one way to do this is
to put a test particle in orbit around it at that same distance R. Universal gravitation then
gives for the mass M the expression M = Rv 2 ~G, where v is the speed of the orbiting test
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1429 e
particle. From v < c, we thus deduce that M < c 2 R~G; since the minimum value for R is
the Planck distance, we get (neglecting again factors of order unity) a limit for the mass
density ρ, namely
ρ < ρ Pl . (682)

In other words, the Planck density is the maximum possible value for mass density. Unsur-
prisingly, a volume of Planck dimensions cannot contain a mass larger than the Planck
mass.
Interesting things happen when we start to determine the error ∆M of a mass meas-
urement in a Planck volume. Let us return to the mass measurement by an orbiting probe.

Dvipsbugw
1014 xi general rel ativity versus quantum mechanics

Dvipsbugw

R R

F I G U R E 381 A Gedanken experiment showing that

at Planck scales, matter and vacuum cannot be
distinguished

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

From the relation GM = rv 2 we deduce by differentiation that G∆M = v 2 ∆r + 2vr∆v A
2vr∆v = 2GM∆v~v. For the error ∆v in the velocity measurement we have the inde-
terminacy relation ∆v E ħ~(m∆r) + ħ~(MR) E ħ~(MR). Inserting this in the previous
inequality, and forgetting again the factor of 2, we find that the mass measurement error
∆M of a mass M enclosed in a volume of size R is subject to the condition

∆M E
ħ
. (683)
cR
Note that for everyday situations, this error is extremely small, and other errors, such as
the technical limits of the balance, are much larger.
To check this result, we can explore another situation. We even use relativistic expres-
sions, in order to show that the result does not depend on the details of the situation or
the approximations. Imagine having a mass M in a box of size R and weighing the box
with a scale. (It is assumed that either the box is massless or that its mass is subtracted by
the scale.) The mass error is given by ∆M = ∆E~c 2 , where ∆E is due to the indeterminacy
in the kinetic energy of the mass inside the box. Using the expression E 2 = m 2 c 4 + p2 c 2 ,
we get that ∆M E ∆p~c, which again reduces to equation (683). Now that we are sure of
Copyright © Christoph Schiller November 1997–May 2006

the result, let us continue.

From equation (683) we deduce that for a box of Planck dimensions, the mass meas-
urement error is given by the Planck mass. But from above we also know that the mass
that can be put inside such a box must not be larger than the Planck mass. Therefore, for a
box of Planck dimensions, the mass measurement error is larger than (or at best equal to)
the mass contained in it: ∆M E M Pl . In other words, if we build a balance with two boxes
of Planck size, one empty and the other full, as shown in Figure 381, nature cannot decide
which way the balance should hang! Note that even a repeated or a continuous measure-
ment will not resolve the situation: the balance will only randomly change inclination,
staying horizontal on average.
The argument can be rephrased as follows. The largest mass that we can put in a box

Dvipsbugw
d oes matter differ from vacuum? 1015

of size R is a black hole with a Schwarzschild radius of the same value; the smallest mass
present in such a box – corresponding to what we call vacuum – is due to the indeterm-
inacy relation and is given by the mass with a Compton wavelength that matches the size
of the box. In other words, inside any box of size R we have a mass m, the limits of which
are given by:
c2 R
AmA
ħ
(full box) (empty box) . (684)
G cR
We see directly that for sizes R of the order of the Planck scale, the two limits coincide; Dvipsbugw
in other words, we cannot distinguish a full box from an empty box in that case.
To be sure of this strange result, we check whether it also occurs if, instead of measur-
ing the gravitational mass, as we have just done, we measure the inertial mass. The inertial
mass for a small object is determined by touching it, i.e. physically speaking, by perform-
ing a scattering experiment. To determine the inertial mass inside a region of size R, a
probe must have a wavelength smaller than R, and thus a correspondingly high energy.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

A high energy means that the probe also attracts the particle through gravity. (We thus
find the intermediate result that at Planck scales, inertial and gravitational mass cannot be
distinguished. Even the balance experiment shown in Figure 381 illustrates this: at Planck
scales, the two types of mass are always inextricably linked.) Now, in any scattering ex-
periment, e.g. in a Compton-type experiment, the mass measurement is performed by
measuring the wavelength change δλ of the probe before and after the scattering experi-
ment. The mass indeterminacy is given by

=
∆M ∆δλ
. (685)
M δλ
In order to determine the mass in a Planck volume, the probe has to have a wavelength of
the Planck length. But we know from above that there always is a minimum wavelength
indeterminacy, given by the Planck length l Pl . In other words, for a Planck volume the
mass error is always as large as the Planck mass itself: ∆M E M Pl . Again, this limit is a
direct consequence of the limit on length and space measurements.
This result has an astonishing consequence. In these examples, the measurement error
is independent of the mass of the scatterer, i.e. independent of whether or not we start
with a situation in which there is a particle in the original volume. We thus find that in
a volume of Planck size, it is impossible to say whether or not there is something there
Copyright © Christoph Schiller November 1997–May 2006

when we probe it with a beam!

In short, all arguments lead to the same conclusion: vacuum, i.e. empty space-time,
cannot be distinguished from matter at Planck scales. Another, often used way to express
this state of affairs is to say that when a particle of Planck energy travels through space
it will be scattered by the fluctuations of space-time itself, thus making it impossible to
say whether it was scattered by empty space-time or by matter. These surprising results
rely on a simple fact: whatever definition of mass we use, it is always measured via com-
bined length and time measurements. (This is even the case for normal weighing scales:
mass is measured by the displacement of some part of the machine.) The error in these
measurements makes it impossible to distinguish vacuum from matter.
We can put this result in another way. If on one hand, we measure the mass of a piece

Dvipsbugw
 1016 xi general rel ativity versus quantum mechanics

of vacuum of size R, the result is always at least ħ~cR; there is no possible way to find a
perfect vacuum in an experiment. On the other hand, if we measure the mass of a particle,
we find that the result is size dependent; at Planck dimensions it approaches the Planck
mass for every type of particle, be it matter or radiation.
If we use another image, when two particles approach each other to a separation of
the order of the Planck length, the indeterminacy in the length measurements makes
it impossible to say whether there is something or nothing between the two objects. In
short, matter and vacuum are interchangeable at Planck dimensions. This is an important
result: since both mass and empty space-time cannot be differentiated, we have confirmed Dvipsbugw
that they are made of the same ‘fabric’. This approach, already suggested above, is now
commonplace in all attempts to find a unified description of nature.
This approach is corroborated by the attempts to apply quantum mechanics in highly
curved space-time, where a clear distinction between vacuum and particles is impossible.
Ref. 1056 This has already been shown by Fulling–Davies–Unruh radiation. Any accelerated ob-
server and any observer in a gravitational field detects particles hitting him, even if he is

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 858 in vacuum. The effect shows that for curved space-time the idea of vacuum as a particle-
free space does not work. Since at Planck scales it is impossible to say whether space is
flat or not, it again follows that it is impossible to say whether it contains particles or not.

Curiosities and fun challenges on Planck scales

The strange results at Planck scales imply many other consequences.

**
Observes are made of matter. Observer are thus biased, because they take a specific stand-
point. But at Planck scale, vacuum, radiation and matter cannot me distinguished. Two
conclusions result: first, only at those scales would a description be free of any bias in
favour of matter; but secondly, observers do not exist at all at Planck energy.
**
The Planck energy is rather large. Imagine that we want to impart this amount of energy
Challenge 1430 n to protons using a particle accelerator. How large would that accelerator have to be? In
contrast, in everyday life, the Planck energy is rather small. Measured in litres of gasoline,
Challenge 1431 n how much fuel does it correspond to?
**
Copyright © Christoph Schiller November 1997–May 2006

The usual concepts of matter and of radiation are not applicable at Planck dimensions.
Usually, it is assumed that matter and radiation are made up of interacting elementary
particles. The concept of an elementary particle is one of an entity that is countable, point-
like, real and not virtual, that has a definite mass and a definite spin, that is distinct from its
antiparticle, and, most of all, that is distinct from vacuum, which is assumed to have zero
mass. All these properties are found to be incorrect at Planck scales. At Planck dimensions,
it does not make sense to use the concepts of ‘mass’, ‘vacuum’, ‘elementary particle’, ‘radiation’
and ‘matter’.

**

Dvipsbugw
 d oes matter differ from vacuum? 1017

Do the large mass measurement errors make it possible to claim that mass can be negative
Challenge 1432 n at Planck energy?

**
We now have a new answer to the old question: why is there anything rather than nothing?
Well, we now see that at Planck scales there is no difference between anything and nothing.
In addition, we now can honestly say about ourselves that we are made of nothing.
** Dvipsbugw
If vacuum and matter or radiation cannot be distinguished, then it is incorrect to claim
that the universe appeared from nothing. The impossibility of making this distinction
thus shows that naive creation is a logical impossibility. Creation is not a description of
reality. The term ‘creation’ turns out to be a result of lack of imagination.
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Special relativity implies that no length or energy can be invariant. Since we have come
to the conclusion that the Planck energy and the Planck length are invariant, there must
be deviations from Lorentz invariance at high energy. Can you imagine what the effects
Challenge 1433 r would be? In what experiment could they be measured? If you find an answer, publish it;
Ref. 1058 you might get known. First attempts are appearing in the research papers. We return to
the issue in the third part, with some interesting insights.
**
Quantum mechanics alone gives, via the Heisenberg indeterminacy relation, a lower limit
on the spread of measurements, but strangely enough not on their precision, i.e. not on
Ref. 1057 the number of significant digits. Jauch gives the example that atomic lattice constants are
known much more precisely than the position indeterminacy of single atoms inside the
Ref. 1038 crystal.
It is sometimes claimed that measurement indeterminacies smaller than the Planck
values are possible for large enough numbers of particles. Can you show why this is in-
Challenge 1434 ny correct, at least for space and time?
**
Of course, the idea that vacuum is not empty is not new. More than two thousand years
ago, Aristotle argued for a filled vacuum, even though he used incorrect arguments, as
Copyright © Christoph Schiller November 1997–May 2006

seen from today’s perspective. In the fourteenth century the discussion on whether empty
space was composed of indivisible entities was rather common, but died down again later.

**
A Planck energy particle falling in a gravitational field would gain energy. However, this
is impossible, as the Planck energy is the highest energy in nature. What does this imply
Challenge 1435 n for this situation?
**
One way to generalize the results presented here is to assume that, at Planck energy, nature
Ref. 1027 is event symmetric, i.e. nature is symmetric under exchange of any two events. This idea,

Dvipsbugw
 1018 xi general rel ativity versus quantum mechanics

developed by Phil Gibbs, provides an additional formulation of the strange behaviour of

nature at extreme scales.

**
Page 489 Because there is a minimum length in nature, so-called naked singularities do not exist.
The issue, so hotly debated in the twentieth century, becomes uninteresting, thus ending
decades of speculation.
** Dvipsbugw
Since mass density and thus energy density are limited, we know that the number of
Page 870 degrees of freedom of any object of finite volume is finite. The entropy of black holes has
shown us already that entropy values are always finite. This implies that perfect baths do
not exist. Baths play an important role in thermodynamics (which is thus found to be only
an approximation) and also in recording and measuring devices: when a device measures,
it switches from a neutral state to a state in which it shows the result of the measurement.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 1059 In order to avoid the device returning to the neutral state, it must be coupled to a bath.
Without a bath, a reliable measuring device cannot be made. In short, perfect clocks and
length measuring devices do not exist because nature puts a limit on their storage ability.
**
If vacuum and matter cannot be distinguished, we cannot distinguish between objects
Page 37 and their environment. However, this was one the starting points of our journey. Some
interesting adventures thus still await us!

**
We have seen earlier that characterizing nature as made up of particles and vacuum cre-
ates problems when interactions are included, since on one hand interactions are the dif-
ference between the parts and the whole, while on the other hand, according to quantum
Page 676 theory, interactions are exchanges of particles. This apparent contradiction can be used
to show either that vacuum and particles are not the only components of nature, or that
something is counted twice. However, since matter and space-time are both made of the
same ‘stuff,’ the contradiction is resolved.
**
Challenge 1436 d Is there a smallest possible momentum? And a smallest momentum error?
Copyright © Christoph Schiller November 1997–May 2006

**
There is a maximum acceleration in nature. Can you deduce the value of this so-called
Challenge 1437 n Planck acceleration? Does it require quantum theory?
**
Given that time becomes an approximation at Planck scales, can we still say whether
nature is deterministic? Let us go back to the beginning. We can define time, because in
nature change is not random, but gradual. What is the situation now that we know that
time is only approximate? Is non-gradual change possible? Is energy conserved? In other
words, are surprises possible?

Dvipsbugw
 d oes matter differ from vacuum? 1019

To say that time is not defined at Planck scales and that therefore determinism is an
undefinable concept is correct, but not a satisfying answer. What happens at daily life
scales? The first answer is that at our everyday scales, the probability of surprises is so
small that the world indeed is effectively deterministic. The second answer is that nature
is not deterministic, but that the difference is not measurable, since every measurement
and observation, by definition, implies a deterministic world. The lack of surprises would
be due to the limitations of our human nature, and more precisely to the limitations of
our senses and brain. The third answer is that the lack of surprises is only apparent, and
that we have not yet experienced them yet. Dvipsbugw
Challenge 1438 n Can you imagine any other possibility? To be honest, it is not possible to answer at this
point. But we need to keep the alternatives in mind. We have to continue searching, but
with every step we take, we have to consider carefully what we are doing.
**
If matter and vacuum cannot be distinguished, matter and vacuum each has the proper-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ties of the other. For example, since space-time is an extended entity, matter and radiation
are also extended entities. Furthermore, as space-time is an entity that reaches the bor-
ders of the system under scrutiny, particles must also do so. This is the first hint at the
extension of matter; in the following, we will examine this argument in more detail.
**
Vacuum has zero mass density at large scales, but Planck mass density at Planck scales.
Cosmological measurements show that the cosmos is flat or almost flat at large scales,
i.e. its energy density is quite low. In contrast, quantum field theory maintains that va-
cuum has a high energy density (or mass density) at small scales. Since mass is scale
dependent, both viewpoints are right, providing a hint to the solution of what is usually
Ref. 1060 called the cosmological constant problem. The contradiction is only apparent; more about
this issue later on.
**
Challenge 1439 n When can matter and vacuum be distinguished? At what energy?
**
If matter and vacuum cannot be distinguished, there is a lack of information, which in
turn produces an intrinsic basic entropy associated with any part of the universe. We will
Copyright © Christoph Schiller November 1997–May 2006

come back to this topic shortly, in the discussion of the entropy of black holes.

**
Can we distinguish between liquids and gases by looking at a single atom? No, only by
looking at many. In the same way, we cannot distinguish between matter and vacuum
by looking at one point, but only by looking at many. We must always average. However,
even averaging is not completely successful. Distinguishing matter from vacuum is like
distinguishing clouds from the clear sky; like clouds, matter also has no defined boundary.

**
In our exploration we have found that there is no argument which shows that space and

Dvipsbugw
 1020 xi general rel ativity versus quantum mechanics

time are either continuous or made up of points. Indeed, in contrast, we have found that
the combination of relativity and quantum theory makes this impossible. In order to pro-
ceed in our ascent of Motion Mountain, we need to leave behind us the usual concept of
space-time. At Planck dimensions, the concept of ‘space-time point’ or ‘mass point’ is not
applicable in the description of nature.

Farewell to the big bang

A minimum length, or equivalently,* a minimum action, both imply that there is a Dvipsbugw
maximum curvature for space-time. Curvature can be measured in several ways; for ex-
ample, surface curvature is an inverse area. A minimum length thus implies a maximum
curvature. Within a factor of order one, we find

c3
K< = 0.39 ë 1070 m−2 . (686)
Għ

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

as limit for surface curvature K in nature. In other words, the universe never has been a
point, never had zero age, never had infinite density and never had infinite curvature. It
Challenge 1440 n is not difficult to get a similar limit for temperature or any other physical quantity.
In short, since events do not exist, also the big bang cannot have been an event. There
never was an initial singularity or a beginning of the universe.

The baggage left behind

In this rapid journey, we have destroyed all the experimental pillars of quantum the-
ory: the superposition principle, space-time symmetry, gauge symmetry, renormaliza-
tion symmetry and permutation symmetry. We also have destroyed the foundations of
general relativity, namely the existence of the space-time manifold, the field concept, the
particle concept and the concept of mass. We have even seen that matter and space-time
cannot be distinguished.
All these conclusions can be drawn in a simpler manner, by using the minimum action
Page 1068 of quantum theory and the maximum force of general relativity. All the mentioned results
above are confirmed. It seems that we have lost every concept used for the description of
motion, and thus made its description impossible. We naturally ask whether we can save
the situation.
First of all, since matter is not distinguishable from vacuum, and since this is true for all
Copyright © Christoph Schiller November 1997–May 2006

types of particles, be they matter or radiation, we have an argument which demonstrates

that the quest for unification in the description of elementary particles is correct and
necessary.
Moreover, since the concepts ‘mass’, ‘time’ and ‘space’ cannot be distinguished from
each other, we also know that a new, single entity is necessary to define both particles and
space-time. To find out more about this new entity, three approaches are being pursued
at the beginning of the twenty-first century. The first, quantum gravity, especially the ap-
Ref. 1012 proach using the loop representation and Ashtekar’s new variables, starts by generalizing
Ref. 1013 space-time symmetry. The second, string theory, starts by generalizing gauge symmetries

* The big bang section was added in summer 2002.

Dvipsbugw
 some experimental predictions 1021

and interactions, while the third, the algebraic quantum group approach, looks for gen-
Ref. 1014 eralized permutation symmetries. We will describe these approaches in more detail later
on.
Before we go on however, we should check with experiments what we have deduced
so far.

S ome experimental predictions

At present, there is a race going on both in experimental and in theoretical physics: which Dvipsbugw
will be the first experiment that will detect quantum gravity effects, i.e. effects sensitive
to the Planck energy?*
One might think that the fluctuations of space and time might make images from far
away galaxies unsharp or destroy the phase relation between the photons. However, this
Ref. 1061 effect has been shown to be unmeasurable in all possible cases.
A better candidate is measurement of the speed of light at different frequencies in

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

far away light flashes. There are natural flashes, called gamma ray bursts, which have an
extremely broad spectrum, from 100 GeV down to visible light of about 1 eV. These flashes
often originate at cosmological distances d. From the difference in arrival time ∆t for two
frequencies we can define a characteristic energy by setting

ħ(ω 1 − ω 2 )d
E char = . (687)
c∆t

Ref. 1062 This energy value is 8 ë 1016 GeV for the best measurement to date. This value is not far
from the Planck energy; in fact, it is even closer when the missing factors of order unity
are included. It is expected that the Planck scale will be reached in a few years, so that
tests will become possible on whether the quantum nature of space-time influences the
Ref. 1063, Ref. 1064 dispersion of light signals. Planck scale effects should produce a minimum dispersion,
different from zero. Detecting it would confirm that Lorentz symmetry is not valid at
Planck scales.
Another candidate experiment is the direct detection of distance fluctuations between
Ref. 1065, Ref. 1064 bodies. Gravitational wave detectors are sensitive to extremely small noise signals in
length measurements. There should be a noise signal due to the distance fluctuations
induced near Planck energies. The length indeterminacy with which a length l can be
Ref. 1066 measured is predicted to be
l Pl 2~3
Copyright © Christoph Schiller November 1997–May 2006

E‹ 
δl
(688)
l l

The expression is deduced simply by combining the measurement limit of a ruler in

quantum theory with the requirement that the ruler cannot be a black hole. We will dis-
Page 1036 cuss this result in more detail in the next section. The sensitivity to noise of the detectors
might reach the required level in the early twenty-first century. The noise induced by
Ref. 1067 quantum gravity effects is also predicted to lead to detectable quantum decoherence and
vacuum fluctuations.

* As more candidates appear, they will be added to this section.

Dvipsbugw
 1022 xi general rel ativity versus quantum mechanics

A third candidate for measurable quantum gravity is the detection of the loss of CPT
symmetry at high energies. Especially in the case of the decay of certain elementary
Ref. 1064 particles, such as neutral kaons, the precision of experimental measurement is approach-
ing the detection of Planck scale effects.
A fourth candidate is the possibility that quantum gravity effects may change the
Ref. 1068 threshold energy at which certain particle reactions become possible. It may be that ex-
tremely high energy photons or cosmic rays will make it possible to prove that Lorentz
invariance is indeed broken near the Planck scale.
A fifth candidate is the possibility that the phase of light that travels over long distances Dvipsbugw
gets washed out. However, the first tests show that this is not the case; light form extremely
distant galaxies still interferes. The precise prediction of the phase washing effect is still
Ref. 1069 in discussion; most probably the effect is too small to be measured.
In the domain of atomic physics, it has also been predicted that quantum gravity effects
Ref. 1067 will induce a gravitational Stark effect and a gravitational Lamb shift in atomic transitions.
Either effect could be measurable.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

A few candidates for quantum gravity effects have also been predicted by the author.
To get an overview, we summarize and clarify the results found so far.* Special relativ-
ity starts with the discovery that observable speeds are limited by the speed of light c.
Quantum theory starts with the result that observable actions are limited by ħ~2. Gravit-
ation shows that for every system with length L and mass M, the observable ratio L~M is
limited by the constant 4G~c 2 . Combining these results, we have deduced that all physical
observables are bound, namely by what are usually called the Planck values, though mod-
ified by a factor of square root of 2 (or several of them) to compensate for the numerical
factors omitted from the previous sentence.
We need to replace ħ by ħ~2 and G by 4G in all the defining expressions for Planck
quantities, in order to findº the corresponding measurement limits. In particular, the limit
for lengths and timesº is 2 times the Planck value and the limit for energy is the Planck
value divided by 8 .** Interestingly, the existence of bounds on all observables makes
it possible to deduce several experimentally testable predictions for the unification of
quantum theory and general relativity. These predictions do not depend on the detailed
final theory.
However, first we need to correct the argument that we have just presented. The argu-
ment is only half the story, because we have cheated. The (corrected) Planck values do not
seem to be the actual limits to measurements. The actual measurement limits are even
stricter still.
Copyright © Christoph Schiller November 1997–May 2006

First of all, for any measurement, we need certain fundamental conditions to be real-
ized. Take the length measurement of an object. We need to be able to distinguish between
matter and radiation, since the object to be measured is made up of matter, and since ra-
diation is the measurement tool that is used to read distances from the ruler. For a meas-
urement process, we need an interaction, which implies the use of radiation. Note that
even the use of particle scattering to determine lengths does not invalidate this general
requirement.
* This subsection, in contrast to the ones so far, is speculative; it was added in February 2001.
º half the minimum area. Of
** The entropy of a black hole is thus given by the ratio between its horizon and
course, a detailed investigation also shows that the Planck mass (divided by 8 ) is the limit for elementary
particles from below and for black holes from above. For everyday systems, there is no limit.

Dvipsbugw
 some experimental predictions 1023

In addition, for the measurement of wavelengths we need to distinguish between mat-

ter and radiation, because matter is necessary to compare two wavelengths. In fact, all
Page 1104 length measurements require the distinction between matter and radiation.* However,
this distinction is impossible at the energy of grand unification, when the electroweak
Ref. 1070 and the strong nuclear interactions are unified. At and above this energy, particles of
matter and particles of radiation transform into each other; in practice they cannot be
distinguished from each other.
If all matter and radiation particles were the same or mixtures of each other, mass
could not be defined. Similarly, spin, charge or any other quantum numbers could be Dvipsbugw
defined. To sum up, no measurement can be performed at energies equal to or greater
than the GUT unification energy.
In other words, the particle concept (and thus the matter concept) does not run into
trouble at the Planck scale, it has already done so at the unification scale. Only below the
unification scale do our standard particle and space-time concepts apply. Only below the
º
unification scale can particles and vacuum be effectively distinguished.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

As a result, the smallest length in natureº is 2 times the Planck length reduced by
the ratio between the maximal energy E Pl ~ 8 and the unification energy E GUT . Present
Ref. 1071 estimates give E GUT = 1016 GeV, implying that the smallest accessible length L min is

º
L min = 2 l Pl º  10−32 m  800 l Pl .
E Pl
(689)
8 E GUT

It is unlikely that measurements at these dimensions will ever be possible. Anyway, the
smallest measurable length is significantly larger than the Planck scale of nature discussed
above. The reason for this is that the Planck scale is that length for which particles and
vacuum cannot be distinguished, whereas the minimal measurable length is the distance
at which particles of matter and particles of radiation cannot be distinguished. The latter
happens at lower energy than the former. We thus have to correct our previous statement
to: the minimum measurable length cannot be smaller than L min .
The experimentally determined factor of about 800 is one of the great riddles of phys-
ics. It is the high-energy equivalent of the quest to understand why the electromagnetic
coupling constant is about 1~137, or more simply, why all things have the colours they
have. Only the final theory of motion will provide the answer.
In particular, the minimum length puts a bound on the electric dipole moment d of
elementary particles, i.e. on any particles without constituents. We get the limit
Copyright © Christoph Schiller November 1997–May 2006

d A d min = e L min = e 10−32 m = 1.5 ë 10−51 Cm . (690)

We saw that this result is in contradiction with one of the predictions deduced from the
Page 1012 standard model, but not with others. More interestingly, the prediction is in the reach of
Ref. 1072 future experiments. This improved limit may be the simplest possible measurement of
yet unpredicted quantum gravity effects. Measuring the dipole moment could thus be a
way to determine the unification energy (the factor 800) independently of high-energy
physics experiments and possibly to a higher precision.

* To speak in modern high energy concepts, all measurements require broken supersymmetry.

Dvipsbugw
 1024 xi general rel ativity versus quantum mechanics

Inverse of
coupling

electromagnetism

Dvipsbugw

1/26 weak interaction

strong interaction

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

105 1010 1015 "1020" Energy
in GeV
GUT Planck
energy energy

F I G U R E 382 Coupling constants and their spread as a function of energy

Interestingly, the bound on the measurability of observables also puts a bound on the
measurement precision for each observable. This bound is of no importance in every-
day life, but it is important at high energy. What is the precision with which a coupling
constant can be measured? We can illustrate this by taking the electromagnetic coupling
constant as an example. This constant α, also called the fine structure constant, is related
to the charge q by ¼
q= 4πε 0 ħcα . (691)

Now, any electrical charge itself is defined and measured by comparing, in an electrical
field, the acceleration to which the charged object is subjected with the acceleration of
Copyright © Christoph Schiller November 1997–May 2006

some unit charge q unit . In other words, we have

=
q ma
. (692)
q unit m unit a unit

Therefore any error in mass and acceleration measurements implies errors in measure-
ments of charge and the coupling constant.
Page 860 We found in the part on quantum theory that the electromagnetic, the weak and the
strong interactions are characterized by coupling constants, the inverse of which depend
linearly on the logarithm of the energy. It is usually assumed that these three lines meet
at the unification energy already mentioned. Measurements put the unification coupling

Dvipsbugw
 some experimental predictions 1025

Ref. 1071 value at about 1/26.

We know from the above discussions that the minimum measurement error for any
energy measurement at high energies is given by the ratio between the energy to be meas-
ured and the limit energy. Inserting this into the graph of the coupling constants ‘running’
with energy – as physicist like to say – we get the result shown in Figure 382. The search
for the consequences of this fan-out effect is delightful. One way to put the result is to say
that coupling constants are by definition subject to an error. However, all measurement
devices, be they clocks, metre rules, scales or any other device, use electromagnetic ef-
fects at energies of around 1 eV plus the electron rest energy. This is about 10−16 times the Dvipsbugw
GUT energy. As a consequence, the measurement precision of any observable is limited
to about 19 digits. The maximum precision presently achieved is 15 digits, and, for the
electromagnetic coupling constant, about 9 digits. It will thus be quite some time before
this prediction can be tested.
The fun is thus to find a system in which the spreading coupling constant value appears
more clearly in the measurements. For example, it may be that high-precision measure-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ments of the д-factor of elementary particles or of high-energy cosmic ray reactions will
show some effects of the fan-out. The lifetime of elementary particles could also be af-
Challenge 1441 n fected. Can you find another effect?
In summary, the experimental detection of quantum gravity effects should be possible,
despite their weakness, at some time during the twenty-first century. The successful de-
tection of any such effect will be one of the highlights of physics, as it will challenge the
usual description of space and time even more than general relativity did.
We now know that the fundamental entity describing space-time and matter that we
are looking for is not point-like. What does it look like? To get to the top of Motion
Mountain as rapidly as possible, we will make use of some explosive to blast away a few
disturbing obstacles.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
1026 xi general rel ativity versus quantum mechanics

Biblio graphy

1003 Steven Weinberg, The cosmological constant problem, Reviews of Modern Physics 61,
pp. 1–23, 1989. Cited on page 996.
1004 Steven Weinberg, The quantum theory of fields, Cambridge University Press, volume
I, 1995, and II, 1996. Cited on page 996.
1005 The difficulties are summarised by B.S. de Witt, Quantum field theory in curved space-
time, Physics Reports 19, pp. 295–357, 1975. Cited on page 997. Dvipsbugw
1006 C.W. Misner & K.S. Thorne, J.A. Wheeler, Gravitation, Freeman, 1973. Cited on
pages 998, 1000, and 1004.
1007 J.A. Wheeler, in Relativity, Groups and Topology, edited by C. DeWitt & B. DeWitt,
Gordon and Breach, 1994. See also J.A. Wheeler, Physics at the Planck length, Interna-
tional Journal of Modern Physics A 8, pp. 4013–4018, 1993. Cited on page 998.
1008 J.L. Friedman & R.D. Sorkin, Spin 1/2 from gravity, Physical Review Letters 44,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pp. 1100–1103, 1980. Cited on page 998.
1009 A.P. Bal achandran, G. Bimonte, G. Marmo & A. Simoni, Topology change
and quantum physics, Nuclear Physics B 446, pp. 299–314, 1995, http://www.arxiv.org/abs/
hep-th/9503046. Cited on page 998.
1010 Many themes discussed here resulted from an intense exchange with Luca Bombelli, now
at the Department of Physics and Astronomy of the University of Mississippi. These pages
are a reworked version of those published in French as C. Schiller, Le vide diffère-t-il
de la matière? in E. Gunzig & S. Diner, editeurs, Le vide – Univers du tout et du rien
– Des physiciens et des philosophes s’interrogent, Les Éditions de l’Université de Bruxelles,
1998. An older English version is also available as C. Schiller, Does matter differ from
vacuum? http://www.arxiv.org/abs/gr-qc/9610066. Cited on page 998.
1011 See for example R ichard P. Feynman, R.B. Leighton & M. Sands, The Feynman
Lectures on Physics, Addison Wesley, 1977. Cited on page 998.
1012 A. Ashtekar, Recent mathematical developments in quantum general relativity, pre-
print found at http://www.arxiv.org/abs/gr-qc/9411055. C. Rovelli, Ashtekar Formula-
tion of general relativity and loop-space non-perturbative quantum gravity: a report, Clas-
sical and Quantum Gravity 8, pp. 1613–1675, 1991. Cited on pages 999 and 1020.
1013 M.B. Green, J.H. Schwarz & E. Witten, Superstring theory, volume 1 and 2 (Cam-
bridge University Press, 1987). G. Veneziano, A stringy nature needs just two constants,
Europhysics Letters 2, pp. 199–204, 1986. Cited on pages 999 and 1020.
1014 S. Majid, Introduction to Braided Geometry and q-Minkowski Space, preprint down-
Copyright © Christoph Schiller November 1997–May 2006

loadable at http://www.arxiv.org/abs/hep-th/9410241. See also S. Majid, Beyond su-

persymmetry and quantum symmetry (an introduction to braided groups and braided
matrices), in Quantum Groups, Integrable Statistical Models and Knot Theory, edited by
M.-L. Ge & H.J. de Vega, World Scientific, 1993, pp. 231–282. Cited on pages 999
and 1021.
1015 C.J. Isham, Prima facie questions in quantum gravity, http://www.arxiv.org/abs/gr-qc/
9310031. Cited on page 999.
1016 Steven Weinberg, Gravitation and Cosmology, Wiley, 1972. Cited on pages 1000, 1003,
and 1004.
1017 The observations of black holes at the centre of galaxies and elsewhere are summarised by
R. Bl andford & N. Gehrels, Revisiting the black hole, Physics Today 52, June 1999.

Dvipsbugw
bibliography 1027

Cited on page 1000.

1018 M. Pl anck, Über irreversible Strahlungsvorgänge, Sitzungsberichte der Kaiserlichen
Akademie der Wissenschaften zu Berlin, pp. 440–480, 1899. Today it is commonplace to
use Dirac’s ħ instead of Planck’s h, which Planck originally called b. Cited on page 1001.
1019 The argument is given e.g. in E.P. Wigner, Relativistic invariance and quantum phenom-
ena, Reviews of Modern Physics 29, pp. 255–258, 1957. Cited on page 1001.
1020 The starting point for the following arguments is taken from M. Schön, Operative time
definition and principal indeterminacy, preprint http://www.arxiv.org/abs/gr-qc/9304024,
and from T. Padmanabhan, Limitations on the operational definition of space-time
Dvipsbugw
events and quantum gravity, Classical and Quantum Gravity 4, pp. L107–L113, 1987; see
also Padmanabhan’s earlier papers referenced there. Cited on page 1001.
1021 W. Heisenberg, Über den anschaulichen Inhalt der quantentheoretischen Kinematik
und Mechanik, Zeitschrift für Physik 43, pp. 172–198, 1927. Cited on page 1002.
1022 E.H. Kennard, Zur Quantenmechanik einfacher Bewegungstypen, Zeitschrift für Physik
44, pp. 326–352, 1927. Cited on page 1002.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1023 M.G. R aymer, Uncertainty principle for joint measurement of noncommuting variables,
American Journal of Physics 62, pp. 986–993, 1994. Cited on page 1002.
1024 H. Salecker & E.P. Wigner, Quantum limitations of the measurement of space-time
distances, Physical Review 109, pp. 571–577, 1958. Cited on page 1003.
1025 E.J. Zimmerman, The macroscopic nature of space-time, American Journal of Physics
30, pp. 97–105, 1962. Cited on page 1003.
1026 S.W. Hawking, Particle creation by black holes, Communications in Mathematical Phys-
ics 43, pp. 199–220, 1975; see also S.W. Hawking, Black hole thermodynamics, Physical
Review D13, pp. 191–197, 1976. Cited on page 1003.
1027 P. Gibbs, The small structure of space-time: a bibliographical review, unpublished pre-
print found at http://www.arxiv.org/abs/hep-th/9506171. Cited on pages 1002 and 1017.
1028 C.A. Mead, Possible connection between gravitation and fundamental length, Physical
Review B 135, pp. 849–862, 1964. Cited on pages 1003, 1004, 1006, and 1027.
1029 P.K. Townsend, Small-scale structure of space-time as the origin of the gravitational
constant, Physical Review D 15, pp. 2795–2801, 1977. Cited on page 1004.
1030 M.-T. Jaekel & S. R enaud, Gravitational quantum limit for length measurement, Phys-
ics Letters A 185, pp. 143–148, 1994. Cited on page 1004.
1031 D.V. Ahluwalia, Quantum measurement, gravitation and locality, Physics Letters B
339, pp. 301–303, 1994, also preprint http://www.arxiv.org/abs/gr-qc/9308007. Cited on
page 1004.
Copyright © Christoph Schiller November 1997–May 2006

1032 L. Garay, Quantum gravity and minimum length, International Journal of Modern Phys-
ics A 10, pp. 145–165, 1995, also preprint http://www.arxiv.org/abs/gr-qc/9403008; this pa-
per also includes an extensive bibliography. Cited on page 1004.
1033 G. Amelino–Camelia, Limits on the measurability of space-time distances in (the
semiclassical approximation of) quantum gravity, Modern Physics Letters A 9, pp. 3415–
3422, 1994, also preprint http://www.arxiv.org/abs/gr-qc/9603014. See also Y.J. Ng & H.
Van Dam, Limit to space-time measurement, Modern Physics Letters A 9, pp. 335–340,
1994. Cited on page 1004.
1034 The generalised indeterminacy relation is an implicit part of Ref. 1028, especially on page
852, but the issue is explained rather unclearly. Obviously the author found the result too
simple to be mentioned explicitly. Cited on pages 1004 and 1005.

Dvipsbugw
1028 xi general rel ativity versus quantum mechanics

1035 C. Rovelli & L. Smolin, Discreteness of area and volume in quantum gravity, Nuclear
Physics B 442, pp. 593–619, 1995. R. Loll, The volume operator in discretized quantum
gravity, preprint http://www.arxiv.org/abs/gr-qc/9506014. Cited on page 1005.
1036 D. Amati, M. Ciafaloni & G. Veneziano, Superstring collisions at Planckian en-
ergies, Physics Letters B 197, pp. 81–88, 1987. D.J. Gross & P.F. Mende, The high en-
ergy behavior of string scattering amplitudes, Physics Letters B 197, pp. 129–134, 1987. K.
Konishi, G. Paffu ti & P. Provero, Minimum physical length and the generalised
uncertainty principle, Physics Letters B 234, pp. 276–284, 1990. P. Aspinwall, Minimum
Distances in Non-Trivial String Target Spaces, Nuclear Physics B 431, pp. 78–96, 1994, also Dvipsbugw
preprint http://www.arxiv.org/abs/hep-th/9404060. Cited on page 1005.
1037 M. Maggiore, A generalised uncertainty principle in quantum mechanics, Physics Let-
ters B 304, pp. 65–69, 1993. Cited on page 1005.
1038 A simple approach is S. Doplicher, K. Fredenhagen & J.E. Roberts, Space-time
quantization induced by classical gravity, Physics Letters B 331, pp. 39–44, 1994, Cited on
pages 1005 and 1017.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1039 A. Kempf, Uncertainty relation in quantum mechanics with quantum group symmetry,
Journal of Mathematical Physics 35, pp. 4483–4496, 1994. A. Kempf, Quantum Groups
and quantum field theory with Nonzero minimal uncertainties in positions and momenta,
Czechoslovak Journal of Physics 44, pp. 1041–1048, 1994. Cited on page 1005.
1040 E.J. Hellund & K. Tanaka, Quantized space-time, Physical Review 94, pp. 192–195,
1954. Cited on page 1005.
1041 The impossibility to determine temporal ordering in quantum theory is discussed by J.
Oppenheimer, B. R eznik & W.G. Unruh, Temporal ordering in quantum mechan-
ics, Journal of Physics A 35, pp. 7641–7652, 2001, or http://www.arxiv.org/abs/quant-ph/
0003130. Cited on page 1006.
1042 A. Peres & N. Rosen, Quantum Limitations on the measurement of gravitational fields,
Physical Review 118, pp. 335–336, 1960. Cited on page 1006.
1043 It is the first definition in Euclid’s Elements, c. 300 bce. For an English translation see T.
Heath, The Thirteen Books of the Elements, Dover, 1969. Cited on page 1007.
1044 A beautiful description of the Banach–Tarski paradox is the one by Ian Stewart, Para-
dox of the spheres, New Scientist, 14 January 1995, pp. 28–31. Cited on page 1008.
1045 H.S. Snyder, Quantized space-time, Physical Review 71, pp. 38–41, 1947. H.S. Snyder,
The electromagnetic field in quantised of space-time, Physical Review 72, pp. 68–74, 1947.
A. Schild, Discrete space-time and integral Lorentz transformations, Physical Review
73, pp. 414–415, 1948. E.L. Hill, Relativistic theory of discrete momentum space and dis-
Copyright © Christoph Schiller November 1997–May 2006

crete space-time, Physical Review 100, pp. 1780–1783, 1950. H.T. Flint, The quantization
of space-time, Physical Review 74, pp. 209–210, 1948. A. Das, Cellular space-time and
quantum field theory, Il Nuovo Cimento 18, pp. 482–504, 1960. Cited on page 1009.
1046 D. Finkelstein, ‘Superconducting’ causal nets, International Journal of Theoretical Phys-
ics 27, pp. 473–519, 1985. Cited on page 1009.
1047 N.H. Christ, R. Friedberg & T.D. Lee, Random lattice field theory: general formu-
lation, Nuclear Physics B 202, pp. 89–125, 1982. G. ’t Hooft, Quantum field theory for
elementary particles – is quantum field theory a theory?, Physics Reports 104, pp. 129–142,
1984. Cited on page 1009.
1048 See for example L. B ombelli, J. Lee, D. Meyer & R.D. Sorkin, Space-time as
a causal set, Physical Review Letters 59, pp. 521–524, 1987. G. Brightwell & R.

Dvipsbugw
bibliography 1029

Gregory, Structure of random space-time, Physical Review Letters 66, pp. 260–263, 1991.
Cited on page 1010.
1049 For a discussion, see R. Sorabji, Time, Creation and the continuum: theories in antiquity
and the early Middle Ages, Duckworth, London 1983. Cited on page 1009.
1050 For quantum gravity, see for example A. Ashtekar, Quantum geometry in action: big
bang and black holes, http://www.arxiv.org/abs/math-ph/0202008. Cited on page 1010.
1051 The false belief that particles like quarks or electrons are composite is slow to die out. See,
e.g, S. Fredriksson, Preon prophecies by the standard Model, http://www.arxiv.org/
Dvipsbugw
abs/hep-ph/0309213. Preon models gained popularity in the 1970s and 1980s, in particular
through the papers by J.C. Pati & A. Sal am, Physical Review D 10,, p. 275, 1974, H.
Harari, Physics Letters B 86,, p. 83, 1979, M.A. Shupe, Physics Letters B 86,, p. 87, 1979,
and H. Fritzsch & G. Mandelbaum, Physics Letters B 102,, p. 319, 1981. Cited on
page 1011.
1052 A.D. Sakharov, Vacuum quantum fluctuations in curved space and the theory of grav-
itation, Soviet Physics – Doklady, 12, pp. 1040–1041, 1968. Cited on page 1012.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1053 C. Wolf, Upper limit for the mass of an elementary particle due to discrete time quantum
mechanics, Il Nuovo Cimento B 109, pp. 213–218, 1994. Cited on page 1012.
1054 N.F. R amsey & A. Weis, Suche nach permanenten elektrischen Dipolmomenten: ein
Test der Zeitumkehrinvarianz, Physikalische Blätter 52, pp. 859–863, 1996. The paper by
E.D. Commins, S.B. Ross, D. DeMille & B.C. R egan, Improved experimental
limit on the electric dipole moment of the electron, Physical Review A 50, p. 2960, 1994,
gives an upper experimental limit to dipole moment of the electron of 3.4 ë 10−29 e m. W.
Bernreu ther & M. Suzuki, The electric dipole moment of the electron, Reviews of
Modern Physics 63, pp. 313–340, 1991. See also the musings in Hans Dehmelt, Is the
electron a composite particle?, Hyperfine Interactions 81, pp. 1–3, 1993. Cited on page 1012.
1055 K. Akama, T. Hattori & K. Katsuura, Naturalness bounds on dipole moments
from new physics, http://www.arxiv.org/abs/hep-ph/0111238 Cited on page 1012.
1056 W.G. Unruh, Notes on black hole evaporation, Physical Review D 14, pp. 870–875, 1976.
W.G. Unruh & R.M. Wald, What happens when an accelerating observer detects a
Rindler particle, Physical Review D 29, pp. 1047–1056, 1984. Cited on page 1016.
1057 W. Jauch, Heisenberg’s uncertainty relation and thermal vibrations in crystals, American
Journal of Physics 61, pp. 929–932, 1993. Cited on page 1017.
1058 One way out is proposed in J. Magueijo & L. Smolin, Lorentz invariance with an
invariant energy scale, Physical Review Letters 88, p. 190403, 26 April 2002. They propose
a modification of the mass energy relation of the kind
Copyright © Christoph Schiller November 1997–May 2006

γmc 2
E= p=
γmv
γmc 2
and γmc 2
. (693)
1+ E Pl
1+ E Pl

Another, similar approach of recent years, with a different proposal, is called doubly special
relativity. A recent summary is G. Amelino–Camelia, Doubly-special relativity: first
results and key open problems, http://www.arxiv.org/abs/gr-qc/0210063. The paper shows
how conceptual problems hinder the advance of the field. Cited on page 1017.
1059 H.D. Zeh, On the interpretation of measurement in quantum theory, Foundations of Phys-
ics 1, pp. 69–76, 1970. Cited on page 1018.
1060 Steven Weinberg, The cosmological constant problem, Reviews of Modern Physics 61,
pp. 1–23, 1989. Cited on page 1019.

Dvipsbugw
1030 xi general rel ativity versus quantum mechanics

1061 See Y.J. Ng, W.A. Christiansen & H. van Dam, Probing Planck-scale physics with
extragalactic sources?, Astrophysical Journal 591, pp. L87–L90, 2003, also downloadable
electronically at http://www.arxiv.org/abs/astro-ph/0302372; D.H. Coule, Planck scale
still safe from stellar images, Classical and Quantum Gravity 20, pp. 3107–3112, 2003, also
available as electronic preprint at http://www.arxiv.org/abs/astro-ph/0302333. Negative ex-
perimental results (and not always correct calculations) are found in R. Lieu & L. Hill-
man, The phase coherence of light from extragalactic sources - direct evidence against first
order Planck scale fluctuations in time and space, Astrophysical Journal 585, pp. L77–L80,
2003, and R. R agazzoni, M. Turatto & W. Gaessler, The lack of observational Dvipsbugw
evidence for the quantum structure of spacetime at Planck scales, Astrophysical Journal
587, pp. L1–L4, 2003. Cited on page 1021.
1062 B.E. Schaefer, Severe limits on variations of the speed of light with frequency, Physical
Review Letters 82, pp. 4964–4966, 21 June 1999. Cited on page 1021.
1063 G. Amelino–Camelia, J. Ellis, N.E. Mavromatos, D.V. Nanopoulos & S.
Sakar, Potential sensitivity of gamma-ray-burster observations to wave dispersion in
vacuo, Nature 393, p. 763, 1998, and preprint http://www.arxiv.org/abs/astro-ph/9712103.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Cited on page 1021.
1064 G. Amelino–Camelia, Phenomenological description of space-time foam, preprint
found at http://www.arxiv.org/abs/gr-qc/0104005, 2001. The paper includes a clearly writ-
ten overview of present experimental approaches to detecting quantum gravity effects.
See also his update G. Amelino–Camelia, Quantum-gravity phenomenology: status
and prospects, preprint found at http://www.arxiv.org/abs/gr-qc/0204051, 2002. Cited on
pages 1021 and 1022.
1065 G. Amelino–Camelia, An interferometric gravitational wave detector as a quantum
gravity apparatus, Nature 398, p. 216, 1999, and preprint http://www.arxiv.org/abs/gr-qc/
9808029. Cited on page 1021.
1066 F. Karolyhazy, Gravitation and quantum mechanics of macroscopic objects, Il Nuovo
Cimento A42, pp. 390–402, 1966, Y.J. Ng & H. van Dam, Limit to space-time measure-
ment, Modern Physics Letters A 9, pp. 335–340, 1994, Y.J. Ng & H. Van Dam, Modern
Physics Letters A Remarks on gravitational sources, 10, pp. 2801–2808, 1995. The discussion
is neatly summarised in Y.J. Ng & H. van Dam, Comment on ‘Uncertainty in measure-
ments of distance’, http://www.arxiv.org/abs/gr-qc/0209021. See also Y.J. Ng, Spacetime
foam, preprint found at http://www.arxiv.org/abs/gr-qc/0201022. Cited on page 1021.
1067 L.J. Garay, Spacetime foam as a quantum thermal bath, Physics Review Letters 80,
pp. 2508–2511, 1998, also downloadable as http://www.arxiv.org/abs/gr-qc/9801024. Cited
on pages 1021 and 1022.
Copyright © Christoph Schiller November 1997–May 2006

1068 G. Amelino–Camelia & T. Piran, Planck scale-deformation of Lorentz symmetry

as a solution to the UHECR and the TeV γ paradoxes, preprint http://www.arxiv.org/
astro-ph/0008107, 2000. Cited on page 1022.
1069 Experimental tests were described by R. Lieu & L.W. Hillman, The phase coher-
ence of light from extragalactic sources: direct evidence against first-order Planck-scale
fluctuations in time and space, Astrophysical Journal 585, pp. L77–L80, 2003, and R.
R agazzoni & al., The lack of observational evidence for the quantum structure of space-
time at Planck scales, Astrophysical Journal 587, pp. L1–L4, 2003, Discussions on the pre-
dicted phase values are found in http://www.arxiv.org/abs/astro-ph/0302372 and http://
www.arxiv.org/abs/astro-ph/0302333. Cited on page 1022.
1070 At the unification scale, supersymmetry is manifest; all particles can transform into each
other. See for example ... Cited on page 1023.

Dvipsbugw
bibliography 1031

1071 See the paper by U. Amaldi, W. de B oer & H. Fürstenau, Comparison of grand
unified theories with elektroweak and strong coupling constants measured at LEP, Physics
Letters 260, pp. 447–455, 1991. This is the standard reference on the topic. Cited on pages
1023 and 1025.
1072 M.V. Romalis, W.C. Griffith, J.P. Jacobs & E.N. Fortson, New limit on the
permanent electric dipole moment of 199 Hg, Physical Review Letters 86, pp. 2505–2508, 19
March 2001, or http://www.arxiv.org/abs/hep-ex/0012001. Their upper limit on the dipole
moment of mercury atoms is 2.1 ë 1030 e m. Research for better measurement methods is
progressing; a modern proposal is by S.K. L amoreaux, Solid state systems for electron Dvipsbugw
electric dipole moment and other fundamental measurements, http://www.arxiv.org/abs/
nucl-ex/0109014. Cited on page 1023.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 1032 xi general rel ativity versus quantum mechanics

34. nature at l arge scales – is the universe

something or nothing?
Die Grenze ist der eigentlich fruchtbare Ort der

“ Erkenntnis.*
Paul Tillich, Auf der Grenze.

”
T
his strange question is the topic of the current leg of our mountain ascent. In
Ref. 1073 he last section we explored nature in the vicinity of Planck dimensions; but Dvipsbugw
he other limit, namely to study the description of motion at large, cosmological
scales, is equally fascinating. As we proceed, many incredible results will appear, and at
the end we will discover a surprising answer to the question in the section title.
This section is not standard textbook material; a large part of it is original** and thus
speculative and open to question. Even though this section aims at explaining in simple
words the ongoing research in the domains of quantum gravity and superstring theory,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

be warned. With every sentence of this section you will find at least one physicist who
disagrees!
We have discussed the universe as a whole several times already. In classical physics
we enquired about the initial conditions of the universe and whether it is isolated. In the
first intermezzo we asked whether the universe is a set or a concept and indeed, whether
it exists at all. In general relativity we gave the classical definition of the term, as the sum
of all matter and space-time, we studied the expansion of the universe and we asked about
its size and topology. In quantum theory we asked whether the universe has a wave func-
tion, whether it is born from a quantum fluctuation, and whether it allows the number
of particles to be defined.
Here we will settle all these issues by combining general relativity and quantum theory
at cosmological scales. That will lead us to some of the strangest results we will encounter
in our journey.

Cosmological scales
Hic sunt leones.***

“
The description of motion requires the application of general relativity whenever the scale
Antiquity

”
d of the situation are of the order of the Schwarzschild radius, i.e. whenever
Copyright © Christoph Schiller November 1997–May 2006

d  r S = 2Gm~c 2 . (694)

It is straightforward to confirm that, with the usually quoted mass m and size d of
Challenge 1442 n everything visible in the universe, this condition is indeed fulfilled. We do need general
relativity and thus curved space-time when talking about the whole of nature.

* ‘The frontier is the really productive place of understanding’. Paul Tillich (1886–1965), German theologian,
socialist and philosopher.
** Written between June and December 2000.
*** ‘Here are lions.’ Written across unknown and dangerous regions on ancient maps.

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1033

Similarly, quantum theory is required for the description of motion of an object

whenever we approach it within a distance d of the order of the Compton wavelength
λ C , i.e. whenever
d  λC =
h
. (695)
mc
Obviously, for the total mass of the universe this condition is not fulfilled. However, we
are not interested in the motion of the universe itself; we are interested in the motion
of its components. In the description of these components, quantum theory is required Dvipsbugw
whenever pair production and annihilation play a role. This is especially the case in the
early history of the universe and near the horizon, i.e. for the most distant events that
we can observe in space and time. We are thus obliged to include quantum theory in any
precise description of the universe.
Since at cosmological scales we need both quantum theory and general relativity, we
start our investigation with the study of time, space and mass, by asking at large scales

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the same questions that we asked above at Planck scales.

Maximum time
Is it possible to measure time intervals of any imaginable size? General relativity shows
that in nature there is a maximum time interval, with a value of about fourteen thou-
sand million years or 430 Ps, providing an upper limit to the measurement of time. It is
called the ‘age’ of the universe and has been deduced from two sets of measurements: the
expansion of space-time and the age of matter.
We all know of clocks that have been ticking for a long time: the hydrogen atoms in
our body. All hydrogen atoms were formed just after the big bang. We can almost say
that the electrons in these atoms have been orbiting the nuclei since the dawn of time.
In fact, inside the protons in these atoms, the quarks have been moving already a few
hundred thousand years longer. Anyway, we thus get a common maximum time limit for
Challenge 1443 n any clock made of atoms. Even ‘clocks’ made of radiation (can you describe one?) yield
a similar maximum time. Also the study of the spatial expansion of the universe leads
to the same maximum. No real or imaginable clock or measurement device was ticking
before this maximum time and no clock could provide a record of having done so.
In summary, it is not possible to measure time intervals greater than the maximum one,
either by using the history of space-time or by using the history of matter or radiation.*
The maximum time is thus rightly called the ‘age’ of the universe. Of course, all this is not
Copyright © Christoph Schiller November 1997–May 2006

new, although looking at the issue in more detail does provide some surprises.

Does the universe have a definite age?

One should never trust a woman who tells one

“ her real age. A woman who would tell one that,

would tell one anything.
Oscar Wilde

big bang there were other phenomena, cannot be based on nature or on observations. They are based on
beliefs.
”
* This conclusion implies that so-called ‘oscillating’ universe models, in which it is claimed that ‘before’ the

Dvipsbugw
 1034 xi general rel ativity versus quantum mechanics

Asking about the age of the universe may seem a silly question, because we have just
discussed it. Furthermore, the value is found in many books and tables, including that of
Page 1167 Appendix B, and its precise determination is actually one of the most important quests
in modern astrophysics. But is this quest reasonable?
In order to measure the duration of a movement or the age of a system, we need a
clock. The clock has to be independent of that movement or system and thus has to be
outside the system. However, there are no clocks outside the universe and, inside it, a clock
cannot be independent. In fact we have just seen that inside the universe, no clock can
run throughout its complete history. Indeed, time can be defined only once it is possible to Dvipsbugw
distinguish between matter and space-time. Once this distinction can be made, only the
two possibilities just discussed remain: we can either talk about the age of space-time, as
is done in general relativity, by assuming that matter provides suitable and independent
clocks; or we can talk about the age of matter, such as stars or galaxies, by assuming that
the extension of space-time or some other matter provides a good clock. Both possibilities
are being explored experimentally in modern astrophysics; both give the same result of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

about fourteen thousand million years that was mentioned previously. However, for the
universe as a whole, an age cannot be defined.
The issue of the starting point of time makes this difficulty even more apparent. We
may imagine that going back in time leads to only two possibilities: either the starting
instant t = 0 is part of time or it is not. (Mathematically, this means that the segment
describing time should be either closed or open.) Both cases assume that it is possible to
measure arbitrarily small times, but we know from the combination of general relativity
and quantum theory that this is not the case. In other words, neither possibility is incor-
rect: the beginning cannot be part of time, nor can it not be part of it. To this situation
there is only one solution: there was no beginning at all.
In other words, the situation is consistently muddled. Neither the age of the universe
nor its origin makes sense. What is going wrong? Or, more correctly, how are things going
wrong? In other words, what happens if instead of jumping directly to the big bang, we
approach it as closely as possible? The best way to clarify the issue is to ask about the
measurement error we are making when we say that the universe is fourteen thousand
million years old. This turns out to be a fascinating topic.

How precise can age measurements be?

No woman should ever be quite accurate about

“ her age. It looks so calculating.

Copyright © Christoph Schiller November 1997–May 2006

Oscar Wilde

Ref. 1075
The first way to measure the age of the universe* is to look at clocks in the usual sense of
the term, namely at clocks made of matter. As explained in the part on quantum theory,
Salecker and Wigner showed that a clock built to measure a total time T with a precision
”
* Note that the age t 0 is not the same as the Hubble time T = 1~H 0 . The Hubble time is only a computed
quantity and (almost) always larger than the age; the relation between the two depends on the value of the
cosmological constant, on the density and on other parameters of the universe. For example, for the standard
hot big bang scenario, i.e. for the matter-dominated Einstein–de Sitter model, we have the simple relation
Ref. 1074 T = (3~2) t 0 .

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1035

∆t has a minimum mass m given by

mA
ħ T
.
c 2 (∆t)2
(696)

A simple way to incorporate general relativity into this result was suggested by Ng and
Ref. 1076 van Dam. Any clock of mass m has a minimum resolution ∆t due to the curvature of
space that it introduces, given by
Dvipsbugw
∆t A 3 .
Gm
(697)
c
If m is eliminated, these two results imply that any clock with a precision ∆t can only
measure times T up to a certain maximum value, namely

(∆t)3
T< , (698)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2
t Pl
»
where t Pl = ħG~c 5 = 5.4 ë 10−44 s is the already familiar Planck time. (As usual, we
have omitted factors of order one in this and in all the following results of this section.)
In other words, the higher the accuracy of a clock, the shorter the time during which
the clock works dependably! The precision of a clock is not (only) limited by the budget
spent on building it, but by nature itself. Nevertheless, it does not take much to check that
Challenge 1444 e for clocks in daily life, this limit is not even remotely approached. For example, you may
want to deduce how precisely your own age can be specified.
As a consequence of (698), a clock trying to achieve an accuracy of one Planck time
can do so for at most one single Planck time! Simply put, a real clock cannot achieve Planck
time accuracy. If we try to go beyond limit (698), fluctuations of space-time hinder the
working of the clock and prevent higher precision. With every Planck time that passes,
the clock accumulates a measuring error of at least one Planck time. Thus, the total meas-
urement error is at least as large as the measurement itself. The conclusion is also valid
Challenge 1445 ny for clocks based on radiation, for example on background radiation.
In short, measuring age with a clock always involves some errors; whenever we try
to reduce these errors to Planck level, the clock becomes so imprecise that age measure-
ments become impossible.
Copyright © Christoph Schiller November 1997–May 2006

Does time exist?

Time is waste of money.

Page 45
“ Oscar Wilde

From the origins of physics onwards, the concept of ‘time’ has designated what is meas-
ured by a clock. Since equation (698) expresses the non-existence of perfect clocks, it also
”
implies that time is only an approximate concept, and that perfect time does not exist.
Thus there is no ‘idea’ of time, in the Platonic sense. In fact, all discussion in the previous
and present sections can be seen as proof that there are no perfect or ‘ideal’ examples of
any classical or everyday concept.

Dvipsbugw
 1036 xi general rel ativity versus quantum mechanics

Time does not exist. Despite this conclusion, time is obviously a useful concept in
everyday life. A simple explanation is provided when we focus on the importance of en-
ergy. Any clock, in fact any system of nature, is characterized by a simple number, namely
the highest ratio of the kinetic energy to the rest energy of its components. In daily life,
Challenge 1446 e this fraction is about 1 eV~10 GeV = 10−10 . Such low-energy systems are well suited to
building clocks. The more precisely the motion of the main moving part – the pointer of
the clock – can be kept constant and can be monitored, the higher the precision of the
clock becomes. To achieve the highest possible precision, the highest possible mass of
the pointer is required; indeed, both the position and the speed of the pointer must be Dvipsbugw
measured, and the two measurement errors are related by the quantum mechanical inde-
terminacy relation ∆v ∆x A ħ~m. High mass implies low intrinsic fluctuations. In order
to screen the pointer from outside influences, even more mass is needed. This connection
might explain why better clocks are usually more expensive than less accurate ones.
The usually quoted indeterminacy relation is valid only at everyday energies. Increas-
ing the mass does not allow to reach arbitrary small time errors, since general relativity
changes the indeterminacy relation to ∆v ∆x A ħ~m + G(∆v)2 m~c 3 . The additional term

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 1073
on the right hand side, negligible at everyday scales, is proportional to energy. Increas-
Page 1004 ing it by too large an amount limits the achievable precision of the clock. The smallest
Challenge 1447 e measurable time interval turns out to be the Planck time.
In summary, time exists as a good approximation only for low-energy systems. Any
increase in precision beyond a certain limit will require an increase in the energy of the
components; at Planck energy, this energy increase will prevent an increase in precision.

What is the error in the measurement of the age of the universe?

Applying the discussion about the measurement of time to the age of the universe is now
straightforward. Expression (698) implies that the highest precision possible for a clock
Challenge 1448 e is about 10−23 s, or about the time light takes to move across a proton. The finite age of the
universe also yields a maximum relative measurement precision. Expression (698) can be
written as
A ( )2~3
∆t t Pl
(699)
T T
which shows that no time interval can be measured with a precision of more than about
40 decimals.
In order to clarify the issue, we can calculate the error in measurement as a function
Copyright © Christoph Schiller November 1997–May 2006

of the observation energy E meas . There are two limit cases. For small energies, the error
is given by quantum effects as

∆t 1
(700)
T E meas

and thus decreases with increasing measurement energy. For high energies, however, the
error is given by gravitational effects as


∆t E meas
(701)
T E Pl

Dvipsbugw
nature at l arge scales – is the universe something or nothing? 1037

Relative
measurement error
1

quantum
total
error
error

quantum
gravity
Dvipsbugw
∆Emin error
E

Energy Eopt EPl

F I G U R E 383 Measurement errors as a function of measurement energy

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

so that the total result is as shown in Figure 383. In particular, energies that are too high
do not reduce measurement errors, because any attempt to reduce the measurement error
for the age of the universe below 10−23 s would require energies so high that the limits of
space-time would be reached, making the measurement itself impossible.
We reached this conclusion through an argument based on clocks made of particles.
Below we will find out that even by determining the age of the universe using space-time
expansion leads to the same limit.
Imagine to observe a tree which, as a result of some storm
or strong wind, has fallen towards second tree, touching it at
the very top, as shown in Figure 384. It is possible to determ-
ine the height of both trees by measuring their separation and
the angles at the base. The error in height will depend on the er- v
rors in separation and angles. Similarly, the age of the universe
follows from the present distance and speed of objects – such
as galaxies – observed in the night sky. The present distance d
T
corresponds to separation of the trees at ground level and the
speed v to the angle between the two trees. The Hubble time T
of the universe – which, as has already been mentioned, is usu-
ally assumed to be larger than the age of the universe – then cor- d
Copyright © Christoph Schiller November 1997–May 2006

responds to the height at which the two trees meet. This time
since the universe ‘started’, in a naive sense since the galaxies F I G U R E 384 Trees and
galaxies
‘separated’, is then given, within a factor of order one, by

T=
d
. (702)
v
This is in simple words the method used to determine the age of the universe from the
expansion of space-time, for galaxies with red-shifts below unity.* Of interest in the fol-

* At higher red-shifts, the speed of light, as well as the details of the expansion, come into play; if we continue

Dvipsbugw
 1038 xi general rel ativity versus quantum mechanics

lowing is the (positive) measurement error ∆T, which becomes

=
∆T ∆d ∆v
+ . (703)
T d v
Exploring this in more detail is worthwhile. For any measurement of T we have to choose
the object, i.e. a distance d, as well as an observation time ∆t, or, equivalently, an observa-
tion energy ∆E = 2πħ~∆t. We will now investigate the consequences of these choices for
expression (703), always taking into account both quantum theory and general relativity. Dvipsbugw
At everyday energies, the result of the determination of the age t 0 is about
14  2 ë 109 Ga. This value is deduced by measuring red-shifts, i.e. velocities, and dis-
tances, using stars and galaxies in distance ranges from some hundred thousand light
years up to a red-shift of about 1. Measuring red-shifts does not produce large velocity er-
rors. The main source of experimental error is the difficulty in determining the distances
of galaxies.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

What is the smallest possible error in distance? Obviously, equation (699) implies

2~3
∆d l Pl
A 2~3 (704)
T d
thus giving the same indeterminacy in the age of the universe as found above in the case
Challenge 1449 e of material clocks.
We can try to reduce this error in two ways: by choosing objects at either small or large
distances. Let us start with the smallest possible distances. In order to get high precision at
Challenge 1450 e small distances, we need high observation energies. It is fairly obvious that at observation
energies near the Planck value, the value of ∆T~T approaches unity. In fact, both terms
on the right-hand side of expression (703) become of order one. At these energies, ∆v
approaches c and the maximum value for d approaches the Planck length, for the same
reason that at Planck energies the maximum measurable time is the Planck time. In short,
at Planck scales it is impossible to say whether the universe is old or young.
Let us continue with the other extreme, namely objects extremely far away, say with
a red-shift of z Q 1. Relativistic cosmology requires the diagram of Figure 384 to be
replaced by the more realistic diagram of Figure 385. The ‘light onion’ replaces the familiar
Ref. 1074 light cone of special relativity: light converges near the big bang.
In this case the measurement error for the age of the universe also depends on the
Copyright © Christoph Schiller November 1997–May 2006

distance and velocity errors. At the largest possible distances, the signals an object must
send out must be of high energy, because the emitted wavelength must be smaller than
the universe itself. Thus, inevitably we reach Planck energies. However, we saw that in
such high-energy situations, the emitted radiation, as well as the object itself, are indis-
tinguishable from the space-time background. In other words, the red-shifted signal we
would observe today would have a wavelength as large as the size of the universe, with a
correspondingly small frequency.

with the image of inclined trees, we find that the trees are not straight all the way up to the top and that they
grow on a slope, as shown in Figure 385.

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1039

Another way to describe the situation is the fol- big bang

lowing. At Planck energies or near the horizon, the space
original signal has an error of the same size as the
signal itself. When measured at the present time,
the red-shifted signal still has an error of the same 4ct0/9
size as the signal. As a result, the error on the ho- light cone:
rizon distance becomes as large as the value to be what we
measured. time
can see
In short, even if space-time expansion and large Dvipsbugw
scales are used, the instant of the so-called begin- other galaxies
ning of the universe cannot be determined with an in the night sky
error smaller than the age of the universe itself, a
result we also found at Planck distances. Whenever t0 our galaxy
we aim for perfect precision, we find that the uni-
verse is 14  14 thousand million years old! In other F I G U R E 385 The speed and distance of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

words, at both extremal situations it is impossible to remote galaxies
say whether the universe has a non-vanishing age.
We have to conclude that the anthropomorphic concept of ‘age’ does not make any
sense for the universe as a whole. The usual textbook value is useful only for domains in
time, space and energy for which matter and space-time are clearly distinguished, namely
at everyday, human-scale energies; however, this anthropocentric value has no overall
meaning.
Challenge 1451 ny By the way, you may like to examine the issue of the fate of the universe using the
same arguments. In the text, however, we continue on the path outlined at the start of
this section; the next topic is the measurement of length.

Maximum length
General relativity shows that in the standard cosmological model, for hyperbolic (open)
and parabolic (marginal) evolutions of the universe, the actual size of the universe is in-
finite. It is only the horizon distance, i.e. the distance of objects with infinite red-shift,
which is finite. In a hyperbolic or parabolic universe, even though the size is infinite, the
most distant visible events (which form the horizon) are at a finite distance.* For elliptical
evolution, the total size is finite and depends on the curvature. However, in this case also
Page 460 the present measurement limit yields a minimum size for the universe many times larger
Copyright © Christoph Schiller November 1997–May 2006

than the horizon distance. At least, this is what general relativity says.
On the other hand, quantum field theory is based on flat and infinite space-time. Let
us see what happens when the two theories are combined. What can we say about meas-
urements of length in this case? For example, would it be possible to construct and use a
metre rule to measure lengths larger than the distance to the horizon? It is true that we
would have no time to push it up to there, since in the standard Einstein–de Sitter big

* In cosmology, we need to distinguish between the scale factor R, the Hubble radius c~H = cR~Ṙ, the
horizon distance h and the size d of the universe. The Hubble radius is a computed quantity giving the
distance at which objects move away with the speed of light. It is always smaller than the horizon distance,
at which in the standard Einstein–de Sitter model, for example, objects move away with twice the speed of
Ref. 1074 light. However, the horizon itself moves away with three times the speed of light.

Dvipsbugw
 1040 xi general rel ativity versus quantum mechanics

Ref. 1074 bang model the horizon moves away from us faster than the speed of light. We should
have started using the metre rule right at the big bang.
For fun, let us assume that we have actually managed to do this. How far away can
we read off distances? In fact, since the universe was smaller in the past and since every
observation of the sky is an observation of the past, Figure 385 shows that the maximum
Ref. 1074 spatial distance an object can be seen away from us is only (4~9)ct 0 . Obviously, for space-
time intervals, the maximum remains ct 0 .
Thus, in all cases it turns out to be impossible to measure lengths larger than the ho-
rizon distance, even though general relativity predicts such distances. This unsurprising Dvipsbugw
result is in obvious agreement with the existence of a limit for measurements of time
intervals. The real surprises come now.

Is the universe really a big place?

Ref. 1077 Astronomers and Hollywood movies answer this question in the affirmative. Indeed, the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 1167 distance to the horizon of the universe is usually included in tables. Cosmological models
specify that the scale factor R, which fixes the distance to the horizon, grows with time
t; for the case of the usually assumed mass-dominated Einstein–de Sitter model, i.e. for
a vanishing cosmological constant and flat space, we have

R(t) = C t 2~3 , (705)

where the numerical constant C relates the commonly accepted horizon distance to the
commonly accepted age. Indeed, observation shows that the universe is large and is still
getting larger. But let us investigate what happens if we add to this result from general
relativity the limitations of quantum theory. Is it really possible to measure the distance
to the horizon?
We look first at the situation at high energies. We saw above that space-time and mat-
ter are not distinguishable at Planck scales. Therefore, at Planck energies we cannot state
whether objects are localized or not. At Planck scales, a basic distinction of our thinking,
namely the one between matter and vacuum, becomes obsolete. Equivalently, it is not
possible to claim that space-time is extended at Planck scales. Our concept of extension
derives from the possibility of measuring distances and time intervals, and from obser-
vations such as the ability to align several objects behind one another. Such observations
are not possible at Planck scales. In fact, none of the observations in daily life from which
Copyright © Christoph Schiller November 1997–May 2006

we deduce that space is extended are possible at Planck scales. At Planck scales, the basic
distinction between vacuum and matter, namely the opposition between extension and loc-
alization, disappears. As a consequence, at Planck energies the size of the universe cannot
be measured. It cannot even be called larger than a match box.
At cosmological distances, the situation is even easier. All the arguments given above
on the errors in measurement of the age can be repeated for the distance to the hori-
Challenge 1452 ny zon. Essentially, at the largest distances and at Planck energies, the measurement errors
are of the same magnitude as the measured value. All this happens because length meas-
urements become impossible at nature’s limits. This is corroborated by the lack of any
standard with which to compare the size of the universe.
Studying the big bang also produces strange results. At Planck energies, whenever we

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1041

try to determine the size of the big bang, we cannot claim that the universe was smaller
than the present size. At Planck energies, there is no way to distinguish length values.
Somehow, Planck dimensions and the size of the universe get confused.
There are also other confirmations. Let us go back to the example above. If we had a
metre rule spanning the whole universe, even beyond the horizon, with zero at the place
where we live, what measurement error would it produce for the horizon? It does not take
Challenge 1453 ny long to discover that the expansion of space-time from Planck scales to the present also
expands the indeterminacy in the Planck size into one of the order of the distance to the
horizon. The error is as large as the measurement result. Dvipsbugw
Since this also applies when we try to measure the diameter of the universe instead
of its radius, it becomes impossible to state whether the antipodes in the sky really are
distant from each other!
We can summarize the situation by noting that anything said about the size of the
universe is as limited as anything said about its age. The height of the sky depends on the
observation energy. At Planck energies, it cannot be distinguished from the Planck length.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

If we start measuring the sky at standard observation energies, trying to increase the
precision of measurement of the distance to the horizon, the measurement error increases
beyond all bounds. At Planck energies, the volume of the universe is indistinguishable
from the Planck volume!

The boundary of space-time – is the sky a surface?

The horizon of the universe, essentially the black part of the night sky, is a fascinating
entity. Everybody interested in cosmology wants to know what happens there. In news-
papers the horizon is sometimes called the boundary of space-time. Some surprising in-
sights, not yet common in newspapers, appear when the approaches of general relativity
and quantum mechanics are combined.
We saw above that the errors in measuring the distance of the horizon are substantial.
They imply that we cannot pretend that all points of the sky are equally far away from
us. Thus we cannot say that the sky is a surface. There is even no way to determine the
dimensionality of the horizon or the dimensionality of space-time near the horizon.*
Measurements thus do not allow us to determine whether the boundary is a point, a
surface, or a line. It may be an arbitrary complex shape, even knotted. In fact, quantum
theory tells us that it must be all of these from time to time, in short, that the sky fluctuates
in height and shape. In short, it is impossible to determine the topology of the sky. But
Copyright © Christoph Schiller November 1997–May 2006

that is nothing new. As is well known, general relativity is unable to describe pair creation
particles with spin 1/2. The reason for this is the change in space-time topology required
by the process. On the other hand, the universe is full of such processes, implying that it is
impossible to define a topology for the universe and, in particular, to talk of the topology
Challenge 1455 ny of the horizon itself. Are you able to find at least two other arguments to show this?
Worse still, quantum theory shows that space-time is not continuous at a horizon, as

* In addition, the measurement errors imply that no statement can be made about translational symmetry at
Challenge 1454 ny cosmological scales. Are you able to confirm this? In addition, at the horizon it is impossible to distinguish
between spacelike and timelike distances. Even worse, concepts such as ‘mass’ or ‘momentum’ are muddled
at the horizon. This means that, as at Planck energies, we are unable to distinguish between objects and the
background, and between state and intrinsic properties. We will come back to this important point shortly.

Dvipsbugw
 1042 xi general rel ativity versus quantum mechanics

Ref. 1073 can easily be deduced by applying the Planck-scale arguments from the previous section.
Time and space are not defined there.
Finally, there is no way to decide whether the various boundary points are different
from each other. The distance between two points in the night sky is undefined. In other
words, it is unclear what the diameter of the horizon is.
In summary, the horizon has no specific distance or shape. The horizon, and thus the
universe, cannot be shown to be manifolds. This leads to the next question:

Does the universe have initial conditions? Dvipsbugw

Ref. 1078 One often reads about the quest for the initial conditions of the universe. But before join-
ing this search, we should ask whether and when such initial conditions make any sense.
Obviously, our everyday description of motion requires them. Initial conditions describe
the state of a system, i.e. all those aspects that differentiate it from a system with the same
intrinsic properties. Initial conditions, like the state of a system, are attributed to a system

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

by an outside observer.
More specifically, quantum theory tells us that initial conditions or the state of a system
can only be defined by an outside observer with respect to an environment. It is already
difficult to be outside the universe. In addition, quite independently of this issue, even
inside the universe a state can only be defined if matter can be distinguished from vacuum.
However, this is impossible at Planck energies, near the big bang, or at the horizon. Thus
the universe has no state. No state also means that there is no wave function of the universe.
Ref. 1073 The limits imposed by the Planck values confirm this conclusion in other ways. First
of all, they show that the big bang was not a singularity with infinite curvature, density or
temperature, because infinitely large values do not exist in nature. Second, since instants
of time do not exist, it is impossible to define the state of any system at a given time. Third,
as instants of time do not exist, neither do events exist, and thus the big bang was not an
event, so that for this more prosaic reason, neither an initial state nor an initial wave
function can be ascribed to the universe. (Note that this also means that the universe
cannot have been created.)
In short, there are no initial conditions for the universe. Initial conditions make sense
only for subsystems and only far away from Planck scales. Thus, for initial conditions to
exist, the system must be far away from the horizon and it must have evolved for some
time ‘after’ the big bang. Only when these two requirements are fulfilled can objects move
in space. Of course, this is always the case in everyday life.
Copyright © Christoph Schiller November 1997–May 2006

At this point in our mountain ascent, where time and length are unclearly defined
at cosmological scales, it should come as no surprise that there are similar difficulties
concerning the concept of mass.

Does the universe contain particles and stars?

The number of stars, about 10231 , is included in every book on cosmology, as it is in the
table of Appendix B.1167 A subset of this number can be counted on clear nights. If we ask
the same question about particles instead of stars, the situation is similar. The commonly
quoted number of baryons is 10811 , together with 10901 photons. However, this does not
settle the issue. Neither quantum theory nor general relativity alone make predictions

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1043

about the number of particles, either inside or outside the horizon. What happens if we
combine them?
In order to define the number of particles in a region, quantum theory first of all re-
Page 773 quires a vacuum state to be defined. The number of particles is defined by comparing the
system with the vacuum. If we neglect or omit general relativity by assuming flat space-
time, this procedure poses no problem. However, if we include general relativity and thus
a curved space-time, especially one with such a strangely behaved horizon as the one we
have just found, the answer is simple: there is no vacuum state with which we can com-
pare the universe, for two reasons. First, nobody can explain what an empty universe Dvipsbugw
would look like; second, and more importantly, there is no way to define a state of the
universe at all. The number of particles in the universe thus becomes undefinable. Only
at everyday energies and for finite dimensions are we able to speak of an approximate
number of particles.
Comparison between a system and the vacuum is also impossible in the case of the
universe for purely practical reasons. The requirement for such a comparison effectively

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

translates into the requirement that the particle counter be outside the system. (Can you
Challenge 1456 ny confirm the reason for this connection?) In addition, it is impossible to remove particles
from the universe. The impossibility of defining a vacuum state, and thus the number
of particles in the universe, is not surprising. It is an interesting exercise to investigate
the measurement errors that appear when we try to determine the number of particles
Challenge 1457 ny despite this fundamental impossibility.
Can we count the stars? In principle, the same conclusion applies as for particles. How-
ever, at everyday energies the stars can be counted classically, i.e. without taking them out
of the volume in which they are enclosed. For example, this is possible if the stars are dif-
ferentiated by mass, colour or any other individual property. Only near Planck energies
or near the horizon are these methods inapplicable. In short, the number of stars is only
defined as long as the observation energy is low, i.e. as long as we stay away from Planck
energies and from the horizon.
In short, despite what appears to be the case on human scales, there is no definite num-
ber of particles in the universe. The universe cannot be distinguished from vacuum by
counting particles. Even though particles are necessary for our own existence and func-
tioning, a complete count of them cannot be made.
This conclusion is so strange that we should not accept it too easily. Let us try an-
other method of determining the content of matter in the universe: instead of counting
particles, let us weigh them.
Copyright © Christoph Schiller November 1997–May 2006

Does the universe contain masses and objects?

Page 1167 The average density of the universe, about 10−26 kg~m3 , is frequently cited in texts. Is it
different from a vacuum? Quantum theory shows that, as a result of the indeterminacy
relation, even an empty volume of size R has a mass. For a zero-energy photon inside
such a vacuum, we have E~c = ∆p A ħ~∆x, so that in a volume of size R, we have a
minimum mass of at least m mi n (R) = h~cR. For a spherical volume of radius R there is

Dvipsbugw
 1044 xi general rel ativity versus quantum mechanics

thus a minimal mass density given approximately by

m mi n (R)
ρ min  = 4 .
ħ
(706)
R 3 cR
For the universe, if the standard horizon distance R 0 of 14 000 million light years is in-
serted, the value becomes about 10−142 kg~m3 . This describes the density of the vacuum.
In other words, the universe, with a density of about 10−26 kg~m3 , seems to be clearly
different from vacuum. But are we sure? Dvipsbugw
We have just deduced that the radius of the horizon is undefined: depending on the
observation energy, it can be as small as the Planck length. This implies that the density
of the universe lies somewhere between the lowest possible value, given by the density
of vacuum just mentioned, and the highest possible one, namely the Planck density.* In
short, relation (706) does not really provide a clear statement.
Another way to measure the mass of the universe would be to apply the original defini-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 310 tion of mass, as given by Mach and modified by special relativity. Thus, let us try to collide
a standard kilogram with the universe. It is not hard to see that whatever we do, using
either low or high energies for the standard kilogram, the mass of the universe cannot be
Challenge 1459 ny constrained by this method. We would need to produce or to measure a velocity change
∆v for the rest of the universe after the collision. To hit all the mass in the universe at the
same time, we need high energy; but then we are hindered by Planck energy effects. In
addition, a properly performed collision measurement would require a mass outside the
universe, a rather difficult feat to achieve.
Yet another way to measure the mass would be to determine the gravitational mass
of the universe through straightforward weighing. But the lack of balances outside the
universe makes this an impractical solution, to say the least.
A way out might be to use the most precise definition of mass provided by general
relativity, the so-called ADM mass. However, for definition this requires a specified beha-
viour at infinity, i.e. a background, which the universe lacks.
We are then left with the other general relativistic method: determining the mass of
the universe by measuring its average curvature. Let us take the defining expressions for
Page 417 average curvature κ for a region of size R, namely

3 4πR 2 − S 15 4πR 3 ~3 − V
κ= = =
1
. (708)
2
r curvature 4π R4 4π R5
Copyright © Christoph Schiller November 1997–May 2006

We have to insert the horizon radius R 0 and either its surface area S 0 or its volume V0 .
However, given the error margins on the radius and the volume, especially at Planck
Challenge 1460 ny energies, for the radius of curvature we again find no reliable result.

* In fact, at everyday energies the density of the universe lies almost exactly between the two values, yielding
Challenge 1458 ny the strange relation
m 0 ~R 0  m Pl ~R Pl = c ~G .
2 2 2 2 4 2
(707)
Page 869 But this is nothing new. The approximate equality can be deduced from equation 16.4.3 (p. 620) of Steven
Weinberg, Gravitation and Cosmology, Wiley, 1972, namely Gn b m p = 1~t 02 . The relation is required by
several cosmological models.

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1045

An equivalent method starts with the usual expression for the indeterminacy ∆κ in
Ref. 1079 the scalar curvature for a region of size R provided by Rosenfeld, namely
2
16πl Pl
∆κ A . (709)
R4
However, this expression also shows that the error in the radius of curvature behaves like
the error in the distance to the horizon.
In summary, at Planck energies, the average radius of curvature of nature turns out to lie Dvipsbugw
between infinity and the Planck length. This implies that the density of matter lies between
the minimum value and the Planck value. There is thus no method to determine the mass
Challenge 1461 ny of the universe at Planck energies. (Can you find one?) The concept of mass cannot be
applied to the universe as a whole. Thus, the universe has no mass.

Do symmetries exist in nature?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We have already seen that at the horizon, space-time translation symmetry breaks down.
Let us have a quick look at the other symmetries.
What happens to permutation symmetry? Exchange is an operation on objects in
space-time. Exchange thus automatically requires a distinction between matter, space and
time. If we cannot distinguish positions, we cannot talk about exchange of particles. How-
ever, this is exactly what happens at the horizon. In short, general relativity and quantum
theory together make it impossible to define permutation symmetry at the horizon.
CPT symmetry suffers the same fate. As a result of measurement errors or of limiting
maximum or minimum values, it is impossible to distinguish between the original and
the transformed situations. It is therefore impossible to maintain that CPT is a symmetry
of nature at horizon scales. In other words, matter and antimatter cannot be distinguished
at the horizon.
Challenge 1462 ny The same happens with gauge symmetry, as you may wish to check in detail yourself.
For its definition, the concept of gauge field requires a distinction between time, space
and mass; at the horizon this is impossible. We therefore also deduce that at the horizon
also concepts such as algebras of observables cannot be used to describe nature. Renor-
malization also breaks down.
All symmetries of nature break down at the horizon. The complete vocabulary we use
when we talk about observations, including terms such as such as magnetic field, electric
field, potential, spin, charge, or speed, cannot be used at the horizon. And that is not all.
Copyright © Christoph Schiller November 1997–May 2006

Does the universe have a boundary?

It is common to take ‘boundary’ and ‘horizon’ as synonyms in the case of the universe,
because they are the same for all practical purposes. To study this concept, knowledge of
mathematics does not help us; the properties of mathematical boundaries, e.g. that they
themselves have no boundary, are not applicable in the case of nature, since space-time
is not continuous. We need other, physical arguments.
The boundary of the universe is obviously supposed to represent the boundary
between something and nothing. This gives three possibilities:

Dvipsbugw
 1046 xi general rel ativity versus quantum mechanics

— ‘Nothing’ could mean ‘no matter’. But we have just seen that this distinction cannot be
made at Planck scales. As a consequence, the boundary will either not exist at all or it
will encompass the horizon as well as the whole universe.
— ‘Nothing’ could mean ‘no space-time’. We then have to look for those domains where
space and time cease to exist. These occur at Planck scales and at the horizon. Again,
the boundary will either not exist or it will encompass the whole universe.
— ‘Nothing’ could mean ‘neither space-time nor matter.’ The only possibility is a bound-
ary that encloses domains beyond the Planck scale and beyond the horizon; but again,
such a boundary would also encompass all of nature. Dvipsbugw

This result is puzzling. When combining quantum theory and relativity, we do not
seem to be able to find a conceptual definition of the horizon that distinguishes the hori-
Challenge 1463 ny zon from what it includes. In fact, if you find one, publish it! A distinction is possible in
general relativity; and equally, a distinction is possible in quantum theory. However, as
soon as we combine the two, the boundary becomes indistinguishable from its content.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The interior of the universe cannot be distinguished from its horizon. There is no boundary.
A difficulty to distinguish the horizon and its contents is definitely interesting; it sug-
gests that nature may be symmetric under transformations that exchange interiors and
boundaries. This connection is nowadays called holography because it vaguely recalls the
working of credit card holograms. It is a busy research field in present-day high-energy
Ref. 1080 physics. However, for the time being, we shall continue with our original theme, which
directly leads us to ask:

Is the universe a set? – Again

We are used to calling the universe the sum of all matter and all space-time. In other
words, we imply that the universe is a set of components, all different from each other.
This idea was introduced in three situations: it was assumed that matter consists of
particles, that space-time consists of events (or points) and that the set of states consists
of different initial conditions. However, our discussion so far shows that the universe is
not a set of such distinguishable elements. We have encountered several proofs: at the ho-
rizon, at the big bang and at Planck scales distinction between events, between particles,
between observables and between space-time and matter becomes impossible. In those
domains, distinctions of any kind become impossible. We have found that any distinction
between two entities, such as between a toothpick and a mountain, is only approximately
possible. The approximation is possible because we live at energies much smaller than
Copyright © Christoph Schiller November 1997–May 2006

the Planck energy. Obviously, we are able to distinguish cars from people and from tooth-
picks; the approximation is so good that we do not notice the error when we perform
it. Nevertheless, the discussion of the situation at Planck energies shows that a perfect
distinction is impossible in principle. It is impossible to split the universe into separate
entities.
Another way to reach this result is the following. Distinguishing between two entit-
ies requires different measurement results, such as different positions, masses, sizes, etc.
Whatever quantity we choose, at Planck energies the distinction becomes impossible.
Only at everyday energies is it approximately possible.
In short, since the universe contains no distinguishable entities, the universe is not a

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1047

Page 681 set. We have already envisaged this possibility in the first intermezzo; now it is confirmed.
The concept of ‘set’ is already too specialized to describe the universe. The universe must
be described by a mathematical concept that does not contain any set.
This is a powerful result: it means that the universe cannot be described precisely if
any of the concepts used for its description presuppose the use of sets. But all concepts
we have used so far to describe nature, such as space-time, phase space, Hilbert space and
its generalizations, namely Fock space and particle space, are based on sets. They all must
be abandoned at Planck energies, as well as in any precise description.
Furthermore, many speculations about unified descriptions do not satisfy the criterion Dvipsbugw
that sets must not be included. In particular, all studies of quantum fluctuations, mathem-
atical categories, posets, complex mathematical spaces, computer programs, Turing ma-
chines, Gödel’s theorem, creation of any sort, space-time lattices, quantum lattices and
Bohm’s unbroken wholeness fail to satisfy this requirement. In addition, almost none of
the speculations about the origin of the universe can be correct. For example, you may
Challenge 1464 ny wish to check the religious explanations you know against this result. In fact, no approach

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

used by theoretical physics in the year 2000 satisfies the requirement that sets must be
abandoned; perhaps a future version of string or M theory will do so. The task is not easy;
Challenge 1465 n do you know of a single concept not based on a set?
Note that this radical conclusion is deduced from only two statements: the necessity
of using quantum theory whenever the dimensions are of the order of the Compton
wavelength, and the necessity to use general relativity whenever the dimensions are of
the order of the Schwarzschild radius. Together, they mean that any precise description
of nature cannot contain sets. We have reached this result after a long and interesting,
but in a sense unnecessary, digression. The difficulties in complying with this result may
explain why the unification of the two theories has not so far been successful. Not only
does unification require that we stop using space, time and mass for the description of
nature; it also requires that all distinctions, of any kind, should be only approximate. But
all physicists have been educated on the basis of exactly the opposite creed!
Note that, because it is not a set, the universe is not a physical system. Specifically, it has
no state, no intrinsic properties, no wave function, no initial conditions, no density, no
entropy and no cosmological constant. The universe is thus neither thermodynamically
closed nor open; and it contains no information. All thermodynamic quantities, such as
entropy, temperature and free energy, are defined using ensembles. Ensembles are limits
of systems which are thermodynamically either open or closed. As the universe is neither
of these two, no thermodynamic quantity can be defined for it.* All physical properties
Copyright © Christoph Schiller November 1997–May 2006

are defined only for parts of nature which are approximated or idealized as sets, and thus
are physical systems.

* Some people knew this long before physicists; for example, the belief that the universe is or contains in-
formation was ridiculed most thoroughly in the popular science fiction parody by Douglas Adams, The
Hitchhiker’s Guide to the Galaxy, 1979, and its sequels.

Dvipsbugw
 1048 xi general rel ativity versus quantum mechanics

Curiosities and fun challenges about the universe

Insofern sich die Sätze der Mathematik auf die

“ Wirklichkeit beziehen, sind sie nicht sicher, und

sofern sie sicher sind, beziehen sie sich nicht auf
die Wirklichkeit.*
Albert Einstein

The contradictions between the term ‘universe’ and the concept of ‘set’ lead to numerous
fascinating issues. Here are a few.
” Dvipsbugw
**
In mathematics, 2 + 2 = 4. This statement is an idealization of statements such as ‘two
apples plus two apples makes four apples.’ However, we now know that at Planck energies
this is not a correct statement about nature. At Planck energies, objects cannot be counted.
Challenge 1466 ny Are you able to confirm this?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
In 2002, Seth Lloyd estimated how much information the universe can contain, and how
Ref. 1081 many calculations it has performed since the big bang. This estimate is based on two
ideas: that the number of particles in the universe is a well-defined quantity, and that
the universe is a computer, i.e. a physical system. We now know that neither assumption
is correct. This example shows the power of the criteria that we deduced for the final
description of motion.
**
People take pictures of the cosmic background radiation and its variations. Is it possible
that the photographs will show that the spots in one direction of the sky are exactly the
Challenge 1467 ny same as those in the diametrically opposite direction?

**
In 1714, Leibniz published his Monadologie. In it he explores what he calls a simple sub-
stance, which he defined to be a substance that has no parts. He called it a monade and
Ref. 1082 explores some of its properties. However, due mainly to its incorrect deductions, the term
has not been taken over by others. Let us forget the strange deductions and focus only on
Challenge 1468 n the definition: what is the physical concept most related to that of monade?
Copyright © Christoph Schiller November 1997–May 2006

**
We usually speak of the universe, implying that there is only one of them. Yet there is a
simple case to be made that ‘universe’ is an observer-dependent concept, since the idea
Challenge 1469 ny of ‘all’ is observer-dependent. Does this mean that there are many universes?
**
If all particles would be removed – assuming one would know where to put them – there
Challenge 1470 ny wouldn’t be much of a universe left. True?

* In so far as mathematical statements describe reality, they are not certain, and as far as they are certain,
they are not a description of reality.

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1049

**
At Planck energies, interactions cannot be defined. Therefore, ‘existence’ cannot be
Challenge 1471 ny defined. In short, at Planck energies we cannot say whether particles exist. True?

Hilbert’s sixth problem settled

Page 657 In the year 1900, David Hilbert gave a famous lecture in which he listed 23 of the great chal-
Ref. 1083 lenges facing mathematics in the twentieth century. Most of these problems provided chal-
lenges to many mathematicians for decades afterwards. A few are still unsolved, among Dvipsbugw
them the sixth, which challenged mathematicians and physicists to find an axiomatic
treatment of physics.
Since the universe is not even a set, we can deduce that such an axiomatic description
of nature is impossible. The reasoning is simple; all mathematical systems, be they algeb-
raic systems, order systems or topological systems, are based on sets. Mathematics does
not have axiomatic systems that do not contain sets. The reason for this is simple: any

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

(mathematical) concept contains at least one set. However, nature does not.
The impossibility of an axiomatic system for physics is also confirmed in another way.
Physics starts with a circular definition: space-time is defined with the help of objects and
Page 157 objects are defined with the help of space-time. Physics thus has never been modelled on
the basis of mathematics. Physicists have always had to live with logical problems.
The situation is similar to a child’s description of the sky as ‘made of air and clouds’.
Looking closely, we discover that clouds are made up of water droplets. However, there is
air inside clouds, and there is also water vapour elsewhere in the air. When clouds and air
are viewed through a microscope, there is no clear boundary between the two. We cannot
define either of the terms ‘cloud’ and ‘air’ without the other. No axiomatic definition is
possible.
Objects and vacuum also behave in this way. Virtual particles are found in vacuum,
and vacuum is found inside objects. At Planck scales there is no clear boundary between
the two; we cannot define either of them without the other. Despite the lack of a precise
definition and despite the logical problems that ensue, in both cases the description works
well at large, everyday scales.
We note that, since the universe is not a set and since it contains no information, the
Page 265 paradox of the physics book containing a full description of nature disappears. Such a
book can exist, as it does not contradict any property of the universe. But then a question
arises naturally:
Copyright © Christoph Schiller November 1997–May 2006

Does the universe make sense?

Drum hab ich mich der Magie ergeben,

“ [...]
Daß ich erkenne, was die Welt
Im Innersten zusammenhält.*
Goethe, Faust.

Is the universe really the sum of matter–energy and space-time? Or of particles and va-
cuum? We have heard this so often up to now that we may be lulled into forgetting to
”
* Thus I have devoted myself to magic, [...] that I understand how the innermost world is held together.

Dvipsbugw
 1050 xi general rel ativity versus quantum mechanics

TA B L E 75 Physical statements about the universe

The universe has no age. The universe has no beginning.

The universe has no size. The universe has no volume.
The universe has no shape. The universe’s particle number is undefined.
The universe has no mass. The universe has no energy.
The universe has no density. The universe contains no matter.
The universe has no cosmological constant. The universe has no initial conditions. Dvipsbugw
The universe has no state. The universe has no wave function.
The universe is not a physical system. The universe contains no information.
The universe is not isolated. The universe is not open.
The universe has no boundaries. The universe does not interact.
The universe cannot be measured. The universe cannot be said to exist.
The universe cannot be distinguished from The universe cannot be distinguished from a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

nothing. single event.
The universe contains no moments. The universe is not composite.
The universe is not a set. The universe is not a concept.
The universe cannot be described. There is no plural for ‘universe’.
The universe cannot be distinguished from va- The universe was not created.
cuum.

check the statement. To find the answer, we do not need magic, as Faust thought; we only
need to list what we have found so far, especially in this section, in the section on Planck
Page 631 scales, and in the intermezzo on brain and language. Table 75 shows the result.
Not only are we unable to state that the universe is made of space-time and matter;
in fact, we are unable to say anything about the universe at all!* It is not even possible
to say that it exists, since it is impossible to interact with it. The term ‘universe’ does not
Challenge 1473 r allow us to make a single sensible statement. (Can you find one?) We are only able to say
which properties it does not have. We are unable to find any property the universe does
have. Thus, the universe has no properties! We cannot even say whether the universe is
something or nothing. The universe isn’t anything in particular. In other words, the term
‘universe’ is not useful at all for the description of motion.
Page 644 We can obtain a confirmation of this strange conclusion from the first intermezzo.
Copyright © Christoph Schiller November 1997–May 2006

There we found that any concept needs defined content, defined limits and a defined
domain of application. In this section, we have found that for the term ‘universe’, not one
of these three aspects is defined; there is thus no such concept. If somebody asks: ‘why
does the universe exist?’ the answer is: not only does the use of ‘why’ wrongly suggest that
something may exist outside the universe, providing a reason for it and thus contradicting
the definition of the term ‘universe’ itself; most importantly of all, the universe simply
does not exist. In summary, any sentence containing the word universe makes no sense.

* There is also another well-known, non-physical concept about which nothing can be said. Many scholars
Challenge 1472 n have explored it in detail. Can you see what it is?

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1051

The term ‘universe’ only seems to express something, even if it doesn’t.*

The conclusion that the term universe makes no sense may be interesting, even
strangely beautiful; but does it help us to understand motion more precisely? Interest-
ingly so, it does.

A concept without a set eliminates contradictions

The discussion about the term ‘universe’ also shows that the term does not contain any set.
In other words, this is the first term that will help us on the way to a precise description Dvipsbugw
of nature. We will see later on how this happens.
By taking into account the limits on length, time, mass and all other quantities we
have encountered, we have deduced a number of almost painful conclusions about nature.
However, we also received something in exchange: all the contradictions between general
relativity and quantum theory that we mentioned at the beginning of this chapter are now
Challenge 1474 e resolved. Although we have had to leave behind us many cherished habits, in exchange

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

we have the promise of a description of nature without contradictions. But we get even
more.

Extremal scales and open questions in physics

Page 959 In the chapter Quantum physics in a nutshell we listed all the unexplained properties of
nature left open either by general relativity or by quantum theory. The present conclusions
provide us with new connections among them. Indeed, many of the cosmological results
of this section sound surprisingly familiar; let us compare them systematically with those
of the section on Planck scales. Both sections explored topics – some in more detail than
others – from the list of unexplained properties of nature.
First, Table 76 shows that none of the unexplained properties makes sense at both limits
of nature, the small and the large. All open questions are open at both extremes. Second
and more importantly, nature behaves in the same way at horizon scales and at Planck
scales. In fact, we have not found any difference between the two cases. (Are you able to
Challenge 1475 r discover one?) We are thus led to the hypothesis that nature does not distinguish between
the large and the small. Nature seems to be characterized by extremal identity.

Is extremal identity a principle of nature?

The principle of extremal identity incorporates some rather general points:
Copyright © Christoph Schiller November 1997–May 2006

— all open questions about nature appear at its two extremes;

— a description of nature requires both general relativity and quantum theory;
— nature or the universe is not a set;
— initial conditions and evolution equations make no sense at nature’s limits;
— there is a relation between local and global issues in nature;
— the concept of ‘universe’ has no content.

* Of course, the term ‘universe’ still makes sense if it is defined more restrictively, such as ‘everything inter-
acting with a particular human or animal observer in everyday life’. But such a definition is not useful for
our quest, as it lacks the precision required for any description of motion.

Dvipsbugw
 1052 xi general rel ativity versus quantum mechanics

TA B L E 76 Properties of nature at maximal, everyday and minimal scales

P h ys i c a l p r o p e rt y o f n at u r e At h ori - At At P l a n c k
zon scale every - scale
day
scale
requires quantum theory and relativity true false true
intervals can be measured precisely false true false
Dvipsbugw
length and time intervals are limited unlimited limited
space-time is not continuous true false true
points and events cannot be distinguished true false true
space-time is not a manifold true false true
space is 3 dimensional false true false
space and time are indistinguishable true false true

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

initial conditions make sense false true false
space-time fluctuates true false true
Lorentz and Poincaré symmetry does not apply applies does not apply
CPT symmetry does not apply applies does not apply
renormalization does not apply applies does not apply
permutation symmetry does not apply applies does not apply
interactions do not exist exist do not exist
number of particles undefined defined undefined
algebras of observables undefined defined undefined
matter indistinguishable from vacuum true false true
boundaries exist false true false
nature is a set false true false

Extremal identity thus looks like a good candidate tool for use in the search for a unified
description of nature. To be a bit more provocative, it may be the only known principle
incorporating the idea that the universe is not a set, and thus might be the only candidate
tool for use in the quest of unification. Extremal identity is beautiful in its simplicity, in
its unexpectedness and in the richness of its consequences. You might enjoy exploring it
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1476 ny by yourself.

The study of the consequences of extremal identity is currently the focus of much activ-
Ref. 1084 ity in high energy particle physics, although often under different names. The simplest
approach to extremal identity – in fact one that is too simple to be correct – is inversion.
It looks as if extremal identity implies a connection such as

2 2
l Pl l Pl xµ
r or xµ  (710)
r xµ x µ

relating distances r or coordinates x µ with their inverse values using the Planck length l Pl .
Can this mapping, called inversion, be a symmetry of nature? At every point of space? For

Dvipsbugw
 nature at l arge scales – is the universe something or nothing? 1053

example, if the horizon distance is inserted, equation (710) implies that lengths smaller
Challenge 1477 ny than l Pl ~1061  10−96 m never appear in physics. Is this the case? What would inversion
imply for the big bang?
Numerous fascinating questions are contained in the simple hypothesis of extremal
identity. They lead to two main directions for investigation.
First, we have to search for some stronger arguments for the validity of extremal iden-
tity. We will discover a number of simple arguments, all showing that extremal identity
is indeed a property of nature and producing many beautiful insights.
The other quest then follows. We need to find the correct version of equation (710). Dvipsbugw
That oversimplified expression is neither sufficient nor correct. It is not sufficient because
it does not explain any of the issues left open by general relativity and quantum theory.
It only relates some of them, thus reducing their number, but it does not solve any of
Challenge 1478 ny them. You may wish to check this for yourself. In other words, we need to find the precise
description of quantum geometry and of elementary particles.
However, inversion is also simply wrong. Inversion is not the correct description of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

extremal identity because it does not realize a central result discovered above: it does
not connect states and intrinsic properties. Inversion keeps them distinct. This means that
inversion does not take interactions into account. And most open issues at this point of
our mountain ascent are properties of interactions.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
1054 xi general rel ativity versus quantum mechanics

Biblio graphy

1073 Available at http://www.dse.nl/motionmountain/C10-QMGR.pdf. Those pages are a re-

worked version of the ones published in French as C. Schiller, Le vide diffère-t-il de la
matière? in E. Gunzig & S. Diner, editeurs, Le vide – Univers du tout et du rien – Des
physiciens et des philosophes s’interrogent, Les Éditions de l’Université de Bruxelles, 1998.
See also C. Schiller, Does matter differ from vacuum? http://www.arxiv.org/abs/gr-qc/
9610066. Cited on pages 1032, 1036, and 1042.
Dvipsbugw
1074 See the lucid discussion by G.F.R. Ellis & T. Rothman, Lost horizons, American
Journal of Physics 61, pp. 883–893, 1993. Cited on pages 1034, 1038, 1039, and 1040.
1075 H. Salecker & E.P. Wigner, Quantum limitations of the measurement of space-time
distances, Physical Review 109, pp. 571–577, 1958. See also the popular account of this paper
by E.J. Zimmerman, The macroscopic nature of space-time, American Journal of Physics
30, pp. 97–105, 1962. Cited on page 1034.
1076 F. Karolyhazy, Gravitation and quantum mechanics of macroscopic objects, Il Nuovo

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Cimento A42, pp. 390–402, 1966, Y.J. Ng & H. Van Dam, Limit to space-time measure-
ment, Modern Physics Letters A 9, pp. 335–340, 1994, Y.J. Ng & H. Van Dam, Remarks on
gravitational sources, Modern Physics Letters A 10, pp. 2801–2808, 1995. Newer summaries
are Y.J. Ng & H. van Dam, Comment on “Uncertainty in measurements of distance”,
http://www.arxiv.org/abs/gr-qc/0209021 and Y.J. Ng, Spacetime foam, http://www.arxiv.
org/abs/gr-qc/0201022. Cited on page 1035.
1077 See, for example, the Hollywood movie Contact, based on the book by Carl Sagan,
Contact, 1985. Cited on page 1040.
1078 See, for example, the international bestseller by S. Hawking, A Brief History of Time –
From the Big Bang to Black Holes, 1988. Cited on page 1042.
1079 L. Rosenfeld, in H.-J. Treder, Entstehung, Entwicklung und Perspektiven der Ein-
steinschen Gravitationstheorie, Springer Verlag, 1966. Cited on page 1045.
1080 Holography in high-energy physics is connected with the work by ’t Hooft and by Susskind.
See for example G. ’t Hooft, Dimensional reduction in quantum gravity, pp. 284–296,
in A. Ali, J. Ellis & S. R andjbar-Daemi, Salaamfeest, 1993, or the often cited paper
by L. Susskind, The world as a hologram, Journal of Mathematical Physics 36, pp. 6377–
6396, 1995, or as http://www.arxiv.org/abs/hep-th/9409089 A good modern overview is
R. B ousso, The holographic principle, Review of Modern Physics 74, pp. 825–874, 2002,
also available as http://www.arxiv.org/abs/hep-th/0203101, even though it has some argu-
mentation errors, as explained on page 871. Cited on page 1046.
Copyright © Christoph Schiller November 1997–May 2006

1081 S. Lloyd, Computational capacity of the universe, Physical Review Letters 88, p. 237901,
2002. Cited on page 1048.
1082 Gottfried Wilhelm Leibniz, La Monadologie, 1714. Written in French, it is avail-
able freely at http://www.uqac.uquebec.ca/zone30/Classiques_des_sciences_sociales and
in various other languages on other websites. Cited on page 1048.
1083 See, for example, H. Wussing & P.S. Alexandrov, (eds.), Die Hilbertschen Probleme,
Akademische Verlagsgesellschaft Geest & Portig, 1983, or Ben H. Yandell, The Honours
Class: Hilbert’s Problems and their Solvers, A.K. Peters, 2002. Cited on page 1049.
1084 Large part of the study of dualities in string and M theory can be seen as investigations into
the detailed consequences of extremal identity. For a review of duality, see ... A classical ver-
sion of duality is discussed by M.C.B. Abdall a, A.L. Gadelka & I.V. Vancea, Du-

Dvipsbugw
 Dvipsbugw

Dvipsbugw
Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net Copyright © Christoph Schiller November 1997–May 2006
ality between coordinates and the Dirac field, http://www.arxiv.org/abs/hep-th/0002217.
1055

Cited on page 1052.

bibliography
 35. the physics of love – a summary of the first t wo
and a half parts
Sex is the physics urge sublimated.

“ Graffito

”
M
aybe you have once met a physicist who has told you, in one of those
oments of confidentiality, that studying physics is even more beautiful than
Dvipsbugw
aking love. At this statement, many will simply shake their head in disbelief
and strongly disapprove. In this section we shall argue that it is possible to learn so much
about physics while making love that discussions about their relative beauty can be put
aside altogether.
Imagine to be with your partner on a beautiful tropical island, just after sunset, and to
look together at the evening sky. Imagine as well that you know little of what is taught at
school nowadays, e.g. that your knowledge is that of the late Renaissance, which probably

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

is a good description of the average modern education level anyway.
Imagine being busy enjoying each other’s company. The most important results of
physics can be deduced from the following experimental facts:*

Love is communication. Love is tiring.

Love is an interaction between moving bodies. Love takes time.
Love is attractive. Love is repulsive.
Love makes noise. In love, size matters.
Love is for reproduction. Love can hurt.
Love needs memory. Love is Greek.
Love uses the sense of sight. Love is animalic.
Love is motion. Love is holy.
Love is based on touch. Love uses motion again.
Love is fun. Love is private.
Love makes one dream.

Let us see what these observations imply for the description of nature.
**
Copyright © Christoph Schiller November 1997–May 2006

Love is communication. Communication is possible because nature looks similar from dif-
ferent standpoints and because nature shows no surprises. Without similarity we could
not understand each other, and a world of surprises would even make thinking im-
possible; it would not be possible to form concepts to describe observations. But fortu-
nately, the world is regular; it thus allows to use concepts such as time and space for its
description.

* Here we deduce physics from love. We could also deduce physics from sexuality. The modern habit of
saying ‘sex’ instead of ‘sexuality’ mixes up two completely different concepts. In fact, studying the influences
Ref. 1085 of sex on physics is almost fully a waste of time. We avoid it. Maybe one day we shall understand why there
do not seem to be any female crackpots proposing pet physical theories.

Dvipsbugw
 the physics of love – a summary of the first two and a half parts 1057

**
Love is an interaction between moving bodies. Together with the previous result, this im-
plies that we can and need to describe moving bodies with mass, energy and momentum.
That is not a small feat. For example, it implies that the Sun will rise tomorrow if the sea
Challenge 1479 n level around the island is the usual one.
**
Love is attractive. When feeling attracted to your partner, you may wonder if this attrac-
tion is the same which keeps the Moon going around the Earth. You make a quick calcu- Dvipsbugw
lation and find that applying the expression for universal gravity

E pot = −
GMm
(711)
r
to both of you, the involved energy is about as much as the energy added by the leg of a

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

fly on the skin. (M and m are your masses, r your distance, and the gravitational constant
has the value G = 6.7 ë 10−11 m3 ~kg s2 .) In short, your partner teaches you that in nature
there are other attractive interactions apart from gravity; the average modern education
is incomplete.
Nevertheless, this first equation is important: it allows to predict the position of the
planets, the time of the tides, the time of eclipses, the return of comets, etc., to a high
accuracy for thousands of years in advance.
**
Love makes noise. That is no news. However, even after making love, even when every-
body and everything is quiet, in a completely silent environment, we do hear something.
The noises we hear are produced within the ear, partly by the blood flowing through the
head, partly by the electrical noise generated in the nerves. That is strange. If matter were
continuous, there would be no noise even for low signal levels. In fact, all proofs for the
discreteness of matter, of electric current, of energy, or of light are based on the increase of
fluctuations with the smallness of systems under consideration. The persistence of noise
thus makes us suspect that matter is made of smallest entities. Making love confirms this
suspicion in several ways.
**
Copyright © Christoph Schiller November 1997–May 2006

Love is for reproduction. Love is what we owe our life to, as we all are results of reproduc-
tion. But the reproduction of a structure is possible only if it can be constructed, in other
words if the structure can be built from small standard entities. Thus we again suspect
ourselves to be made of smallest, discrete entities.
Love is also a complicated method of reproduction. Mathematics provides a much sim-
pler one. If matter objects were not made of particles, but were continuous, it would be
possible to perform reproduction by cutting and reassembling. A famous mathematical
theorem by Banach and Tarski proves that it is possible to take a continuous solid, cut it
into five pieces and rearrange the pieces in such a way that the result are two copies of
the same size and volume as the original. In fact, even volume increases can be produced
in this way, thus realizing growth without any need for food. Mathematics thus provides

Dvipsbugw
 1058 xi general rel ativity versus quantum mechanics

some interesting methods for growth and reproduction. However, they assume that mat-
ter is continuous, without a smallest length scale. The observation that these methods do
not work in nature is compatible with the idea that matter is not continuous.
**
Ref. 1086 Love needs memory. If you would not recognize your partner among all possible ones,
your love life would be quite complicated. A memory is a device which, in order to store
information, must have small internal fluctuations. Obviously, fluctuations in systems get
smaller as their number of components increase. Since our memory works so well, we can Dvipsbugw
follow that we are made of a large number of small particles.
In summary, love shows that we are made of some kind of lego bricks: depending
on the level of magnification, these bricks are called molecules, atoms, or elementary
particles. It is possible to estimate their size using the sea around the tropical island, as
Challenge 1480 n well as a bit of oil. Can you imagine how?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Love uses the sense of sight. Seeing each other is only possible because we are cold whereas
the Sun is hot. If we and our environment all had the same temperature as the Sun, we
Ref. 1087 would not see each other. This can be checked experimentally by looking into a hot oven:
Inside a glowing oven filled with glowing objects it is impossible to discern them against
the background.
**
Love is motion. Bodies move against each other. Moreover, their speed can be measured.
Since measurement is a comparison with a standard, there must be a velocity standard in
nature, some special velocity standing out. Such a standard must either be the minimum
or the maximum possible value. Now, daily life shows that for velocity a finite minimum
value does not exist. We are thus looking for a maximum value. To estimate the value of
the maximum, just take your mobile phone and ring home from the island to your family.
From the delay in the line and the height of the satellite, you can deduce the telephone
speed c and get 3 ë 108 m~s.
The existence of a maximum speed c implies that time is different for different observ-
ers. Looking into the details, we find that this effect becomes noticeable at energies

E different time  mc 2 , (712)

Copyright © Christoph Schiller November 1997–May 2006

where m is the mass of the object. For example, this applies to electrons inside a television
tube.
**
Love is based on touching. When we touch our partner, sometimes we get small shocks.
The energies involved are larger than than those of touching fly legs. In short, people are
electric.
In the dark, we observe that discharges emit light. Light is thus related to electricity. In
addition, touching proves that light is a wave: simply observe the dark lines between two
Challenge 1481 n fingers near your eye in front of a bright background. The lines are due to interference

Dvipsbugw
 the physics of love – a summary of the first two and a half parts 1059

effects. Light thus does not move with infinite speed. In fact, it moves with the same speed
as that of telephone calls.

**
Love is fun. People like to make love in different ways, such as in a dark room. But rooms
get dark when the light is switched off only because we live in a space of odd dimensions.
In even dimensions, a lamp would not turn off directly after the switch is flipped, but dim
Page 210 only slowly.
Love is also fun because with our legs, arms and bodies we can make knots. Knots Dvipsbugw
are possible only in three dimensions. In short, love is real fun only because we live in 3
dimensions.
**
Love is tiring. The reason is gravity. But what is gravity? A little thinking shows that since
there is a maximum speed, gravity is the curvature of space-time. Curved space also

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

means that a horizon can appear, i.e. a largest possible visible distance. From equations
(711) and (712), we deduce that this happens when distances are of the order of

R horizon  Gm~c 2 . (713)

For example, only due a horizon, albeit one appearing in a different way, the night sky is
dark.

**
Love takes time. It is known that men and women have different opinions on durations.
It is also known that love happens between your ears. Indeed, biological research has
Ref. 1091 shown that we have a clock inside the brain, due to circulating electrical currents. This
clock provides our normal sense of time. Since such a brain clock can be built, there must
be a time standard in nature. Again, such a standard must be either a minimum or a
maximum time interval. We shall discover it later on.

**
Love is repulsive. And in love, size matters. Both facts turn out to be the two sides of the
same coin. Love is based on touch, and touch needs repulsion. Repulsion needs a length
scale, but neither gravity nor classical electrodynamics provide one. Classical physics only
Copyright © Christoph Schiller November 1997–May 2006

allows for the measurement of speed. Classical physics cannot explain that the measure-
ment of length, time, or mass is possible.* Classically, matter cannot be hard; it should
be possible to compress it. But love shows us that this is not the case. Love shows us that
lengths scales do exist in nature and thus that classical physics is not sufficient for the
description of nature.
**
Love can hurt. For example, it can lead to injuries. Atoms can get ripped apart. That hap-

* Note that the classical electron radius is not an exception: it contains the elementary charge e, which
contains a length scale, as shown on page 492.

Dvipsbugw
 1060 xi general rel ativity versus quantum mechanics

pens when energies are concentrated on small volumes, such as a few aJ per atom. In-
vestigating such situations more precisely, we finds that strange phenomena appear at
distances r if energies exceed the value

E
ħc
; (714)
r
in particular, energy becomes chunky, things become fuzzy, boxes are not tight, and
particles get confused. These are called quantum phenomena. The new constant ħ = Dvipsbugw
10−34 Js is important: it determines the size of things, because it allows to define distance
and time units. In other words, objects tear and break because in nature there is a min-
imum action, given roughly by ħ.
If even more energy is concentrated in small volumes, such as energies of the order of
mc 2 per particle, one even observes transformation of energy into matter, or pair produc-
tion. From equations (712) and (714), we deduce that this happens at distances of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

r pair production 
ħ
. (715)
mc
At such small distances we cannot avoid using the quantum description of nature.
**
Love is not only Greek. The Greeks were the first to make theories above love, such as
Plato in his Phaedrus. But they also described it in another way. Already before Plato,
Democritus said that making love is an example of particles moving and interacting in va-
cuum. If we change ‘vacuum’ to ‘curved 3+1-dimensional space’ and ‘particle’ to ‘quantum
particle’, we do indeed make love in the way Democritus described 2500 years ago.
It seems that physics has not made much progress in the mean time. Take the statement
made in 1939 by the British astrophysicist Arthur Eddington:

I believe there are 15,747,724,136,275,002,577,605,653,961,181,555,468,044,717,914-

,527,116,709,366,231,425,076,185,631,031,296 protons in the universe and the same
Ref. 1088 number of electrons.

Compare it with the version of 2006:

Copyright © Christoph Schiller November 1997–May 2006

Baryons in the universe: 10811 ; total charge: near zero.

The second is more honest, but which of the two is less sensible? Both sentences show that
there are unexplained facts in the Greek description nature, in particular the number of
involved particles.
**
Love is animalic. We have seen that we can learn a lot about nature from the existence of
love. We could be tempted to see this approach of nature as a special case of the so-called

Dvipsbugw
 the physics of love – a summary of the first two and a half parts 1061

anthropic principle. However, some care is required here. In fact, we could have learned
exactly the same if we had taken as starting point the observation that apes or pigs have
love. There is no ‘law’ of nature which distinguishes between them and humans. In fact,
there is a simple way to determine whether any ‘anthropic’ statement makes sense: the
reasoning must be equally true for humans, apes, and pigs.
A famous anthropic deduction was drawn by the British astrophysicist Fred Hoyle.
While studying stars, he predicted a resonance in the carbon-12 nucleus. If it did not
exist, he argued, stars could not have produced the carbon which afterwards was spread
out by explosions into interstellar space and collected on Earth. Also apes or pigs could Dvipsbugw
reason this way; therefore Hoyle’s statement does make sense.
On the other hand, claiming that the universe is made especially for people is not sens-
ible: using the same arguments, pigs would say it is made for pigs. The existence of either
requires all ‘laws’ of nature. In summary, the anthropic principle is true only in so far as
its consequences are indistinguishable from the porcine or the simian principle. In short,
the animalic side of love puts limits to the philosophy of physics.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Love is holy. Following the famous definition by the theologian Rudolf Otto, holiness
Ref. 1089 results from a mixture of a mysterium tremendum and a mysterium fascinans. Tremendum
means that it makes one tremble. Indeed, love produces heat and is a dissipative process.
All systems in nature which produce heat have a finite lifetime. That is true for machines,
stars, animals, lightning, fire, lamps and people. Through heat, love shows us that we are
going to die. Physicists call this the second principle of thermodynamics.
But love also fascinates. Everything which fascinates has a story. Indeed, this is a prin-
ciple of nature: every dissipative structure, every structure which appears or is sustained
through the release of energy, tells us that it has a story. Take atoms, for example. All the
protons we are made of formed during the big bang. Most hydrogen we are made of is
also that old. The other elements were formed in stars and then blown into the sky during
nova or supernova explosions. They then regrouped during planet formation. We truly
are made of stardust.
Why do such stories fascinate? If you only think about how you and your partner have
met, you will discover that it is through a chain of incredible coincidences. If only one of
all these coincidences had not taken place, you and your partner would not be together.
And of course, we all owe our existence to such a chain of coincidences, which brought
our parents together, our grandparents, and made life appear on Earth.
Copyright © Christoph Schiller November 1997–May 2006

The realization of the importance of coincidences automatically produces two kinds

of questions: why? and what if? Physicists have now produced a list of all the answers to
repeated why questions and many are working at the list of what-if questions. The first list,
the why-list of Table 77, gives all facts still unexplained. It can also be called the complete
list of all surprises in nature. (Above, it was said that there are no surprises in nature about
what happens. However, so far there still are a handful of surprises on how all these things
happen.)

Dvipsbugw
1062 xi general rel ativity versus quantum mechanics

TA B L E 77 Everything quantum field theory and general relativity do not explain; in other words, a list of
the only experimental data and criteria available for tests of the unified description of motion
O b s e r va b l e P r o p e r t y u n e x p l a i n e d s o f a r

Local quantities, from quantum theory

α em the low energy value of the electromagnetic coupling constant
αw the low energy value of the weak coupling constant
αs the low energy value of the strong coupling constant
Dvipsbugw
mq the values of the 6 quark masses
ml the values of 3 lepton masses
mW the values of the independent mass of the W vector boson
θW the value of the Weinberg angle
β1 , β2 , β3 three mixing angles
θ CP the value of the CP parameter

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

θ st the value of the strong topological angle
3 the number of particle generations
0.5 nJ~m3 the value of the observed vacuum energy density or cosmological constant
3+1 the number of space and time dimensions
Global quantities, from general relativity
1.2(1) ë 1026 m ? the distance of the horizon, i.e. the ‘size’ of the universe (if it makes sense)
1082 ? the number of baryons in the universe, i.e. the average matter density in the
universe (if it makes sense)
A 1092 ? the initial conditions for more than 1092 particle fields in the universe, includ-
ing those at the origin of galaxies and stars (if they make sense)

Local structures, from quantum theory

S(n) the origin of particle identity, i.e. of permutation symmetry
Ren. group the renormalization properties, i.e. the existence of point particles
SO(3,1) the origin of Lorentz (or Poincaré) symmetry
(i.e. of spin, position, energy, momentum)
C‡ the origin of the algebra of observables
Gauge group the origin of gauge symmetry
(and thus of charge, strangeness, beauty, etc.)
Copyright © Christoph Schiller November 1997–May 2006

in particular, for the standard model:

U(1) the origin of the electromagnetic gauge group (i.e. of the quantization of elec-
tric charge, as well as the vanishing of magnetic charge)
SU(2) the origin of weak interaction gauge group
SU(3) the origin of strong interaction gauge group
Global structures, from general relativity
maybe R  S3 ? the unknown topology of the universe (if it makes sense)

This why-list fascinates through its shortness, which many researchers are still trying

Dvipsbugw
the physics of love – a summary of the first two and a half parts 1063

to reduce. But it is equally interesting to study what consequences appear if any of the
values from Table 77 were only a tiny bit different. It is not a secret that small changes in
nature would lead to completely different observations, as shown in Table 78.

TA B L E 78 A small selection of the consequences when changing aspects of nature

O b s e r va b l e C h a n g e R e su lt

Moon size smaller small Earth magnetic field; too much cosmic radiation; Dvipsbugw
widespread child cancers.
Moon size larger large Earth magnetic field; too little cosmic radiation; no
evolution into humans.
Jupiter smaller too many comet impacts on Earth; extinction of animal
life.
Jupiter larger too little comet impacts on Earth; no Moon; no dinosaur
extinction.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Oort belt smaller no comets, no irregular asteroids, no Moon; still dino-
saurs.
Star distance smaller irregular planet motion; supernova dangers.
Strong coupling smaller proton decay; leucemia.
constant

The large number of coincidences of life force our mind to realize that we are only a
tiny part of nature. We are a small droplet shaken around in the ocean of nature. Even the
tiniest changes in nature would prevent the existence of humans, apes and pigs. In other
words, making love tells us that the universe is much larger than we are and tells us how
much we are dependent and connected to the rest of the universe.
**
We said above that love uses motion. It contains a remarkable mystery, worth a second
look:
- Motion is the change of position with time of some bodies.
- Position is what we measure with a ruler. Time is what we measure with a clock. Both
rulers and clocks are bodies.
- A body is an entity distinct from its environment by its shape or its mass. Shape is
Copyright © Christoph Schiller November 1997–May 2006

the extension of a body in space (and time). Mass is measured by measuring speed or
acceleration, i.e. by measuring space and time.
This means that we define space-time with bodies – as done in detail in general relativ-
ity – and that we define bodies with space-time – as done in detail in quantum theory.
This circular reasoning shows that making love is truly a mystery. The circular reason-
ing has not yet been eliminated yet; at present, modern theoretical physicists are busy
attempting to do so. The most promising approach seems to be M-theory, the modern
extension of string theory. But any such attempt has to overcome important difficulties
which can also be experienced while making love.

**

Dvipsbugw
 1064 xi general rel ativity versus quantum mechanics

Dvipsbugw

R R

F I G U R E 386 A Gedanken experiment showing that

at Planck scales, matter and vacuum cannot be
distinguished

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Love is private. But is it? Privacy assumes that a person can separate itself from the rest,
without important interactions, at least for a given time, and come back later. This is
possible if the person puts enough empty space between itself and others. In other words,
privacy is based on the idea that objects can be distinguished from vacuum. Let us check
whether this is always possible.
Ref. 1090 What is the smallest measurable distance? This question has been almost, but only
almost answered by Max Planck in 1899. The distance δl between two objects of mass m is
surely larger than their position indeterminacy ħ~∆p; and the momentum indeterminacy
must be smaller that the momentum leading to pair production, i.e. ∆p < mc. This means
that
δl E ∆l E
ħ
. (716)
mc
In addition, the measurements require that signals leave the objects; the two masses must
not be black holes. Their masses must be so small that the Schwarzschild radius is smaller
than the distance to be measured. This means that r S  Gm~c 2 < δl or that
¾
δl E = l Pl = 1.6 ë 10−35 m .
ħG
Copyright © Christoph Schiller November 1997–May 2006

(717)
c3

This expression defines a minimum length in nature, the so-called Planck length. Every
other Gedanken experiment leads to this characteristic length as well. In fact, this min-
imum distance (and the corresponding minimum time interval) provides the measure-
ment standard we were looking for at the beginning of our musings about length and
time measurements.
A more detailed discussion shows that the smallest measurable distance is somewhat
larger, a multiple of the Planck length, as measurements require the distinction of matter
and radiation. This happens at scales about 800 times the Planck length.
In other words, privacy has its limits. In fact, the issue is even more muddled when

Dvipsbugw
 the physics of love – a summary of the first two and a half parts 1065

we explore the consequences for bodies. A body, also a human one, is something we can
touch, throw, hit, carry or weigh. Physicists say that a body is something with energy or
momentum. Vacuum has none of it. In addition, vacuum is unbounded, whereas objects
are bounded.
What happens if we try to weigh objects at Planck scales? Quantum theory makes a
simple prediction. If we put an object of mass M in a box of size R onto a scale – as in
Figure 386 – equation (714) implies that there is a minimal mass error ∆M given by
Dvipsbugw
∆M 
ħ
. (718)
cR
If the box has Planck size, the mass error is the Planck mass
»
∆M = M Pl = ħc~G  22 µg . (719)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

How large is the mass we can put into a box of Planck size? Obviously it is given by the
maximum possible mass density. To determine it, imagine a planet and put a satellite in
orbit around it, just skimming its surface. The density ρ of the planet with radius r is given
by
v2
ρ 3 = 2 .
M
(720)
r Gr
Using equation (716) we find that the maximum mass density in nature, within a factor
of order one, is the so-called Planck density, given by

c5
ρ Pl = = 5.2 ë 1096 kg~m3 . (721)
G2ħ
Therefore the maximum mass that can be contained inside a Planck box is the Planck
mass. But that was also the measurement error for that situation. This implies that we
cannot say whether the original box we measured was empty or full: vacuum cannot be
distinguished from matter at Planck scales. This astonishing result is confirmed by every
Ref. 1090 other Gedanken experiment exploring the issue.
It is straightforward to deduce with similar arguments that objects are not bound in
size at Planck scales, i.e. that they are not localized, and that the vacuum is not necessarily
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1482 n extended at those scales. In addition, the concept of particle number cannot be defined
at Planck scales.
So, why is there something instead of nothing? Making love shows that there is no
difference between the two options!

**
Love makes us dream. When we dream, especially at night, we often look at the sky. How
far is it away? How many atoms are enclosed by it? How old is it? These questions have
an answer for small distances and for large distances; but for the whole of the sky or the
whole of nature they cannot have one, as there is no way to be outside of the sky in order
to measure it. In fact, each of the impossibilities to measure nature at smallest distances

Dvipsbugw
 1066 xi general rel ativity versus quantum mechanics

are found again at the largest scales. There seems to be a fundamental equivalence, or, as
Ref. 1092 physicists say, a duality between the largest and the smallest distances.
The coming years will hopefully show how we can translate these results into an even
more precise description of motion and of nature. In particular, this description should
allow us to reduce the number of unexplained properties of nature.
**
In summary, making love is a good physics lesson. Enjoy the rest of your day.
Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
bibliography 1067

Biblio graphy

1085 An attempt to explain the lack of women in physics is made in Margaret Wertheim,
Pythagoras’ Trousers – God, Physics and the Gender Wars, Fourth Estate, 1997. Cited on
page 1056.
1086 The consequences of memory loss in this case are already told by Voltaire, Aventure de
la mémoire, 1775. Cited on page 1058.
1087 A picture of objects in a red hot oven and at room temperature is shown in C.H. Bennett, Dvipsbugw
Demons, Engines and the Second Law, Scientific American 255, pp. 108–117, November
1987. Cited on page 1058.
1088 The famous quote is found at the beginning of chapter XI, ‘The Physical Universe’, in Ar-
thur Eddington, The Philosophy of Physical Science, Cambridge, 1939. Cited on page
1060.
1089 See the first intermezzo for details and references. Cited on page 1061.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1090 Details can be found in Chapter XI of this text, a reworked version of the pages published
in French as C. Schiller, Le vide diffère-t-il de la matière? in E. Gunzig & S. Diner,
editeurs, Le vide – Univers du tout et du rien – Des physiciens et des philosophes s’interrogent,
Les Éditions de l’Université de Bruxelles, 1998. Cited on pages 1064 and 1065.
1091 An introduction to the sense of time as result of signals in the ganglia and in the temporal
lobe of the brain is found in R.B. Ivry & R. Spencer, The neural representation of time,
Current Opinion in Neurobiology 14, pp. 225–232, 2004. Cited on page 1059.
1092 For details of fthe arguments leading to duality, see section 34. It also includes suggestions
supporting the notion that the universe is not a even a set, thus proposing a solution for
Hilbert’s sixth problem. Cited on page 1066.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
1068 xi general rel ativity versus quantum mechanics

36. maximum force and minimum distance – physics

in limit statements
The principle of maximum force or power allows us to summarize special relativity,
quantum theory and general relativity in one fundamental limit principle each. The three
principles are fully equivalent to the standard formulation of the theories. In particular,
we show that the maximum force c 4 ~4G implies and is equivalent to the field equations
of general relativity. With the speed limit of special relativity and the action–angular mo-
Dvipsbugw
mentum limit of quantum theory, the three fundamental principles imply a bound for
every physical observable, from acceleration to size. The new, precise limit values differ
from the usual Planck values by numerical prefactors of order unity.* Continuing this
approach, we show that every observable in nature has both a lower and an upper limit
value.
As a result, a maximum force and thus a minimum length imply that the non-
continuity of space-time is an inevitable consequence of the unification of quantum the-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ory and relativity. Furthermore, the limits are shown to imply the maximum entropy
bound, including the correct numerical prefactor. The limits also imply a maximum meas-
urement precision possible in nature, thus showing that any description by real numbers
is approximate. Finally, the limits show that nature cannot be described by sets. As a res-
ult, the limits point to a solution to Hilbert’s sixth problem. The limits also show that
vacuum and matter cannot be fully distinguished in detail; they are two limit cases of the
same entity. These fascinating results provide the basis for any search for a unified theory
of motion.

Fundamental limits to all observables

At dinner parties physicists are regularly asked to summarize physics in a few sentences.
It is useful to be able to present a few simple statements answering the request.** Here we
propose such a set of statements. We are already familiar with each statement from what
we have found out so far in our exploration of motion. But by putting them all together
we will get direct insight into the results of modern research on unification.
The main lesson of modern physics is the following: Nature limits the possibilities of
motion. These limits are the origin of special relativity, general relativity and quantum
theory. In fact, we will find out that nature poses limits to every aspect of motion; but let
us put first things first.
Copyright © Christoph Schiller November 1997–May 2006

Special relativity in one statement

The step from everyday or Galilean physics to special relativity can be summarized in a
single limit statement on motion. It was popularized by Hendrik Antoon Lorentz: There
is a maximum energy speed in nature. For all physical systems and all observers,

vDc. (722)

* These limits were given for the first time in 2003, in this section of the present text.
** Stimulating discussions with Saverio Pascazio, Corrado Massa and Steve Carlip helped shaping this sec-
tion.

Dvipsbugw
 fundamental limits to all observables 1069

A few well-known remarks set the framework for the discussion that follows. The speed v
is smaller than or equal to the speed of light c for all physical systems;* in particular, this
limit is valid both for composed systems as well as for elementary particles. The speed
limit statement is also valid for all observers; no exception to the statement is known. In-
deed, only a maximum speed ensures that cause and effect can be distinguished in nature,
or that sequences of observations can be defined. The opposite statement, implying the
existence of (real) tachyons, has been explored and tested in great detail; it leads to nu-
merous conflicts with observations.
The maximum speed forces us to use the concept of space-time to describe nature. Dvipsbugw
Maximum speed implies that space and time mix. The existence of a maximum speed in
Page 282 nature also implies observer-dependent time and space coordinates, length contraction,
time dilation, mass–energy equivalence and all other effects that characterize special re-
lativity. Only the existence of a maximum speed leads to the principle of maximum ageing
that governs special relativity; only a maximum speed leads to the principle of least ac-
tion at low speeds. In addition, only a finite speed limit makes it to define a unit of speed

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that is valid at all places and at all times. If a global speed limit did not exist, no natural
measurement standard for speed, independent of all interactions, would exist in nature;
speed would not then be a measurable quantity.
Special relativity also limits the size of systems, independently of whether they are
composed or elementary. Indeed, the limit speed implies that acceleration a and size l
cannot be increased independently without bounds, because the two ends of a system
must not interpenetrate. The most important case concerns massive systems, for which
we have
c2
lD . (723)
a

This size limit is induced by the speed of light c; it is also valid for the displacement d of
a system, if the acceleration measured by an external observer is used. Finally, the limit
implies an ‘indeterminacy’ relation

∆l ∆a D c 2 (724)

for the length and acceleration indeterminacies. You might want to take a minute to de-
Challenge 1483 n duce it from the time-frequency indeterminacy. All this is standard knowledge. Copyright © Christoph Schiller November 1997–May 2006

Quantum theory in one statement

In the same way, the difference between Galilean physics and quantum theory can be
Ref. 1093 summarized in a single statement on motion, due to Niels Bohr: There is a minimum
Page 704 action in nature. For all physical systems and all observers,

SE
ħ
. (725)
2

* A physical system is a region of space-time containing mass–energy, the location of which can be followed
over time and which interacts incoherently with its environment. With this definition, images, geometrical
points or incomplete entangled situations are excluded from the definition of system.

Dvipsbugw
 1070 xi general rel ativity versus quantum mechanics

Half the Planck constant ħ is the smallest observable action or angular momentum. This
statement is valid for both composite and elementary systems. The action limit is used
less frequently than the speed limit. It starts from the usual definition of the action, S =
∫ (T − U)dt, and states that between two observations performed at times t and t + ∆t,
even if the evolution of a system is not known, the action is at least ħ~2. Physical action is
defined to be the quantity that measures the amount of change in the state of a physical
system. In other words, there is always a minimum change of state taking place between
two observations of a system. In this way, the quantum of action expresses the well-known
fundamental fuzziness of nature at a microscopic scale. Dvipsbugw
It can easily be checked that no observation results in a smaller value of action, ir-
respective of whether photons, electrons or macroscopic systems are observed. No ex-
ception to the statement is known. A minimum action has been observed for fermions,
bosons, laser beams and matter systems and for any combination of these. The opposite
statement, implying the existence of change that is arbitrary small, has been explored in
detail; Einstein’s long discussion with Bohr, for example, can be seen as a repeated attempt

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

by Einstein to find experiments that make it possible to measure arbitrarily small changes
in nature. In every case, Bohr found that this aim could not be achieved.
Ref. 1094 The minimum value of action can be used to deduce the indeterminacy relation, the
tunnelling effect, entanglement, permutation symmetry, the appearance of probabilities
in quantum theory, the information theory aspect of quantum theory and the existence
of elementary particle reactions. The minimum value of action implies that, in quantum
theory, the three concepts of state, measurement operation and measurement result need
to be distinguished from each other; a so-called Hilbert space needs to be introduced. The
minimal action is also part of the Einstein–Brillouin–Keller quantization. The details of
Page 704 these connections can be found in the chapter on quantum theory.
Obviously, the existence of a quantum of action has been known right from the be-
ginning of quantum theory. The quantum of action is at the basis of all descriptions of
quantum theory, including the many-path formulation and the information-theoretic de-
scriptions. The existence of a minimum quantum of action is completely equivalent to all
standard developments of quantum theory.
We also note that only a finite action limit makes it possible to define a unit of action.
If an action limit did not exist, no natural measurement standard for action would exist
in nature; action would not then be a measurable quantity.
The action bound S D pd D mcd, together with the quantum of action, implies a limit
on the displacement d of a system between any two observations:
Copyright © Christoph Schiller November 1997–May 2006

dE
ħ
. (726)
2mc
In other words, (half) the (reduced) Compton wavelength of quantum theory is recovered
as lower limit on the displacement of a system, whenever gravity plays no role. Since the
quantum displacement limit applies in particular to an elementary system, the limit is
also valid for the size of a composite system. However, the limit is not valid for the size of
elementary particles.
The action limit of quantum theory also implies Heisenberg’s well-known indetermin-
Page 708 acy relation for the displacement d and momentum p of systems:

Dvipsbugw
 fundamental limits to all observables 1071

∆d ∆p E
ħ
. (727)
2
This is valid for both massless and massive systems. All this is textbook knowledge, of
course.

General relativity in one statement

Least known of all is the possibility of summarizing the step from Galilean physics to
Dvipsbugw
general relativity in a single statement on motion: There is a maximum force or power in
nature. For all physical systems and all observers,

c4 c5
FD = 3.0 ë 1043 N or PD = 9.1 ë 1051 W . (728)
4G 4G
The limit statements contain both the speed of light c and the constant of gravitation

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

G; they thus indeed qualify as statements concerning relativistic gravitation. Like the
previous limit statements, they are valid for all observers. This formulation of general
relativity is not common; in fact, it seems that it was only discovered 80 years after the
Ref. 1095 theory of general relativity had first been proposed.* A detailed discussion is given in the
Page 349 chapter on general relativity.
The value of the force limit is the energy of a Schwarzschild black hole divided by its
diameter; here the ‘diameter’ is defined as the circumference divided by π. The power
limit is realized when such a black hole is radiated away in the time that light takes to
travel along a length corresponding to the diameter.
Force is change of momentum; power is the change of energy. Since momentum and
energy are conserved, force or power can be pictured as the flow of momentum or energy
through a given surface. The value of the maximum force is the mass–energy of a black
hole divided by its diameter. It is also the surface gravity of a black hole times its mass.
The force limit thus means that no physical system of a given mass can be concentrated
in a region of space-time smaller than a (non-rotating) black hole of that mass. In fact,
Challenge 1484 e the mass–energy concentration limit can be easily transformed by simple algebra into the
force limit; both are equivalent.
It is easily checked that the maximum force is valid for all systems observed in nature,
whether they are microscopic, macroscopic or astrophysical. Neither the ‘gravitational
force’ (as long as it is operationally defined) nor the electromagnetic or the nuclear inter-
Copyright © Christoph Schiller November 1997–May 2006

Challenge 1485 e actions are ever found to exceed this limit.

The next aspect to check is whether a system can be imagined that does exceed the limit.
Page 349 An extensive discussion shows that this is impossible, if the proper size of observers or
test masses is taken into account. Even for a moving observer, when the force value is
increased by the (cube of the) relativistic dilation factor, or for an accelerating observer,
when the observed acceleration is increased by the acceleration of the observer itself, the
force limit must still hold. However, no situations allow the limit to be exceeded, because
for high accelerations a, horizons appear at distance a~c 2 ; since a mass m has a minimum

* It might be that the present author was the first to point it out, in this very textbook. Also Gary Gibbons
found this result independently.

Dvipsbugw
 1072 xi general rel ativity versus quantum mechanics

diameter given by l E 4Gm~c 2 , we are again limited by the maximum force.

The exploration of the force and power limits shows that they are achieved only at
horizons; the limits are not reached in any other situation. The limits are valid for all
observers and all interactions.
As an alternative to the maximum force and power limits, we can use as basic principle
the statement: There is a maximum mass change in nature. For all systems and observers,
one has
dm c 3
D = 1.0 ë 1035 kg~s . (729) Dvipsbugw
dt 4G
Equivalently, we can state: There is a maximum mass per length ratio in nature. For all
systems and observers, one has

dm c 2
D = 3.4 ë 1026 kg~m . (730)
dl 4G

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In detail, both the force and the power limits state that the flow of momentum or of
energy through any physical surface – a term defined below – of any size, for any observer,
in any coordinate system, never exceeds the limit values. Indeed, as a result to the lack
of nearby black holes or horizons, neither limit value is exceeded in any physical system
found so far. This is the case at everyday length scales, in the microscopic world and
in astrophysical systems. In addition, even Gedanken experiments do not allow to the
Page 349 limits to be exceeded. However, the limits become evident only when in such Gedanken
experiments the size of observers or of test masses is taken into account. If this is not
Page 349 done, apparent exceptions can be constructed; however, they are then unphysical.

Deducing general relativity*

In order to elevate the force or power limit to a principle of nature, we have to show that,
in the same way that special relativity results from the maximum speed, general relativity
results from the maximum force.
Ref. 1096 The maximum force and the maximum power are only realized at horizons. Horizons
are regions of space-time where the curvature is so high that it limits the possibility of
observation. The name ‘horizon’ is due to a certain analogy with the usual horizon of
everyday life, which also limits the distance to which one can see. However, in general
relativity horizons are surfaces, not lines. In fact, we can define the concept of horizon in
Copyright © Christoph Schiller November 1997–May 2006

general relativity as a region of maximum force; it is then easy to prove that it is always a
two-dimensional surface, and that it has all properties usually associated with it.
The connection between horizons and the maximum force or power allows a simple
deduction of the field equations. We start with a simple connection. All horizons show
energy flow at their location. This implies that horizons cannot be planes. An infinitely
extended plane would imply an infinite energy flow. To characterize the finite extension
of a given horizon, we use its radius R and its total area A.
The energy flow through a horizon is characterized by an energy E and a proper length
L of the energy pulse. When such an energy pulse flows perpendicularly through a hori-

* This section can be skipped at first reading.

Dvipsbugw
 fundamental limits to all observables 1073

zon, the momentum change dp~dt = F is given by

F=
E
. (731)
L
For a horizon, we need to insert the maximum possible values. With the horizon area A
and radius R, we can rewrite the limit case as

c4
= 4πR 2
E 1 Dvipsbugw
(732)
4G A L

where the maximum force and the maximum possible area 4πR 2 of a horizon of (max-
imum local) radius R were introduced. The fraction E~A is the energy per area flowing
through the horizon. Often, horizons are characterized by the so-called surface gravity a
instead of the radius R. In the limit case, two are related by a = c 2 ~2R. This leads to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

E=
1
a2 A L . (733)
4πG
Ref. 1097 Special relativity shows that horizons limit the product aL between proper length and
acceleration to the value c 2 ~2. This leads to the central relation for the energy flow at
horizons:
c2
E= aA. (734)
8πG
This equation makes three points. First, the energy flowing through a horizon is limited.
Second, this energy is proportional to the area of the horizon. Third, the energy flow is
proportional to the surface gravity. These results are fundamental statements of general
relativity. No other part of physics makes comparable statements.
For the differential case the last relation can be rewritten as

c2
δE = a δA . (735)
8πG
In this way, the result can also be used for general horizons, such as horizons that are
curved or time-dependent.*
Copyright © Christoph Schiller November 1997–May 2006

Ref. 1098 In a well known paper, Jacobson has given a beautiful proof of a simple connection:
if energy flow is proportional to horizon area for all observers and all horizons, then

* Relation (735) is well known, though with different names for the observables. Since no communication
is possible across a horizon, the detailed fate of energy flowing through a horizon is also unknown. Energy
whose detailed fate is unknown is often called heat. Relation (735) therefore states that the heat flowing
through a horizon is proportional to the horizon area. When quantum theory is introduced into the discus-
sion, the area of a horizon can be called ‘entropy’ and its surface gravity can be called ‘temperature’; relation
(735) can then be rewritten as
δQ = T δS . (736)
However, this translation of relation (735), which requires the quantum of action, is unnecessary here. We
only cite it to show the relation between horizon behaviour and quantum gravity.

Dvipsbugw
1074 xi general rel ativity versus quantum mechanics

general relativity holds. To see the connection to general relativity, we generalize relation
(735) to general coordinate systems and general energy-flow directions. This is achieved
by introducing tensor notation.
To realize this at horizons, one introduces the general surface element dΣ and the local
boost Killing vector field k that generates the horizon (with suitable norm). Jacobson uses
them to rewrite the left hand side of relation (735) as

δE = ∫T ab k
a
dΣ b , (737) Dvipsbugw

where Tab is the energy–momentum tensor. This rewrites the energy for arbitrary co-
ordinate systems and arbitrary energy flow directions. Jacobson’s main result is that the
essential part of the right hand side of relation (735) can be rewritten, using the (purely
geometric) Raychaudhuri equation, as

aδA = c 2 ∫R

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

a
ab k dΣ b , (738)

where R ab is the Ricci tensor describing space-time curvature.

Combining these two steps, we find that the energy–area relation (735) for horizons
can be rewritten as
c4
∫ Tab k a dΣ b = ∫
R ab k a dΣ b . (739)
8πG
Jacobson shows that this equation, together with local conservation of energy (i.e., van-
ishing divergence of the energy–momentum tensor), can only be satisfied if

c4
Tab = ‹R ab − ( R + Λ)дab  ,
1
(740)
8πG 2

where Λ is a constant of integration whose value is not specified by the problem. These are
the full field equations of general relativity, including the cosmological constant Λ. The
field equations are thus shown to be valid at horizons. Since it is possible, by choosing a
suitable coordinate transformation, to position a horizon at any desired space-time event,
the field equations must be valid over the whole of space-time.
It is possible to have a horizon at every event in space-time; therefore, at every event
Copyright © Christoph Schiller November 1997–May 2006

in nature there is the same maximum possible force (or power). This maximum force (or
power) is a constant of nature.
In other words, we just showed that the field equations of general relativity are a direct
consequence of the limited energy flow at horizons, which in turn is due to the existence
of a maximum force or power. Maximum force or power implies the field equations. One
can thus speak of the maximum force principle. In turn, the field equations imply max-
imum force. Maximum force and general relativity are equivalent.
The bounds on force and power have important consequences. In particular, they im-
ply statements on cosmic censorship, the Penrose inequality, the hoop conjecture, the
non-existence of plane gravitational waves, the lack of space-time singularities, new ex-
perimental tests of the theory, and on the elimination of competing theories of relativistic

Dvipsbugw
 fundamental limits to all observables 1075

Ref. 1099 gravitation. These consequences are presented elsewhere.

Deducing universal gravitation

Universal gravitation can be derived from the force limit in the case where forces
º speeds are2 much smaller than the maximum
and values. The first condition implies
4GMa  c , the second v  c and al  c 2 . Let us apply this to a specific case. We
study a satellite circling a central mass M at distance R with acceleration a. This system,
with length l = 2R, has only one characteristic speed. Whenever º this speed v is much Dvipsbugw
smaller than c, v must be proportional both to al = 2aR and to 4GMa . If they are
2

taken together, they imply that a = f GM~R 2 , where the numerical factor f is not yet
Challenge 1486 e fixed. A quick check, for example using the observed escape velocity values, shows that
f = 1. Forces and speeds much smaller than the limit values thus imply that the inverse
square law of gravity describes the interaction between systems. In other words, nature’s
limit on force implies the universal law of gravity, as is expected.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The size of physical systems in general relativity
General relativity, like the other theories of modern physics, provides a limit on the size
of systems: there is a limit to the amount of matter that can be concentrated into a small
volume.
lE 2 .
4Gm
(741)
c
The size limit is only achieved for black holes, those well-known systems which swallow
everything that is thrown into them. It is fully equivalent to the force limit. All composite
systems in nature comply with the lower size limit. Whether elementary particles fulfil
or even achieve this limit remains one of the open issues of modern physics. At present,
neither experiment nor theory allow clear statements on their size. More about this issue
below.
Ref. 1100 General relativity also implies an ‘indeterminacy relation’:

c4
D
∆E
. (742)
∆l 4G
Since experimental data are available only for composite systems, we cannot say yet
whether this inequality also holds for elementary particles. The relation is not as pop-
Copyright © Christoph Schiller November 1997–May 2006

ular as the previous. In fact, testing the relation, for example with binary pulsars, may
lead to new tests that would distinguish general relativity from competing theories.

A mechanical analogy for the maximum force

The maximum force is central to the theory of general relativity. That is the reason why the
value of the force (adorned with a factor 2π) appears in the field equations. The import-
ance of a maximum force becomes clearer when we return to our old image of space-time
as a deformable mattress. Like any material body, a mattress is characterized by a mater-
ial constant that relates the deformation values to the values of applied energy. Similarly,
a mattress, like any material, is characterized by the maximum stress it can bear before

Dvipsbugw
 1076 xi general rel ativity versus quantum mechanics

breaking. Like mattresses, crystals also have these two values. In fact, for perfect crystals
(without dislocations) these two material constants are the same.
Empty space-time somehow behaves like a perfect crystal or a perfect mattress: it has
a deformation-energy constant that at the same time is the maximum force that can be ap-
plied to it. The constant of gravitation thus determines the elasticity of space-time. Now,
crystals are not homogeneous, but are made up of atoms, while mattresses are made up
of foam bubbles. What is the corresponding structure of space-time? This is a central
question in the rest of our adventure. One thing is sure: vacuum has no preferred direc-
tions, in complete contrast to crystals. In fact, all these analogies even suggest that the Dvipsbugw
appearance of matter might be nature’s way of preventing space-time from ripping apart.
We have to patient for a while, before we can judge this option. A first step towards the
answer to the question appears when we put all limits together.

Units and limit values for all physical observables

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The existence of a maximum force in nature is equivalent to general relativity. As a result,
physics can now be seen as making three simple statements on motion that is found in
nature:

S E
ħ
quantum theory on action:
2
special relativity on speed: v Dc
c4
general relativity on force: F D . (743)
4G
The limits are valid for all physical systems, whether composed or elementary, and are
valid for all observers. We note that the limit quantities of special relativity, quantum
theory and general relativity can also be seen as the right-hand sides of the respective
indeterminacy relations. Indeed, the set (724, 727, 742) of indeterminacy relations or the
set (723, 726, 741) of length limits is each fully equivalent to the three limit statements
Challenge 1487 e (743). Each set of limits can be taken as a (somewhat selective) summary of twentieth
century physics.
If the three fundamental limits are combined, a limit on a number of physical observ-
ables arise. The following limits are valid generally, for both composite and elementary
systems:
¾
Copyright © Christoph Schiller November 1997–May 2006

tE = 7.6 ë 10−44 s
2Għ
time interval: (744)
c5
td E 4 = 1.7 ë 10−78 sm
2Għ
time distance product: (745)
¾
c
c7
acceleration: aD = 4.0 ë 1051 m~s2 (746)
¾
2Għ
c5
angular frequency: ω D 2π = 8.2 ë 1043 ~s (747)
2Għ

Dvipsbugw
 units and limit values for all physical observables 1077

With the additional knowledge that, in nature, space and time can mix, we get
¾
dE = 2.3 ë 10−35 m
2Għ
distance: (748)
c3
AE 3 = 5.2 ë 10−70 m2
2Għ
area: (749)
c
2Għ 3~2
volume V E‹ 3  = 1.2 ë 10−104 m3 (750) Dvipsbugw
c
c3
curvature: KD = 1.9 ë 1069 ~m2 (751)
2Għ
c5
mass density: ρD = 6.5 ë 1095 kg~m3 (752)
8G 2 ħ
Of course, speed, action, angular momentum, power and force are also limited, as has

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

already been stated. Except for a small numerical factor, for every physical observable
these limits correspond to the Planck value. (The limit values are deduced from the com-
monly used Planck values simply by substituting 4G for G and ħ~2 for ħ.) These values
are the true natural units of nature. In fact, the most aesthetically pleasing solution is
to redefine the usual Planck values for every observable to these extremal values by ab-
sorbing the numerical factors into the respective definitions. In the following, we call the
redefined limits the (corrected) Planck limits and assume that the factors have been prop-
erly included. In other words, every natural unit or (corrected) Planck unit is at the same
time the limit value of the corresponding physical observable.
Most of these limit statements are found throughout the literature, though the numer-
ical factors are often different. Each limit has a string of publications attached to it. The
Ref. 1101, Ref. 1102 existence of a smallest measurable distance and time interval of the order of the Planck
values is discussed in quantum gravity and string theory. The largest curvature has been
Ref. 1103 studied in quantum gravity; it has important consequences for the ‘beginning’ of the uni-
verse, where it excludes any infinitely large or small observable. The maximum mass dens-
ity appears regularly in discussions on the energy of vacuum.
With the present deduction of the limits, two results are achieved. First of all, the vari-
ous arguments used in the literature are reduced to three generally accepted principles.
Second, the confusion about the numerical factors is solved. During the history of Planck
units, the numerical factors have varied greatly. For example, the fathers of quantum the-
Copyright © Christoph Schiller November 1997–May 2006

ory forgot the 1/2 in the definition of the quantum of action. Similarly, the specialists of
relativity did not emphasize the factor 4. With the present framework, the issue of the
correct factors in the Planck units can be considered as settled.
We also note that the dimensional independence of the three limits in nature also
means that quantum effects cannot be used to overcome the force limit; similarly, the
power limit cannot be used to overcome the speed limit. The same is valid for any other
combination of limits: they are independent and consistent at the same time.

Dvipsbugw
 1078 xi general rel ativity versus quantum mechanics

Limits to space and time

The three limits of nature (743) result in a minimum distance and a minimum time in-
terval. These minimum intervals directly result from the unification of quantum theory
and relativity. They do not appear if the theories are kept separate. In short, unification
implies that there is a smallest length in nature. The result is important: the formulation of
physics as a set of limit statements shows that the continuum description of space and time
is not correct. Continuity and manifolds are only approximations; they are valid for large
values of action, low speeds and low values of force. The reformulation of general relativ- Dvipsbugw
ity and quantum theory with limit statements makes this result especially clear. The result
is thus a direct consequence of the unification of quantum theory and general relativity.
No other assumptions are needed.
Ref. 1104 In fact, the connection between minimum length and gravity is not new. In 1969, An-
drei Sakharov pointed out that a minimum length implies gravity. He showed that regular-
izing quantum field theory on curved space with a cut-off will induce counter-terms that

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

include to lowest order the cosmological constant and then the Einstein Hilbert action.
The existence of limit values for the length observable (and all others) has numer-
Ref. 1102, Ref. 1105 ous consequences discussed in detail elsewhere. In particular, the existence of a smallest
length – and a corresponding shortest time interval – implies that no surface is physical
if any part of it requires a localization in space-time to dimensions smaller that the min-
imum length. (In addition, a physical surface must not cross any horizon.) Only by stip-
ulation of this condition can unphysical examples that contravene the force and power
limits be eliminated. For example, this condition has been overlooked in Bousso’s earlier
Ref. 1106 discussion of Bekenstein’s entropy bound – though not in his more recent ones.
The corrected value of the Planck length should also be the expression that appears
Ref. 1107 in the so-called theories of ‘doubly special relativity’. These then try to expand special
relativity in such a way that an invariant length appears in the theory.
A force limit in nature implies that no physical system can be smaller than a Schwarz-
schild black hole of the same mass. The force limit thus implies that point particles do
not exist. So far, this prediction has not been contradicted by observations, as the pre-
dicted sizes are so small that they are outside experimental reach. If quantum theory is
taken into account, this bound is sharpened. Because of the minimum length, elementary
particles are now predicted to be larger than the corrected Planck length. Detecting the
sizes of elementary particles would thus make it possible to check the force limit directly,
Ref. 1102 for example with future electric dipole measurements.
Copyright © Christoph Schiller November 1997–May 2006

Mass and energy limits

Mass plays a special role in all these arguments. The set of limits (743) does not make it
possible to extract a limit statement on the mass of physical systems. To find one, the aim
has to be restricted.
The Planck limits mentioned so far apply for all physical systems, whether they are
composed or elementary. Additional limits can only be found for elementary systems. In
quantum theory, the distance limit is a size limit only for composed systems. A particle is

Dvipsbugw
 units and limit values for all physical observables 1079

elementary if the system size l is smaller than any conceivable dimension:

lD
ħ
for elementary particles: . (753)
2mc
By using this new limit, valid only for elementary particles, the well-known mass, energy
and momentum limits are found:
¾
Dvipsbugw
for elementary particles: m D = 7.7 ë 10−9 kg = 0.42 ë 1019 GeV~c2
ħc

¾
8G
ħc 5
for elementary particles: E D = 6.9 ë 108 J = 0.42 ë 1019 GeV
¾
8G
ħc 3
for elementary particles: p D = 2.3 kg m~s = 0.42 ë 1019 GeV~c (754)
8G

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

These single-particle limits, corresponding to the corrected Planck mass, energy and mo-
mentum, were already discussed in 1968 by Andrei Sakharov, though again with different
Ref. 1108 numerical factors. They are regularly cited in elementary particle theory. Obviously, all
known measurements comply with the limits. It is also known that the unification of the
electromagnetic and the two nuclear interactions takes place at an energy near, but still
clearly below the maximum particle energy.

Virtual particles – a new definition

In fact, there are physical systems that exceed all three limits. Nature does contain systems
that move faster than light, that show action values below half the quantum of action
and that experience forces larger than the force limit. The systems in question are called
virtual particles.
Page 315 We know from special relativity that the virtual particles exchanged in collisions move
faster than light. We know from quantum theory that virtual particle exchange implies
Page 759 action values below the minimum action. Virtual particles also imply an instantaneous
change of momentum; they thus exceed the force limit. Virtual particles are thus those
particles that exceed all the limits that hold for (real) physical systems.

Limits in thermodynamics
Copyright © Christoph Schiller November 1997–May 2006

Thermodynamics can also be summarized in a single statement on motion: There is a

smallest entropy in nature.
SE .
k
(755)
2
The entropy S is limited by half the Boltzmann constant k. The result is almost 100 years
Ref. 1109 old; it was stated most clearly by Leo Szilard, though with a different numerical factor.
In the same way as in the other fields of physics, this result can also be phrased as a

Dvipsbugw
 1080 xi general rel ativity versus quantum mechanics

indeterminacy relation:
∆U E .
1 k
∆ (756)
T 2
Ref. 1110 This relation was given by Bohr and discussed by Heisenberg and many others (though
with k instead of k~2). It is mentioned here in order to complete the list of indeterminacy
relations and fundamental constants. With the single-particle limits, the entropy limit
leads to an upper limit for temperature:
¾
Dvipsbugw
5
TD = 1.0 ë 1032 K .
ħc
(757)
2Gk 2

This corresponds to the temperature at which the energy per degree of freedom is given
by the (corrected) Planck energy. A more realistic value would have to take into account
the number of degrees of freedom of a particle at Planck energy. This would change the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

numerical factor.

Electromagnetic limits and units

The discussion of limits can be extended to include electromagnetism. Using the (low-
energy) electromagnetic coupling constant α, one gets the following limits for physical
systems interacting electromagnetically:
¼
electric charge: qE 4πε 0 αcħ = e = 0.16 aC (758)
¾
c7 c4
electric field: ED = = 2.4 ë 1061 V~m (759)
64πε o αħG 2 4Ge
¾
c5 c3
magnetic field: BD = = 7.9 ë 1052 T (760)
64πε 0 αħG 2 4Ge
¾ ¾
c4 ħc 5
UD = = 1.5 ë 1027 V
1
voltage: (761)
32πε 0 αG 8G
¾
e
¾
ħ3 G
LE = = 4.4 ë 10−40 H
1 2ħG 1
inductance: (762)
8πε o α c7 e2 2c 5
Copyright © Christoph Schiller November 1997–May 2006

With the additional assumption that in nature at most one particle can occupy one Planck
volume, one gets

½ ¾
πε o α c 5 c9
charge density: ρe D = e = 1.3 ë 1085 C~m3 (763)
¾ ¾
2G 3 ħ 8G 3 ħ 3

C E 8πε 0 α =e = 1.0 ë 10−46 F

2ħG 2 8G
capacitance: (764)
c 3 c5 ħ

Dvipsbugw
 units and limit values for all physical observables 1081

For the case of a single conduction channel, one gets

RE = 2 = 2.1 kΩ
1 ħ
electric resistance: (765)
8πε 0 αc 2e
2e 2
electric conductivity: G D 8πε 0 αc = = 0.49 mS (766)
¾ ¾
ħ
2πε 0 αc 6 c5
electric current: ID =e = 7.4 ë 1023 A (767) Dvipsbugw
G 2ħG

Many of these limits have been studied already. The magnetic field limit plays a role in
the discussion of extreme stars and black holes. The maximum electric field plays a role
Ref. 1111 in the theory of gamma ray bursters. The studies of limit values for current, conductivity
and resistance in single channels are well known; the values and their effects have been
Ref. 1112 studied extensively in the 1980s and 1990s. They will probably win a Nobel prize in the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

future.
The observation of quarks and of collective excitations in semiconductors with charge
e~3 does not necessarily invalidate the charge limit for physical systems. In neither case
is there is a physical system – defined as localized mass–energy interacting incoherently
with the environment – with charge e~3.

Vacuum and mass–energy – two sides of the same coin

In this discussion we have found that there is a fixed limit to every physical observable.
Page 996 Many consequences have been discussed already in previous sections. There we saw that
Page 1032 the existence of a limit to all observables implies that at Planck scales no physical observ-
able can be described by real numbers and that no low-energy symmetry is valid.
One conclusion is especially important for the rest of our adventure. We saw that there
is a limit for the precision of length measurements in nature. The limit is valid both for
the length measurements of empty space and for the length measurements of matter (or
radiation). Now let us recall what we do when we measure the length of a table with a
ruler. To find the ends of the table, we must be able to distinguish the table from the sur-
rounding air. In more precise terms, we must be able to distinguish matter from vacuum.
But we have no way to perform this distinction at Planck energy. In these domains, the
intrinsic measurement limitations of nature do not allow us to say whether we are meas-
uring vacuum or matter. There is no way, at Planck scales, to distinguish the two.
Copyright © Christoph Schiller November 1997–May 2006

Ref. 1102, Ref. 1105

Page 349 We have explored this conclusion in detail above and have shown that it is the only
consistent description at Planck scales. The limitations in length measurement precision,
in mass measurement precision and in the precision of any other observable do not allow
to tell whether a box at Planck scale is full or empty. You can pick any observable you
want to distinguish vacuum from matter. Use colour, mass, size, charge, speed, angular
momentum, or anything you want. At Planck scales, the limits to observables also lead
to limits in measurement precision. At Planck scales, the measurement limits are of the
same size as the observable to be measured. As a result, it is impossible to distinguish
matter and vacuum at Planck scales.
We put the conclusion in the sharpest terms possible: Vacuum and matter do not differ

Dvipsbugw
 1082 xi general rel ativity versus quantum mechanics

at Planck scales. This counter-intuitive result is one of the charms of theoretical high en-
ergy physics. This result alone inspired many researchers in the field and induced them
to write best-sellers. Brian Greene was especially successful in presenting this side of
Ref. 1113 quantum geometry to the wider public.
However, at this point of our adventure, most issues are still open. The precise manner
in which a minimum distance leads to a homogeneous and isotropic vacuum is unclear.
The way to describe matter and vacuum with the same concepts has to be found. And of
course, the list of open questions in physics, given above, still waits. However, the concep-
tual results give hope; there are interesting issues awaiting us. Dvipsbugw

Curiosities and fun challenges about Planck limits

The (corrected) Planck limits are statements about properties of nature. There is no way
to measure values exceeding these limits, whatever experiment is performed. As can be
expected, such a claim provokes the search for counter-examples and leads to many para-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

doxes.

**
The minimum angular momentum may surprise at first, especially when one thinks about
particles with spin zero. However, the angular momentum of the statement is total angular
momentum, including the orbital part with respect to the observer. The total angular
momentum is never smaller than ħ~2.

**
If any interaction is stronger than gravity, how can the maximum force be determined
by gravity alone, which is the weakest interaction? It turns out that in situations near the
maximum force, the other interactions are negligible. This is the reason that gravity must
be included in a unified description of nature.
**
At first sight, it seems that electric charge can be used in such a way that the acceleration
of a charged body towards a charged black hole is increased to a value exceeding the force
Challenge 1488 ny limit. However, the changes in the horizon for charged black holes prevent this.
**
Copyright © Christoph Schiller November 1997–May 2006

The general connection that to every limit value in nature there is a corresponding inde-
terminacy relation is also valid for electricity. Indeed, there is an indeterminacy relation
for capacitors of the form
∆C ∆U E e (768)

where e is the positron charge, C capacity and U potential difference, and one between
electric current I and time t
∆I ∆t E e (769)

Ref. 1114 and both relations are found in the literature.

Dvipsbugw
 units and limit values for all physical observables 1083

**
The gravitational attraction between two masses never yields force values sufficiently high
to exceed the force limit. Why? First of all, masses m and M cannot come closer than the
sum of their horizon radii. Using F = GmM~r 2 with the distance r given by the (naive)
sum of the two black hole radii as r = 2G(M + m)~c 2 , one gets

c4
FD
Mm
4G (M + m)2
, (770)
Dvipsbugw
which is never larger than the force limit. Even two attracting black holes thus do not
exceed the force limit – in the inverse square approximation of universal gravity. The
minimum size of masses does not allow to exceed the a maximum force.
**
It is well known that gravity bends space. To be fully convincing, the calculation needs to

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

be repeated taking space curvature into account. The simplest way is to study the force
generated by a black hole on a test mass hanging from a wire that is lowered towards a
Ref. 1115 black hole horizon. For an unrealistic point mass, the force would diverge on the horizon.
Indeed, for a point mass m lowered towards a black hole of mass M at (conventionally
defined radial) distance d, the force would be

F= ¼
GMm
. (771)
d 2 1 − 2G M
d c2

The expression diverges at d = 0, the location of the horizon. However, even a test mass
cannot be smaller than its own gravitational radius. If we want to reach the horizon with
a realistic test mass, we need to chose a small test mass m; only a small – and thus light
– mass can get near the horizon. For vanishingly small masses however, the resulting
force tends to zero. Indeed, letting the distance tend to the smallest possible value by
letting d = 2G(m + M)~c 2 d = 2GM~c 2 requires m 0, which makes the force
F(m, d) vanish. If on the other hand, we remain away from the horizon and look for the
maximum force by using a mass as large as can possibly fit into the available distance
(the calculation is straightforward algebra) again the force limit is never exceeded. In
other words, for realistic test masses, expression (771) is never larger than c 4 ~4G. Taking
Copyright © Christoph Schiller November 1997–May 2006

into account the minimal size of test masses thus prevents that the maximum force is
exceeded in gravitational systems.
**
An absolute power limit implies a limit on the energy that can be transported per time
unit through any imaginable surface. At first sight, it may seem that the combined power
emitted by two radiation sources that each emit 3/4 of the maximum value should give 3/2
times the upper value. However, the combination forms a black hole or at least prevents
part of the radiation to be emitted by swallowing some of it between the sources.
**

Dvipsbugw
 1084 xi general rel ativity versus quantum mechanics

One possible system that actually achieves the power limit is the final stage of black hole
evaporation. But even in this case the power limit is not exceeded.

**
The maximum force limit states that the stress–energy tensor, when integrated over any
Ref. 1095 physical surface, does not exceed the limit value. No such integral, over any physical sur-
face whatsoever, of any tensor component in any coordinate system, can exceed the force
limit, provided that it is measured by a nearby observer or a test body with a realistic
proper size. The maximum force limit thus applies to any component of any force vector, Dvipsbugw
as well as to its magnitude. It applies to gravitational, electromagnetic, and nuclear forces.
It applies to all realistic observers. Whether the forces are real or fictitious is not import-
ant. It also plays no role whether we discuss 3-forces of Galilean physics or 4-forces of
special relativity. Indeed, the force limit applied to the 0-th component of the 4-force is
the power limit.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

**
Translated to mass flows, the power limit implies that flow of water through a tube is
limited in throughput. Also this limit seems unknown in the literature.
**
The force limit cannot be overcome with Lorentz boosts. A Lorentz boost of any non-
vanishing force value seems to allow exceeding the force limit at high speeds. However,
such a transformation would create a horizon that makes any point with a potentially
higher force value inaccessible.
**
The power limits is of interest if applied to the universe as a whole. Indeed, it can be used
to explain Olber’s paradox. The sky is dark at night because the combined luminosity of
all light sources in the universe cannot be brighter than the maximum value.

**
One notes that the negative energy volume density −Λc 4 ~4πG introduced by the positive
cosmological constant Λ corresponds to a negative pressure (both quantities have the
same dimensions). When multiplied with the minimum area it yields a force value Copyright © Christoph Schiller November 1997–May 2006

F= = 4.8 ë 10−79 N .
Λħc
(772)
2π

»
This is also the gravitational force between two corrected Planck masses located at the
cosmological distance π~4Λ . If we make the (wishful) assumption that this is the smal-
lest possible force in nature (the numerical prefactor is not finalized yet), we get the fas-
cinating conjecture that the full theory of general relativity, including the cosmological
constant, is defined by the combination of a maximum and a minimum force in nature.

Another consequence of the limits merits a dedicated section.

Dvipsbugw
 upper and lower limits to observables 1085

Upper and lower limits to observables

In our quest to understand motion, we have focused our attention to the limitations it is
subjected to. Special relativity poses a limit to speed, namely the speed of light c. General
Page 349 relativity limits force and power respectively by c 4~4G and c 5 ~4G, and quantum theory in-
troduces a smallest value ħ~2 for angular momentum or action. We saw that nature poses
a limit on entropy and a limit on electrical charge. Together, all these limits provide an
extreme value for every physical observable. The extremum is given by the correspond-
ing (corrected) Planck value. We have explored the list of all Planck limits in the previous Dvipsbugw
Page 1068 section. The question may now arise whether nature provides a limit for physical observ-
ables also on the opposite end of the measurement scale. For example, there is a highest
force and a highest power in nature; is there also a lowest force and a lowest power? Is
there also a lowest speed?
We show in the following that there indeed are such limits, for all observables. We
give the general method to generate such bounds and explore several examples.* The

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

exploration will lead us along an interesting scan across modern physics.

Size and energy dependence

Looking for additional limits in nature, we directly note a fundamental property. Any up-
per limit for angular momentum or any lower limit for power must be system dependent.
Such limits will not be absolute, but will depend on properties of the system. Now, any
physical system is a part of nature characterized by a boundary and its content.** The
simplest properties all systems share are thus their size (characterized in the following by
the diameter) L and energy E. With these characteristics we can enjoy deducing system-
dependent limits for every physical observable. The general method is straightforward.
We take the known inequalities for speed, action, power, charge and entropy and then
extract a limit for any observable, by inserting length and energy as required. We then
have to select the strictest of the limits we find.

Angular momentum, action and speed

It only takes a moment to note that the ratio of angular momentum D to mass times
length has the dimension of speed. Since speeds are limited by the speed of light, we get

DD
1
LE . (773)
Copyright © Christoph Schiller November 1997–May 2006

c
Indeed, there do not seem to be any exceptions to this limit in nature. No known system
has a larger angular momentum value, from atoms to molecules, from ice skaters to galax-
ies. For example, the most violently rotating objects, the so-called extremal black holes,
Ref. 1116 are also limited in angular momentum by D D LE~c. (In fact, this limit is correct only if
the energy is taken as the irreducible mass times c 2 ; if the usual mass is used, the limit is
too large by a factor 4.) One remarks that the limit deduced from general relativity, given

* This section was added in June 2004.

** We mention here that quantum theory narrows down this definition as a part of nature that in addition
Page 800 interacts incoherently with its environment. We assume that this condition is realized in the following.

Dvipsbugw
 1086 xi general rel ativity versus quantum mechanics

by D D L 2 c 3 ~4G is not stricter than the one just given. In addition, no system-dependent
lower limit for angular momentum can be deduced.
The maximum angular momentum value is also interesting when it is seen as action
limit. Action is the time integral of the difference between kinetic and potential energy.
In fact, since nature always minimizes action W, we are not used to search for systems
which maximize its value. You might check by yourself that the action limit W D LE~c is
Challenge 1489 ny not exceeded in any physical process.
Similarly, speed times mass times length is an action. Since action values in nature are
limited from below by ħ~2, we get Dvipsbugw

ħc 2 1
vE . (774)
2 LE
This relation is a rewritten form of the indeterminacy relation of quantum theory and is
no news. No system of energy E and diameter L has a smaller speed than this limit. Even

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the slowest imaginable processes show this speed value. For example, the extremely slow
Challenge 1490 ny radius change of a black hole by evaporation just realizes this minimal speed. Continuing
with the method just outlined, one also finds that the limit deduced from general relativity,
Challenge 1491 e v D (c 2 ~4G)(L~E), gives no new information. Therefore, no system-dependent upper
speed limit exists.
Incidentally, the limits are not unique. Additional limits can be found in a system-
atic way. Upper limits can be multiplied, for example, by factors of (L~E)(c 4 ~4G) or
(LE)(2~ħc) yielding additional, but less strict upper limits. A similar rule can be given
for lower limits.*
With the same approach we can now systematically deduce all size and energy depend-
ent limits for physical observables. We have a tour of the most important ones.

Force, power and luminosity

We saw that force and power are central to general relativity. Due to the connection W =
FLT between action W, force, distance and time, we can deduce

FE
ħ 1
. (775)
2c T 2
Experiments do not reach this limit. The smallest forces measured in nature are those
Copyright © Christoph Schiller November 1997–May 2006

in atomic force microscopes, where values as small as 1 aN are observed. However, the
values are all above the lower force limit.
The power P emitted by a system of size L and mass M is limited by

E P E 2ħG 3
M M
c3 (776)
L L
The left, upper limit gives the upper limit for any engine or lamp deduced from relativity;
not even the universe exceeds it. The right, lower limit gives the minimum power emit-
* The strictest upper limits are thus those with the smallest possible exponent for length, and the strictest
lower limits are those with the largest sensible exponent of length.

Dvipsbugw
 upper and lower limits to observables 1087

ted by any system due to quantum gravity effects. Indeed, no system is completely tight.
Even black holes, the systems with the best ability in nature to keep components inside
their enclosure, nevertheless radiate. The power radiated by black holes should just sat-
urate this limit, provided the length L is taken to be the circumference of the black hole.
(However, present literature values of the numerical factors in the black hole power are
not yet consistent). The claim of the quantum gravity limit is thus that the power emitted
by a black hole is the smallest power that is emitted by any composed system of the same
surface gravity.
Dvipsbugw
Acceleration
When the acceleration of a system is measured by a nearby inertial observer, the accele-
ration a of a system of size L and mass M is limited by

E a E 4G 2
1 M
c2 (777)

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

L L
The lower, right limit gives the acceleration due to universal gravity. Indeed, the relative
acceleration between a system and an observer has at least this value. The left, upper
limit to acceleration is the value due to special relativity. No exception to either of these
limits has ever been observed. Using the limit to the size of masses, the upper limit can
Ref. 1117 be transformed into the equivalent acceleration limit

2c 3
MEa (778)
ħ
which has never been approached either, despite many attempts. The upper limit to acce-
leration is thus a quantum limit, the lower one a gravitational limit.
The acceleration of the radius of a black hole due to evaporation can be much slower
Challenge 1492 ny than the limit a E 4GM~L 2 . Why is this not a counter-example?

Momentum
The momentum p of a system of size L is limited by

c3
LEpE
ħ 1
(779)
Copyright © Christoph Schiller November 1997–May 2006

4G 2 L
The lower limit is obviously due to quantum theory; experiments confirmed it for all
radiation and matter. The upper limit for momentum is due to general relativity. It has
never been exceeded.

Dvipsbugw
1088 xi general rel ativity versus quantum mechanics

Lifetime, distance and curvature

Time is something special. What are the limits to time measurements? Like before, we
find that any measured time interval t in a system in thermal equilibrium is limited by

ML 2 E t E 3 M
2 4G
(780)
ħ c
The lower time limit is the gravitational time. No clock can measure a smaller time than Dvipsbugw
this value. Similarly, no system can produce signals shorter than this duration. Black holes,
for example, emit radiation with a frequency given by this minimum value. The upper
time limit is expected to be the exact lifetime of a black hole. (There is no consensus in
the literature on the numerical factor yet.)
The upper limit to time measurements is due to quantum theory. It leads to a question:
What happens to a single atom in space after the limit time has passed by? Obviously, an
atom is not a composed system comparable with a black hole. The lifetime expression

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

assumes that decay can take place in the most tiny energy steps. As long as there is no
decay mechanism, the life-time formula does not apply. The expression (780) thus does
not apply to atoms.
Distance limits are straightforward.

ML 2 E d E 2 M
2c 4G
(781)
ħ c
Since curvature is an inverse square distance, curvature of space-time is also limited.

Mass change
The mass change dM~dt of a system of size L and mass M is limited by

c 5 L dM ħ 1
E E (782)
16G 2 M dt 2 L2
The limits apply to systems in thermal equilibrium. The left, upper limit is due to general
relativity; it is never exceeded. The right, lower limit is due to quantum theory. Again, all
experiments are consistent with the limit values.
Copyright © Christoph Schiller November 1997–May 2006

Mass and density

Limits for rest mass make only sense if the system is characterized by a size L only. We
then have
c2
LEME
ħ 1
(783)
4G 2c L
The upper limit for mass was discussed in general relativity. Adding mass or energy to
a black hole always increases its size. No system can show higher values than this value,
and indeed, no such system is known or even imaginable. The lower limit on mass is
obviously due to quantum theory; it follows directly from the quantum of action.

Dvipsbugw
 upper and lower limits to observables 1089

The mass density a system of size L is limited by

c2 1
EρE
ħ 1
(784)
4G L 2 2c L 4
The upper limit for mass density, due to general relativity, is only achieved for black holes.
The lower limit is the smallest density of a system due to quantum theory. It also applies to
the vacuum, if a piece of vacuum of site L is taken as a physical system. We note again that
many equivalent (but less strict) limits can be formulated by using the transformations Dvipsbugw
Page 1086 rules mentioned above.

The strange charm of the entropy bound

Ref. 1118 In 1973, Bekenstein discovered a famous limit that connects the entropy S of a physical
system with its size and mass.
No system has a larger entropy than one bounded by a horizon. The larger the horizon

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

surface, the larger the entropy. We write

D
S A
(785)
S limit A limit

which gives
kc 3
SD A, (786)
4Għ
where A is the surface of the system. Equality is realized only for black holes. The old
question of the origin of the factor 4 in the entropy of black holes is thus answered here in
he following way: it is due to the factor 4 in the force or power bound in nature (provided
Ref. 1119 that the factors from the Planck entropy and the Planck action cancel). The future will
Page 245 tell whether this explanation will stand the winds of time. Stay tuned.
We can also derive a more general relation if we use a mysterious assumption that we
discuss afterwards. We assume that the limits for vacuum are opposite to those for matter.
We can then write c 2 ~4G D M~L for the vacuum. This gives

SD ML =
πkc 2πkc
MR . (787)
ħ ħ
Copyright © Christoph Schiller November 1997–May 2006

In other words, we used

D
S M A L corr. Planck
. (788)
S corr. Planck M corr. Planck A corr. Planck L

Expression (787) is called Bekenstein’s entropy bound. Up to today, no exception has been
Ref. 1106 found or constructed, despite many attempts. Again, the limit value itself is only realized
for black holes.
We still need to explain the strange assumption used above. We are exploring the en-
tropy of a horizon. Horizons are not matter, but limits to empty space. The entropy of

Dvipsbugw
 1090 xi general rel ativity versus quantum mechanics

horizons is due to the large amount of virtual particles found there. In order to deduce
the maximum entropy of expression (788) one therefore has to use the properties of the
vacuum. In other words, either (1) we use a mass to length ratio for vacuum above the
Planck limit or (2) we use the Planck entropy as maximum value for vacuum.
Other, equivalent limits for entropy can be found if other variables are introduced. For
Ref. 1120 example, since the ratio of shear viscosity η to the volume density of entropy (times k)
has the dimensions of an action, we can directly write
Dvipsbugw
SD
k
ηV (789)
ħ
Again, equality is only reached in the case of black holes. With time, the list of similar
bounds will grow longer and longer.
Is there also a smallest entropy limit? So far, there does not seem to be a system-
dependent minimum value for entropy; the approach gives no expression that is larger

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1493 ny than k.
The entropy limit is an important step in making the description of motion consistent.
If space-time can move, as general relativity maintains, it also has an entropy. How could
entropy be limited if space-time is continuous? Clearly, due to the minimum distance
and a minimum time in nature, space-time is not continuous, but has a finite number
of degrees of freedom. The number of degrees of freedom and thus the entropy of space-
time is thus finite.
In addition, the Bekenstein limit also allows some interesting speculations. Let us spec-
ulate that the universe itself, being surrounded by a horizon, saturates the Bekenstein
bound. The entropy bound gives a bound to all degrees of freedom inside a system; it
Challenge 1494 e tells us that the number N d.o.f . of degrees of freedom of the universe is roughly

N d.o.f .  10132 . (790)

This compares with the number N Pl. vol. of Planck volumes in the universe

N Pl. vol.  10183 (791)

and with the number N part. of particles in the universe

N part.  1091 .
Copyright © Christoph Schiller November 1997–May 2006

(792)

In other words, particles are only a tiny fraction of what moves around. Most motion
must be that of space-time. At the same time, space-time moves much less than naively
expected. Finding out how all this happens is the challenge of the unified description of
motion.

Temperature
A lower limit for the temperature of a thermal system can be found using the idea that
the number of degrees of freedom of a system is limited by its surface, more precisely, by

Dvipsbugw
 upper and lower limits to observables 1091

the ratio between the surface and the Planck surface. One gets the limit

TE
4Għ M
(793)
πkc L 2
Alternatively, using the method given above, one can use the limit on the thermal energy
kT~2 E ħc~(2πL) (the thermal wavelength must be smaller than the size of the system)
together with the limit on mass c 2 ~4G E M~L and deduce the same result.
We know the limit already: when the system is a black hole, it gives the temperature of Dvipsbugw
the emitted radiation. In other words, the temperature of black holes is the lower limit for
all physical systems for which a temperature can be defined, provided they share the same
boundary gravity. As a criterion, boundary gravity makes sense: boundary gravity is ac-
cessible from the outside and describes the full physical system, since it makes use both of
its boundary and its content. So far, no exception to this claim is known. All systems from
everyday life comply with it, as do all stars. Even the coldest known systems in the uni-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

verse, namely Bose–Einstein condensates and other cold matter in various laboratories,
are much hotter than the limit, and thus much hotter than black holes of the same surface
Challenge 1495 n gravity. (Since a consistent Lorentz transformation for temperature is not possible, as we
Page 310 saw earlier, the limit of minimum temperature is only valid for an observer at the same
gravitational potential and at zero relative speed to the system under consideration.)
There seems to be no consistent way to define an upper limit for a system-dependent
Challenge 1496 ny temperature. However, limits for other thermodynamic quantities can be found, but are
not discussed here.

Electromagnetic observables
When electromagnetism plays a role, the involved system also needs to be characterized
by a charge Q. The method used so far then gives the following lower limit for the electric
field E:
M2
E E 4Ge 2 2 (794)
Q L

We write the limitºusing the elementary charge e, though writing it using the fine structure
constant via e = 4πε 0 αħc would be more appropriate. Experimentally, this limit is not
exceeded in any system in nature. Can you show whether it is achieved by maximally
Challenge 1497 ny charged black holes?
Copyright © Christoph Schiller November 1997–May 2006

For the magnetic field we get

4Ge M 2
BE (795)
c Q 2 L2

Again, this limit is satisfied all known systems in nature.

Similar limits can be found for the other electromagnetic observables. In fact, several
of the earlier limits are modified when electrical charge is included. Can you show how
Challenge 1498 ny the size limit changes when electric charge is taken into account? In fact, a dedicated
research field is concerned only with the deduction of the most general limits valid in

Dvipsbugw
 1092 xi general rel ativity versus quantum mechanics

nature.

Curiosities and fun challenges about limits to observables

The limits yield a plethora of interesting paradoxes that can be discussed in lectures and
student exercises. All paradoxes can be solved by carefully taking into account the com-
bination of effects from general relativity and quantum theory. All apparent violations
only appear when one of the two aspects is somehow forgotten. We study a few examples.
Dvipsbugw
**
Several black hole limits are of importance to the universe itself. For example, the ob-
served average mass density of the universe is not far from the corresponding limit. Also
the lifetime limit is obviously valid for the universe and provides an upper limit for its
age. However, the age of the universe is far from that limit by a large factor. In fact, since
the universe’s size and age still increase, the age limit is pushed further into the future

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

with every second that passes. The universe evolves in a way to escape its own decay.

**
The content of a system is not only characterized by its mass and charge, but also by its
strangeness, isospin, colour charge, charge and parity. Can you deduce the limits for these
Challenge 1499 r quantities?
**
In our discussion of black hole limits, we silently assumed that they interact, like any
thermal system, in an incoherent way with the environment. What changes in the results
Challenge 1500 n of this section when this condition is dropped? Which limits can be overcome?
**
Challenge 1501 e Can you find a general method to deduce all limits?

**
Brushing some important details aside, we can take the following summary of our study
of nature. Galilean physics is that description for which the difference between composed
and elementary systems does not exist. Quantum theory is the description of nature with
no (really large) composed systems; general relativity is the description of nature with
Copyright © Christoph Schiller November 1997–May 2006

no elementary systems. This distinction leads to the following interesting conclusion. A

unified theory of nature has to unify quantum theory and general relativity. Since the
first theory affirms that no (really large) composed systems exist, while the other that no
elementary systems exist, a unified theory should state that no systems exist. This strange
result indeed seems to be one way to look at the issue. The conclusion is corroborated by
the result that in the unified description of nature, the observables time, space and mass
Page 1068 cannot be distinguished clearly from each other, which implies that systems cannot be
clearly distinguished from their surroundings. To be precise, systems thus do not really
exist at unification energy.

Dvipsbugw
 limits to measurement precision and their challenge to thought 1093

L imits to measurement precision and their challenge to

thought
We now know that in nature, every physical measurement has a lower and an upper
bound. One of the bounds is size-dependent, the other is absolute. So far, no violation
of these claims is known. The smallest relative measurement error possible in nature is
thus the ratio of the two bounds. In short, a smallest length, a highest force and a smallest
action, when taken together, imply that all measurements are limited in precision.
A fundamental limit to measurement precision is not a new result any more. But it Dvipsbugw
raises many issues. If the mass and the size of any system are themselves imprecise, can
Challenge 1502 r you imagine or deduce what happens then to the limit formulae given above?
Due to the fundamental limits to measurement precision, the measured values of phys-
ical observables do not require the full set of real numbers. In fact, limited precision implies
that no observable can be described by real numbers. We thus recover again a result that
appears whenever quantum theory and gravity are brought together.
In addition, we found that measurement errors increase when the characteristic meas-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ref. 1102
urement energy approaches the Planck energy. In that domain, the measurement errors
of any observable are comparable with the measurement values themselves.
Limited measurement precision thus implies that at the Planck energy it is impossible
to speak about points, instants, events or dimensionality. Limited measurement precision
also implies that at the Planck length it is impossible to distinguish positive and negative
time values: particle and anti-particles are thus not clearly distinguished at Planck scales.
A smallest length in nature thus implies that there is no way to define the exact bound-
aries of objects or elementary particles. However, a boundary is what separates matter
from vacuum. In short, a minimum measurement error means that, at Planck scales, it is
Ref. 1102 impossible to distinguish objects from vacuum with complete precision. To put it bluntly,
at Planck scales, time and space do not exist.
The mentioned conclusions are the same as those that are drawn by modern research
on unified theories. The force limit, together with the other limits, makes it possible to
reach the same conceptual results found by string theory and the various quantum grav-
ity approaches. To show the power of the maximum force limit, we now explore a few
conclusions which go beyond present approaches.

Measurement precision and the existence of sets

The impossibility of completely eliminating measurement errors has an additional and im-
Copyright © Christoph Schiller November 1997–May 2006

portant consequence. In physics, it is assumed that nature is a set of components. These

components are assumed to be separable from each other. This tacit assumption is intro-
duced in three main situations: it is assumed that matter consists of separable particles,
that space-time consists of separable events or points, and that the set of states consists of
separable initial conditions. So far, all of physics has thus built its complete description
of nature on the concept of set.
A fundamentally limited measurement precision implies that nature is not a set of
Ref. 1102, Ref. 1105 such separable elements. A limited measurement precision implies that distinguishing
physical entities is possible only approximately. The approximate distinction is only pos-
sible at energies much lower than the Planck energy. As humans we do live at such smaller
energies; thus we can safely make the approximation. Indeed, the approximation is excel-

Dvipsbugw
 1094 xi general rel ativity versus quantum mechanics

lent; we do not notice any error when performing it. But the discussion of the situation at
Planck energies shows that a perfect separation is impossible in principle. In particular, at
the cosmic horizon, at the big bang, and at Planck scales any precise distinction between
two events or two particles becomes impossible.
Another way to reach this result is the following. Separation of two entities requires dif-
ferent measurement results, such as different positions, different masses, different velocit-
ies, etc. Whatever observable is chosen, at the Planck energy the distinction becomes im-
possible, due to the large measurements errors. Only at everyday energies is a distinction
approximately possible. Any distinction between two physical systems, such as between a Dvipsbugw
toothpick and a mountain, is thus possible only approximately; at Planck scales, a bound-
Ref. 1102 ary cannot be drawn.
A third argument is the following. In order to count any entities in nature – a set of
particles, a discrete set of points, or any other discrete set of physical observables – the en-
tities have to be separable. The inevitable measurement errors, however, contradict separ-
Ref. 1102 ability. At the Planck energy, it is thus impossible to count physical objects with precision.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Nature has no parts.
In short, at Planck energies a perfect separation is impossible in principle. We cannot
distinguish observations at Planck energies. In short, at Planck scale it is impossible to split
nature into separate entities. There are no mathematical elements of any kind – or of any
set – in nature. Elements of sets cannot be defined. As a result, in nature, neither discrete
Ref. 1105 nor continuous sets can be constructed. Nature does not contain sets or elements.
Since sets and elements are only approximations, the concept of ‘set’, which assumes
separable elements, is already too specialized to describe nature. Nature cannot be de-
scribed at Planck scales – i.e., with full precision – if any of the concepts used for its de-
scription presupposes sets. However, all concepts used in the past twenty-five centuries to
describe nature – space, time, particles, phase space, Hilbert space, Fock space, particle
space, loop space or moduli space – are based on sets. They all must be abandoned at
Planck energy. No approach used so far in theoretical physics, not even string theory or the
various quantum gravity approaches, satisfies the requirement to abandon sets. Nature is
one and has no parts. Nature must be described by a mathematical concept that does not
contain any set. This requirement must guide us in the future search for the unification
of relativity and quantum theory.
Es ist fast unmöglich, die Fackel der Wahrheit

“ durch ein Gedränge zu tragen, ohne jemandem

den Bart zu sengen.*
Georg Christoph Lichtenberg (1742–1799)
Copyright © Christoph Schiller November 1997–May 2006

Why are observers needed? ”

Certain papers on quantum theory give the impression that observers are indispensable
for quantum physics. We have debunked this belief already, showing that the observer in
quantum theory is mainly a bath with a definite interaction. Often, humans are observers.
At Planck scales, observers also play a role. At Planck scales, quantum theory is man-
datory. In these domains, observers must realize an additional requirement: they must

* ‘It is almost impossible to carry the torch of truth through a crowd without scorching somebody’s beard.’

Dvipsbugw
 limits to measurement precision and their challenge to thought 1095

function at low energies. Only at low energy, an observer can introduce sets for the de-
scription of nature. Introducing observers is thus the same as introducing sets.
To put it in another way, the limits of human observers is that they cannot avoid us-
ing sets. However, human observers share this limitation with video recorders, cameras,
computers and pencil and paper. Nothing singles out humans in this aspect.
In simple terms, observers are needed to describe nature at Planck scales only in so
far as they are and use sets. We should not get too bloated about our own importance.

A solution to Hilbert’s sixth problem Dvipsbugw

In the year 1900, David Hilbert gave a well-known lecture in which he listed twenty-
Ref. 1121 three of the great challenges facing mathematics in the twentieth century. Most prob-
lems provided challenges to many mathematicians for decades afterwards. Of the still
unsolved ones, Hilbert’s sixth problem challenges mathematicians and physicists to find
an axiomatic treatment of physics. The challenge has remained in the minds of many

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

physicists since that time.
Since nature does not contain sets, we can deduce that such an axiomatic description
Ref. 1105 of nature does not exist! The reasoning is simple; we only have to look at the axiomatic
systems found in mathematics. Axiomatic systems define mathematical structures. These
structures are of three main types: algebraic systems, order systems or topological sys-
tems. Most mathematical structures – such as symmetry groups, vector spaces, mani-
folds or fields – are combinations of all three. But all mathematical structures contain
sets. Mathematics does not provide axiomatic systems that do not contain sets. The un-
derlying reason is that every mathematical concept contains at least one set.
Furthermore, all physical concepts used so far in physics contain sets. For humans, it is
difficult even simply to think without first defining a set of possibilities. However, nature
is different; nature does not contain sets. Therefore, an axiomatic formulation of physics
is impossible. Of course, this conclusion does not rule out unification in itself; however, it
does rule out an axiomatic version of it. The result surprises, as separate axiomatic treat-
ments of quantum theory or general relativity (see above) are possible. Indeed, only their
unification, not the separate theories, must be approached without an axiomatic systems.
Axiomatic systems in physics are always approximate. The requirement to abandon axio-
matic systems is one of the reasons for the difficulties in reaching the unified description
of nature. Copyright © Christoph Schiller November 1997–May 2006

Outlook
Physics can be summarized in a few limit statements. They imply that in nature every
physical observable is limited by a value near the Planck value. The speed limit is equi-
valent to special relativity, the force limit to general relativity, and the action limit to
quantum theory. Even though this summary could have been made (or at least conjec-
tured) by Planck, Einstein or the fathers of quantum theory, it is much more recent. The
numerical factors for most limit values are new. The limits provoke interesting Gedanken
experiments, none of which leads to violations of the limits. On the other hand, the force
limit is not yet within direct experimental reach.
The existence of limit values to all observables implies that the description of space-
time with a continuous manifold is not correct at Planck scales; it is only an approx-

Dvipsbugw
 1096 xi general rel ativity versus quantum mechanics

imation. For the same reason, is predicted that elementary particles are not point-like.
Nature’s limits also imply the non-distinguishability of matter and vacuum. As a result,
the structure of particles and of space-time remains to be clarified. So far, we can conclude
that nature can be described by sets only approximately. The limit statements show that
Hilbert’s sixth problem cannot be solved and that unification requires fresh approaches,
taking unbeaten paths into unexplored territory.
We saw that at Planck scales there is no time, no space, and there are no particles. Mo-
tion is a low energy phenomenon. Motion only appears for low-energy observers. These
Page 28 are observers who use sets. The (inaccurate) citation of Zeno at the beginning of our walk, Dvipsbugw
stating that motion is an illusion, turns out to be correct! Therefore, we now need to find
out how motion actually arises.
The discussion so far hints that motion appears as soon as sets are introduced. To check
this hypothesis, we need a description of nature without sets. The only way to avoid the
use of sets seems a description of empty space-time, radiation and matter as being made
of the same underlying entity. The inclusion of space-time dualities and of interaction du-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

alities is most probably a necessary step. Indeed, both string theory and modern quantum
gravity attempt this, though possibly not yet with the full radicalism necessary. Realizing
this unification is our next task.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
bibliography 1097

Biblio graphy

1093 Bohr explained the indivisibilty of the quantum of action in his famous Como lecture. N.
B ohr, Atomtheorie und Naturbeschreibung, Springer, Berlin, 1931. More statements about
the indivisibility of the quantum of action can be found in N. B ohr, Atomic Physics and
Human Knowledge, Science Editions, New York, 1961. For summaries of Bohr’s ideas by
others see Max Jammer, The Philosophy of Quantum Mechanics, Wiley, first edition, 1974,
pp. 90–91, and John Honner, The Description of Nature – Niels Bohr and the Philosophy
of Quantum Physics, Clarendon Press, 1987, p. 104. Cited on page 1069. Dvipsbugw
1094 For an overview of the quantum of action as basis of quantum theory, see the introduction
to quantum theory in this textbook, available as by C. Schiller, An appetizer – quantum
theory for poets and lawyers, http://www.motionmountain.net/C-5-QEDA.pdf. Cited on
page 1070.
1095 The first published statements of the principle of maximum force were in the present text-
book and independently in the paper by G.W. Gibbons, The maximum tension principle

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

in general relativity, Foundations of Physics 32, pp. 1891–1901, 2002, or http://www.arxiv.
org/abs/hep-th/0210109. Cited on pages 1071 and 1084.
1096 Maximal luminosity of often mentioned in studies around gravitational wave detection;
nevertheless, the general maximum is never mentioned. See for example L. Ju, D.G.
Bl air & C. Zhao, Detection of gravitational waves, Reports on Progress in Physics 63,
pp. 1317–1427, 2000. See also C. Misner, K. Thorne & J.A. Wheeler, Gravitation,
Freeman, 1973, page 980. Cited on page 1072.
1097 See for example Wolf gang R indler, Relativity – Special, General and Cosmological,
Oxford University Press, 2001, p. 70 ff, or R ay d’Inverno Introducing Einstein’s Relativ-
ity, Clarendon Press, 1992, p. 36 ff. Cited on page 1073.
1098 T. Jacobson, Thermodynamics of spacetime: the Einstein equation of state, Physical Re-
view Letters 75, pp. 1260–1263, 1995, or preprint http://www.arxiv.org/abs/gr-qc/9504004
Cited on page 1073.
1099 C. Schiller, Maximum force: a simple principle encompassing general relativity, part
of this text and downloadable as http://www.motionmountain.net/C-2-MAXF.pdf. Cited
on page 1075.
1100 Indeterminacy relations in general relativity are discussed in C.A. Mead, Possible con-
nection between gravitation and fundamental length, Physical Review B 135, pp. 849–862,
1964. See also P.K. Townsend, Small-scale structure of space-time as the origin of the
gravitational constant, Physical Review D 15, pp. 2795–2801, 1977, or the paper by M.-
T. Jaekel & S. R enaud, Gravitational quantum limit for length measurement, Physics
Copyright © Christoph Schiller November 1997–May 2006

Letters A 185, pp. 143–148, 1994. Cited on page 1075.

1101 Minimal length and minimal time intervals are discussed for example by G. Amelino–
Camelia, Limits on the measurability of space-time distances in (the semiclassical ap-
proximation of) quantum gravity, Modern Physics Letters A 9, pp. 3415–3422, 1994, and by
Y.J. Ng & H. Van Dam, Limit to space-time measurement, Modern Physics Letters A 9,
pp. 335–340, 1994. Many other authors have explored the topic. Cited on page 1077.
1102 A pedagogical summary is C. Schiller, http://www.motionmountain.net/C10-QMGR.
pdf or C. Schiller, Le vide diffère-t-il de la matière? in E. Gunzig & S. Diner, ed-
iteurs, Le vide – Univers du tout et du rien – Des physiciens et des philosophes s’interrogent,
Les Editions de l’Université de Bruxelles, 1998. An older English version is also available as
C. Schiller, Does matter differ from vacuum? http://www.arxiv.org/abs/gr-qc/9610066.

Dvipsbugw
1098 xi general rel ativity versus quantum mechanics

Cited on pages 1077, 1078, 1081, 1093, and 1094.

1103 Maximal curvature, as well as area and volume quantization are discussed by A.
Ashtekar, Quantum geometry and gravity: recent advances, http://www.arxiv.org/
abs/gr-qc/0112038 and by A. Ashtekar, Quantum geometry in action: big bang and
black holes, preprint downloadable at http://www.arxiv.org/abs/math-ph/0202008. Cited
on page 1077.
1104 This was kindly pointed out by Achim Kempf. The story is told in A.D. Sakharov, Gen-
eral Relativity and Gravitation 32, p. 365, 2000, a reprint of a paper of 1969. Cited on page
1078. Dvipsbugw
1105 About set definition and Hilbert’s sixth problem see C. Schiller, Nature at large scales
– is the universe something or nothing? http://www.motionmountain.net/C11-LGSM.pdf.
Cited on pages 1078, 1081, 1093, 1094, and 1095.
1106 Several incorrect counterclaims to the entropy limit were made in R. B ousso, The holo-
graphic principle, Review of Modern Physics 74, pp. 825–874, 2002, also available as down-
loadable preprint at http://www.arxiv.org/abs/hep-th/0203101. Bousso has changed pos-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ition in the meantime. Cited on pages 1078 and 1089.
1107 G. Amelino–Camelia, Doubly-special relativity: first results and key open problems,
International Journal of Modern Physics 11, pp. 1643–1669, 2002, electronic preprint avail-
able as http://www.arxiv.org/abs/gr-qc/0210063. See also J. Magueijo & L. Smolin,
Lorentz invariance with an invariant energy scale, Physical Review Letters 88, p. 190403,
2002, or http://www.arxiv.org/abs/hep-th/0112090. Cited on page 1078.
1108 Maximons, elementary particles of Planck mass, are discussed by A.D. Sakharov, Va-
cuum quantum fluctuations in curved space and the theory of gravitation, Soviet Physics
– Doklady, 12, pp. 1040–1041, 1968. Cited on page 1079.
1109 Minimal entropy is discussed by L. Szil ard, Über die Entropieverminderung in einem
thermodynamischen System bei Eingriffen intelligenter Wesen, Zeitschrift für Physik 53,
pp. 840–856, 1929. This classic paper can also be found in English translation in his collec-
ted works. Cited on page 1079.
1110 See for example A.E. Shalyt-Margolin & A.Ya. Tregubovich, Generalized
uncertainty relation in thermodynamics, http://www.arxiv.org/abs/gr-qc/0307018, or J.
Uffink & J. van Lith-van Dis, Thermodynamic uncertainty relations, Foundations
of Physics 29, pp. 655–692, 1999. Cited on page 1080.
1111 Gamma ray bursts are discussed by G. Preparata, R. Ruffini & S.-S. Xue, The dya-
dosphere of black holes and gamma-ray bursts, Astronomy and Astrophysics 338, pp. L87–
L90, 1998, and C.L. Bianco, R. Ruffini & S.-S. Xue, The elementary spike produced
by a pure e+ e− pair-electromagnetic pulse from a black hole: the PEM pulse, Astronomy
Copyright © Christoph Schiller November 1997–May 2006

and Astrophysics 368, pp. 377–390, 2001. Cited on page 1081.

1112 See for example the review in C.W.J. Beenakker & al., Quantum transport in semicon-
ductor nanostructures, pp. 1–228, in H. Ehrenreich & D. Turnbull, eds., Solid State
Physics, 44, Academic Press, Cambridge, 1991. The prediction of a future Nobel Prize was
made to the authors by C. Schiller in 1992. Cited on page 1081.
1113 Brian Greene, The Elegant Universe - Superstrings, Hidden Dimensions, and the Quest
for the Ultimate Theory, Vintage, 2000, Cited on page 1082.
1114 A discussion of a different electric uncertainty relation, namely between current and
charge, can be found in Y-Q. Li & B. Chen, Quantum theory for mesoscopic electronic
circuits and its applications, http://www.arxiv.org/abs/cond-mat/9907171. Cited on page
1082.

Dvipsbugw
bibliography 1099

1115 Wolf gang R indler, Relativity – Special, General and Cosmological, Oxford University
Press, 2001, p. 230. Cited on page 1083.
1116 Hans C. Ohanian & R emo Ruffini, Gravitation and Spacetime, W.W. Norton & Co.,
1994. Cited on page 1085.
1117 A maximal acceleration for microscopic systems was stated by E.R. Caianiello, Lettere
al Nuovo Cimento 41, p. 370, 1984. It is discussed by G. Papini, Shadows of a maximal
acceleration, http://www.arxiv.org/abs/gr-qc/0211011. Cited on page 1087.
1118 The entropy limit for black holes is discussed by J.D. Bekenstein, Entropy bounds
Dvipsbugw
and black hole remnants, Physical Review D 49, pp. 1912–1921, 1994. See also J.D. Bek-
enstein, Universal upper bound on the entropy-to-energy ration for bounded systems,
Physical Review D 23, pp. 287–298, 1981. Cited on page 1089.
1119 Private communications with Jos Uffink and Bernard Lavenda. Cited on page 1089.
1120 P. Kovtun, D.T. Son & A.O. Starinets, A viscosity bound conjecture, preprint
found at http://www.arxiv.org/abs/hep-th/0405231. Cited on page 1090.
H. Wussing & P.S. Alexandrov, (eds.) Die Hilbertschen Probleme, Akademische Ver-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1121
lagsgesellschaft Geest & Portig, 1983, or Ben H. Yandell, The Honours Class: Hilbert’s
Problems and their Solvers, A.K. Peters, 2002. Cited on page 1095.

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
the shape of points – extension in nature 1101

37. the shape of p oints – extension in nature

Nil tam difficile est, quin quaerendo investigari

“ possiet.

The expressions for the Compton wavelength λ = h~mc and for the Schwarzschild ra-
Terence*

”
dius r s = 2Gm~c 2 imply a number of arguments which lead to the conclusion that at
Planck energies, space-time points and point particles must be described, in contrast to Dvipsbugw
their name, by extended entities. These arguments point towards a connection between
microscopic and macroscopic scales, confirming the present results of string theory and
quantum gravity. At the same time, they provide a pedagogical summary of this aspect
of present day theoretical physics.**
1 – It is shown that any experiment trying to measure the size or the shape of an ele-
mentary particle with high precision inevitably leads to the result that at least one dimen-
sion of the particle is of macroscopic size.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2 – There is no data showing that empty space-time is continuous, but enough data
showing that it is not. It is then argued that in order to build up an entity such as the
vacuum, that is extended in three dimensions, one necessarily needs extended building
blocks
3 – The existence of minimum measurable distances and time intervals is shown to im-
ply the existence of space-time duality, i.e. a symmetry between very large and very small
distances. Space-time duality in turn implies that the fundamental entities that make up
vacuum and matter are extended.
4 – It is argued that the constituents of the universe and thus of space-time, matter and
radiation cannot form a set. As a consequence any precise description of nature must use
extended entities.
5 – The Bekenstein–Hawking expression for the entropy of black holes, in particular
its surface dependence, confirms that both space-time and particles are composed of ex-
tended entities.
6 – Trying to extend statistical properties to Planck scales shows that both particles and
space-time points behave as braids at high energies, a fact which also requires extended
entities.
7 – The Dirac construction for spin provides a model for fermions, without contradic-
tion with experiments, that points to extended entities.
An overview of other arguments in favour of extended entities provided by present
Copyright © Christoph Schiller November 1997–May 2006

research efforts is also given. To complete the discussion, experimental and theoretical
checks for the idea of extended building blocks of nature are presented.

* ‘Nothing is so difficult that it could not be investigated.’ Terence is Publius Terentius Afer (c. 190–159 bce ),
important roman poet. He writes this in his play Heauton Timorumenos, verse 675.
** This section describes a research topic and as such is not a compendium of generally accepted results (yet).
It was written between December 2001 and May 2002.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1102

Introduction – vacuum and particles

Our greatest pretenses are built up not to hide

“ the evil and the ugly in us, but our emptyness.

The hardest thing to hide is something that is
not there.
Eric Hoffer, The Passionate State of Mind.

Only the separation of the description of nature into general relativity and quantum the-
ory allows us to use continuous space-time and point particles as basic entities. When
the two theories are united, what we use to call ‘point’ turns out to have quite counter-
” Dvipsbugw

intuitive properties. We explore a few of them in the following.

Above, we have given a standard argument showing that points do not exist in nature.
The Compton wavelength and the Schwarzschild radius together determine a minimal
Ref. 1122, Ref. 1123 length and a minimal time interval in nature. They are given (within a factor of order
one, usually omitted in the following) by the Planck length and the Planck time, with the
»

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

values
l Pl = »ħG~c 3 = 1.6 ë 10−35 m
t Pl = ħG~c 5 = 5.4 ë 10−44 s .
(796)

The existence of a minimal length and space interval in nature implies that points in
space, time or space-time have no experimental backing and that we are forced to part
from the traditional idea of continuity. Even though, properly speaking, points do not
exist, and thus space points, events or point particles do not exist either, we can still ask
what happens when we study these entities in detail. The results provide many fascinating
surprises.
Page 1013 Using a simple Gedanken experiment, we have found above that particles and space-
time cannot be distinguished from each other at Planck scales. The argument was the
following. The largest mass that can be put in a box of size R is a black hole with a Schwarz-
schild radius of the same value. That is also the largest possible mass measurement error.
But any piece of vacuum also has a smallest mass measurement error.
The issue of a smallest mass measurement error is so important that it merits spe-
cial attention. Mass measurement errors always prevent humans to state that a region of
space has zero mass. In exactly the same way, also nature cannot ‘know’ that the mass
of a region is zero, provided that this error is due to quantum indeterminacy. Otherwise
nature would circumvent quantum theory itself. Energy and mass are always unsharp in
Copyright © Christoph Schiller November 1997–May 2006

quantum theory. We are not used to apply this fact to vacuum itself, but at Planck scales
we have to. We remember from quantum field theory that the vacuum, like any other sys-
tem, has a mass; of course, its value is zero for long time averages. For finite measuring
times, the mass value will be uncertain and thus different from zero. Not only limitations
in time, but also limitations in space lead to mass indeterminacy for the vacuum. These
indeterminacies in turn lead to a minimum mass errors for vacuum regions of finite size.
Quantum theory implies that nobody, not even nature, knows the exact mass value of a
system or of a region of empty space.
A box is empty if it does not contain anything. But emptiness is not well defined for
photons with wavelength of the size R of the box or larger. Thus the mass measurement
error for an ‘empty’ box – corresponding to what we call vacuum – is due to the indeterm-

Dvipsbugw
 the shape of points – extension in nature 1103

inacy relation and is given by that mass whose Compton wavelength matches the size of
Page 1013 the box. As shown earlier on, the same mass value is found by every other Gedanken ex-
periment: trying to determine the gravitational mass by weighing the ‘piece’ of vacuum or
Ref. 1122 by measuring its kinetic energy gives the same result. Another, but in the end equivalent
way to show that a region of vacuum has a finite mass error is to study how the vacuum
energy depends on the position indeterminacy of the border of the region. Any region is
defined through its border. The position indeterminacy of the border will induce a mass
error for the contents of the box, in the same way that a time limit does. Again, the res-
ulting mass error value for a region of vacuum is the one for which the box size is the Dvipsbugw
Compton wavelength.
Summarizing, for a box of size R, nature allows only mass values and mass measure-
ment error value m between two limits:

c2 R
EmE
ħ
(full box) (empty box) . (797)
G cR

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

We see directly that for sizes R of the order of the Planck length, the two limits coincide;
they both give the Planck mass
¾
M Pl = =  10−8 kg  1019 GeV~c2 .
ħ ħc
(798)
c l Pl G

In other words, for boxes of Planck size, we cannot distinguish a full box from an empty
one. This means that there is no difference between vacuum and matter at Planck scales.
Of course, a similar statement holds for the distinction between vacuum and radiation.
At Planck scales, vacuum and particles cannot be distinguished.

How else can we show that matter and vacuum cannot be

distinguished?
The impossibility to distinguish vacuum from particles is a strong statement. A strong
statement needs additional proof.
Mass can also be measured by probing its inertial aspect, i.e. by colliding the unknown
mass M with known velocity V with a known probe particle of mass m and momentum
p = mv. We then have
M=
∆p
,
Copyright © Christoph Schiller November 1997–May 2006

(799)
∆V
where the differences are taken between the values before and after the collision. The error
δM of such a measurement is simply given by

=
δM δ∆v δm δ∆V
+ + . (800)
M ∆v m ∆V

At Planck scales we have δ∆v~∆v  1, because the velocity error is always, like the velo-
cities themselves, of the order of the speed of light. In other words, at Planck scales the
mass measurement error is so large that we cannot determine whether a mass is different

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1104

from zero: vacuum is indistinguishable from matter.

The same conclusion arises if we take light with a wavelength λ as the probe particle.
In this case, expression (799) leads to a mass error

=
δM δ∆λ δ∆V
+ . (801)
M ∆λ ∆V
In order that photon scattering can probe Planck dimensions, we need a wavelength of
the order of the Planck value; but in this case the first term is approximately unity. Again Dvipsbugw
we find that at Planck scales the energy indeterminacy is always of the same size as the
energy value to be measured. Measurements cannot distinguish between vanishing and
non-vanishing mass M at Planck scales. In fact, this result appears for all methods of mass
Challenge 1503 ny measurement that can be imagined. At Planck scales, matter cannot be distinguished
from vacuum.
Incidentally, the same arguments are valid if instead of the mass of matter we meas-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

ure the energy of radiation. In other words, it is also impossible to distinguish radiation
from vacuum at high energies. In short, no type of particle differs from vacuum at high
energies.
The indistinguishability of particles and vacuum, together with the existence of min-
imum space-time intervals, suggest that space, time, radiation and matter are macro-
scopic approximations of an underlying, common and discrete structure. This structure
Ref. 1124 is often called quantum geometry. How do the common constituents of the two aspects
of nature look like? We will provide several arguments showing that these constituents
are extended and fluctuating.
Also, die Aufgabe ist nicht zu sehen, was noch

“ nie jemand gesehen hat, sondern über dasjenige

was jeder schon gesehen hat zu denken was
noch nie jemand gedacht hat.*
Erwin Schrödinger

Argument 1: The size and shape of elementary particles

Size is the length of vacuum taken by an object. This definition comes natural in everyday
”
life, quantum theory and relativity. However, approaching Planck energy, vacuum and
matter cannot be distinguished: it is impossible to define the boundary between the two,
and thus it is impossible to define the size of an object. As a consequence, every object
becomes as extended as the vacuum. Care is therefore required.
Copyright © Christoph Schiller November 1997–May 2006

What happens if Planck energy is approached, advancing step by step to higher energy?
Every measurement requires comparison with a standard. A standard is made of matter
and comparison is performed using radiation (see Figure 387). Thus any measurement
requires to distinguish between matter, radiation and space-time. However, the distinc-
tion between matter and radiation is possible only up to the (grand) unification energy,
which is about an 800th of the Planck energy. Measurements do not allow us to prove that
particles are point-like. Let us take a step back and check whether measurements allow
us to say whether particles can at least be contained inside small spheres.

* ‘Our task is not to see what nobody has ever seen, but to think what nobody has ever thought about that
which everybody has seen already.’

Dvipsbugw
argument 1: the size and shape of elementary particles 1105

object

comparison

standard

F I G U R E 387 Measurement requires matter

and radiation Dvipsbugw

Do boxes exist?

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The first and simplest way to determine the size of a compact particle such as a sphere or
something akin to it, is to measure the size of a box it fits in. To be sure that the particle fits
inside, we first of all must be sure that the box is tight. This is done by checking whether
something, such as matter or radiation, can leave the box. However, nature does not
provide a way to ensure that a box has no holes. Potential hills cannot get higher than
the maximum energy, namely the Planck energy. The tunnelling effect cannot be ruled
out. In short, there is no way to make fully tight boxes.
In addition, already at the unification energy there is no way to distinguish between
the box and the object enclosed in it, as all particles can be transformed from any one
type into any other.
Let us cross-check this result. In everyday life, we call particles ‘small’ because they can
be enclosed. Enclosure is possible because in daily life walls are impenetrable. However,
walls are impenetrable for matter particles only up to roughly 10 MeV and for photons
only up to 10 keV. In fact, boxes do not even exist at medium energies. We thus cannot
extend the idea of ‘box’ to high energies at all.
In summary, we cannot state that particles are compact or of finite size using boxes.
We need to try other methods.

Can the Greeks help? – The limits of knives

Copyright © Christoph Schiller November 1997–May 2006

The Greeks deduced the existence of atoms by noting that division of matter must end. In
contrast, whenever we think of space (or space-time) as made of points, we assume that
it can be subdivided without end. Zeno noted this already long time ago and strongly cri-
ticized this assumption. He was right: at Planck energy, infinite subdivision is impossible.
Any attempt to divide space stops at Planck dimensions at the latest. The process of cut-
ting is the insertion of a wall. Knifes are limited in the same ways that walls are. The limits
of walls imply limits to size determination.
In particular, the limits to walls and knives imply that at Planck energies, a cut does not
necessarily lead to two separate parts. One cannot state that the two parts have been really
separated; a thin connection between can never be excluded. In short, cutting objects at
Planck scales does not prove compactness.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1106

Are cross-sections finite?

Particles are not point like. Particles are not compact. Are particles are at least of finite
size?
To determine particle size, we can take try to determine their departure from point-
likeness. At high energy, detecting this departure requires scattering. For example, we can
suspend the particle in some trap and then shoot some probe at it. What happens in a
scattering experiment at high energies? The question has been studied already by Leonard
Ref. 1125 Susskind and his coworkers. When shooting at the particle with a high energy probe, the Dvipsbugw
scattering process is characterized by an interaction time. Extremely short interaction
times imply sensitivity to the size and shape fluctuations due to the quantum of action.
An extremely short interaction time provides a cut-off for high energy shape and size
fluctuations and thus determines the measured size. As a result, the size measured for any
microscopic, but extended object increases when the probe energy is increased towards
the Planck value.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In summary, even though at experimentally achievable energies the size is always smal-
ler than measurable, when approaching the Planck energy, size increases above all bounds.
As a result, at high energies we cannot give a limit to sizes of particles! In other words,
since particles are not point-like at everyday energy, at Planck energies they are enorm-
ous: particles are extended.
That is quite a deduction. Right at the start of our mountain ascent, we distinguished
objects from their environment. Objects are by definition localized, bounded and com-
pact. All objects have a boundary, i.e. a surface which itself does not have a boundary.
Objects are also bounded in abstract ways; boundedness is also a property of the symmet-
ries of any object, such as its gauge group. In contrast, the environment is not localized,
but extended and unbounded. All these basic assumptions disappear at Planck scales. At
Planck energy, it is impossible to determine whether something is bounded or compact.
Compactness and localisation are only approximate properties; they are not correct at
high energies. The idea of a point particle is a low energy, approximated concept.
Particles at Planck scales are as extended as the vacuum. Let us perform another check
of this conclusion.

Can one take a photograph of a point?

“
Καιρὸν γνῶθι.*
Pittacus.

”
Copyright © Christoph Schiller November 1997–May 2006

Humans or any other types of observers can only observe a part of the world with finite
resolution in time and in space. In this, humans resemble a film camera. The highest pos-
Ref. 1126 sible resolution has (almost) been discovered in 1899: the Planck time and the Planck
length. No human, no film camera and no measurement apparatus can measure space or
Ref. 1122, Ref. 1127 time intervals smaller than the Planck values. But what would happen if we took photo-
graphs with shutter times approaching the Planck time?
Imagine that you have the world’s best shutter and that you are taking photographs
at increasingly shorter times. Table 79 gives a rough overview of the possibilities. For
check the greek * ‘Recognize the right moment.’ also rendered as: ‘Recognize thine opportunity.’ Pittacus (Πιττακος) of
Mitylene, (c. 650–570 BCE ) was the Lesbian tyrant that was also one of the ancient seven sages.

Dvipsbugw
argument 1: the size and shape of elementary particles 1107

TA B L E 79 Effects of various camera shutter times on photographs

D u r at i o n B lu r O b s e r va t i o n p o s s i b i l i t i e s

1h high Ability to see faint quasars at night if motion is compensated

1s high Everyday motion is completely blurred
20 ms lower Interruption by eyelids; impossibility to see small changes
10 ms lower Effective eye/brain shutter time; impossibility to see tennis ball when
hitting it Dvipsbugw
0.25 ms lower Shortest commercial photographic camera shutter time; allows to
photograph fast cars
1 µs very low Ability to photograph flying bullets; requires strong flashlight
c. 10 ps lowest Study of molecular processes; ability to photograph flying light
pulses; requires laser light to get sufficient illumination
10 fs higher Light photography becomes impossible due to wave effects

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

100 zs high X-ray photography becomes impossible; only γ-ray imaging is left
over
shorter times very high Photographs get darker as illumination gets dimmer; gravitational
effects start playing a role
10−41 s highest imaging makes no sense

shorter and shorter shutter times, photographs get darker and darker. Once the shutter
time reaches the oscillation time of light, strange things happen: light has no chance to
pass undisturbed; signal and noise become impossible to distinguish; in addition, the
moving shutter will produce colour shifts. In contrast to our intuition, the picture would
get blurred at extremely short shutter times. Photography is not only impossible at long
but also at short shutter times.
The difficulty of taking photographs is independent of the used wavelength. The limits
move, but do not disappear. A short shutter time τ does not allow photons of energy lower
than ħ~τ to pass undisturbed. The blur is small when shutter times are those of everyday
life, but increases when shutter times are shortened towards Planck times. As a result,
there is no way to detect or confirm the existence of point objects by taking pictures.
Points in space, as well as instants of time, are imagined concepts; they do not allow a
precise description of nature.
Copyright © Christoph Schiller November 1997–May 2006

At Planck shutter times, only signals with Planck energy can pass through the shutter.
Since at these energies matter cannot be distinguished from radiation or from empty
space, all objects, light and vacuum look the same. As a result, it becomes impossible to
say how nature looks at shortest times.
But the situation is much worse: a Planck shutter does not exist at all, as it would
need to be as small as a Planck length. A camera using it could not be built, as lenses
do not work at this energy. Not even a camera obscura – without any lens – would work,
as diffraction effects would make image production impossible. In other words, the idea
that at short shutter times, a photograph of nature shows a frozen version of everyday life,
like a stopped film, is completely wrong! Zeno criticized this image already in ancient
Greece, in his discussions about motion, though not so clearly as we can do now. Indeed,

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1108

at a single instant of time nature is not frozen at all.* At short times, nature is blurred and
fuzzy. This is also the case for point particles.
In summary, whatever the intrinsic shape of what we call a ‘point’ might be, we know
that being always blurred, it is first of all a cloud. Whatever method to photograph of a
point particle is used, it always shows an extended entity. Let us study its shape in more
detail.

What is the shape of an electron?

Dvipsbugw
Given that particles are not point-like, they have a shape. How can we determine it? Every-
day object shape determination is performed by touching the object from all sides. This
works with plants, people or machines. It works with molecules, such as water molecules.
We can put them (almost) at rest, e.g. in ice, and then scatter small particles off them. Scat-
tering is just a higher energy version of touching. However, scattering cannot determine
shapes smaller than the wavelength of the used probes. To determine a size as small as

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

that of an electron, we need the highest energies available. But we already know what
happens when approaching Planck scales: the shape of a particle becomes the shape of
all the space surrounding it. Shape cannot be determined in this way.
Another method to determine the shape is to build a tight box filled of wax around
the system under investigation. We let the wax cool and and observe the hollow part.
However, near Planck energies boxes do not exist. We are unable to determine the shape
in this way.
A third way to measure shapes is cutting something into pieces and then study the
pieces. But cutting is just a low-energy version of a scattering process. It does not work at
high energies. Since the term ‘atom’ means ‘uncuttable’ or ‘indivisible’, we have just found
out that neither atoms nor indivisible particles can exist. Indeed, there is no way to prove
this property. Our everyday intuition leads us completely astray at Planck energies.
A fourth way to measure shapes could appear by distinguishing transverse and lon-
gitudinal shape, with respect to the direction of motion. However, for transverse shape
we get the same issues as for scattering; transverse shape diverges for high energy. To de-
termine longitudinal shape, we need at least two infinitely high potential walls. Again, we
already know that this is impossible.
A further, indirect way of measuring shapes is the measurement of the moment of
inertia. A finite moment of inertia means a compact, finite shape. However, when the
measurement energy is increased, rotation, linear motion and exchange become mixed
Copyright © Christoph Schiller November 1997–May 2006

Ref. 1122 up. We do not get meaningful results.

Still another way to determine shapes is to measure the entropy of a collection of
particles we want to study. This allows to determine the dimensionality and the number
of internal degrees of freedom. At high energies, a collection of electrons would become
a black hole. We study the issue separately below, but again we find no new information.
Are these arguments water-tight? We assumed three dimensions at all scales, and as-
sumed that the shape of the particle itself is fixed. Maybe these assumptions are not valid
at Planck scales. Let us check the alternatives. We have already shown above that due

* In fact, a shutter does not exist even at medium energy, as shutters, like walls, stop existing at around
10 MeV.

Dvipsbugw
 argument 1: the size and shape of elementary particles 1109

to the fundamental measurement limits, the dimensionality of space-time cannot be de-

Ref. 1122 termined at Planck scales. Even if we could build perfect three-dimensional boxes, holes
could remain in other dimensions. It does not take long to see that all the arguments
against compactness work even if space-time has additional dimensions.

Is the shape of an electron fixed?

Only an object composed of localized entities, such as a house or a molecule, can have
a fixed shape. The smaller a system gets, the more quantum fluctuations play a role. Any Dvipsbugw
entity with a finite size, thus also an elementary particle, cannot have a fixed shape. Every
Gedanken experiment leading to finite shape also implies that the shape itself fluctuates.
But we can say more.
The distinction between particles and environment resides in the idea that particles
have intrinsic properties. In fact, all intrinsic properties, such as spin, mass, charge and
parity, are localized. But we saw that no intrinsic property is measurable or definable at

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Planck scales. It is impossible to distinguish particles from the environment, as we know
already. In addition, at Planck energy particles have all properties the environment also
has. In particular, particles are extended.
In short, we cannot prove by experiments that at Planck energy elementary particles
are finite in size in all directions. In fact, all experiments one can think of are compatible
with extended particles, with ‘infinite’ size. We can also say that particles have tails. More
precisely, a particle always reaches the borders of the region of space-time under explor-
ation.
Not only are particles extended; in addition, their shape cannot be determined by the
methods just explored. The only possibility left over is also suggested by quantum theory:
The shape of particles is fluctuating.
We note that for radiation particles we reach the same conclusions. The box argument
shows that also radiation particles are extended and fluctuating.
In our enthusiasm we have also settled an important detail about elementary particles.
We saw above that any particle which is smaller than its own Compton wavelength must
Page 761 be elementary. If it were composite, there would be a lighter component inside it; this
lighter particle would have a larger Compton wavelength than the composite particle.
This is impossible, since the size of a composite particle must be larger than the Compton
wavelength of its components.*
However, an elementary particle can have constituents, provided that they are not com-
Copyright © Christoph Schiller November 1997–May 2006

pact. The difficulties of compact constituents were already described by Sakharov in the
Ref. 1128 1960s. But if the constituents are extended, they do not fall under the argument, as exten-
ded entities have no localized mass. As a result, a flying arrow, Zeno’s famous example,
cannot be said to be at a given position at a given time, if it is made of extended entities.
Shortening the observation time towards the Planck time makes an arrow disappear in
the same cloud that also makes up space-time.**

* Examples are the neutron, positronium, or the atoms. Note that the argument does not change when the
elementary particle itself is unstable, such as the muon. Note also that the possibility that all components be
heavier than the composite, which would avoid this argument, does not seem to lead to satisfying physical
properties; e.g. it leads to intrinsically unstable composites.
Ref. 1129 ** Thus at Planck scales there is no quantum Zeno effect any more.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1110

In summary, only the idea of points leads to problems at Planck scales. If space-time
and matter are imagined to be made, at Planck scales, of extended and fluctuating entit-
ies, all problems disappear. We note directly that for extended entities the requirement
Ref. 1122 of a non-local description is realized. Similarly, the entities being fluctuating, the require-
ment of a statistical description of vacuum is realized. Finally, the argument forbidding
composition of elementary particles is circumvented, as extended entities have no clearly
defined mass. Thus the concept of Compton wavelength cannot be defined or applied.
Elementary particles can thus have constituents if they are extended. But if the compon-
Challenge 1504 e ents are extended, how can compact ‘point’ particles be formed with them? A few options Dvipsbugw
will be studied shortly.

Argument 2: The shape of points in vacuum

Thus, since there is an impossibility that [finite]

“ quantities are built from contacts and points, it

is necessary that there be indivisible material

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

elements and [finite] quantities.
Aristotle, Of Generation and Corruption.

”
Ref. 1130

We are used to think that empty space is made of spatial points. Let us check whether
this is true at high energy. At Planck scales no measurement can give zero length, zero
mass, zero area or zero volume. There is no way to state that something in nature is a
point without contradicting experimental results. In addition, the idea of a point is an
extrapolation of what is found in small empty boxes getting smaller and smaller. However,
we just saw that at high energies small boxes cannot be said to be empty. In fact, boxes do
not exist at all, as they are never tight and do not have impenetrable walls at high energies.
Also the idea of a point as a continuous subdivision of empty space is untenable. At
small distances, space cannot be subdivided, as division requires some sort of dividing
wall, which does not exist.
Even the idea of repeatedly putting a point between two others cannot be applied. At
high energy, it is impossible to say whether a point is exactly on the line connecting the
outer two points; and near Planck energy, there is no way to find a point between them
at all. In fact, the term ‘in between’ makes no sense at Planck scales.
We thus find that space points do not exist, in the same way that point particles do not
exist. But there is more; space cannot be made of points for additional reasons. Common
sense tells us that points need to be kept apart somehow, in order to form space. Indeed,
Copyright © Christoph Schiller November 1997–May 2006

mathematicians have a strong argument stating why physical space cannot be made of
mathematical points: the properties of mathematical spaces described by the Banach–
Ref. 1122 Tarski paradox are quite different from that of the physical vacuum. The Banach–Tarski
paradox states states that a sphere made of mathematical points can be cut into 5 pieces
which can be reassembled into two spheres each of the same volume as the original sphere.
Mathematically, volume makes no sense. Physically speaking, we can say that the concept
of volume does not exist for continuous space; it is only definable if an intrinsic length
exists. This is the case for matter; it must also be the case for vacuum. But any concept
with an intrinsic length, also the vacuum, must be described by one or several extended
components.* In summary, we need extended entities to build up space-time!
Ref. 1131 * Imagining the vacuum as a collection of entities with Planck size in all directions, such as spheres, would

Dvipsbugw
 argument 2: the shape of points in vacuum 1111

Not only is it impossible to generate a volume with mathematical points; it is also im-
possible to generate exactly three physical dimensions with mathematical points. Math-
ematics shows that any compact one-dimensional set has as many points as any compact
three-dimensional set. And the same is true for any other pair of dimension values. To
build up the physical three-dimensional vacuum we need entities which organize their
neighbourhood. This cannot be done with purely mathematical points. The fundamental
entities must possess some sort of bond forming ability. Bonds are needed to construct or
fill three dimensions instead of any other number. Bonds require extended entities. But
also a collection of tangled entities extending to the maximum scale of the region under Dvipsbugw
consideration would work perfectly. Of course the precise shape of the fundamental en-
tities is not known at this point in time. In any case we again find that any constituents
of physical three-dimensional space must be extended.
In summary, we need extension to define dimensionality and to define volume. We are
not surprised. Above we deduced that the constituents of particles are extended. Since
vacuum is not distinguishable from matter, we expect the constituents of vacuum to be

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

extended as well. Stated simply, if elementary particles are not point-like, then space-time
points cannot be either.

Measuring the void

To check whether space-time constituents are extended, let us perform a few additional
Gedanken experiments. First, let us measure the size of a point of space-time. The clearest
definition of size is through the cross-section. How can we determine the cross-section
of a point? We can determine the cross section of a piece of vacuum and determine the
number of points inside it. From the two determinations we can deduce the cross-section
of a single point. At Planck energies however, we get a simple result: the cross-section of
a volume of empty space is depth independent. At Planck energies, vacuum has a surface,
but no depth. In other words, at Planck energy we can only state that a Planck layer covers
the surface of a volume. We cannot say anything about its interior. One way to picture
the result is to say that space points are long tubes.
Another way to determine the size of a point is to count the points found in a given
volume of space-time. One approach is to count the possible positions of a point particle
in a volume. However, point particles are extended at Planck energies and indistinguish-
able from vacuum. At Planck energy, the number of points is given by surface area of the
volume divided by the Planck area. Again, the surface dependence suggests that particles
Copyright © Christoph Schiller November 1997–May 2006

are long tubes.

Another approach to count the number of points in a volume is to fill a piece of vacuum
with point particles.

What is the maximum number of particles that fits inside a piece

of vacuum?
The maximum mass that fits into a piece of vacuum is a black hole. But also in this case, the
maximum mass depends only on the surface of the given vacuum piece. The maximum

avoid the Banach–Tarski paradox, but would not allow to deduce the numbers of dimensions of space and
Challenge 1505 n time. It would also contradict all other results of this section. We therefore do not explore it further.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1112

mass increases less rapidly than the volume. In other words, the number of points in a
volume is only proportional to the surface area of that volume. There is only one solution:
vacuum must be made of extended entities crossing the whole volume, independently of
the shape of the volume.
Page 223 Two thousand years ago, the Greek argued that matter must be made of particles be-
Ref. 1132 cause salt can be dissolved in water and because fish can swim through water. Now that
we know more about Planck scales, we have to reconsider the argument. Like fish through
water, particles can move through vacuum; but since vacuum has no bounds and since
it cannot be distinguished from matter, vacuum cannot be made of particles. However, Dvipsbugw
there is another possibility that allows for motion of particles through vacuum: both va-
cuum and particles can be made of a web of extended entities. Let us study this option in
more detail.

Argument 3: The l arge, the small and their connection

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

I could be bounded in a nutshell and count

“ myself a king of infinite space, were it not that I

have bad dreams.
William Shakespeare, Hamlet.

If two observables cannot be distinguished, there is a symmetry transformation connect-

ing them. For example, when switching observation frame, an electric field may change
into a magnetic one. A symmetry transformation means that we can change the view-
”
point (i.e. the frame of observation) with the consequence that the same observation is
described by one quantity from one viewpoint and by the other quantity from the other
viewpoint.
When measuring a length at Planck scales it is impossible to say whether we are meas-
uring the length of a piece of vacuum, the Compton wavelength of a body, or the Schwarz-
schild diameter of a body. For example, the maximum size for an elementary object is its
Compton wavelength. The minimum size for an elementary object is its Schwarzschild
radius. The actual size of an elementary object is somewhere in between. If we want to
measure the size precisely, we have to go to Planck energy: but then all these quantities are
the same. In other words, at Planck scales, there is a symmetry transformation between
Compton wavelength and Schwarzschild radius. In short, at Planck scales there is a sym-
metry between mass and inverse mass.
As a further consequence, at Planck scales there is a symmetry between size and in-
Copyright © Christoph Schiller November 1997–May 2006

verse size. Matter–vacuum indistinguishability means that there is a symmetry between

Ref. 1135 length and inverse length at Planck energies. This symmetry is called space-time duality
or T-duality in the literature of superstrings.* Space-time duality is a symmetry between
situations at scale n l Pl and at scale f l Pl ~n, or, in other words, between R and ( f l Pl )2 ~R,
where the experimental number f has a value somewhere between 1 and 1000.
Duality is a genuine non-perturbative effect. It does not exist at low energy, since dual-
ity automatically also relates energy E and energy E Pl 2
~E = ħc 3 ~GE, i.e. it relates energies
below and above Planck scale. Duality is a quantum symmetry. It does not exist in every-
day life, as Planck’s constant appears in its definition. In addition, it is a general relativistic
* There is also an S-duality, which connects large and small coupling constants, and a U-duality, which is
the combination of S- and T-duality.

Dvipsbugw
 argument 3: the l arge, the small and their connection 1113

effect, as it includes the gravitational constant and the speed of light. Let us study duality
in more detail.

Small is large?
[Zeno of Elea maintained:] If the existing are

“ many, it is necessary that they are at the same

time small and large, so small to have no size,
and so large to be without limits.
Simplicius, Commentary on the Physics of Dvipsbugw
Aristotle, 140, 34.

”
Ref. 1133

To explore the consequences of duality, we can compare it to the 2π rotation symmetry

in everyday life. Every object in daily life is symmetrical under a full rotation. For the
rotation of an observer, angles make sense only as long as they are smaller than 2π. If a
rotating observer would insist on distinguishing angles of 0, 2π, 4π etc., he would get a
new copy of the universe at each full turn.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2
Similarly, in nature, scales R and l Pl ~R cannot be distinguished. Lengths make no sense
when they are smaller than l Pl . If however, we insist on using even smaller values and
insist on distinguishing them from large ones, we get a new copy of the universe at those
small scales. Such an insistence is part of the standard continuum description of motion,
where it is assumed that space and time are described by the real numbers, which are
defined for arbitrary small intervals. Whenever the (approximate) continuum description
with infinite extension is used, the R  l Pl 2
~R symmetry pops up.
Duality implies that diffeomorphism invariance is only valid at medium scales, not at
extremal ones. At extremal scales, quantum theory has to be taken into account in the
proper manner. We do not know yet how to do this.
Space-time duality means that introducing lengths smaller than the Planck length (like
when one defines space points, which have size zero) means at the same time introducing
things with very large (‘infinite’) value. Space-time duality means that for every small
enough sphere the inside equals outside.
Duality means that if a system has a small dimension, it also has a large one. And vice
versa. There are thus no small objects in nature. As a result, space-time duality is consistent
with the idea that the basic entities are extended.

Unification and total symmetry

Copyright © Christoph Schiller November 1997–May 2006

So far, we have shown that at Planck energy, time and length cannot be distinguished.
Duality has shown that mass and inverse mass cannot be distinguished. As a consequence,
length, time and mass cannot be distinguished from each other. Since every observable is
a combination of length, mass and time, space-time duality means that there is a symmetry
between all observables. We call it the total symmetry.*

* A symmetry between size and Schwarzschild radius, i.e. a symmetry between length and mass, will lead to
general relativity. Additionally, at Planck energy there is a symmetry between size and Compton wavelength.
In other words, there is a symmetry between length and 1/mass. It means that there is a symmetry between
coordinates and wave functions. Note that this is a symmetry between states and observables. It leads to
quantum theory.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1114

Total symmetry implies that there are many types of specific dualities, one for each pair
of quantities under investigation. Indeed, in string theory, the number of duality types dis-
Page 556 covered is increasing every year. It includes, among others, the famous electric–magnetic
Ref. 1134 duality we first encountered in the chapter on electrodynamics, coupling constant duality,
surface–volume duality, space-time duality and many more. All this confirms that there
is an enormous symmetry at Planck scales. Similar statements are also well-known right
Ref. 1135 from the beginning of string theory.
Most importantly, total symmetry implies that gravity can be seen as equivalent to all
other forces. Space-time duality shows that unification is possible. Physicist have always Dvipsbugw
dreamt about unification. Duality tells us that this dream can indeed be realized.
It may seem that total symmetry is in complete contrast with what was said in the
previous section, where we argued that all symmetries are lost at Planck scales. Which
result is correct? Obviously, both of them are.
At Planck scales, all low energy symmetries are indeed lost. In fact, all symmetries
that imply a fixed energy are lost. Duality and its generalizations however, combine both

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

small and large dimensions, or large and small energies. Most symmetries of usual phys-
ics, such as gauge, permutation and space-time symmetries, are valid at each fixed energy
separately. But nature is not made this way. The precise description of nature requires to
take into consideration large and small energies at the same time. In everyday life, we do
not do that. Everyday life is a low and fixed energy approximation of nature. For most of
the twentieth century, physicists aimed to reach higher and higher energies. We believed
that precision increases with increasing energy. But when we combine quantum theory
and gravity we are forced to change this approach; to achieve high precision, we must
take both high and low energy into account at the same time.*
The large differences in phenomena at low and high energies are the main reason why
unification is so difficult. So far, we were used to divide nature along the energy scale.
We thought about high energy physics, atomic physics, chemistry, biology, etc. The differ-
ences between these sciences is the energy of the processes involved. But now we are not
allowed to think in this way any more. We have to take all energies into account at the
same time. That is not easy, but we do not have to despair. Important conceptual progress
has been achieved in the last decade of the twentieth century. In particular, we now know
that we need only one single concept for all things which can be measured. Since there is
only one concept, there are many ways to study it. We can start from any (low-energy)
concept in physics and explore how it looks and behaves when we approach Planck scales.
In the present section, we are looking at the concept of point. Obviously, the conclusions
Copyright © Christoph Schiller November 1997–May 2006

must be the same, independently of the concept we start with, be it electric field, spin, or
Challenge 1506 d any other. Such studies thus provide a check for the results in this section.
Unification thus implies to think using duality and using concepts which follow from
it. In particular, we need to understand what exactly happens to duality when we restrict
Challenge 1507 e ourselves to low energy only, as we do in everyday life. This question is left for the next
section.

* Renormalization energy does connect different energies, but not in the correct way; in particular, it does
not include duality.

Dvipsbugw
 argument 4: d oes nature have parts? 1115

Argument 4: D oes nature have parts?

Pluralitas non est ponenda sine necessitate.*

“ William of Occam

Another argument, independent of the ones given above, underlines the correctness of a
model of nature made of extended entities. Let us take a little broader view. Any concept
”
for which we can distinguish parts is described by a set. We usually describe nature as
a set of objects, positions, instants, etc. The most famous set description of nature is the Dvipsbugw
oldest known, given by Democritus: ‘The world is made of indivisible particles and void.’
This description was extremely successful in the past; there were no discrepancies with
observations yet. However, after 2500 years, the conceptual difficulties of this approach
are obvious.
We now know that Democritus was wrong, first of all, because vacuum and matter
cannot be distinguished at Planck scales. Thus the word ‘and’ in his sentence is already
mistaken. Secondly, due to the existence of minimal scales, the void cannot be made of

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

‘points,’ as we usually assume nowadays. Thirdly, the description fails because particles
are not compact objects. Fourth, the total symmetry implies that we cannot distinguish
parts in nature; nothing can really be distinguished from anything else with complete
precision, and thus the particles or points in space making up the naive model of void
cannot exist.
In summary, quantum theory and general relativity together show that in nature, all
differences are only approximate. Nothing can really be distinguished from anything else
with complete precision. In other words, there is no way to define a ‘part’ of nature, neither
for matter, nor for space, nor for time, nor for radiation. Nature cannot be a set.
The conclusion that nature is not a set does not come as a surprise. We have already
encountered another reason to doubt that nature is a set. Whatever definition we use for
the term ‘particle’, Democritus cannot be correct for a purely logical reason. The descrip-
tion he provided is not complete. Every description of nature defining nature as a set of
parts necessarily misses certain aspects. Most importantly, it misses the number of these
parts. In particular, the number of particles and the number of dimensions of space-time
Page 681 must be specified if we describe nature as made from particles and vacuum. Above we
saw that it is rather dangerous to make fun of the famous statement by Arthur Eddington

Ref. 1136 I believe there are 15,747,724,136,275,002,577,605,653,961,181,555,468,044,717,914-

,527, 116,709,366,231,425,076,185,631,031,296 protons in the universe and the same
Copyright © Christoph Schiller November 1997–May 2006

number of electrons.

In fact, practically all physicists share this belief; usually they either pretend to favour
some other number, or worse, they keep the number unspecified. We have seen during

* ‘Multitude should not be introduced without necessity.’ This famous principle is commonly called Occam’s
razor. William of Ockham (b. 1285/1295 Ockham, d. 1349/50 München), or Occam in the common Latin
spelling, was one of the great thinkers of his time. In his famous statement he expresses that only those
concepts which are strictly necessary should be introduced to explain observations. It can be seen as the
requirement to abandon beliefs when talking about nature. In addition, at this stage of our mountain ascent
it has an even more direct interpretation.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1116

our walk that in modern physics many specialized sets are used to describe nature. We
have used vector spaces, linear spaces, topological spaces and Hilbert spaces. But very
consistently we refrained, like all physicists, from asking about the origin of their sizes
(mathematically speaking of their dimensionality or their cardinality). In fact, it is equally
unsatisfying to say that the universe contains some specific number of atoms as it is to say
that space-time is made of point-like events arranged in 3+1 dimensions. Both statements
are about set sizes in the widest sense. In a complete, i.e. in a unified description of nature
the number of smallest particles and the number of space-time points must not be added
to the description, but must result from the description. Only in this case is unification Dvipsbugw
achieved.
Requiring a complete explanation of nature leads to a simple consequence. Any part
of nature is by definition smaller than the whole of nature and different from other parts.
As a result, any description of nature by a set cannot possibly yield the number of particles
nor space-time dimensionality. As long as we insist in using space-time or Hilbert spaces
for the description of nature, we cannot understand the number of dimensions or the

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

number of particles.
Well, that is not too bad, as we know already that nature is not made of parts. We know
that parts are only approximate concepts. In short, if nature were made of parts, it could
not be a unity, or a ‘one.’ If however, nature is a unity, a one, it cannot have parts.* Nature
cannot be separable exactly. It cannot be made of particles.
To sum up, nature cannot be a set. Sets are lists of distinguishable elements. When
general relativity and quantum theory are unified, nature shows no elements: nature stops
being a set at Planck scales. The result confirms and clarifies a discussion we have started in
classical physics. There we had discovered that matter objects were defined using space
and time, and that space and time were defined using objects. Including the results of
quantum theory, this implies that in modern physics particles are defined with the help
of the vacuum and the vacuum with particles. That is not a good idea. We have just seen
that since the two concepts are not distinguishable from each other, we cannot define
them with each other. Everything is the same; in fact, there is no ‘every’ and no ‘thing.’
Since nature is not a set, the circular reasoning is dissolved.
Space-time duality also implies that space is not a set. Duality implies that events can-
not be distinguished from each other. They thus do not form elements of some space.
Ref. 1138 Phil Gibbs has given the name event symmetry to this property of nature. This thought-
provoking term, even though still containing the term ‘event’, underlines that it is im-
possible to use a set to describe space-time.
Copyright © Christoph Schiller November 1997–May 2006

In summary, nature cannot be made of vacuum and particles. That is bizarre. People
propagating this idea have been persecuted for 2000 years. This happened to the atomists
from Democritus to Galileo. Were their battles it all in vain? Let us continue to clarify our
thoughts.

* As a curiosity, practically the same discussion can already be found, in Plato’s Parmenides, written in the
Ref. 1137 fourth century bce. There, Plato musically ponders different arguments on whether nature is or can be a
unity or a multiplicity, i.e. a set. It seems that the text is based on the real visit by Parmenides and Zeno in
Athens, where they had arrived from their home city Elea, which lies near Naples. Plato does not reach a
conclusion. Modern physics however, does.

Dvipsbugw
 argument 4: d oes nature have parts? 1117

Does the universe contain anything?

Stating that the universe contains something implies that we are able to distinguish
the universe from its contents. However, we now know that precise distinctions are im-
possible. If nature is not made of parts, it is wrong to say that the universe contains some-
thing.
Let us go further. As nothing can be distinguished, we need a description of nature
which allows to state that at Planck energies nothing can be distinguished from anything
else. For example, it must be impossible to distinguish particles from each other or from Dvipsbugw
the vacuum. There is only one solution: everything, or at least what we call ‘everything’ in
everyday life, must be made of the same single entity. All particles are made of one same
‘piece.’ Every point in space, every event, every particle and every instant of time must be
made of the same single entity.

An amoeba

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

A theory of nothing describing everything is

“ ” better than a theory of everything describing

nothing.

We found that parts are approximate concepts. The parts of nature are not strictly smaller
than nature itself. As a result, any ‘part’ must be extended. Let us try to extract more
information about the constituents of nature.
The search for a unified theory is the search for a description in which all concepts
appearing are only approximately parts of the whole. Thus we need an entity Ω, describing
nature, which is not a set but which can be approximated by one. This is unusual. We all
are convinced very early in our life that we are a part of nature. Our senses provide us
with this information. We are not used to think otherwise. But now we have to.
Let us eliminate straight away a few options for Ω. One concept without parts is the
empty set. Perhaps we need to construct a description of nature from the empty set? We
Page 651 could be inspired by the usual construction of the natural numbers from the empty set.
However, the empty set makes only sense as the opposite of some full set. That is not the
case here. The empty set is not a candidate for Ω.
Another possibility to define approximate parts is to construct them from multiple
copies of Ω. But in this way we would introduce a new set through the back door. In
addition, new concepts defined in this way would not be approximate.
We need to be more imaginative. How can we describe a whole which has no parts,
Copyright © Christoph Schiller November 1997–May 2006

but which has parts approximately? Let us recapitulate. The world must be described by
a single entity, sharing all properties of the world, but which can be approximated into a
set of parts. For example, the approximation should yield a set of space points and a set of
particles. We also saw that whenever we look at any ‘part’ of nature without any approxim-
ation, we should not be able to distinguish it from the whole world. In other words, com-
posed entities are not always larger than constituents. On the other hand, composed en-
tities must usually appear to be larger than their constituents. For example, space ‘points’
or ‘point’ particles are tiny, even though they are only approximations. Which concept
without boundaries can be at their origin? Using usual concepts the world is everywhere
at the same time; if nature is to be described by a single constituent, this entity must be

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1118

extended.
The entity has to be a single one, but it must seem to be multiple, i.e. it has to be multiple
approximately, as nature shows multiple aspects. The entity must be something folded. It
must be possible to count the folds, but only approximately. (An analogy is the question
of how many tracks are found on an LP or a CD; depending on the point of view, local or
global, one gets different answers.) Counting folds would correspond to a length meas-
urement.
The simplest model would be the use of a single entity which is extended, fluctuating,
going to infinity and allowing approximate localization, thus allowing approximate defin- Dvipsbugw
ition of parts and points.* In more vivid imagery, nature could be described by some
deformable, folded and tangled up entity: a giant, knotted amoeba. An amoeba slides
between the fingers whenever one tries to grab a part of it. A perfect amoeba flows around
any knife trying to cut it. The only way to hold it would be to grab it in its entirety. How-
ever, for an actor himself made of amoeba strands this is impossible. He can only grab
it approximately, by catching part of it and approximately blocking it, e.g. using a small

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

hole so that the escape takes a long time.
To sum up, nature is modelled by an entity which is a single unity (to eliminate distin-
guishability), extended (to eliminate localizability) and fluctuating (to ensure approxim-
ate continuity). A far-reaching, fluctuating fold, like an amoeba. The tangled branches of
the amoeba allow a definition of length via counting of the folds. In this way, discreteness
of space, time and particles could also be realized; the quantization of space-time, matter
and radiation thus follows. Any flexible and deformable entity is also a perfect candidate
for the realization of diffeomorphism invariance, as required by general relativity.
A simple candidate for the extended fold is the image of a fluctuating, flexible tube.
Counting tubes implies to determine distances or areas. The minimum possible count
of one gives the minimum distance, and thus allows us to deduce quantum theory. In
fact, we can use as model any object which has flexibility and a small dimension, such
as a tube, a thin sheet, a ball chain or a woven collection of rings. These options give
the different but probably equivalent models presently explored in simplicial quantum
gravity, in Ashtekar’s variables and in superstrings.

Argument 5: The entropy of bl ack holes

We are still collecting arguments to determine particle shape. For a completely different
way to explore the shape of particles it is useful to study situations where they appear in
Copyright © Christoph Schiller November 1997–May 2006

large numbers. Collections of high numbers of constituents behave differently if they are
point-like or extended. In particular, their entropy is different. Studying large-number en-
tropy thus allows to determine component shape. The best approach is to study situations
in which large numbers of particles are crammed in a small volume. This leads to study
the entropy of black holes. A black hole is a body whose gravity is so strong that even light
cannot escape. Black holes tell us a lot about the fundamental entities of nature. It is eas-
ily deduced from general relativity that any body whose mass m fits inside the so-called
Schwarzschild radius
r S = 2Gm~c 2 (802)
Challenge 1508 ny * Is this the only method to describe nature? Is it possible to find another description, in particular if space
and time are not used as background? The answers are unclear at present.

Dvipsbugw
 argument 5: the entropy of bl ack holes 1119

is a black hole. A black hole can be formed when a whole star collapses under its own
weight. A black hole is thus a macroscopic body with a large number of constituents. For
black holes, like for every macroscopic body, an entropy can be defined. The entropy S of
Ref. 1139, Ref. 1140 a macroscopic black hole was determined by Bekenstein and Hawking and is given by

4πGm 2
S= S=k
k A
2 4
or (803)
l Pl ħc

where k is the Boltzmann constant and A = 4πr S2 is the surface of the black hole horizon.
Dvipsbugw
Ref. 1141 This important result has been derived in many different ways. The various derivations
also confirm that space-time and matter are equivalent, by showing that the entropy value
can be seen both as an entropy of matter and as one of space-time. In the present context,
the two main points of interest are that the entropy is finite, and that it is proportional to
the area of the black hole horizon.
In view of the existence of minimum lengths and times, the finiteness of entropy is not

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

surprising any more. A finite black hole entropy confirms the idea that matter is made of
a finite number of discrete entities per volume. The existence of an entropy also shows
that these entities behave statistically; they fluctuate. In fact, quantum gravity leads to
Ref. 1142 a finite entropy for any object, not only for black holes; Bekenstein has shown that the
entropy of any object is always smaller than the entropy of a (certain type of) black hole
of the same mass.
The entropy of a black hole is also proportional to its horizon area. Why is this the
Ref. 1143 case? This question is the topic of a stream of publications up to this day.* A simple way
to understand the entropy–surface proportionality is to look for other systems in nature
with the property that entropy is proportional to system surface instead of system volume.
In general, the entropy of any collection of one-dimensional flexible objects, such as poly-
Ref. 1145 mer chains, shows this property. Indeed, the expression for the entropy of a polymer chain
made of N monomers, each of length a, whose ends are kept a distance r apart, is given
Ref. 1146 by
3r 2 º
S(r) = k for Na Q Na Q r . (804)
2Na 2
The formula is derived in a few lines from the properties of a random walk on a lattice,
using only two assumptions: the chains are extended, and they have a characteristic in-
ternal length a given by the smallest straight segment. Expression (804) is only valid if
Copyright © Christoph Schiller November 1997–May 2006

º
the polymers are effectively infinite, i.e. if the length Na of the chain and their effective
average size, the elongation a N , are much larger than the radius r of the region of in-
terest; if the chain length is comparable or smaller than the region of interest, one gets
the usual extensive entropy, fulfilling S  r 3 . Thus only flexible extended entities yield a
S  r 2 dependence.
However, there is º a difficulty. From ºthe entropy expression of a black hole we deduce
that the elongation a N is given by a N  l Pl ; thus it is much smaller than the radius
of a general, macroscopic black hole which can have diameters of several kilometres. On

Ref. 1144 * The result can be derived from quantum statistics alone. However, this derivation does not yield the pro-
portionality coefficient.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1120

the other hand, the formula for long entities is only valid when the chains are longer than
the distance r between the end points.
This difficulty disappears once we remember that space near a black hole is strongly
curved. All lengths have to be measured in the same coordinate system. It is well known
that for an outside observer, any object of finite size falling into a black hole seems to cover
the complete horizon for long times (whereas it falls into the hole in its original size for
Ref. 1125 an observer attached to the object). In short, an extended entity can have a proper length
of Planck size but still, when seen by an outside observer, be as long as the horizon of the
black hole in question. We thus find that black holes are made of extended entities. Dvipsbugw
Another viewpoint can confirm the result. Entropy is (proportional to) the number of
yes/no questions needed to know the exact state of the system. This view of black holes
has been introduced by Gerard ’t Hooft. But if a system is defined by its surface, like a
black hole is, its components must be extended.
Finally, imagining black holes as made of extended entities is also consistent with the
so-called no-hair theorem: black holes’ properties do not depend on what material falls

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

into them, as all matter and radiation particles are made of the same extended compon-
ents. The final state only depends on the number of entities and on nothing else. In short,
the entropy of a black hole is consistent with the idea that it is made of a big tangle of
extended entities, fluctuating in shape.

Argument 6: Exchanging space points or particles at

Pl anck scales
We are still collecting arguments for the extension of fundamental entities in nature. Let
us focus on their exchange behaviour. We saw above that points in space have to be elim-
inated in favour of continuous, fluctuating entities common to space, time and matter. Is
such a space ‘point’ or space entity a boson or a fermion? If we exchange two points of
empty space in everyday life, nothing happens. Indeed, quantum field theory is based –
among others – on the relation

[x, y] = x y − yx = 0 (805)

between any two points with coordinates x and y, making them bosons. But at Planck
scale, due to the existence of minimal distances and areas, this relation is at least changed
to
[x, y] = l Pl
2
+ ... . (806)
Copyright © Christoph Schiller November 1997–May 2006

This means that ‘points’ are neither bosons nor fermions.* They have more complex ex-
change properties. In fact, the term on the right hand side will be energy dependent, with
Ref. 1122 an effect increasing towards Planck scales. In particular, we saw that gravity implies that
double exchange does not lead back to the original situation at Planck scales. Entities
following this or similar relations have been studied in mathematics for many decades:
Ref. 1147 braids. In summary, at Planck scales space-time is not made of points, but of braids or
some of their generalizations. Thus quantum theory and general relativity taken together
Ref. 1147 again show that vacuum must be made of extended entities.
* The same reasoning destroys the fermionic or Grassmann coordinates used in supersymmetry.

Dvipsbugw
 argument 7: the meaning of spin 1121

Particles behave in a similar way. We know that at low, everyday energies, particles of
the same type are identical. Experiments sensitive to quantum effects show that there is
no way to distinguish them: any system of several identical particles obeys permutation
symmetry. On the other hand we know that at Planck energy all low-energy symmetries
disappear. We also know that, at Planck energy, permutation cannot be carried out, as it
implies exchanging positions of two particles. At Planck energy, nothing can be distin-
guished from vacuum; thus no two entities can be shown to have identical properties.
Indeed, no two particles can be shown to be indistinguishable, as they cannot even be
shown to be separate. Dvipsbugw
What happens when we slowly approach Planck energy? At everyday energies, per-
mutation symmetry is defined by commutation or anticommutation relations of any two
particle creation operators
a† b†  b† a† = 0 . (807)

At Planck energies this cannot be correct. At those energies, quantum gravity effects ap-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pear and modify the right hand side; they add an energy dependent term that is negligible
at experimentally accessible energies, but which becomes important at Planck energy. We
Ref. 1122 know from our experience with Planck scales that exchanging particles twice cannot lead
back to the original situation, in contrast to everyday life. It is impossible that a double
exchange at Planck energy has no effect, because at planck energy such statements are
impossible. The simplest extension of the commutation relation (807) satisfying the re-
Ref. 1147 quirement that the right side does not vanish is again braid symmetry. Thus Planck scales
suggest that particles are also made of extended entities.

Argument 7: The meaning of spin

In the last argument, we will show that even at everyday energy, the extension of particles
makes sense. Any particle is a part of the universe. A part is something which is differ-
ent from anything else. Being ‘different’ means that exchange has some effect. Distinction
means possibility of exchange. In other words, any part of the universe is described by its
exchange behaviour. Everyday life tells us that exchange can be seen as composed of ro-
tation. In short, distinguishing parts are described by their rotation behaviour. For this
reason, for microscopic particles, exchange behaviour is specified by spin. Spin distin-
guishes particles from vacuum.*
We note that volume does not distinguish vacuum from particles; neither does rest
Copyright © Christoph Schiller November 1997–May 2006

mass or charge: there are particles without rest mass or without charge, such as photons.
The only candidate observables to distinguish particles from vacuum are spin and mo-
mentum. In fact, linear momentum is only a limiting case of angular momentum. We
again find that rotation behaviour is the basic aspect distinguishing particles from vacuum.
If spin is the central property distinguishing particles from vacuum, finding a model
for spin is of central importance. But we do not have to search for long. An well-known
model for spin 1/2 is part of physics folklore. Any belt provides an example, as we dis-

* With a flat (or other) background, it is possible to define a local energy–momentum tensor. Thus particles
can be defined. Without background, this is not possible, and only global quantities can be defined. Without
background, even particles cannot be defined! Therefore, we assume that we have a slowly varying space-
time background in this section.

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1122

flexible bands
in unspecified
number
particle
reaching the
position
border
of space

Dvipsbugw
J=1/2

F I G U R E 388 A possible model for a spin 1/2

particle

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 771 cussed in detail in chapter VI on permutation symmetry. Any localized structure with
any number of long tails attached to it – and reaching the border of the region of space
under consideration – has the same properties as a spin 1/2 particle. It is a famous exercise
to show that such a model, shown in Figure 388, is indeed invariant under 4π rotation
but not under 2π rotations, that two such particles get entangled when exchanged, but
get untangled when exchanged twice. The model has all properties of spin 1/2 particles,
independently of the precise structure of the central region, which remains unknown at
this point. The tail model even has the same problems with highly curved space as real
spin 1~2 particles have. We explore the idea in detail shortly.
The tail model thus confirms that rotation is partial exchange. More interestingly, it
Ref. 1148 shows that rotation implies connection with the border of space-time. Extended particles
can be rotating. Particles can have spin 1~2 provided that they have tails going to the
border of space-time. If the tails do not reach the border, the model does not work. Spin
1~2 thus even seems to require extension. We again reach the conclusion that extended
entities are a good description for particles.

Present research
To understand is to perceive patterns.

“ Isaiah Berlin

”
Copyright © Christoph Schiller November 1997–May 2006

Ref. 1132 The Greek deduced the existence of atoms because fish can swim through water. They
argued that only if water is made of atoms, can a fish find its way through it by pushing
the atoms aside. We can ask a similar question when a particle flies through vacuum: why
are particles able to move through vacuum at all? Vacuum cannot be a fluid or a solid of
small entities, as this would not fix its dimensionality. Only one possibility remains: both
vacuum and particles are made of a web of extended entities.
Describing matter as composed of extended entities is an idea from the 1960s. Describ-
ing nature as composed of ‘infinitely’ extended entities is an idea from the 1980s. Indeed,
in addition to the arguments presented so far, present research provides several other
approaches that arrive at the same conclusion.

Dvipsbugw
 present research 1123

**
Bosonization, the construction of fermions using an infinite number of bosons, is a cent-
Ref. 1149 ral property of modern unification attempts. It also implies coupling duality, and thus the
extension of fundamental constituents.
**
String theory and in particular its generalization to membranes are explicitly based on ex-
Ref. 1150 tended entities, as the name already states. The fundamental entities are indeed assumed
to reach the limits of space-time. Dvipsbugw

**
Ref. 1151 Research into quantum gravity, in particular the study of spin networks and spin foams,
has shown that the vacuum must be thought as a collection of extended entities.
**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

In the 1990s, Dirk Kreimer has shown that high-order QED diagrams are related to knot
Ref. 1152 theory. He thus proved that extension appears through the back door even when electro-
magnetism is described using point particles.
**
A recent discovery in particle physics, ‘holography’, connects the surface and volume of
Ref. 1153 physical systems at high energy. Even if it were not part of string theory, it would still
imply extended entities.

**
Other fundamental nonlocalities in the description of nature, such as wave function
Page 805 collapse, can be seen as the result of extended entities.
**
The start of the twenty-first century has brought forwards a number of new approaches,
such as string net condensation or knotted particle models. All these make use of exten-
ded entities.

Conceptual checks of extension Copyright © Christoph Schiller November 1997–May 2006

Is nature really described by extended entities? The idea is taken for granted by all present
approaches in theoretical physics. How can we be sure about this result? The arguments
Challenge 1509 ny presented above provide several possible checks.
— As Ed Witten likes to say, any unified model of nature must be supersymmetric and
dual. The idea of extended entities would be dead as soon as it is shown not to be
compatible with these requirements.
— Any model of nature must be easily extendible to a model for black holes. If not, it
cannot be correct.
— Showing that the results on quantum gravity, such as the results on the area and volume
quantization, are in contradiction with extended entities would directly invalidate the

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1124

model.
— The same conclusion of extended entities must appear if one starts from any physical
(low-energy) concept – not only from length measurements – and continues to study
how it behaves at Planck scales. If the conclusion were not reached, the idea of exten-
sion would not be consistent and thus incorrect.
— Showing that any conclusion of the idea of extension is in contrast with string theory
or with M-theory would lead to strong doubts.
— Showing that the measurement of length cannot be related to the counting of folds
would invalidate the model. Dvipsbugw
— Finding a single Gedanken experiment invalidating the extended entity idea would
prove it wrong.

Experimental falsification of models based on extended entities

Physics is an experimental science. What kind of data could show that extension is incor-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

rect?
— Observing a single particle in cosmic rays with energy above the Planck energy would
bring this approach to a sudden stop. (The present record is a million times lower.)
— Finding any property of nature not consistent with extended entities would spell the
end of the idea.
— Finding an elementary particle of spin 0 would invalidate the idea. In particular, find-
ing the Higgs boson and showing that it is elementary, i.e. that its size is smaller than
its own Compton wavelength, would invalidate the model.
— Most proposed experimental checks of string theory can also yield information on the
ideas presented. For example, Paul Mende has proposed a number of checks on the
Ref. 1154 motion of extended objects in space-time. He argues that an extended object moves
differently from a mass point; the differences could be noticeable in scattering or dis-
persion of light near masses.
— In July 2002, the Italian physicist Andrea Gregori has made a surprising prediction
valid for any model using extended entities that reach the border of the universe: if
Ref. 1155 particles are extended in this way, their mass should depend on the size of the universe.
Particle masses should thus change with time, especially around the big bang. This
completely new point is still a topic of research.
Incidentally, most of these scenarios would spell the death penalty for almost all present
Copyright © Christoph Schiller November 1997–May 2006

unification attempts.

Possibilities for confirmation of extension models

— The best confirmation would be to find a concrete model for the electron, muon, tau
and for their neutrinos. In addition, a concrete model for photons and gravitons is
needed. With these models, finding a knot-based definition for the electrical charge
and the lepton number would be a big step ahead. All quantum numbers should be
topological quantities deduced from these models and should behave as observed.
— Estimating the coupling constants and comparing them with the experimental values
is of course the main dream of modern theoretical physics.

Dvipsbugw
 present research 1125

Nature's energy scale

Eeveryday EPl

F I G U R E 389 Planck effects make the energy

axis an approximation

Dvipsbugw
— Proving in full detail that extended entities imply exactly three plus one space-time
dimensions is still necessary.
— Estimating the total number of particles in the visible universe would provide the final
check of any extended entity model.
Generally speaking, the only possible confirmations are those from the one-page table of
unexplained properties of nature given in Chapter X. No other confirmations are possible.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Page 959
The ones mentioned here are the main ones.

Curiosities and fun challenges about extension

No problem is so formidable that you can’t walk

“ away from it.

Charles Schultz

Even though this section already provided sufficient food for thought, here is some more.
”
**
If measurements become impossible near Planck energy, we cannot even draw a diagram
Challenge 1510 ny with an energy axis reaching that value. (See Figure 389) Is this conclusion valid in all
cases?

**
Quantum theory implies that even if tight walls would exist, the lid of such a box can
Challenge 1511 n never be tightly shut. Can you provide the argument?
**
Challenge 1512 n Is it correct that a detector able to detect Planck mass particles would be of infinite size?
Copyright © Christoph Schiller November 1997–May 2006

What about a detector to detect a particle moving with Planck energy?

**
Challenge 1513 e Can you provide an argument against the idea of extended entities?*
**
Does duality imply that the cosmic background fluctuations (at the origin of galaxies and
Challenge 1514 ny clusters) are the same as vacuum fluctuations?

* If so, please email it to the author

Dvipsbugw
 xi general rel ativity versus quantum mechanics • the shape of points
1126

**
Does duality imply that a system with two small masses colliding is the same as one with
Challenge 1515 ny two large masses gravitating?
**
It seems that in all arguments so far we assumed and used continuous time, even though
Challenge 1516 d we know it is not. Does this change the conclusions so far?
** Dvipsbugw
Duality also implies that large and small masses are equivalent in some sense. A mass
2
m in a radius r is equivalent to a mass m Pl ~m in a radius l Pl
2
~r. In other words, duality
transforms mass density from ρ to ρ Pl ~ρ. Vacuum and maximum density are equivalent.
2

Vacuum is thus dual to black holes.

**

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Duality implies that initial conditions for the big bang make no sense. Duality again
shows the uselessness of the idea, as minimal distance did before. As duality implies a
symmetry between large and small energies, the big bang itself becomes an unclearly
defined concept.
**
The total symmetry, as well as space-time duality, imply that there is a symmetry between
Challenge 1517 n all values an observable can take. Do you agree?

**
Can supersymmetry be an aspect or special case of total symmetry or is it something
Challenge 1518 n else?
**
Any description is a mapping from nature to mathematics, i.e. from observed differences
(and relations) to thought differences (and relations). How can we do this accurately, if
Challenge 1519 n differences are only approximate? Is this the end of physics?
**
Challenge 1520 d What do extended entities imply for the big-bang?
Copyright © Christoph Schiller November 1997–May 2006

**
Can you show that going to high energies or selecting a Planck size region of space-time
Challenge 1521 d is equivalent to visiting the big-bang?
**
Additionally, one needs a description for the expansion of the universe in terms of exten-
Ref. 1155, Ref. 1156 ded entities. First approaches are being explored; no final conclusions can be drawn yet.
Challenge 1522 ny Can you speculate about the solution?

Dvipsbugw
 present research 1127

An intermediate status report

Wir müssen wissen, wir werden wissen.*

“ David Hilbert, 1930.

Many efforts for unification advance by digging deeper and deeper into details of ”
quantum field theory and general relativity. Here we took the opposite approach: we took
a step back and looked at the general picture. Guided by this idea we found several argu-
ments, all leading to the same conclusion: space-time points and point particles are made Dvipsbugw
of extended entities.
Somehow it seems that the universe is best described by a fluctuating, multi-branched
entity, a crossing between a giant amoeba and a heap of worms. Another analogy is a big
pot of boiling and branched spaghetti. Such an extended model of quantum geometry
is beautiful and simple, and these two criteria are often taken as indication, before any
experimental tests, of the correctness of a description. We toured topics such as the exist-
ence of Planck limits, 3-dimensionality, curvature, renormalization, spin, bosonization,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

the cosmological constant problem, as well as the search for a background free descrip-
Ref. 1157 tion of nature. We will study and test specific details of the model in the next section. All
these tests concern one of only three possible topics: the construction of space-time, the
construction of particles and the construction of the universe. These are the only issues
remaining on our mountain ascent of Motion Mountain. We will discuss them in the next
section. Before we do so, we enjoy two small thoughts.

Sexual preferences in physics

Fluctuating entities can be seen to answer an old and not so serious question. When
Page 223 nature was defined as made of tiny balls moving in vacuum, we described this as a typic-
ally male idea. Suggesting that it is male implies that the female part is missing. Which
part would that be?
From the present point of view, the female part of physics might be the quantum de-
scription of the vacuum. The unravelling of the structure of the vacuum, as extended
container of localized balls, could be seen as the female half of physics. If women had
Ref. 1158 developed physics, the order of the discoveries would surely have been different. Instead
of studying matter, as men did, women might have studied the vacuum first.
In any case, the female and the male approaches, taken together, lead us to the descrip-
tion of nature by extended entities. Extended entities, which show that particles are not
Copyright © Christoph Schiller November 1997–May 2006

balls and that the vacuum is not a container, transcend the sexist approaches and lead
to the unified description. In a sense, extended entities are thus the politically correct
approach to nature.

A physical aphorism
To ‘show’ that we are not far from the top of Motion Mountain, we give a less serious
Ref. 1159, Ref. 1160 argument as final curiosity. Salecker and Wigner and then Zimmerman formulated the
fundamental» limit for the measurement precision τ attainable by a clock of mass M. It is
given by τ = ħT~Mc 2 , where T is the time to be measured. We can then ask what time
* ‘We must know, we will know.’ This was Hilbert’s famous personal credo.

Dvipsbugw
 1128 xi general rel ativity versus quantum mechanics • string theory

T can be measured with a precision of a Planck time t Pl , given a clock of the mass of the
whole universe. We get a maximum time of
2 2
T=
t Pl c
M. (808)
ħ
Inserting numbers, we find rather precisely that the time T is the present age of the uni-
Challenge 1523 e verse. With the right dose of humour we can see the result as an omen for the belief that
time is now ripe, after so much waiting, to understand the universe down to the Planck Dvipsbugw
scale. We are getting nearer to the top of Motion Mountain. Be prepared for even more
fun.

38. string theory – a web of dualities

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

String theory, superstring theory or membrane theory – all are names for related ap-
proaches – can be characterized as the most investigated theory ever. String theory ex-
plores the idea of particles as extended entities. String theory contains a maximum speed,
a minimum action and a maximum force. It thus incorporates special relativity, quantum
theory and general relativity.
In its attempt to achieve a unified description, string theory uses four ideas that go
beyond usual general relativity and quantum theory: extension, supersymmetry, higher
dimensions and dualities.
— String theory describes particles as extended entities. In a further generalization, M-
theory describes particles as higher-dimensional membranes.
— String theory uses supersymmetry. Supersymmetry is the symmetry that relates matter
to radiation, or equivalently, fermions to bosons. Supersymmetry is the most general
local interaction symmetry that is known.
— String theory uses higher dimensions to introduce extended entities and to unify inter-
actions. A number of dimensions higher than 3 + 1 seems to be necessary for a consist-
ent quantum field theory. At present however, the topology and size of the dimensions
above 4 is still unclear. (In fact, to be honest, the topology of the first 4 dimensions is
also unknown.)
Ref. 1161 — String theory makes heavy use of dualities. Dualities, in the context of high energy
Copyright © Christoph Schiller November 1997–May 2006

physics, are symmetries that map large to small values of physical observables. Import-
ant examples are space-time duality and coupling constant duality. Dualities are global
interaction and space-time symmetries. They are essential to include gravitation in the
description of nature. With dualities, string theory integrates the equivalence between
space-time and matter-radiation.
By incorporating these four ideas, string theory acquires a number of appealing prop-
erties. First of all, the theory is unique. It has no adjustable parameters. Furthermore, as
expected from any theory with extended entities,string theory contains gravity. In addi-
tion, string theory describes interactions. It does describe gauge fields such as the elec-
tromagnetic field. The theory is thus an extension of quantum field theory. All essential
points of quantum field theory are retained.

Dvipsbugw
strings and membranes – why string theory is so difficult 1129

String theory has many large symmetries, a consequence of its many dualities. These
symmetries connect many wildly different situations; that makes the theory fascinating
but also difficult to picture. Finally, string theory shows special cancellations of anom-
alies and inconsistencies. Historically, the first example was the Green-Schwarz anomaly
cancellation; however, many other mathematical problems of quantum field theory are
solved.
In short, string theory research has shown that general relativity and quantum field
theory can be unified using extended entities.
Dvipsbugw
Strings and membranes – why string theory is so difficult
What is string theory? There are two answers. The first answer is to follow its historical
development. Unfortunately, that is almost the hardest of all possible ways to learn and
understand string theory.* Thus we only give a short overview of this approach here.
The full description of string theory along historical lines starts from classical strings,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

then proceeds to string field theory, incorporating dualities, higher dimensions and su-
persymmetry on the way. Each step can be described by a Lagrangian. Let us have a look
at them.
The first idea of string theory, as the name says, is the use of strings to describe particles.
Strings replace the idea of point particles. The simplest form of string theory can be given
if the existence of a background space-time is already taken for granted.

– CS – more to be added in the future – CS –

Matrix models and M-theory

One of the most beautiful results of string theory research is the fact that string theory
can be deduced from an extremely small number of fundamental ideas. The approach is
to take a few basic ideas and to deduce M-theory form them. M-theory is the most recent
formulation of string theory.
M-theory assumes that the basic entities of nature are membranes of different
dimensions.M-theory incorporates dualities and holography. The existence of interac-
tions can been deduced. Supersymmetry and, in particular, super Yang–Mills theory can
be deduced. Also the entropy of black holes follows from M-theory. Finally, supergravity
has been deduced from M-theory. This means that the existence of space-time follows
Copyright © Christoph Schiller November 1997–May 2006

from M-theory. M-theory is thus a promising candidate for a theory of motion.

– CS – the rest of the chapter will be made available in the future – CS –

In other words, M-theory complies with all conceptual requirements that a unified theory
must fulfil. Let us now turn to experiment.
* The history of string theory was characterized by short periods of excitement followed by long periods of
disappointment. The main reason for this ineffective evolution was that most researchers only studied topics
that everybody else was also studying. Due to the fear of unemployment of young researchers and out of the
fear of missing out something, people were afraid to research topics that nobody else was looking at. Thus
string theory had an extremely difficult birth.

Dvipsbugw
 1130 xi general rel ativity versus quantum mechanics • string theory

Masses and couplings

One of the main results of QCD, the theory of strong interactions, is the explanation of
Ref. 1166 mass relations such as

m proton  e−k~α unif m Pl and k = 11~2π , α unif = 1~25 . (809)

Here, the value of the coupling constant α unif is taken at the unifying energy. This energy
is known to be a factor of about 800 below the Planck energy. In other words, a general Dvipsbugw
understanding of masses of bound states of the strong interaction, such as the proton,
requires almost only a knowledge of the unification energy and the coupling constant at
that energy. The approximate value α unif = 1~25 is an extrapolation from the low energy
value, using experimental data.
Any unified theory must allow one to calculate the coupling constants as function of
energy, including the value α unif at the unification energy itself. At present, researchers

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

state that the main difficulty of string theory and of M-theory is the search for the vacuum
state. Without the vacuum state, no calculations of the masses and coupling constants are
possible.
The vacuum state of string theory or M-theory is expected by researchers to be one of
an extremely involved set of topologically distinct manifolds. At present, is it estimated
that there are around 10750250 candidate vacuum states. One notes that the error alone is
as large as the number of Planck 4-volumes in the history of the universe. This poorness
of the estimate reflects the poor understanding of the theory.
On can describe the situation also in the following way. String and M-theory predict
states with Planck mass and with zero mass. The zero mass particles are then thought to
get a tiny mass (compared to the Planck mass) due to the Higgs mechanism. However,
the Higgs mechanism, with all its coupling constants, has not yet been deduced from
string theory. In other words, so far, string theory predicts no masses for elementary
particles. This disappointing situation is the reason that several scientists are dismissing
Ref. 1167 string theory altogether. Future will show.

Outlook
It is estimated that 10 000 man-years have been invested in the search for string theory;
compare this with about 5 for Maxwell’s electrodynamics, 10 for general relativity and a
similar number for the foundation of quantum theory.
Copyright © Christoph Schiller November 1997–May 2006

Historically, the community of string theorists took over 10 years to understand that
strings were not the basic entities of string theory. The fundamental entities are mem-
branes. Then string theorists took another 10 years and more to understand that mem-
branes are not the most practical fundamental entities for calculation. However, the
search for the most practical entities is still ongoing. It is probable that knotted mem-
branes must be taken into consideration.
Why is M-theory so hard? Doing calculations on knotted membranes is difficult. In
fact, it is extremely difficult. Mathematicians and physicists all over the world are still
struggling to find simple ways for such calculations.
In short, a new approach for calculations with extended entities is needed. We then can
ask the following question. M-theory is based on several basic ideas: extension, higher di-

Dvipsbugw
 matrix models and m-theory 1131

mensions, supersymmetry and duality. Which of these basic assumptions is confusing?

Dualities seem to be related to extension, for which experimental and theoretical evid-
Page 1101 ence exists. However, supersymmetry and higher dimensions are completely speculative.
We thus leave them aside, and continue our adventure.

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
1132 xi general rel ativity versus quantum mechanics

Biblio graphy

1122 C. Schiller, Does matter differ from vacuum? http://www.arxiv.org/abs/gr-qc/9610066.

A French language version has been published as C. Schiller, Le vide diffère-t-il de la
matière? in E. Gunzig & S. Diner, editeurs, Le vide – Univers du tout et du rien – Des
physiciens et des philosophes s’interrogent, Les Éditions de l’Université de Bruxelles, 1998.
Cited on pages 1102, 1103, 1106, 1108, 1109, 1110, 1120, and 1121.
1123 Most arguments for minimal temporal and spatial intervals are due to M. Schön, Oper- Dvipsbugw
ative time definition and principal uncertainty, preprint http://www.arxiv.org/abs/gr-qc/
9304024, and to T. Padmanabhan, Limitations on the operational definition of space-
time events and quantum gravity, Classical and Quantum Gravity 4, pp. L107–L113, 1987;
see also his earlier papers referenced there. Cited on page 1102.
1124 A similar point of view, often called monism, was proposed by Baruch Spinoza, Ethics
Demonstrated in Geometrical Order, 1677, originally in Latin; an affordable French edition
is B. Spinoza, L’Ethique, Folio-Gallimard, 1954. For a discussion of his ideas, especially

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

his monism, see Don Garret, ed., The Cambridge Companion to Spinoza, Cambridge
University Press, 1996, or any general text on the history of philosophy. Cited on page 1104.
1125 See L. Susskind & J. Uglum, Black holes, interactions, and strings, http://www.arxiv.
org/abs/hep-th/9410074, or L. Susskind, String theory and the principle of black hole
complementarity, Physical Review Letters 71, pp. 2367–2368, 1993, and M. Karliner, I.
Klebanov & L. Susskind, Size and shape of strings, International Journal of Modern
Physics A 3, pp. 1981–1996, 1988, as well as L. Susskind, Structure of hadrons implied by
duality, Physical Review D 1, pp. 1182–1186, 1970. Cited on pages 1106 and 1120.
1126 M. Pl anck, Über irreversible Strahlungsvorgänge, Sitzungsberichte der Kaiserlichen
Akademie der Wissenschaften zu Berlin, pp. 440–480, 1899. Today it is commonplace to
use ħ = h~2π instead of Planck’s h, which he originally called b in his paper. Cited on page
1106.
1127 C.A. Mead, Possible connection between gravitation and fundamental length, Physical
Review B 135, pp. 849–862, 1964. Cited on page 1106.
1128 A.D. Sakharov, Vacuum quantum fluctuations in curved space and the theory of grav-
itation, Soviet Physics - Doklady 12, pp. 1040–1041, 1968. Cited on page 1109.
1129 P. Facchi & S. Pascazio, Quantum Zeno and inverse quantum Zeno effects, pp. 147–
217, in E. Wolf, editor, Progress in Optics, 42, 2001. Cited on page 1109.
1130 Aristotle, Of Generation and Corruption, I, II, 316 a 13. See Jean-Paul Dumont, Les
écoles présocratiques, Folio Essais, Gallimard, p. 427, 1991. Cited on page 1110.
Copyright © Christoph Schiller November 1997–May 2006

1131 See for example the speculative model of vacuum as composed of Planck-size spheres pro-
posed by F. Winterberg, Zeitschrift für Naturforschung 52a, p. 183, 1997. Cited on page
1110.
1132 The Greek salt and fish arguments are given by Lucrece, or Titus Lucretius Carus, De natura
rerum, around 50 bce. Cited on pages 1112 and 1122.
1133 Cited in Jean-Paul Dumont, Les écoles présocratiques, Folio Essais, Gallimard, p. 379,
1991. Cited on page 1113.
1134 D. Olive & C. Montonen, Magnetic monopoles as gauge particles, Physics Letters 72B,
pp. 117–120, 1977. Cited on page 1114.
1135 J.H. Schwarz, The second superstring revolution, http://www.arxiv.org/abs/hep-th/
9607067. More on dualities in ... Cited on pages 1112 and 1114.

Dvipsbugw
bibliography 1133

1136 The famous quote is found at the beginning of chapter XI, ‘The Physical Universe’, in Ar-
thur Eddington, The Philosophy of Physical Science, Cambridge, 1939. Cited on page
1115.
1137 Pl ato, Parmenides, c. 370 bce. It has been translated into most languages. Reading it
aloud, like a song, is a beautiful experience. A pale reflection of these ideas is Bohm’s
concept of ‘unbroken wholeness’. Cited on page 1116.
1138 P. Gibbs, Event-symmetric physics, http://www.arxiv.org/abs/hep-th/9505089; see also
his http://www.weburbia.com/pg/contents.htm website. Cited on page 1116.
Dvipsbugw
1139 J.D. Bekenstein, Black holes and entropy, Physical Review D 7, pp. 2333–2346, 1973.
Cited on page 1119.
1140 S.W. Hawking, Particle creation by black holes, Communications in Mathematical Phys-
ics 43, pp. 199–220, 1975; see also S.W. Hawking, Black hole thermodynamics, Physical
Review D13, pp. 191–197, 1976. Cited on page 1119.
1141 J.B. Hartle & S.W. Hawking, Path integral derivation of black hole radiance, Physical
Review D 13, pp. 2188–2203, 1976. See also A. Strominger & C. Vafa, Microscopic ori-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

gin of Bekenstein–Hawking entropy, Physics Letters B379, pp. 99–104, 1996, or the down-
loadable version found at http://www.arxiv.org/abs/hep-th/9601029. For another deriva-
tion of black hole entropy, see Gary T. Horowitz & J. Polchinski, A correspond-
ence principle for black holes and strings, Physical Review D 55, pp. 6189–6197, 1997, or
http://www.arxiv.org/abs/hep-th/9612146. Cited on page 1119.
1142 J.D. Bekenstein, Entropy bounds and black hole remnants, Physical Review D 49,
pp. 1912–1921, 1994. Cited on page 1119.
1143 J. Madd ox, When entropy does not seem extensive, Nature 365, p. 103, 1993. Other ref-
erences on the proportionality of entropy and surface are ... Cited on page 1119.
1144 L. B ombelli, R.K. Koul, J. Lee & R.D. Sorkin, Quantum source of entropy of black
holes, Physical Review D 34, pp. 373–383, 1986. Cited on page 1119.
1145 The analogy between polymers and black holes is by G. Weber, Thermodynamics at
boundaries, Nature 365, p. 792, 1993. Cited on page 1119.
1146 P.-G. de Gennes, Scaling Concepts in Polymer Physics, Cornell University Press, 1979.
Cited on page 1119.
1147 See for example S. Majid, Introduction to braided geometry and q-Minkowski space, pre-
print found at http://www.arxiv.org/abs/hep-th/9410241 or S. Majid, Duality principle
and braided geometry, preprint http://www.arxiv.org/abs/gr-qc/940957. Cited on pages
1120 and 1121.
1148 The relation between spin and statistics has been studied recently by M.V. Berry & J.M.
Copyright © Christoph Schiller November 1997–May 2006

Robbins, Quantum indistinguishability: spin–statistics without relativity or field theory?,

in R.C. Hilborn & G.M. Tino, editors, Spin–Statistics Connection and Commutation
Relations, American Institute of Physics CP545 2000. Cited on page 1122.
1149 An example is given by A.A. Sl avnov, Fermi–Bose duality via extra dimension, pre-
print http://www.arxiv.org/abs/hep-th/9512101. See also the standard work by Michael
Stone, editor, Bosonization, World Scientific, 1994. Cited on page 1123.
1150 A popular account making the point is the bestseller by the US-american string theorist
Brian Greene, The Elegant Universe – Superstrings, Hidden Dimensions, and the Quest
for the Ultimate Theory, Jonathan Cape 1999. Cited on page 1123.
1151 The weave model of space-time appears in certain approaches of quantum gravity, as told
by A. Ashtekar, Quantum geometry and gravity: recent advances, December 2001,

Dvipsbugw
1134 xi general rel ativity versus quantum mechanics

preprint found at http://www.arxiv.org/abs/gr-qc/0112038. See also A. Ashtekar,

Quantum geometry in action: big bang and black holes, http://www.arxiv.org/abs/
math-ph/0202008. On a slightly different topic, see also S.A. Major, A spin network
primer, http://www.arxiv.org/abs/gr-qc/9905020. Cited on page 1123.
1152 A good introduction into his work is the paper D. Kreimer, New mathematical struc-
tures in renormalisable quantum field theories, http://www.arxiv.org/abs/hep-th/0211136
to appear in Annals of Physics Cited on page 1123.
1153 Introductions to holography are E. Alvarez, J. Conde & L. Hernandez, Rudi-
ments of Holography, http://www.arxiv.org/abs/hep-th/0205075 or R. B ousso, The holo- Dvipsbugw
graphic principle, Review of Modern Physics 74, pp. 825–874, 2002, also at http://www.arxiv.
org/abs/hep-th/0203101. (However, this otherwise good review has some argumentation
errors, as explained on page 871.) The importance of holography in theoretical high en-
ergy physics was underlined by the discovery of J. Maldacena, The large N limit of
superconformal field theories and supergravity, http://www.arxiv.org/abs/hep-th/9711200.
Cited on page 1123.
P.F. Mende, String theory at short distance and the principle of equivalence, preprint

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1154
available at http://www.arxiv.org/abs/hep-th/9210001. Cited on page 1124.
1155 A. Gregori, Entropy, string theory and our world, http://www.arxiv.org/abs/hep-th/
0207195 preprint. Cited on pages 1124 and 1126.
1156 String cosmology is a passtime for many. Examples are N.E. Mavromatos, String
cosmology, http://www.arxiv.org/abs/hep-th/0111275 and N.G. Sanchez, New develop-
ments in string gravity and string cosmology - a summary report, http://www.arxiv.org/
abs/hep-th/0209016. Cited on page 1126.
1157 Searches for background free approaches are described by E. Witten, .... Cited on page
1127.
1158 Not even the most combative feminists have ever said how women-centered physics should
look like. This lack of alternatives pervades also the agressive text by Margaret Wer-
theim, Pythagoras’ Trousers – God, Physics and the Gender Wars, Fourth Estate, 1997, and
thus makes its author look rather silly. Cited on page 1127.
1159 H. Salecker & E.P. Wigner, Quantum limitations of the measurement of space-time
distances, Physical Review 109, pp. 571–577, 1958. Cited on page 1127.
1160 E.J. Zimmerman, The macroscopic nature of space-time, American Journal of Physics
30, pp. 97–105, 1962. Cited on page 1127.
1161 See A. Sen, An introduction to duality symmetries in string theory, in Les Houches Sum-
mer School: Unity from Duality: Gravity, Gauge Theory and Strings, (Les Houches, France,
2001), Springer Verlag, 76, pp. 241–322, 2002. Cited on page 1128.
Copyright © Christoph Schiller November 1997–May 2006

1162 See for example C. Rovelli, Notes for a brief history of quantum gravity, electronic pre-
print available at http://www.arxiv.org/abs/gr-qc/0006061. No citations.
1163 See the http://www.nuclecu.unam.mx/~alberto/physics/stringrev.html website for an up-
to date list of string theory and M theory reviews. No citations.
1164 W.G. Unruh, Notes on black hole evaporation, Physical Review D 14, pp. 870–875, 1976.
W.G. Unruh & R.M. Wald, What happens when an accelerating observer detects a
Rindler particle, Physical Review D 29, pp. 1047–1056, 1984. No citations.
1165 The most prominent proposer of the idea that particles might be knots was William Thom-
son. See W. Thomson, On vortex motion, Transactions of the Royal Society in Edinburgh
pp. 217–260, 1868. This paper stimulated work on knot theory, as it proposed the idea that
different particles might be differently knotted vortices in the aether. No citations.

Dvipsbugw
bibliography 1135

1166 F. Wilczek, Getting its from bits, Nature 397, pp. 303–306, 1999. Cited on page 1130.
1167 See the well-known blog by Peter Woit, Not even wrong, at http://www.math.columbia.
edu/~woit/blog. Cited on page 1130.

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 Dvipsbugw

Dvipsbugw
Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net Copyright © Christoph Schiller November 1997–May 2006
– CS – this chapter will appear in the near future – CS –
(NOT YET AVAI L ABLE)
UNI FIC ATION
C h a p t e r XII
 Dvipsbugw

Dvipsbugw
Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net Copyright © Christoph Schiller November 1997–May 2006
– CS – this chapter will appear in the near future – CS –
THE TOP OF THE MOUNTAI N
(NOT YET AVAI L ABLE)
C h a p t e r XIII
Fourth Part

Appendices

Where necessary reference information

for mountain ascents is provided,
preparing the reader for this
and any other future adventure.
 Appendix A

NOTATION AND C ONV ENTIONS

Dvipsbugw

N
ewly introduced and defined concepts in this text are indicated by italic typeface.
ew definitions are also referred to in the index by italic page numbers. We
naturally use SI units throughout the text; these are defined in Appendix B. Exper-
imental results are cited with limited precision, usually only two digits, as this is almost
always sufficient for our purposes. High-precision reference values can be found in Ap-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

pendices B and C.
In relativity we use the time convention, where the metric has the signature (+ − − − ).
Ref. 1168 This is used in about 70 % of the literature worldwide. We use indices i, j, k for 3-vectors
and indices a, b, c, etc. for 4-vectors. Other conventions specific to general relativity are
explained as they arise in the text.

The L atin alphabet

What is written without effort is in general read

“ without pleasure.
Samuel Johnson

Books are collections of symbols. Writing was probably invented between 3400 and 3300 ”
bce by the Sumerians in Mesopotamia (though other possibilities are also discussed). It
then took over a thousand years before people started using symbols to represent sounds
instead of concepts: this is the way in which the first alphabet was created. This happened
between 2000 and 1600 bce (possibly in Egypt) and led to the Semitic alphabet. The use
of an alphabet had so many advantages that it was quickly adopted in all neighbouring
cultures, though in different forms. As a result, the Semitic alphabet is the forefather of
all alphabets used in the world.
This text is written using the Latin alphabet. At first sight, this seems to imply that its
Copyright © Christoph Schiller November 1997–May 2006

pronunciation cannot be explained in print, in contrast to the pronunciation of other al-

phabets or of the International Phonetic Alphabet (IPA). (They can be explained using the
alphabet of the main text.) However, it is in principle possible to write a text that describes
exactly how to move lips, mouth and tongue for each letter, using physical concepts where
necessary. The descriptions of pronunciations found in dictionaries make indirect use of
this method: they refer to the memory of pronounced words or sounds found in nature.
Historically, the Latin alphabet was derived from the Etruscan, which itself was a de-
rivation of the Greek alphabet. There are two main forms.

Dvipsbugw
 a notation and conventions 1143

The ancient Latin alphabet,

used from the sixth century bce onwards:
A B C D E F Z H I K L M N O P Q R S T V X

The classical Latin alphabet,

used from the second century bce until the eleventh century:
A B C D E F G H I K L M N O P Q R S T V X Y Z Dvipsbugw

The letter G was added in the third century bce by the first Roman to run a fee-paying
school, Spurius Carvilius Ruga. He added a horizontal bar to the letter C and substituted
the letter Z, which was not used in Latin any more, for this new letter. In the second
century bce, after the conquest of Greece, the Romans included the letters Y and Z from
the Greek alphabet at the end of their own (therefore effectively reintroducing the Z) in
order to be able to write Greek words. This classical Latin alphabet was stable for the next

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

thousand years.*
The classical Latin alphabet was spread around Europe, Africa and Asia by the Ro-
mans during their conquests; due to its simplicity it began to be used for writing in nu-
merous other languages. Most modern ‘Latin’ alphabets include a few other letters. The
letter W was introduced in the eleventh century in French and was then adopted in most
European languages. The letters J and U were introduced in the sixteenth century in Italy,
to distinguish certain sounds which had previously been represented by I and V. In other
languages, these letters are used for other sounds. The contractions æ and œ date from
the Middle Ages. Other Latin alphabets include more letters, such as the German sharp s,
written ß, a contraction of ‘ss’ or ‘sz’, the and the Nordic letters thorn, written Þ or þ, and
Ref. 1169 eth, written Ð or ð, taken from the futhorc,** and other signs.
Lower-case letters were not used in classical Latin; they date only from the Middle
Ages, from the time of Charlemagne. Like most accents, such as ê, ç or ä, which were
also first used in the Middle Ages, lower-case letters were introduced to save the then
expensive paper surface by shortening written words.
Outside a dog, a book is a man’s best friend.

“ Inside a dog, it’s too dark to read.

Groucho Marx

” Copyright © Christoph Schiller November 1997–May 2006

The Greek alphabet

The Latin alphabet was derived from the Etruscan; the Etruscan from the Greek. The
Greek alphabet was itself derived from the Phoenician or a similar northern Semitic al-

* To meet Latin speakers and writers, go to http://www.alcuinus.net/.

** The Runic script, also called Futhark or futhorc, a type of alphabet used in the Middle Ages in Germanic,
Anglo–Saxon and Nordic countries, probably also derives from the Etruscan alphabet. The name derives
from the first six letters: f, u, th, a (or o), r, k (or c). The third letter is the letter thorn mentioned above; it is
often written ‘Y’ in Old English, as in ‘Ye Olde Shoppe.’ From the runic alphabet Old English also took the
Ref. 1170 letter wyn to represent the ‘w’ sound, and the already mentioned eth. (The other letters used in Old English
– not from futhorc – were the yogh, an ancient variant of g, and the ligatures æ or Æ, called ash, and œ or Œ,
called ethel.)

Dvipsbugw
 1144 a notation and conventions

TA B L E 80 The ancient and classical Greek alphabets, and the correspondence with Latin and Indian
digits

Anc. Cl ass. Name Corresp. Anc. Cl ass. Name Corresp.

Α Α α alpha a 1 Ν Ν ν nu n 50
Β Β β beta b 2 Ξ Ξ ξ xi x 60
Γ Γ γ gamma g, n1 3 Ο Ο ο omicron o 70
∆ ∆ δ delta d 4 Π Π π pi p 80 Dvipsbugw
Ε Ε ε epsilon e 5 P Ϟ, qoppa3 q 90
F ϝ, Ϛ digamma, w 6 Ρ Ρ ρ rho r, rh 100
stigma2 Σ Σ σ, ς sigma4 s 200
Ζ Ζ ζ zeta z 7 Τ Τ τ tau t 300
Η Η η eta e 8 Υ υ upsilon y, u5 400
Θ Θ θ theta th 9 Φ φ phi ph, f 500

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Ι Ι ι iota i, j 10 Χ χ chi ch 600
Κ Κ κ kappa k 20 Ψ ψ psi ps 700
Λ Λ λ lambda l 30 Ω ω omega o 800
Μ Μ µ mu m 40 Λ Ϡ
v sampi6 s 900

The regional archaic letters yot, sha and san are not included in the table. The letter san was the ancestor of
sampi.
1. Only if before velars, i.e. before kappa, gamma, xi and chi.
2. ‘Digamma’ is the name used for the F-shaped form. It was mainly used as a letter (but also sometimes, in its
lower-case form, as a number), whereas the shape and name ‘stigma’ is used only for the number. Both names
were derived from the respective shapes; in fact, the stigma is a medieval, uncial version of the digamma.
The name ‘stigma’ is derived from the fact that the letter looks like a sigma with a tau attached under it –
though unfortunately not in all modern fonts. The original letter name, also giving its pronunciation, was
‘waw’.
3. The version of qoppa that looks like a reversed and rotated z is still in occasional use in modern Greek.
Unicode calls this version ‘koppa’.
4. The second variant of sigma is used only at the end of words.
5. Uspilon corresponds to ‘u’ only as the second letter in diphthongs.
6. In older times, the letter sampi was positioned between pi and qoppa.

Copyright © Christoph Schiller November 1997–May 2006

Ref. 1171 phabet in the tenth century bce. The Greek alphabet, for the first time, included letters
also for vowels, which the Semitic alphabets lacked (and often still lack). In the Phoeni-
cian alphabet and in many of its derivatives, such as the Greek alphabet, each letter has a
proper name. This is in contrast to the Etruscan and Latin alphabets. The first two Greek
letter names are, of course, the origin of the term alphabet itself.
In the tenth century bce, the Ionian or ancient (eastern) Greek alphabet consisted of
the upper-case letters only. In the sixth century bce several letters were dropped, while
a few new ones and the lower-case versions were added, giving the classical Greek alpha-
bet. Still later, accents, subscripts and breathings were introduced. Table 80 also gives the
values signified by the letters took when they were used as numbers. For this special use,
the obsolete ancient letters were kept during the classical period; thus they also acquired

Dvipsbugw
a notation and conventions 1145

lower-case forms.
The Latin correspondence in the table is the standard classical one, used for writing
Greek words. The question of the correct pronunciation of Greek has been hotly debated
in specialist circles; the traditional Erasmian pronunciation does not correspond either
to the results of linguistic research, or to modern Greek. In classical Greek, the sound that
sheep make was βη–βη. (Erasmian pronunciation wrongly insists on a narrow η; modern
Greek pronunciation is different for β, which is now pronounced ‘v’, and for η, which is
now pronounced as ‘i’ – a long ‘i’.) Obviously, the pronunciation of Greek varied from
region to region and over time. For Attic Greek, the main dialect spoken in the classical Dvipsbugw
period, the question is now settled. Linguistic research has shown that chi, phi and theta
were less aspirated than usually pronounced in English and sounded more like the initial
sounds of ‘cat’, ‘perfect’ and ‘tin’; moreover, the zeta seems to have been pronounced more
like ‘zd’ as in ‘buzzed’. As for the vowels, contrary to tradition, epsilon is closed and short
whereas eta is open and long; omicron is closed and short whereas omega is wide and
long, and upsilon is really a ‘u’ sound as in ‘boot’, not like a French ‘u’ or German ‘ü.’

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

The Greek vowels can have rough or smooth breathings, subscripts, and acute, grave,
circumflex or diaeresis accents. Breathings – used also on ρ – determine whether the letter
is aspirated. Accents, which were interpreted as stresses in the Erasmian pronunciation,
actually represented pitches. Classical Greek could have up to three of these added signs
per letter; modern Greek never has more than one.
Another descendant of the Greek alphabet* is the Cyrillic alphabet, which is used with
slight variations, in many Slavic languages, such as Russian and Bulgarian. However, there
is no standard transcription from Cyrillic to Latin, so that often the same Russian name
is spelled differently in different countries or even in the same country on different occa-
sions.
TA B L E 81 The beginning of the Hebrew abjad

Letter Name Correspondence

ℵ aleph a 1
° beth b 2
‫ג‬ gimel g 3
² daleth d 4
etc. Copyright © Christoph Schiller November 1997–May 2006

* The Greek alphabet is also the origin of the Gothic alphabet, which was defined in the fourth century by
Wulfila for the Gothic language, using also a few signs from the Latin and futhorc scripts.
The Gothic alphabet is not to be confused with the so-called Gothic letters, a style of the Latin alphabet
used all over Europe from the eleventh century onwards. In Latin countries, Gothic letters were replaced in
the sixteenth century by the Antiqua, the ancestor of the type in which this text is set. In other countries,
Gothic letters remained in use for much longer. They were used in type and handwriting in Germany until
1941, when the National Socialist government suddenly abolished them, in order to comply with popular
demand. They remain in sporadic use across Europe. In many physics and mathematics books, Gothic letters
are used to denote vector quantities.

Dvipsbugw
 1146 a notation and conventions

The Hebrew alphabet and other scripts

The phoenician alphabet is also the origin of the Hebrew consonant alphabet or abjad. Its
Ref. 1172 first letters are given in Table 81. Only the letter aleph is commonly used in mathematics,
Page 649 though others have been proposed.
Around one hundred writing systems are in use throughout the world. Experts clas-
sify them into five groups. Phonemic alphabets, such as Latin or Greek, have a sign for
each consonant and vowel. Abjads or consonant alphabets, such as Hebrew or Arabic,
have a sign for each consonant (sometimes including some vowels, such as aleph), and Dvipsbugw
do not write (most) vowels; most abjads are written from right to left. Abugidas, also
called syllabic alphabets or alphasyllabaries, such as Balinese, Burmese, Devanagari, Ta-
galog, Thai, Tibetan or Lao, write consonants and vowels; each consonant has an inherent
vowel which can be changed into the others by diacritics. Syllabaries, such as Hiragana
or Ethiopic, have a sign for each syllable of the language. Finally, complex scripts, such
as Chinese, Mayan or the Egyptian hieroglyphs, use signs which have both sound and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

meaning. Writing systems can have text flowing from right to left, from bottom to top,
and can count book pages in the opposite sense to this book.
Even though there are about 7000 languages on Earth, there are only about one hun-
Ref. 1173 dred writing systems used today. About fifty other writing systems have fallen out of use. *
For physical and mathematical formulae, though, the sign system used in this text, based
on Latin and Greek letters, written from left to right and from top to bottom, is a standard
the world over. It is used independently of the writing system of the text containing it.

Digits and numbers

Both the digits and the method used in this text to write numbers originated in India.
They were brought to the Mediterranean by Arabic mathematicians in the Middle Ages.
The number system used in this text is thus much younger than the alphabet.** The In-
dian numbers were made popular in Europe by Leonardo of Pisa, called Fibonacci,***
in his book Liber Abaci or ‘Book of Calculation’, which he published in 1202. That book
revolutionized mathematics. Anybody with paper and a pen (the pencil had not yet been
invented) was now able to calculate and write down numbers as large as reason allowed,
or even larger, and to perform calculations with them. Fibonacci’s book started:

Novem figure indorum he sunt 9 8 7 6 5 4 3 2 1. Cum his itaque novem fig- Copyright © Christoph Schiller November 1997–May 2006

* A well-designed website on the topic is http://www.omniglot.com. The main present and past writing
systems are encoded in the Unicode standard, which at present contains 52 writing systems. See http://
www.unicode.org.
** The story of the development of the numbers is told most interestingly by G. Ifrah, Histoire universelle
des chiffres, Seghers, 1981, which has been translated into several languages. He sums up the genealogy in ten
beautiful tables, one for each digit, at the end of the book. However, the book contains many factual errors,
as explained in the http://www.ams.org/notices/200201/rev-dauben.pdf and http://www.ams.org/notices/
200202/rev-dauben.pdf review.
It is not correct to call the digits 0 to 9 Arabic. Both the actual Arabic digits and the digits used in Latin
texts such as this one derive from the Indian digits. Only the digits 0, 2, 3 and 7 resemble those used in
Arabic writing, and then only if they are turned clockwise by 90°.
*** Leonardo di Pisa, called Fibonacci (b. c. 1175 Pisa, d. 1250 Pisa), Italian mathematician, and the most
important mathematician of his time.

Dvipsbugw
 a notation and conventions 1147

uris, et cum hoc signo 0, quod arabice zephirum appellatur, scribitur quilibet
numerus, ut inferius demonstratur.*

The Indian method of writing numbers consists of a large innovation, the positional sys-
tem, and a small one, the digit zero. The positional system, as described by Fibonacci, was
so much more efficient that it completely replaced the previous Roman number system,
which writes 1996 as IVMM or MCMIVC or MCMXCVI, as well as the Greek number system,
in which the Greek letters were used for numbers in the way shown in Table 80, thus
writing 1996 as ͵αϠϞϚʹ. Compared to these systems, the Indian numbers are a much bet- Dvipsbugw
ter technology. Indeed, the Indian system proved so practical that calculations done on
paper completely eliminated the need for the abacus, which therefore fell into disuse. The
abacus is still in use only in those countries which do not use a positional system to write
numbers. (The Indian system also eliminated the need fir systems to represent numbers
with fingers. Such systems, which could show numbers up to 10 000 and more, have left
only one trace: the term ‘digit’ itself, which derives from the Latin word for finger.) Simil-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

arly, only the positional number system allows mental calculations and made – and still
makes – calculating prodigies possible.**

The symbols used in the text

To avoide the tediouse repetition of these

“ woordes: is equalle to: I will sette as I doe often

in woorke use, a paire of paralleles, or Gemowe
lines of one lengthe, thus: = , bicause noe .2.
thynges, can be moare equalle.
Robert Recorde***

Besides text and numbers, physics books contain other symbols. Most symbols have been
developed over hundreds of years, so that only the clearest and simplest are now in use.
In this mountain ascent, the symbols used as abbreviations for physical quantities are all
”
taken from the Latin or Greek alphabets and are always defined in the context where
they are used. The symbols designating units, constants and particles are defined in Ap-
pendices B and C. There is an international standard for them (ISO 31), but it is virtually
Ref. 1172 inaccessible; the symbols used in this text are those in common use.
The mathematical symbols used in this text, in particular those for operations and
relations, are given in the following list, together with their origins. The details of their
Ref. 1174 history have been extensively studied in the literature.
Copyright © Christoph Schiller November 1997–May 2006

* ‘The nine figures of the Indians are: 9 8 7 6 5 4 3 2 1. With these nine figures, and with this sign 0 which in
Arabic is called zephirum, any number can be written, as will be demonstrated below.’
** Currently, the shortest time for finding the thirteenth (integer) root of a hundred-digit number, a result
with 8 digits, is 11.8 seconds. For more about the stories and the methods of calculating prodigies, see the
fascinating book by Steven B. Smith, The Great Mental Calculators – The Psychology, Methods and Lives
of the Calculating Prodigies, Columbia University Press, 1983. The book also presents the techniques that
they use, and that anybody else can use to emulate them.
*** Robert Recorde (c. 1510–1558), English mathematician and physician; he died in prison, though not for
Ref. 1174 his false pretence to be the inventor of the ‘equal’ sign, which he took over from his Italian colleagues, but
for a smaller crime, namely debt. The quotation is from his The Whetstone of Witte, 1557. An image showing
the quote in manuscript can be found at the http://members.aol.com/jeff94100/witte.jpg website.

Dvipsbugw
1148 a notation and conventions

Symbol Meaning Origin

+, − plus, minus J. Regiomontanus 1456; the plus sign is

derived from Latin ‘et’
º
read as ‘square root’ used by C. Rudolff in 1525; the sign
stems from a deformation of the letter ‘r’,
initial of the Latin radix
= equal to Italian mathematicians, early sixteenth
century, then brought to England by R. Dvipsbugw
Recorde
{ }, [ ], ( ) grouping symbols use starts in the sixteenth century
A, < larger than, smaller than T. Harriot 1631
 multiplied with, times W. Oughtred 1631
 divided by G. Leibniz 1684
ë multiplied with, times G. Leibniz 1698

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

an power R. Descartes 1637
x, y, z coordinates, unknowns R. Descartes 1637
ax +by + c = 0 constants and equations for unknowns R. Descartes 1637
d~dx, d2 x, derivative, differential, integral G. Leibniz 1675
∫ y dx
φx function of x J. Bernoulli 1718
f x, f (x) function of x L. Euler 1734
∆x, P difference, sum L. Euler 1755
x is different from L. Euler eighteenth century
∂~∂x partial derivative, read like ‘d~dx’ it was derived from a cursive form of ‘d’
or of the letter ‘dey’ of the Cyrillic alpha-
bet by A. Legendre in 1786
∆ Laplace operator R. Murphy 1833
SxS absolute value K. Weierstrass 1841
∇ read as ‘nabla’ (or ‘del’) introduced by William Hamilton in 1853
and P.G. Tait in 1867, named after the
shape of an old Egyptian musical instru-
ment
[x] the measurement unit of a quantity x twentieth century
Copyright © Christoph Schiller November 1997–May 2006

ª infinity J. Wallis 1655

π 4 arctan 1 H. Jones 1706
Pn=0 n! = lim n ª (1 + 1~n)
º
ª 1 n
e L. Euler 1736
i + −1 L. Euler 1777
8, 9 set union and intersection G. Peano 1888
> element of G. Peano 1888
g empty set André Weil as member of the N.
Bourbaki group in the early twentieth
century
`ψS, Sψe bra and ket state vectors Paul Dirac 1930
a dyadic product or tensor product or unknown
outer product

Dvipsbugw
 a notation and conventions 1149

Other signs used here have more complicated origins. The & sign is a contraction of Latin
Ref. 1175 et meaning ‘and’, as is often more clearly visible in its variations, such as &, the common
italic form.
Each of the punctuation signs used in sentences with modern Latin alphabets, such
as , . ; : ! ? ‘ ’ » « – ( ) ... has its own history. Many are from ancient Greece, but the
Ref. 1176 question mark is from the court of Charlemagne, and exclamation marks appear first in
the sixteenth century.* The @ or at-sign probably stems from a medieval abbreviation of Dvipsbugw
Ref. 1177 Latin ad, meaning ‘at’, similarly to how the & sign evolved from Latin et. In recent years,
the smiley :-) and its variations have become popular. The smiley is in fact a new version of
the ‘point of irony’ which had been formerly proposed, without success, by A. de Brahm
(1868–1942).
The section sign § dates from the thirteenth century in northern Italy, as was shown
Ref. 1178 by the German palaeographer Paul Lehmann. It was derived from ornamental versions

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

of the capital letter C for capitulum, i.e. ‘little head’ or ‘chapter.’ The sign appeared first in
legal texts, where it is still used today, and then spread into other domains.
The paragraph sign ¶ was derived from a simpler ancient form looking like the Greek
letter Γ, a sign which was used in manuscripts from ancient Greece until well into the
Middle Ages to mark the start of a new text paragraph. In the Middle Ages it took the
modern form, probably because a letter c for caput was added in front of it.
Ref. 1179 One of the most important signs of all, the white space separating words, was due to
Celtic and Germanic influences when these people started using the Latin alphabet. It
became commonplace between the ninth and the thirteenth century, depending on the
language in question.

Calendars
The many ways to keep track of time differ greatly from civilization to civilization. The
most common calendar, and the one used in this text, is also one of the most absurd, as
it is a compromise between various political forces who tried to shape it.
In ancient times, independent localized entities, such as tribes or cities, preferred lunar
calendars, because lunar timekeeping is easily organized locally. This led to the use of the
month as a calendar unit. Centralized states imposed solar calendars, based on the year.
Solar calendars require astronomers, and thus a central authority to finance them. For
various reasons, farmers, politicians, tax collectors, astronomers, and some, but not all,
Copyright © Christoph Schiller November 1997–May 2006

religious groups wanted the calendar to follow the solar year as precisely as possible. The
compromises necessary between days and years are the origin of leap days. The com-
promises necessary between months and year led to the varying lengths are different in
different calendars. The most commonly used year–month structure was organized over
2000 years ago by Gaius Julius Ceasar, and is thus called the Julian calendar.
The system was destroyed only a few years later: August was lengthened to 31 days
when it was named after Augustus. Originally, the month was only 30 days long; but in
order to show that Augustus was as important as Caesar, after whom July is named, all

* On the parenthesis see the beautiful book by J. L ennard, But I Digress, Oxford University Press, 1991.

Dvipsbugw
 1150 a notation and conventions

month lengths in the second half of the year were changed, and February was shortened
by an additional day.
The week is an invention of Babylonia, from where it was spread through the world
by various religious groups. (The way astrology and astronomy cooperated to determine
Page 144 the order of the weekdays is explained in the section on gravitation.) Although it is about
three thousand years old, the week was fully included into the Julian calendar only around
the year 300, towards the end of the Western Roman Empire. The final change in the
Julian calendar took place between 1582 and 1917 (depending on the country), when more
precise measurements of the solar year were used to set a new method to determine leap Dvipsbugw
days, a method still in use today. Together with a reset of the date and the fixation of
the week rhythm, this standard is called the Gregorian calendar or simply the modern
calendar. It is used by a majority of the world’s population.
Despite its complexity, the modern calendar does allow you to determine the day of
the week of a given date in your head. Just execute the following six steps:
1. take the last two digits of the year, and divide by 4, discarding any fraction;

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

2. add the last two digits of the year;
3. subtract 1 for January or February of a leap year;
4. add 6 for 2000s or 1600s, 4 for 1700s or 2100s,
2 for 1800s and 2200s, and 0 for 1900s or 1500s;
5. add the day of the month;
6. add the month key value, namely 144 025 036 146 for JFM AMJ JAS OND.
The remainder after division by 7 gives the day of the week, with the correspondence 1-2-
3-4-5-6-0 meaning Sunday-Monday-Tuesday-Wednesday-Thursday-Friday-Saturday.*
When to start counting the years is a matter of choice. The oldest method not attached
to political power structures was that used in ancient Greece, when years were counted
from the first Olympic games. People used to say, for example, that they were born in the
first year of the twenty-third Olympiad. Later, political powers always imposed the count-
ing of years from some important event onwards.** Maybe reintroducing the Olympic

* Remembering the intermediate result for the current year can simplify things even more, especially since
the dates 4.4, 6.6, 8.8, 10.10, 12.12, 9.5, 5.9, 7.11, 11.7 and the last day of February all fall on the same day of
the week, namely on the year’s intermediate result plus 4.
** The present counting of years was defined in the Middle Ages by setting the date for the foundation of
Rome to the year 753 bce, or 753 before the Common Era, and then counting backwards, so that the bce
years behave almost like negative numbers. However, the year 1 follows directly after the year 1 bce: there
was no year 0.
Copyright © Christoph Schiller November 1997–May 2006

Some other standards set by the Roman Empire explain several abbreviations used in the text:
- c. is a Latin abbreviation for circa and means ‘roughly’;
- i.e. is a Latin abbreviation for id est and means ‘that is’;
- e.g. is a Latin abbreviation for exempli gratia and means ‘for the sake of example’;
- ibid. is a Latin abbreviation for ibidem and means ‘at that same place’;
- inf. is a Latin abbreviation for infra and means ‘(see) below’;
- op. cit. is a Latin abbreviation for opus citatum and means ‘the cited work’;
- et al. is a Latin abbreviation for et alii and means ‘and others’.
By the way, idem means ‘the same’ and passim means ‘here and there’ or ‘throughout’. Many terms used in
physics, like frequency, acceleration, velocity, mass, force, momentum, inertia, gravitation and temperature,
are derived from Latin. In fact, it is arguable that the language of science has been Latin for over two thousand
years. In Roman times it was Latin vocabulary with Latin grammar, in modern times it switched to Latin
vocabulary with French grammar, then for a short time to Latin vocabulary with German grammar, after

Dvipsbugw
 a notation and conventions 1151

counting is worth considering?

Abbreviations and eponyms or concepts?

Sentences like the following are the scourge of modern physics:

The EPR paradox in the Bohm formulation can perhaps be resolved using the GRW
approach, using the WKB approximation of the Schrödinger equation.
Dvipsbugw
Using such vocabulary is the best way to make language unintelligible to outsiders. First
of all, it uses abbreviations, which is a shame. On top of this, the sentence uses people’s
names to characterize concepts, i.e. it uses eponyms. Originally, eponyms were intended
as tributes to outstanding achievements. Today, when formulating radical new laws or
variables has become nearly impossible, the spread of eponyms intelligible to a steadily
decreasing number of people simply reflects an increasingly ineffective drive to fame.

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Eponyms are a proof of scientist’s lack of imagination. We avoid them as much as pos-
sible in our walk and give common names to mathematical equations or entities wherever
possible. People’s names are then used as appositions to these names. For example, ‘New-
ton’s equation of motion’ is never called ‘Newton’s equation’; ‘Einstein’s field equations’
is used instead of ‘Einstein’s equations’; and ‘Heisenberg’s equation of motion’ is used in-
stead of ‘Heisenberg’s equation’.
However, some exceptions are inevitable: certain terms used in modern physics have
no real alternatives. The Boltzmann constant, the Planck scale, the Compton wavelength,
the Casimir effect, Lie groups and the Virasoro algebra are examples. In compensation,
the text makes sure that you can look up the definitions of these concepts using the index.
Ref. 1181 In addition, it tries to provide pleasurable reading.

Copyright © Christoph Schiller November 1997–May 2006

which it changed to Latin vocabulary with British/American grammar.

Many units of measurement also date from Roman times, as explained in the next appendix. Even the
Ref. 1180 infatuation with Greek technical terms, as shown in coinages such as ‘gyroscope’, ‘entropy’ or ‘proton’, dates
from Roman times.

Dvipsbugw
1152 a notation and conventions

Biblio graphy

1168 For a clear overview of the various sign conventions in general relativity, see the front cover
of Charles W. Misner, Kip S. Thorne & John A. Wheeler, Gravitation, Free-
man, 1973. We use the gravitational sign conventions of Hans C. Ohanian & R emo
Ruffini, Gravitazione e spazio-tempo, Zanichelli, 1997. Cited on page 1142.
1169 For more information about the letters thorn and eth, have a look at the extensive report
to be found on the website http://www.everytype.com/standards/wynnyogh/thorn.html. Dvipsbugw
Cited on page 1143.
1170 For a modern history of the English language, see David Crystal, The Stories of English,
Allen Lane, 2004. Cited on page 1143.
1171 Hans Jensen, Die Schrift, Berlin, 1969, translated into English as Sign, Symbol and Script:
an Account of Man’s Efforts to Write, Putnam’s Sons, 1970. Cited on page 1144.
1172 David R. Lide, editor, CRC Handbook of Chemistry and Physics, 78th edition, CRC Press,

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

1997. This classic reference work appears in a new edition every year. The full Hebrew al-
phabet is given on page 2-90. The list of abbreviations of physical quantities for use in
formulae approved by ISO, IUPAP and IUPAC can also be found there.
However, the ISO 31 standard, which defines these abbreviations, costs around a thou-
sand euro, is not available on the internet, and therefore can safely be ignored, like any
standard that is supposed to be used in teaching but is kept inaccessible to teachers. Cited
on pages 1146 and 1147.
1173 See the mighty text by Peter T. Daniels & William Bright, The World’s Writing
Systems, Oxford University Press, 1996. Cited on page 1146.
1174 See for example the article ‘Mathematical notation’ in the Encyclopedia of Mathematics,
10 volumes, Kluwer Academic Publishers, 1988–1993. But first all, have a look at the in-
formative and beautiful http://members.aol.com/jeff570/mathsym.html website. The main
source for all these results is the classic and extensive research by Florian Cajori, A
History of Mathematical Notations, 2 volumes, The Open Court Publishing Co., 1928–1929.
The square root sign is used in Christoff Rud olff, Die Coss, Vuolfius Cephaleus
Joanni Jung: Argentorati, 1525. (The full title was Behend vnnd Hubsch Rechnung durch
die kunstreichen regeln Algebre so gemeinlicklich die Coss genent werden. Darinnen alles so
treülich an tag gegeben, das auch allein auss vleissigem lesen on allen mündtliche vnterricht
mag begriffen werden, etc.) Cited on page 1147.
1175 J. Tschichold, Formenwamdlungen der et-Zeichen, Stempel AG, 1953. Cited on page
1149.
Copyright © Christoph Schiller November 1997–May 2006

1176 Malcolm B. Parkes, Pause and Effect: An Introduction to the History of Punctuation
in the West, University of California Press, 1993. Cited on page 1149.
1177 This is explained by Berthold Louis Ullman, Ancient Writing and its Influence, 1932.
Cited on page 1149.
1178 Paul Lehmann, Erforschung des Mittelalters – Ausgewählte Abhandlungen und Aufsätze,
Anton Hiersemann, 1961, pp. 4–21. Cited on page 1149.
1179 Bernard Bischoff, Paläographie des römischen Altertums und des abendländischen
Mittelalters, Erich Schmidt Verlag, 1979, pp. 215–219. Cited on page 1149.
1180 The connections between Greek roots and many French words – and thus many English
ones – can be used to rapidly build up a vocabulary of ancient Greek without much study, as
shown by the practical collection by J. Chaineux, Quelques racines grecques, Wetteren

Dvipsbugw
bibliography 1153

– De Meester, 1929. See also Donald M. Ayers, English Words from Latin and Greek
Elements, University of Arizona Press, 1986. Cited on page 1151.
1181 To write well, read William Strunk & E.B. White, The Elements of Style, Macmillan,
1935, 1979, or Wolf Schneider, Deutsch für Kenner – Die neue Stilkunde, Gruner und
Jahr, 1987. Cited on page 1151.

Dvipsbugw

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Copyright © Christoph Schiller November 1997–May 2006

Dvipsbugw
 Appendix B

UNI T S, MEA SUR EMENT S AND

C ONSTANT S
Dvipsbugw

M
easurements are comparisons with standards. Standards are based on a unit.
any different systems of units have been used throughout the world.
ost standards confer power to the organization in charge of them. Such power
can be misused; this is the case today, for example in the computer industry, and was so
in the distant past. The solution is the same in both cases: organize an independent and

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

global standard. For units, this happened in the eighteenth century: to avoid misuse by
authoritarian institutions, to eliminate problems with differing, changing and irreprodu-
cible standards, and – this is not a joke – to simplify tax collection, a group of scientists,
politicians and economists agreed on a set of units. It is called the Système Interna-
tional d’Unités, abbreviated SI, and is defined by an international treaty, the ‘Convention
du Mètre’. The units are maintained by an international organization, the ‘Conférence
Générale des Poids et Mesures’, and its daughter organizations, the ‘Commission Interna-
tionale des Poids et Mesures’ and the ‘Bureau International des Poids et Mesures’ (BIPM),
Ref. 1182 which all originated in the times just before the French revolution.
All SI units are built from seven base units, whose official definitions, translated from
French into English, are given below, together with the dates of their formulation:
‘The second is the duration of 9 192 631 770 periods of the radiation corresponding
to the transition between the two hyperfine levels of the ground state of the caesium 133
atom.’ (1967)*
‘The metre is the length of the path travelled by light in vacuum during a time interval
of 1/299 792 458 of a second.’ (1983)
‘The kilogram is the unit of mass; it is equal to the mass of the international prototype
of the kilogram.’ (1901)*
‘The ampere is that constant current which, if maintained in two straight parallel
conductors of infinite length, of negligible circular cross-section, and placed 1 metre apart
Copyright © Christoph Schiller November 1997–May 2006

in vacuum, would produce between these conductors a force equal to 2 ë 10−7 newton per
metre of length.’ (1948)
‘The kelvin, unit of thermodynamic temperature, is the fraction 1/273.16 of the ther-
modynamic temperature of the triple point of water.’ (1967)*
‘The mole is the amount of substance of a system which contains as many elementary
entities as there are atoms in 0.012 kilogram of carbon 12.’ (1971)*
‘The candela is the luminous intensity, in a given direction, of a source that emits
monochromatic radiation of frequency 540 ë 1012 hertz and has a radiant intensity in that
direction of (1/683) watt per steradian.’ (1979)*

* The respective symbols are s, m, kg, A, K, mol and cd. The international prototype of the kilogram is a

Dvipsbugw
 b units, measurements and constants 1155

Note that both time and length units are defined as certain properties of a standard
example of motion, namely light. This is an additional example making the point that the
observation of motion as the fundamental type of change is a prerequisite for the defin-
ition and construction of time and space. By the way, the use of light in the definitions
had been proposed already in 1827 by Jacques Babinet.*
From these basic units, all other units are defined by multiplication and division. Thus,
all SI units have the following properties:
SI units form a system with state-of-the-art precision: all units are defined with a pre-
cision that is higher than the precision of commonly used measurements. Moreover, the Dvipsbugw
precision of the definitions is regularly being improved. The present relative uncertainty
of the definition of the second is around 10−14 , for the metre about 10−10 , for the kilogram
about 10−9 , for the ampere 10−7 , for the mole less than 10−6 , for the kelvin 10−6 and for
the candela 10−3 .
SI units form an absolute system: all units are defined in such a way that they can
be reproduced in every suitably equipped laboratory, independently, and with high pre-

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

cision. This avoids as much as possible any misuse by the standard-setting organization.
(The kilogram, still defined with help of an artefact, is the last exception to this require-
ment; extensive research is under way to eliminate this artefact from the definition – an
international race that will take a few more years. There are two approaches: counting
particles, or fixing ħ. The former can be achieved in crystals, the latter using any formula
where ħ appears, such as the formula for the de Broglie wavelength or that of the Joseph-
son effect.)
SI units form a practical system: the base units are quantities of everyday magnitude.
Frequently used units have standard names and abbreviations. The complete list includes
the seven base units, the supplementary units, the derived units and the admitted units.
The supplementary SI units are two: the unit for (plane) angle, defined as the ratio of arc
length to radius, is the radian (rad). For solid angle, defined as the ratio of the subtended
area to the square of the radius, the unit is the steradian (sr).
The derived units with special names, in their official English spelling, i.e. without
capital letters and accents, are:

Copyright © Christoph Schiller November 1997–May 2006

platinum–iridium cylinder kept at the BIPM in Sèvres, in France. For more details on the levels of the caesium
Ref. 1183 atom, consult a book on atomic physics. The Celsius scale of temperature θ is defined as: θ~°C = T~K−273.15;
note the small difference with the number appearing in the definition of the kelvin. SI also states: ‘When the
mole is used, the elementary entities must be specified and may be atoms, molecules, ions, electrons, other
particles, or specified groups of such particles.’ In the definition of the mole, it is understood that the carbon
12 atoms are unbound, at rest and in their ground state. In the definition of the candela, the frequency of the
light corresponds to 555.5 nm, i.e. green colour, around the wavelength to which the eye is most sensitive.
* Jacques Babinet (1794–1874), French physicist who published important work in optics.

Dvipsbugw
 1156 b units, measurements and constants

Name A b b r e v i at i o n Name A b b r e v i at i o n

hertz Hz = 1~s newton N = kg m~s2

pascal Pa = N~m2 = kg~m s2 joule J = Nm = kg m2 ~s2
watt W = kg m2 ~s3 coulomb C = As
volt V = kg m2 ~As3 farad F = As~V = A2 s4 ~kg m2
ohm Ω = V~A = kg m2 ~A2 s3 siemens S = 1~Ω
weber Wb = Vs = kg m2 ~As2 tesla T = Wb~m2 = kg~As2 = kg~Cs Dvipsbugw
henry H = Vs~A = kg m2 ~A2 s2 degree Celsius °C (see definition of kelvin)
lumen lm = cd sr lux lx = lm~m2 = cd sr~m2
becquerel Bq = 1~s gray Gy = J~kg = m2 ~s2
sievert Sv = J~kg = m2 ~s2 katal kat = mol~s

We note that in all definitions of units, the kilogram only appears to the powers of 1, 0
and -1. The final explanation for this fact appeared only recently. Can you try to formulate

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Challenge 1524 ny the reason?
The admitted non-SI units are minute, hour, day (for time), degree 1X = π~180 rad,
minute 1′ = π~10 800 rad, second 1′′ = π~648 000 rad (for angles), litre and tonne. All
other units are to be avoided.
All SI units are made more practical by the introduction of standard names and abbre-
viations for the powers of ten, the so-called prefixes:*

Power Name Power Name Power Name Power Name

101 deca da 10−1 deci d 1018 Exa E 10−18 atto a
102 hecto h 10−2 centi c 1021 Zetta Z 10−21 zepto z
103 kilo k 10−3 milli m 1024 Yotta Y 10−24 yocto y
106 Mega M 10−6 micro µ unofficial: Ref. 1184
109 Giga G 10−9 nano n 1027 Xenta X 10−27 xenno x
1012 Tera T 10−12 pico p 1030 Wekta W 10−30 weko w
1015 Peta P 10−15 femto f 1033 Vendekta V 10−33 vendeko v
36
10 Udekta U 10−36 udeko u

SI units form a complete system: they cover in a systematic way the complete set of
Copyright © Christoph Schiller November 1997–May 2006

observables of physics. Moreover, they fix the units of measurement for all other sciences
as well.

* Some of these names are invented (yocto to sound similar to Latin octo ‘eight’, zepto to sound similar to
Latin septem, yotta and zetta to resemble them, exa and peta to sound like the Greek words for six and five,
the unofficial ones to sound similar to the Greek words for nine, ten, eleven and twelve); some are from
Danish/Norwegian (atto from atten ‘eighteen’, femto from femten ‘fifteen’); some are from Latin (from mille
‘thousand’, from centum ‘hundred’, from decem ‘ten’, from nanus ‘dwarf ’); some are from Italian (from piccolo
‘small’); some are Greek (micro is from µικρός ‘small’, deca/deka from δέκα ‘ten’, hecto from ἑκατόν ‘hundred’,
kilo from χίλιοι ‘thousand’, mega from µέγας ‘large’, giga from γίγας ‘giant’, tera from τέρας ‘monster’).
Translate: I was caught in such a traffic jam that I needed a microcentury for a picoparsec and that my
Challenge 1525 e car’s fuel consumption was two tenths of a square millimetre.

Dvipsbugw
 b units, measurements and constants 1157

SI units form a universal system: they can be used in trade, in industry, in commerce,
at home, in education and in research. They could even be used by extraterrestrial civil-
izations, if they existed.
SI units form a coherent system: the product or quotient of two SI units is also an SI
unit. This means that in principle, the same abbreviation, e.g. ‘SI’, could be used for every
unit.
The SI units are not the only possible set that could fulfil all these requirements, but
they are the only existing system that does so.*
Remember that since every measurement is a comparison with a standard, any meas- Dvipsbugw
urement requires matter to realize the standard (yes, even for the speed standard), and
radiation to achieve the comparison. The concept of measurement thus assumes that mat-
Page 1104 ter and radiation exist and can be clearly separated from each other.

Planck’s natural units

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

Since the exact form of many equations depends on the system of units used, theoretical
physicists often use unit systems optimized for producing simple equations. The chosen
units and the values of the constants of nature are related. In microscopic physics, the
system of Planck’s natural units is frequently used. They are defined by setting c = 1, ħ = 1,
G = 1, k = 1, ε 0 = 1~4π and µ 0 = 4π. Planck units are thus defined from combinations
of fundamental constants; those corresponding to the fundamental SI units are given in
Table 83.** The table is also useful for converting equations written in natural units back
Challenge 1526 e to SI units: just substitute every quantity X by X~X Pl .

TA B L E 83 Planck’s (uncorrected) natural units

Name Definition Va l u e

¼
Basic units
= ħG~c 3 = 1.616 0(12) ë 10−35 m
¼
the Planck length l Pl
= = 5.390 6(40) ë 10−44 s
¼
the Planck time t Pl ħG~c 5
= =
¼
the Planck mass m Pl ħc~G 21.767(16) µg
= 4πε 0 c ~G
6
= 3.479 3(22) ë 1025 A
¼
the Planck current I Pl
the Planck temperature TPl = ħc 5 ~Gk 2 = 1.417 1(91) ë 1032 K
Copyright © Christoph Schiller November 1997–May 2006

* Most non-SI units still in use in the world are of Roman origin. The mile comes from milia passum, which
used to be one thousand (double) strides of about 1480 mm each; today a nautical mile, once defined as
minute of arc on the Earth’s surface, is exactly 1852 m). The inch comes from uncia/onzia (a twelfth – now of
a foot). The pound (from pondere ‘to weigh’) is used as a translation of libra – balance – which is the origin
of its abbreviation lb. Even the habit of counting in dozens instead of tens is Roman in origin. These and all
other similarly funny units – like the system in which all units start with ‘f ’, and which uses furlong/fortnight
as its unit of velocity – are now officially defined as multiples of SI units.
** The natural units x Pl given here are those commonly used today, i.e. those defined using the constant
ħ, and not, as Planck originally did, by using the constant h = 2πħ. The electromagnetic units can also be
defined with other factors than 4πε 0 in the expressions: for example, using 4πε 0 α, with the fine structure
constant α, gives q Pl = e. For the explanation of the numbers between brackets, the standard deviations, see
page 1164.

Dvipsbugw
 1158 b units, measurements and constants

Name Definition Va l u e

Trivial units
the Planck velocity v Pl = c = 0.3 Gm~s
the Planck angular momentum L Pl = ħ = 1.1 ë 10−34 Js
the Planck action S aPl = ħ = 1.1 ë 10−34 Js
the Planck entropy S ePl = k = 13.8 yJ~K
Dvipsbugw
Composed units
the Planck mass density ρ Pl = c 5 ~G 2 ħ
¼ = 5.2 ë 1096 kg~m3
= ħc 5 ~G = 2.0 GJ = 1.2 ë 1028 eV
¼
the Planck energy E Pl
the Planck momentum p Pl = ħc 3 ~G = 6.5 Ns
the Planck force FPl = c 4 ~G = 1.2 ë 1044 N
the Planck power PPl = c 5 ~G
¼ = 3.6 ë 1052 W

Motion Mountain – The Adventure of Physics available free of charge at www.motionmountain.net

= c 7 ~ħG = 5.6 ë 1051 m~s2
¼
the Planck acceleration a Pl
= c 5 ~ħG = 1.9 ë 1043 Hz
¼
the Planck frequency f Pl
= = 1.9 aC = 11.7 e
¼
the Planck electric charge q Pl 4πε 0 cħ
the Planck voltage U Pl = c 4 ~4πε 0 G = 1.0 ë 1027 V
the Planck resistance R Pl = 1~4πε¼ 0c = 30.0 Ω
= = 1.8 ë 10−45 F
¼
the Planck capacitance C Pl 4πε 0 ħG~c 3
= (1~4πε 0 ) ħG~c 7 = 1.6 ë 10−42 H
¼
the Planck inductance L Pl
= c 7 ~4πε 0 ħG 2 = 6.5 ë 1061 V~m
¼
the Planck electric field E Pl
the Planck magnetic flux density B Pl = c 5 ~4πε 0 ħG 2 = 2.2 ë 1053 T

The natural units are important for another reason: whenever a quantity is sloppily called
‘infinitely small (or large)’, the correct expression is ‘as small (or as large) as the corres-
ponding corrected Planck unit’. As explained throughout the text, and especially in the
Page 1076 third part, this substitution is possible because almost all Planck units provide, within a
correction factor of order 1, the extremal value for the corresponding observable – some
Copyright © Christoph Schiller November 1997–May 2006

an upper and some a lower limit. Unfortunately, these correction factors are not yet widely
known. The exact extremal value for each observable in nature is obtained when G is sub-
stituted by 4G, ħ by ħ~2, k by k~2 and 4πε 0 by 8πε 0 α in all Planck quantities. These
extremal values, or corrected Planck units, are the true natural units. To exceeding the
extremal values is possible only for some extensive quantities. (Can you find out which
Challenge 1527 n ones?)

Other unit systems

A central aim of research in high-energy physics is the calculation of the strengths of
all interactions; therefore it is not practical to set the gravitational constant G to unity,
as in the Planck system of units. For this reason, high-energy physicists often only set

Dvipsbugw
 b units, measurements and constants 1159

c = ħ = k = 1 and µ 0 = 1~ε 0 = 4π,* leaving only the gravitational constant G in the

equations.
In this system, only one fundamental unit exists, but its choice is free. Often a stand-