Unit R: The La

Frame-I

Third Edition

Thomas A. Moore

The Pri nc p e
of Relativity
I

Chapter Overview
Introduction
In units C and N, we have explored the Newtonian model of mechanics. In this
uni! we will explore a dffirent model, called the special theory of relatiaity, that better
explains the behavior of objects, especially objects moving at close to the speed of
light. This chapter lays the foundations for that exploration by describing the core
idea of the theory and linking it to Newtonian mechanics.

Section R1.1: lntroduction to the Principle

We can

hformally

state this unit's great idea, the principle of relativity, as follows:

The laws of physics are the same inside a laboratory moving at a constant
velocity as they are inside a laboratory at rest.
The theory of special relativity simply spells out the logical consequences of this idea.
This unit is divided into four subsections. The first (this chapter) discusses the
principle itself. The second (chapters R2 through R4) explores the relativistic concept of time. The third (chapters R5 through R7) discusses how observers in different
reference frames view a sequence of events. Finally, the fourth (chapters R8 and R9)
examines the consequences for the laws of mechanics.

Section R1.2: Events, Coordinates, and Reference Frames

What exactly does the principle mean by a "laboratory"? The first step to understanding this better is to describe operationally how we measure a particle's motion.
An event is something that happens at a well-defined place and time. An event's
spacetime coordinates are a set of four numbers that locate the event in space and
time. A particle's motion is a series of events.
Areference frame is a tool for assigning spacetime coordinates to events. We can
visualize a reference frame as being a cubical lattice with a clock at every intersection.
This ensures that there is a clock present at every event, but it also implies that we
must synchronize the clocks somehow. An observer is a person who interprets results
obtained in a reference frame to reconstruct the motions of particles. A real reference
frame does not actually consist of a cubical lattice of clocks, but must be functionally

equivalent.

Section R1.3: lnertial Reference Frames

An inertial reference frame is a frame in which an isolated object is alwys and
to move at a constant velocity. We can check whether a frame is
inertial by distributing first-law detectors around the frame to test for violations of
Newton's first law.
A consequence of this definition is that two inertial frames in the same region of
space must move at a constant velocity relative to each other. Conversely, if a given
frame moves at a constant velocity relative to another inertial frame in the same
region of space, the first must be inertial also.
euerywhere observed

Section R1.4: The Final Principle of Relativity

Note that in our original statement of the principle of relativity, the "laboratory
moving at a constant velocity" and the "laboratory at rest" arcboth inertial frames.
Moreover, the principle itself implies that there is no physical way to distinguish a
frame in motion from one at rest: only the relative velocity between two inertial reference frames is measurable. Our final, polished statement of the principle therefore
expresses the core issue without referring to "moving" or being af "rcst,,:
The laws of physics are the same in all inertial reference frames.

Section R1.5: Newtonian Relativity

What does the phrase "the laws of physics are the same" mean? We can examine
this issue in the context of Newtonian physics if we temporarily embrace Newton's

hypothesis ab o
I and absolute and thus independent
of reference fra
at have constant relative velocity
and which are
tive to each other; that is, the axes of both
point in the same directions in space and the Other (primed) Frame moves along the
axis relative to the Home (unprimed) Frame. The concept of universal time implies
that time measured in both frames is the same (t = t'). This, together with some simple
vector addition and a little bit of calculus, implies that

l't' =@

-t
'(t') -( -
't' : (

(R1.1)
(R1.3)
(R1.4)

o Purpose: These equations describe how to compute an object's position 1,,

velocity d'and acceleration'at any given time f in the Other Frame, given
the object's positior position v , velocity , and acceleration at the same time f
in the Home Frame, where B is the velocity of the other Frame relative to the
Home Frame.
o Limitations: These equations assume that f : f ', which is not true unless both
<< c and 1< c, for reasons we are about to discover.
o Note: Equation R1.1 and the equation = f 'together comprise the Galilean
transformation equations; equation R1.3 represents the Galilean velocitr
transformation equations.

Equation R1.3 implies that Newton's second law is frame-independent: the vector
sum of the physical forces acting on an object will be equal to the object's mass times
its acceleration in all inertial frames. This is an example of the laws of physics being
the same in two frames (the problems explore some others).

Section R1.: The Problem of Electromagnetic Waves

Maxwell's equations of electromagnetism predict that light (electromagnetic waves)
must move with a certain specific speed c. In the 19th century, people assumed that
electromagnetic waves moved through a hypothetical medium called the ether and
that Maxwell's equations only really applied in the ether's rest frame. rnother frames,
then, the speed of light would be more or less than c according to the Galilean velocity
transformation equation. However, 19th-century experiments seeking to check thii
showed that the speed of light seemed to have the same numerical value in aII rcfterence frames and failed to find any evidence supporting the ether hypothesis.
Einstein asserted that the si
e was to reject the
ether hypothesis and instead as
tisfy the principle
of relativity. This means that the
be the same irrall
inertial reference frames. But this contradicts the Galilean transformation equations.

"co

aa

aa

aa

aa

aa

aa

aa

aaaaa

aaataaaaaaa

aaaaaaaaaa

Chapter

R1.1

R1

The Principle of Relativity

lntroduction to the Principle

Everyday experience with the

principle of relativity

to the earth in the time that it takes the cup to reach the floor). Aball thrown
up in the airby a child in the seat in front of you falls straight back into the
child's lap (instead of being swept back toward you at hundreds of miles per
hour). Your watch, the attendants'microwave overL and the plane's instruments behave just as they would if they were at rest on the ground.
Indeed, suppose you were confined to a small, windowless, and soundproofed room in the plane during a stretch of exceptionally smooth flying. Is
there any physical experiment you could perform entirely within the room
(that is, an experiment that would not depend on any information coming
from beyond the room's walls) that would indicate whether or how fast the
plane was moving?
The answer to this question appears to be "no." No one has ever found

in a laboratory moving al a constant velocity than it does when the laboratory is at

a convincing physical experiment that yields a different result

rest. The designers of the plane's electronic instruments do not have to use
different laws of electromagnetism to predict the behavior of those instruments when the plane is in flight than they do when the plane is at rest.
Scientists working to improve the performance of theVoynger 2 space probe
tested out various techniques on an identical model of the probe at rest on
eart confident that if the techniques worked for the earth-based model,
they would work for the actual probe, even though the actual probe was
moving relative to the earth at nearly 72,000 km/h. Astrophysicists are able
to explain and understand the behavior of distant galaxies and quasars by

An informal statement of the

principle of relativity

using physical laws developed in earth-based laboratories, even though

such objects move with respect to the earth at substantial fractions of the
speed of light.
In short, all available evidence suggests that we can make the following
general statement about the way the universe is constructed:
The laws of physics are the same inside a laboratory moving at
constant velocity as they are in a laboratory at rest.

This is an unpolished statement of what we will call the principle of relativity.

This simple idea, based on commory everyday experience, is the foundation

Historical notes

of Einstein's special theory of relativity. All of that theory's mind-bending

predictions about space and time follow as logicl consequences of this simple principle! The remainder of this book is little more than a step-by-step
unfolding of this statement's rich implications.
The principle of relativity is both a very new and a very old idea. It was
not first stated by Einstein but rather by Galileo Galilei in his book Dinlog
Concerning the Two Chief World Systems-(1,632). (Galileo's vivid and entertaining description of the principle of relativity is a wonderful example of a style
of discourse that has, unfortunately, become archaic.) Lr the nearly three centuries that passed between Galileo's statement and Einstein's first paper on
special relativity in 1905, the principle of relativity as it applied to the laws of
mechnics was widely understood and used (in fact, it was generally considered to be a consequence of the particular nature of Newton's laws).
What Einstein did was to assert the applicability of the principle of
relativity to nII tlne laws of physics, and most particularly to the laws of

R1.1 lntroduction to the Principle

electromagnetism (which had just been developed and thus were completely
unknown to Galileo). Thus, Einstein did not inaent the principle of relativity; rather, his main contribution was to reinterpret it asbeingfundamental,
more fundamental than Newton's laws or even than ideas about time that up
to that point had been considered obvious and inescapable, and to explore
insightfully its implications regarding the nature of ligh time, and space.
Our task in this text is to work out the rich and unexpected consequences
of this principle. Figure R1.1 illustrates how we will proceed to do this in the
remaining chapters of this unit.
This unit is divided into four subsections, as shown in figure R1.1. The
first (this chapter) deals with the foundation of relativity theory, the principle
of relativity itself. The second subsection (chapters R2 through R4) explores
the implications of the principle of relativity regarding the nature of time,
with special emphasis on the metric equation, an equation that links space
and time into a geometric unity that we call spacetim. The third subsection
(chapters R5 through R7) examines how the metric equation and the principle of relativity together determine how observers in different reference
frames will interpret the same sequence of physical events; lnow disgreements
between such observers lead to the phenomenon of length contraction; and
how the fact that all observers must agree about cause and effect implies that
nothing can go faster than light. The final subsection (chapters R8 and R9)
explores why and how we must redefine energy and momentum somewhat
to make conservation of energy and conservation of momentum consistent
with the principle of relativity.
Our focus in this chapter is on the principle of relativity itself, and on
developing an understanding of its meaning in the context of Newtonian
physics, before we proceed to explore the changes that Einstein proposes.
It is important before we proceed , however, to understand two important
things about the principle of relativity: (1) it is a postulate and (2) we must
state it more precisely before we can extract any of its logical implications.
The principle of relativity is one of those core physical assumptions (like
Newton's second law or the law of conservation of energy) that must be
accepted on faith: it cannot be proaed experimentally (it is not possible even
in principle to test eaery physical law in every laboratory moving at a constant velocity) or logically derived from more basic ideas. The value of such
a postulate rests entirely on its ability to provide the foundation for a model
of physics that successfully explains and illuminates experimental results.
The principle of relativity has weathered nearly a century of intense critical examination. No contradiction of the principle or its consequences has
eaer been conclusively demonstrated. Moreover, the principle of relativity
has a variety of unusual and unexpected implications ti'rat hae been verified,
(to an extraordinary degree of accuracy) to occur exactly as predicted. Therefore, while it cannot be proaed, it has not yet been disproaed, and physicists
find it to be something in which one can confidently believe. The prnciple
of relativity, simple as it is, is a very rich and powerful idea, and one that the
physics community has found to be not only helpful but crucil in the understanding of much of modern physics.
Turning to the other problem, we see that the principle of relativity, as
we have just stated it, suffers from certain problems of both abstraction and
ambiguity. For example, what do we men by "the laws of physics are the
same"? What exactly do we mean by "a laboratory at rest"? How can we
tell if a laboratory is " at test" or not? If we intend to explore the logical consequences of any idea, it is essential to state that idea in such a way that its
meaning is clear and unambiguous, Our task in the remaining sections of this
chapter is to address this problem.

An overview of the unit

Foundations

The Principle of Relativity

Time

Coordinate Time

o The Spacetime Interval

Proper Time

Comparing Frames

Coordinate Transformations

o Lorentz Contraction
. The Cosmic Speed Limit

Relativistic Dynamics

Four-Momentum

o Conservation of FourMomentum

Figure R1.1
An outline illustrating the four
subsections of unit R.
The principle of relativity is a
postulate

Our informal statement of the

principle needs clarification

Chapter

R1.2
Our task is to specify what we
really mean by "laboratory"

R1

The Principle of Relativity

Events, Coordinates, and Reference Frames

The principle of relativity, as we have stated it so far, asserts that the laws of
physics are the same in a laboratory moving at a constant velocity as they are
in a laboratory at rest. Alaboratory in this context is presumably a place where

one performs experiments that test the laws of physics. How can we more
carefully state what we mean by this term?
The most fundamental physical laws describe how particles interact with
one another and how they move in response to such interactions. So perhaps
what a physicist seeking to specify and test the laws of physics needs most is
a means of mathematically describing the motion of a particle in space.
As we develop the theory of relativity, we need tobe oery careful about

Definition of eaent

We can describe motion

terms of events

in

x
a

a
a

Biink
events
o

Figure R1.2
We can sketch out a graph of an
object's motion (position vs, time)
by plotting the "blink events" that
occur along the object's path.

describing exactly how we will measure the motion of particles (hidden

assumptions about the measurement process have plagued thinkers both
before and after Einstein). In what follows, I will describe how we can measure the motion of a particle in terms of simple and well-defined concepts
that are based on a minimum of supporting assumptions.
The first of these concepts is described by the technical term event. An
eaent is any physical occurrence that we can consider to happen at a definite place in space and at a definite instant in time. The explosion of a small
firecracker at a particular location in space and at a definite instant in time
is a vivid example of an event. The collision of two particles or the decay of
a single particle at a certain place and time also defines an event. Even the
simple passage of a particle through a given mathematical point in space can
be treated as an event (simply imagine that the particle sets off a firecracker
at that point as it passes by).
Because an event occurs at a specific point in space and at a specific
instant of time, four numbers quantify when and where the event occurs:
three numbers that specify the event's location in some three-dimensional
spatial coordinate system and one number that specifies what time the event
happened. We call these four numbers the event's spacetime coordinates.
Note that the exact values of an event's spacetime coordinates depend
on certain arbitrary choices, such as the origin and orientation of the spatial
coordinate axes and what time we define to be f : 0. Once these choices are
made and consistently used, however, specifying the coordinates of physical
events provides a useful method of mathematically describing motion.
Specifically, we can quantify a particle's motion by treating it as a series
of events. For example, imagine an airplane moving along the r axis of some
coordinate system. The airplane carries a blinking light. Each blink of the
light is an event in the sense that we are using the word here: it occuls at a
definite place in space and at a definite instant of time. We can describe the
plane's motion by plotting a graph of the position coordinate of each "blink
event" versus the time coordinate of the same, as illustrated in figure R1.2.
If we decrease the time between blink events, we get an even more detailed
picture of the plane's motion. We can in fact describe the plane's motion to
whatever accuracy we need by listing the spacetime coordinates of a sufficiently large number of blink events distributed along its path.
The preceding is a specific illustration of a general idea: the motion of
any particle can be mathematically described to arbitrary accuracy by specifying the spacetime coordinates of a sufficiently large number of events suitably distributed along its path. Studying the motion of particles is the most
basic way to discover and test the laws of physics. Therefore, the most fundamental task of a "Iaboratory" (as a place in which the laws of physics are to be
tested) is to proaide a means of mesuring the spcetime coordintes of

ersents.

R1

.2

Events, Coordinates, and Reference Frames

T
v

4\ \
A

Figure R1.3
A reference frame visualized as a cubical
lattice with a clock at every lattice intersection
This figure (and indeed the whole approach
in this section) is adapted from E. F. Taylor
and J. A. Wheeler, Spacetime Physics, San
Francisco: Freeman, 1966, pp. 17-18.

Now, we have already discussed in chapter C3 how we can quantify the

spatial coordinates of an object (and thus presumably an event) using a cubical lattice of measuring sticks (or something equivalent). In our past discussions of reference frames, however, we did not really face the issue of how
one might measure th.:e time of an event: we simply assumed this could be
done in some simple and obvious way. To proceed with our discussion of
relativity, we now need to face this issue squarely. In what follows, we will
extend the cubical lattice model of a refernc" itutne to include a mechanism for measuring time in such a way that we can clearly distinguish the
approaches to time implicit in Newtonian mechanics and special relativity,
The trick is to imagine that we attach a clock to every lattice intersection
(see figure R1.3). We can then define tlne time coordinate of an event (such as
a firecracker explosion) to be the time displayed on the lattice clock nearest
the event (relative to some specified time f : 0) and the event's three spntial
coordinates to be the lattice coordinates of that nearest clock, specified in the
usual way by stating the distances along the lattice directions that one has
to travel (from some specified spatial origin) to the clock. We can determine
these four numbers to whatever precision we want by sufficiently decreasing
the lattice spacing and the time between clock ticks.
Why must we have a clock at euery lattice intersection? The point is to
ensure that there is a clock essentially af the location of any event we wish to
measure. If we attempt to read the the event's time by using a clock located
a substantial distance away, we need to make assumptions about how long
it took the information that the event has occurre d to rech that clock. For
example, if we read the time when the sound rrom an event reaches the distant
clock, then we should correct that value by subtracting the time it takes sound
to travel from the event to the clock. However, to do this, we have to know the
speed of sound in our lattice. We can avoid this problem if we require that an
event's time coordinate be measured by a clock present at that event.
Note the clocks must all be synchronized in some meaningful and selfconsistent manner if we are to get meaningful results. If these clocks are not
synchronized, adjacent clocks might differ wildly, thus giving one a totally
incoherent picture of when a particle moving through the lattice passes
various lattice points. What exactly we mean when we say that our "lattice
clocks are synchronlzed" is precisely where Newton's and Einstein's models
diverge. We will discuss this issue later: for now, it is sufficient to recognize
that we must slmchronize the clocks somehow.

The operational definition of

spacetime coordinates

Chapter

R1

The Principle of Relativity

Once we have specified a synchronization method, the image of a clock

lattice completely defines a procedure that we can use (in principle) to determine an event's spacetime coordinates. This amounts to an operational
definition of spacetime coordinates: An operationl definition of a physical
quantity defines that quantity by describing how we measure it. Operational
definitions provide a useful way of anchoring slippery human words to
physical reality by linking the words to specific, repeatable procedures rather

than to vague comparisons or analogies.

The procedure just descibed represents an admittedly idealized method

for determining an event's spacetime coordinates. The actual methods

employed by physicists may well differ from this description, but these

Technical terms involving

reference frames

methods should be equiaalenf to what is described above: the clock lattice

method defines a standard against which actual methods can be compared.
It is such a simple and direct method that it is inconceivable that any actual
technique could yield different results and still be considered correct and
meaningful.
With this in mind, we define the following technical words to aid us in
future discussions:
A reference frame is defined to be a rigid cubical lattice of appropriately synchronized clocks or itsfunctional equiaalent.

An event's spacetime coordinates in a given reference frame are an

ordered set of four numbers, the first specifying tlne time of the event
as registered by the nearest clock in the lattice, followed by three that
specify the spatial coordinates of that clock in the lattice.
(For example, in a frame oriented in the usual way on the earth's surface, a
firecracker explosion whose spacetime coordinates are [3 s, -3 m,6 m, -1 m]
thus happened 3 meters west, 6 metets north, and 1 meters below the frame's
spatial origin, and 3 seconds after whatever event defines f : 0.)

An observer is defined to be a (possibly hypothetical) person who

interprets measurements made in a reference frame (for example, a
person who interprets the spacetime coordinates collected by a central computer receiving information from all the lattice clocks).
Note that the act of "observing" in the last definition is an acL of interpretation
of measurements generated by the frame apparatus, and that act may have
little or nothing to do with what that observer sees with his or her own eyes.
\tVhen we say that "an observer in such-and-such reference frame observes
such-and-such," we are actually referring to conclusions that the observer
draws from measurements performed using the reference frame lattice.
Figure R1.4 illustrates a computer's "observation" of particle tracks from a
collision between two protons in the Large Hadron Collider. The computer
has reconstructed the particle tracks by collecting dala from the functional
equivalent of a clock-lattice in the particle detector.

R1.3
We can in principle attach a

frame to any object

lnertial Reference Frames

While exploring relativity theory, we will often speak of

a reference frame in
connection with some object. For example, one might refer to "the reference
frame of the surface of the earth" or "the reference frame of the cabin of the
plane" or "the reference frame of the particle." In these cases/ we are being
asked to imagine a clock lattice (or equivalent) fixed to the object in question.
Sometimes the actual frame is referred to only obliquely, as in the phrase

R1.3 lnertial Reference Frames

Figure R1.4
The aftermath of a collision between two protons in the Large
Hadron Collider. The computer reconstructs the particle tracks
by collecting and interpreting data about particle detection
events in the detector. This represents an act of "observation." (Credit: @ CERN/Science Source)

"an observer in the plane cabin finds . . . ." Since obseraer in this text refers to
someone using a reference frame to determine event coordinates, the phrase
presumes the existence of a reference frame attached to the plane's cabin.
A reference frame may be moving or at rest, accelerating, or even rotat-

ing about some axis. The beauty of the definition of spacetime coordinates
given earlier is that we can measure the coordinates of events (and thus measure the motion of oects) in a reference frame no matter how it is moving
(provided only that the clocks in the frame can be s)mchronized in some
meaningful marurer).
However, not all reference frames are equally useful for doing physics.
we saw in chapter N8 that we can divide reference frames into two general
classes: inertial frames and noninertial frames . An inertial frnme is one in
which an isolated object is always and everywhere observed to move at a
constant velocity (as required by Newton's first law); in a noninertil frnme,
such an object moves with a nonconstant velocity in at least some situations.
We can operationally distinguish inertial from noninertial frames using
a first-law detector. Figure R1.5 shows a simple first-law detector. Electrically actuate d " fingers" hold an electrically uncharged and nonmagnetic ball
at rest in the center of an evacuated spherical container. \rVhen the ball is
released, it should remain at rest according to Newton's first law. If it does

Distinguishing inertial from

noninertial frames

not, the frame to which the detector container is attached is noninertial.

(Fingers retracted)

Container

Figure R1.5
A cross-sectional view of a
floati n g-ba I first-law detector.
Electrically actuated "fingers"
hold the ball initially at rest in
the spherical container. (b) After
the fingers are retracted, the ball
(a)

Electrically
actuated

"ltnger"

(a)

Touch-sensitive
surface

ft)

should continue to float at rest

in the containe as long as the
frame to which the container is
attached is inertial.

Chapter

10

Inertial frames move with

constant velocities relative to
each other

R1

The Principle of Relativity

(If we want to operate this detector in a gravitational field, we need to figure

out a way to cancel the gravitational force on the ball without inhibiting its
freedom to move, but principle, this can be done.) If we attach such a firstlaw detector to the clock at each lattice location in our reference frame and
none of thesedetectors registers a violation of Newton's first laW thenwe can
say with confidence that our frame is inertial.
This definition of an inertial frame is simple enough to apply to realistic
examples. For example, while a gravity-compensated detector, as shown in
figure R1.5, would register no violation of the first law if it were attached to
a plane at rest, we know without actually trying it that if the plane began to
accelerate for takeoff, the detector's floating ball would be deflected toward
the rear of the plane by the same (fictitious) force that presses us back into
our chairs. Similarly, we might exPect that detectors in a reference frame
floating in deep space (far from any massive objects) would register no violation of Newton's first law; yet we know that detector balls in a similar frame
that is rotating around its center will be deflected outward by the (fictitious)
centrifugal force in that noninertial frame.
The following statement is an important and useful consequence of the
definition of an inertial reference frame:

Any ineftial reference frame will be observed to move at a constnt

aelocity relative to any other inefttal reference frame. Conversely, a
rigid, nonrotating reference frame that moves at a constant velocity with respect to any other inertial reference frame must itself be
inertial.

Home

frame

Other frame

v'

object
T

Figure R1.
An isolated object at rest in the
Other Frame must move at a
constant velocity with respect to
the Home Frame, so the whole
Other Frame must move at the
same constant velocity relative to
the Home Frame.

We first discussed this issue in chapter N8, but the methods we used there
unfortunately employ certain assumptions about the nature of time that turn
out to be inconsistent with the principle of relativity (as we will see). We can,
however, prove that the statement above follows directly from the definition
of an inertial reference frame without having to make any assumptions about
the nature of time. Here is an argument for the first part of the statement
above; I have left proof of the converse statement an exercise'
Consider two inertial reference frames (see figure R1.6), which we
will call the Home Frame and Other Ftame, respectively. (Home Frme and
Other Frnme are phrases I will use in this text as names of inertial reference
frames, which I will emphasizeby the capitalization.) Since these arc inertil
reference frames, observers will measure an isolated object to move with a
constant velocity in either frame by definition Now supPose that a specific
isolated object happens to be at rest rclative to the Other Frame. Because
such an object must move at a constant velocity in the Other Frame if the
frame is inertial, if that object is initially at rest, it must remin at rest in that
frame. Now let us observe the same object from the FIome Frame. Since the
object is isolated and the Flome Frame is inertial, the object must move at a
constant velocity relative to that frame as well. Because that object is at rest
with respect to the Other Frame, we must therefore observe the entire Other
Frame to move relative to the Home Frame with a constant velocity (the
same constant velocity as the object), consistent with the statement above.

Exercise R1X.1
Using a similar approach, prove the converse part of the statement above
(that is, a rigid reference frame that moves at a constant velocity with respect
to any other inertial reference frame must itselfbe inertial).

R1.5 Newtonian Relativity

R1.4

The Final Principle of Relativity

Our first informal statement of the principle of relativity stated that "the laws
of physics are the same in a laboratory moving at a constant velocity as they
are in a laboratory at rest." We have subsequently developed the idea of a
reference frame to express the essence of what we mean by a "laboratory."
Flowever, how can we physically distinguish a reference frame "moving ai a
constant velocity" from one " at test"?
The short answer is that we cannot! The principle of relativity specifically states that a reference frame moving at a constant velocity is physically
equivalent to a frame at rest. Therefore, there can be no physical basis for
distinguishing a laboratory at rest from another frame moving at a constant
velocity. Imagine you and I are in spaceships coasting at a constant velocity
in deep space. You will consider yourself to be at rest, while I am moving by
you at a constant velocity. I, on the other hand, will consider myself to be
at rest, while you are moving by me at a constant velocity. According to the
principle of relativity, there is no physical experiment that can resolve our
argument about who is "realIy" at rest. We could, of course, agree to choose
one or the other of us to be at rest, but this choice is completely arbitrary.
Therefore, if the principle of relativity is true, there is no basis for assigning an absolute velocity to any reference frame: only the relatiae velocity
between two frames is a physically meaningful concept.
On the other hand, it is plausible that what we really mean by a reference
frame "atrest" is an inertial frame. Moreover, we have just seen that a reference frame moving at a constant velocity relative to it must also be an inertial
frame. Therefore, we can remove both the vague word "laboratory" and the
ambiguity of the concepts "at rest" and "moving at a constant velocity" in
our original statement of the principle of relativity by restating it as follows.
The Principle of Relativity

'11

The laws of physics are the same in nll inertial reference frmes,

Resf has no physical meaning

in relativity

Our final statement of the

principle of relativity

This is our final polished statement of the principle of relativity. It replaces

tlne fuzzy and ambiguous ideas in our original statement with the sharply
and operationally defined idea of an inertial rcference frame. What this principle essentially claims is that if Newton's first law (which describes what
happens to an isolated object) is the same in two given reference frames,
then all the laws of physics are the same in both frames. (Note that the unit's
" great idea" on the front cover is a compressed version of this statement.)

R1.5

Newtonian Relativity

But what exactly do we mean when we say that "the laws of physics are the
same" in two frames? In this sectiory we will discuss the Newtonian assumption about how we can synchronize clocks in an inertial reference frame. We
then use this as a framework to explore what the principle of relativity means
in Newtonian mechanics.
Suppose we have an inertial frame floating in deep space, ready to use.
We would like to use this frame to measure the coordinates of events happening in it so as to test the laws of physics. But an important problem remains to
be solved: how do we slmchronize its clocks?
"The solution is easy," says a Newtonian physicist. "Everyone knows, as
Newton himself asserted, that'time is absolute and flows equably without
regard to anything external.' Any good clock will therefore measue the flow

The Newtonian to clock

synchronization

12

Chapter

R'l

The Principle of Relativity

v'

Figure R1.7
A schematic drawing of two
reference frames in standard
orientation. The spatial origins of
the frames coincided at t: t' : 0
just a little while ago. (You
should imagine the frame lattices
intermeshing so that events can
be recorded in both frames.)

Standard orientation for

inertial reference frames

Consequences of the

Newtonian view of time

o'

o
z

x
Spatia origin
of Home Frame

z'

Spatial origin
of Other Frame

of this absolute time. So we can simply designate one clock to be a master

clock, carry it around to each of the lattice clocks, and synchronize that clock
to the master. Since both the master and lattice clocks measure the flow of
immutable absolute time, the master clock's motion as we carry it from place
to place in the lattice is irrelevant. Once a lattice clock is set to agree with the
master clock, it will certainly remain in agreement with it, since both clocks
measure the flow of absolute time. Indeed, if the master clocks in two different reference frames are in agreement at any given event, then all the clocks
in the fwo frames will always agree. It doesn't matter whether the frames are
in motion with respect to each other; it doesn't even matter if they are inertial
or not. This follows from the self-evident absolute nature of time."
This picture of time is straightforward, credible, and consistent with
ideas about time that most of us already hold. But what are its consequences?
Again consider two inertial frames that we call the Home Frame and the
Other Frame. We will often (but not always!) imagine ourselves to be in the
Home Frame (so that this frame appears lo us lo be at rest). The Other Frame
moves at a constant velocity B w]th respect to the Home Frame according to
the proof given in section R1.4 (P is the "boost" in velocity that one needs to
go from being at rest in the Home Frame to being at rest in the Other Frame).
These frames might in principle have any relative orientation, but it is
conventional in special relativity to use our freedom to choose the orientations to put the two frames in standard orientation, where the Home Frame's
x, y, an:.d z axes point in the same directions as the corresponding axes in
the Other Frame. We conventionally distinguish the Home Frame and Other
Frame axes by referring to the Home Frame axes as x,y, andz and the Other
Frame axes as x' , y ' , artd z' (lhe mark is called a prime). It also is conventional
to define the origin event (the event that defines f : 0 in both frames) to be the
instant at which the spatial origin of one frame passes the origin of the other.
We conventionally choose the common x axis so that the frames move relative to each other along this axis. Finally, we alwnys choose to U" the velocity of the Other Frcme relative to the Home Frame (the velocity of the Home
Fiame relative to the Other frame is thus -F). Signs in mny equations in this
text depend on this convention, so it is wise to memorize this. Figure R1.7
illustrates two frames in standard orientation.
Now consider an object moving in space that periodically emits blinks of
light. Let the spatial position of a certainblink event as measured in the Home
Frame be represented by the vector 7 (f) and the same measured in the Other
Frame by V'(t) (we conventionally write symbols for quantities observed in
the Other Frame with an attached prime). According to our assumption that
time is universal and absolute, observers in both frames should agree at what
time this blink event occurs: t : t'.Tlre position of the spatial origin O'of the
Other Frame in the Home Frame at that time is simply pf, since the Other
Frame moves at a constant velocity p with respect to the Home Frame, and

R1.5 Newtonian Relativity

13

v'

p
Position of
particle at
f

'

timef=f'

Figure R1.8
The relationship between i and 7'
for two inertial reference frames
(assuming that time is universal
and absolute).

we conventionally take both frames' origins to coincide at f : 0. The relationship between the object's positionvectors in the two frames at the time of the
blink is (as shown in figuie R1.8) given byi (t) :7'(t) +4 or

'1t'

-t

(R1.1)

For frames in standard orientation, points along the x axis, meaning that
we can write equation R1.1 in component form as follows:

:t
x':x-t
y, : y
z': z
t'

(reminding us that time is absolute)

(R1.2a)
(R1,2b)

The Galilean transformation

equations

(R1.2c)
(R7.2d)

where we define B without the arrow to be the x component

ot.ltis

there-

fore positive if the Other Frame moves in the *r direction relative to the
Homq Frame and negative if it moves in the -x direction.

Physicists call these

The meaning of B without the

arrow

four equations the Galilean transformation

equations. These equations allow us to find the position of the object at a

given time f in the Other Frame if we know its position at time t : t' inthe
Home Frame (assuming, of course, that time is universal and absolute).
Taking the time derivative of both sides of the last three equations yields
a :Ura

:u!

a _

(Note that since f '

1Jz

(R1.3a)

The Galilean uelocity

(R1.3b)

transformation equations

(R1.3c)

t,7t really doesn't matter that we are taking a derivative

with respect to f'on the left side and a derivative with respect to f on the
right.) These equations tell us how to find an object's velocity in the Other
Frame, given that object's velocity in the Home Frame: We call these equations the Galilean velocity transformation equations.
If we take the time derivative again, we get

a*
al: ao
aL: a"

a'r:

(R1.aa)
(R1.4b)

(Rl.ac)

Equations R1.4 tell us that obseraers in both inertial fratnes agree bout nn object's
accelertion t a giaen time, even though they may well disagree about the
object's position and velocity components at that time!

An object's acceleration is the

same in all inertial reference
frames

14

Chapter

R1

The Principle of Relativity

s
I

A child throws a ball vertically

upward and downward in the
(a)

(b)

cabin of a plane flying at a large

constant horizontal velocity.

From your vantage point on a nearby mountaintop, the ball

seems to follow a shallow parabolic trajectory because
of its large initial horizontal veocity (the same as the plane's)

Figure R1.9
An example illustrating the applcation of the principle of relativity in Newtonian physics. Newton's second law describes
the motion of the ball in both frames.

An illustratior of how the laws

of physics can be the same in
different fr-ames

Now we are in a position to discuss more fully what we might mean by

"the laws of physics are the same" in every inertial reference frame in the
Newtonian context at least. Consider the following example. A child on an
airplane throws a ball vertically into the air and catches it again. As measured in the plane cabin (which we will take to be the Other Frame), the
ball appears to travel vertically along the vertical z axis. Now imagine that
you watch this process from a nearby mountaintop as the plane passes by.
Instead of observing the ball travel vertically up and down, you will instead
observe the ball to follow a shallow parabolic trajectory, because in your
frame (which we will take to be the Home Frame), the ball, plane, and child
all have a considerable horizontal velocity (see figure R1.9).
The ball's motion in these two reference frames looks quite different: it is
vertical in the plane's cabin but parabolic (almost horizontall) in your frame.
Even so, you and an observer on the plane would agree thal (1) the ball has a
certain mass rn (which you and the observer could each measure with your
own balances) and thus should experience a gravitational force of magnitude
mlll actngon it, (2) this force must be the net force on the object while it is in
flight (since nothing else is in contact with the ball, ignoring air friction), and
(3) the ball has the same acceleration in your respective reference frames (see
equations R1.4). Since you agree on the value of m, tli.e magnitude and direction of the net force on the ball, and the acceleration that the ball experiences,
you will both agree that Newton's second law

= mhere)
the ball's motion, even if you disagree

Fn",:

(withF"",

(R1.5)

about the ball's

accurately predicts
initial velocity and thus the exact character of its subsequent motion (that is,
whether it is vertical or parabolic).

R1. The Problem of Electromagnetic Waves

15

In a similar fashion, you might imagine observing a game of pool in

the plane cabin. You on the mountaintop and your friend on the plane will
totally disagree about the initial and final velocities of the balls in any given
collision (since you will observe them to have a large horizontal component
of velocity that your friend does not observe). Even so, you both will find
that the total momentum of the balls just before a given collision is equal to
their total momentum just afterward, consistent with the law of conservation
of momentum (see problems R1M.7 and R1D.2).
This is what it means to say that the "laws of physics are the same" in
different inertial frames. Observers in different inertial frames may disagree
about the ualues of various quantities (particularly positions and velocities),
but each observer will agree that if one takes the mathematical equation
describing a physical law (such as Newton's second law) and substitutes in
the values measured in that observer's frame, one will always find that the
equation is satisfied. In other words, the same basic equations will be found
to describe the laws of physics in all inertial reference frames.

Rt.0
kr

The Problem of Electromagnetc Waves

1864, ]ames Clerk Maxwell published a set of equations (now called Maxwell's
equations) that summarized the laws of electromagnetism in a compact and
elegant form. These equations (which are the focus of the latter part of unit E)
were the culmination of decades of intensive work by many physicists, and
represent one of the greatest achievements of 19th-century physics.
Among the many fascinating consequences of these equations was the
prediction that one could set up traaeling laaes in an electromagnetic field,
much as one can create ripples on the surface of a lake. The speed at which
such electromagnetic waves travel is completely determined by various universal constants appearing in Maxwell's equations, constants whose values
were fixed by experiments involving electrical and magnetic phenomena
and were fairly well known in 1864. The predicted speed of such electromagnetic waves turns out to be about 3.00 x 108 m/s. Light was already known
to have wavelike properties (as demonstrated by experiments performed
by Thomas Young and Augustin-Jean Fresnel) and to travel at roughly this
speed (as measured by Ole Rmer in7675 and ]ean-Bernard Leon Foucault
in 1846). On the basis of this information, Maxwell concluded that light consisted of such electromagnetic waves. Later experiments confirmed Maxwell's bold assertion by showing that the value of the speed of light was
indeed indistinguishable from the value predicted on the basis of the constants in Maxwell's equations. The work of Heinrich Hertz, who was able to
directly generate low-frequency electromagnetic waves (that is, radio waves)
and demonstrate that they had the properties predicted by Maxwell's equations, was particularly compelling.
In short, Maxwell's equations predicted that light waves must travel at
a specific speed c : 3.00 x 108 m/s. The question is, relative to what? The
consensus in the physics community at the time (one that Maxwell shared)
was that electromagnetic waves were oscillations of a hypothetical medium
called the ether, just as sound waves are oscillations in air, and water waves
are oscillations in the surface of a body of water. Physicists therefore generally assumed that light waves would travel at the predicted speed c relative
to this ether, and thus have this speed in a frame in which the ether is at rest.
In alL other inertial reference frames, however, light waves must travel at
a speed different from c. To_see this, imagine a spaceship flying away from a
space station at a velocif P. lblinker on the space station emits a pulse of

Review of Newtonian
approach to relativity

The ether hypothesis

16

Chapter

R1

The Principle of Relativity

Spaceship

Space station

Light flash

l. l.z

Figure R1.10

/l\ +

A light flash chasing a departing

z'

Blinker

spaceship.

light waves toward the departing spaceship (see figure R1.10). Suppose the
space station is at rest with respect to the ether and treat this as the Home
Frame. Observers on the space station will then measure the emitted pulse of
light to move away from the blinker at a speed of c. How rapidly would this
flash of light be observed to travel in a frame fixed to the spaceship?
By construction, both the flash and the spaceship move in the *r direction. The r component of the Galilean velocity transformation equation
(equation R1.3) in this case implies that

Tlcnt,r:ots.r'-

: c -

(R1'6)

So the speed of the light flash, as measured in the frame of the spaceship (that
is, the Other Frame), is c B. This makes sense: if the spaceship happened to
travel at the speed of light (so that c), the flash should intuitiaely appear
to be motionless in the frame of the spaceship and thus never catch up with
it, in agreement with equation R1.2. In any frame moving with respect to the
ether, then, light waves should be measured to have a speed c.
But this means that Maxwell's equations strictly apply only in a certain
inertial reference frame (the frame at rest with respect to the ether), since they
do specify that light waves move with a specific speed c. Presumably some
small modifications would have to be made to these equations to make them
work in frames that are not at rest with respect to the ether.

Physicists fail to detect the

ether

Now, even its proponents admitted that this ether was peculiar stuff. It
fill all space and permeate all objects, and yet be virtually undetectable. It had to have virtually zerc density and viscosity, because it did not
significantly impede anything's motion. But it also had to be extraordinarily
"stiff" with regard to oscillations because the speed of waves in a medium
increases with that medium's stiffness, and c is very large (for comparison,
mechanical waves traveling through solid rockhave speeds of only 6000 m/s).
In 1887, U.S. physicists Albert Michelson and Edward Morley performed
a sensitive experiment designed to prove the existence of this problematic
stuff. If this ether filled all space, the earth must (as a result of its orbital
motion around the sun) be moving through the ether at a speed comparable
to its orbital speed of about 30 km/s. This "ether wind" would make the
speed of light depend on direction: a light wave traveling against the ether
wind would move more slowly than a wave moving across it. So Michelson
and Morley constructed a very sensitive experiment that compared the speed
of two beams of light sent in perpendicular directions in a very clever way.
To the surprise of everyone involved, there was no discernible difference
in the speeds of the two light waves. Michelson and Morely repeated the
experiment with different orientations of the apparatus, at different times of
the year (just in case the earth happened to be at rest with respect to the ether
at the time of the first experiment), as did other physicists. In all cases, the
result was that the speed of light seemed independent of the earth's motion.

had to

R1. The Problem of Electromagnetic Waves

In

17

modern version of the Michelson-Morley experiment performed in

Hall (see PhysicalReaiew Letters,vol.42,p.549,1979)
set up a laser that fed light into a Fabry-Perot cavity, which is essentially a
region of space bounded by two mirrors. The laser's frequency was continually adjusted so that a specific integer number of waves fit between the mirrors. Any variation in the speed of light would cause the number of waves
of a given frequency that fit between the mirrors to increase or decrease,
requiring the electronics to adjust the laser's frequency to keep the number
of waves fixed. The laser and cavity were mounted on a granite turntable
so its orientation relative to any ether wind could be varied. While the table
was rotated, Brillet and Hall observed that the fractional change in the laser
frequency was (1.5 + 2.5) x L0-15. If an ether wind comparable in speed to the
earth's orbital speed l | existed, it should cause a variation in th fractional
frequency on the order of magnitude of ll'c', or about 10-s, more than
a

1978, A. Brillet and ]. L.

million times any possible variation consistent with experiment!

The Michelson-Morley result (and other corroborating results) caused a
ruckus in the physics community, as physicists strove to explain away these
results while saving the basic ether concept. Many explanations were offered
but none provided a satisfactory explanation of all known experimental data.
In 1905, Albert Einstein published a short paper in the European journaI Annlen der Pltysik that changed the direction of physics. In that paper,
Einstein proposed that since it seemed to be impossible to demonstrate the
existence of the ether, the whole concept should be rejected: we should simply accept that light can move in a vacuum. But the vacuum of empty space
provides no anchor for defining a special frame where the speed of light is c.
(\Atrhat would such a frame be attached to?) So if we accept that the speed of
light in a vacuum is c, then we must accept that this speed is c in eaery inertial
reference frame, in direct contradiction to equation R1.6! The assumption that
the speed of light is frame-independent quntity is necessary, Einstein argued, to
make Maxwell's equations consistent with the principle of relativity, as they
are laws of physics that predict a specific value for the speed of light.
But how can one measure a pulse of light to move at the same speed c in
two different frames when those frames are not at rest with respect to each
other? Einstein's bold idea, while neatly sidestepping the difficulties with the
ether concept, seemed impossible to most of his contemporaries.
Flowever, we only have three choices: either the principle of relativity is
wrong, Maxwell's equations are wrong, or the Galilean velocity transformation equations are wrong. By accepting the ether concept, physicists before
Einstein had opted to accept the idea that Maxwell's equations would have
to be modified in frames moving with respect to the ether, thus keeping the
Galilean velocity transformation but implicitly rejecting the full principle of
relativity. Even as evidence against the ether hypothesis became firm and
incontrovertible, rejection of the Galilean velocity transformation, so solidly
based on simple and obvious ideas, seemed absurd.
On the other hand, Einstein's suggestion was elegant in its simplicity.
Throw away the ether, he said. It is an unhelpful hypothesis with no experimental support. Throw away the awkward andbizarre theories that arose to
explain our inability to detect the ether. Embrace instead the beautiful simplicity of the principle of relativity and Maxwell's equations. The speed of
light is then the same in all inertial frames automatically, and the null results
of experiments like the Michelson-Morely experiment are trivially explained.
The cost? The Gnlilean transrmtion equations must be wrong. But what
could possibly be wrong with their derivation? It is the idea of universal and
absolute time that is wrong, argued Einstein. In the next chaptel, we will look
at what the principle of relativity implies about the nature of time.
10

Einstein's proposed solution to

the problem

Chapter

18

R1

The Principle of Relativity

TWO.MINUTE PROBLEMS
R1T.L

\^/hich of the following are (at least nearly) inertial

reference frames and which are not? (Respond T if the
frame is inertial, F if it is noninertial, and C if it is inertial
for everyday pu-rposes. The classification could be debatable,
creating an opportunity to discuss the issues involved.)
(a) A nonrotating frame floating in deep space
(b) A frame floating in deep space that rotates aI1' rev /h
(c) A nonrotating frame attached to the sun
(d) A frame attached to the surface of the earth
(e) Aframe attached to a car moving at a constant velocity
(f) A frame attached to a roller-coaster car

R1T.8

You are in a spaceship traveling away from earth'

You and Mission Control on earth agree that the *x direction is the direction in which your ship is traveling relative
to the earth. If you choose your own frame to be the Home
Frame (so that the earth is the Other Frame), what is the
sign of B, according to the convention established in this
chapter?

A.
B.
C.

Positive
Negative
We are free to choose B to have either sign.

R1T.9
R1T.2

Which of the following physical occurrences fit the

physical definition of an eaent?
A. The collision of two point particles
B. A point particle passing a given point in space
C. Afirecracker explosion
D. A party at your dorm

E. Ahurricane

F.
T.

A, B, and C
Any of the above could be an event, depending on the
reference frame's scale and/or how precise the measurments need to be.

R1T.3

Since the laws of physics are the same in every iner-

tial reference frame, there is no meaningful physical distinction between an inertial frame at rest and one moving
at a constant velocity. True (T) or false (F)?

R1T.4

Since the laws of physics are the same in every reference frame, an object must have the same kinetic energy
in all inertial reference frames. T or F?

R1,T.5 Since the laws of physics are the same in every inertial reference frame, an interaction between objects must
be observed to conserve energy in every inertial reference
frame. T or F?

R1T.6

Since the laws of physics are the same in every iner-

tial reference frame, if you perform identical experiments

in two different inertial frames, you should get exctly the
same results. T or F?

Imagine two boats. One travels 5.0 m/s eastward

relative to the earth and the other 3.4 m/s eastward relative to the earth. We set up a reference frame on each boat
with the x axis pointing eastward, and choose the firstboat
(arbitrarily) to be the Home Frame. The second boat is thus
the Other Frame. What is the sign of B, according to the
convention established in this chapter?

RLT.7

A.
B.

Positive

C.

We are free to choose either sign.

Negative

Suppose you observe

a collision of an isolated

system of two particles. A friend observes the same collision in a reference frame moving in the *r direction with
respect to yours. According to the Galilean transformation
equations, on which aspects of the collision will you agree
with your friend? (Answer T or F.)
(a) On the value of the system's total x-momentum
(b) On the value of the system's total y-momentum
(c) On the value of the system's total z-momentum
(d) On the force F that one particle exerts on the other
(e) That the system's total momentum is conserved

R1T.10 Suppose you are in a train traveling at one-half of

the speed of light relative to the earth. Assuming that photons emitted by the train's headlight travel at the speed of
light relative to you, they would (according to the Galilean
velocity transformation) travel at 1.5 times the speed of
light relative to the earth. T or F?
R1T.11 Suppose you are in a spaceship traveling at twice
the speed of light relative to the earth. Assuming that the
Galilean transformation equations are true and the earth
is approximately at rest relative to the ether, light from the
ship's taillight will never reach the ship's bridge at its front
end. T or F?

R1T.12 Suppose the Galilean transformation equations are

true and your spaceship is moving at twice the speed of
light relative to the ether. \4lhat odd things will you observe
in your spaceship? Select all that apply. (If you are using
the back of the book to communicate your answers/ you can
point to multiple letters with several fingers.)
A. You won't be able to see anything behind you.
B. You won't be able to see anything in front of you.
C. The beam from a laser pointer facing forward and a bit
to your right will get curved toward the ship's stern.
D. Light from stars in front of you will become infinitely
blue-shifted.
E. Stars a bit to the right or left of the forward direction
will have their apparent positions shifted dramatically
toward the shiP's stern'

F. You will see none of these effects.

T. You will see all of these effects.

Homework Problems

19

HOMEWORK PROBLEMS
R18.6

Basic Skills

R18.1 A train moving with a speed of 55 m/s

through a railway station at time t = t'

:0.

passes

Fifteen seconds later a firecracker explodes on the track 1.0 km away

from the train station in the direction the train is moving.
Find the coordinates of this event in both the station frame
(consider this to be the Home Frame) and the train frame.
Assume the train's direction of motion relative to the station defines the *x direction in both frames, and assume
the Galilean transformation equations are true.

R18.2

Suppose we select the rear end of a 120-m-long train

to define the origin r' = 0 in the train frame, and we define
a certain track signal light to define the origin x 0 in the
track frame. Suppose the train's rear end passes this light
att t' 0 as the train moves in the f x direction at a constant speed of 25 m/s. Twelve seconds later, the engineer

turns on the train's headlight. Assume the galilean transformation equations are true.
(a) Where does this event occur in the train frame?
(b) Where does this event occur in the track hame?
(Please explain your response in both cases.)

R18.3

Consider a floating-ball first-law detector like the

one shown in figure R1.4. If the ball is 10 cm in diameter and is placed at I : 0 in the center of a spherical shell
whose inside diameter is 12 cm, about how long will it take
the ball to hit the shell if the shell accelerates af 0.7 m / s2?

Modeling
ll1M.1

Read Galileo's 1632 presentation of the principle of

relativity, as summarized on one of these websites:

en.wikipedia .ory / wiki / Galileo's ship

www. relativi ty.Ii / en / ep stein2 / r e ad / a0 _en /

a2

_en

(or in E. F. Taylor and |. A. Wheeler, Spacetime Physics,2nd

ed., San Francisco: Freeman, 1992, pp.53-55, or elsewhere

online: search for "Galileo" and "Shut yourself up.") In

short paragraph, compare and contrast his presentation of

the principle with the presentation in section R1.1.
(a) What for Galileo corresponds to a "laboratory moving
at a constant velocity"?
(b) What for Galileo corresponds to the phrase "the laws
of physics are the same" in such laboratories?

R1M.2 Imagine that Frames R Us, Inc., is constructing an

economy reference frame whose price will be below every
other frame on the market. Placing a clock at every point in

R18.4

the frame lattice is too expensive, so the company decides

to place one clock at the origin. At all other positions, the
company simply places a flag that springs up when an
object goes by. The flag has the lattice location printed on
it, so an observer sitting at the origin can assign spacetime
coordinates to every event by noting whenhe or she sees
the flag spring up (according to the clock at the origin) and
noting the spatial coordinates indicated by the flag. Why
doesn't this method yield the same spacetime coordinates
as having a clock at every location would? Pinpoint the
assumption that the engineers at Frames R Us are making
that is incorrect. (See problem R1R.2 for a further exploration of the problems with this reference frame.)

Suppose that boat,4, is moving relative to the water

velocity of 6 m / s due east and boat B is moving with
a velocity of \2 rn/s due west. Assume that observers on
both boats use reference frames in which the direction
points east. According to the Galilean velocity transformation equations, what is the velocity of boat 24, relative
to boat B? (Hint: Draw a picture! Be sure to define which
object corresponds to which frame.)

with

Suppose that spaceship A is moving relative to the

earth at a speed of 0.5c (where c is the speed of light) in a
direction we define to be the * direction. Ship B is moving at a speed of 0.9c in the same direction. 14/hat is the
velocity of ship,4, relative to ship B, according to the Galilean velocity transformation equations? (Hint : Draw a picture! Be sure to define which object corresponds to which
frame.)

R18.5
gers,

R1M.3 Two firecrackers explode simultaneously 125 m

in an effort to attract more passenAmtrak trains now offer free bowling in a specially
Suppose that

constructed "bowling alley" car.Imagine that such a train

is traveling at a constant speed of 35 m/s relative to the
ground. A bowling ball is hurled by a passenger on the
train in the same direction as the train is traveling. Assume
the Galilean transformation equations are true.
(a) The ball is measured in the ground frame to have a
speed of 42 m/ s. What is its speed in the frame of the
train according to the Galilean velocity transformation?
(b) Suppose the ball's velocity is 8 m/s in the train frame.
What is its speed relative to the ground?
(Hints: Draw a picture and be sure to define a *x direction
and specify which frame is the Home Frame.)

apart along a railroad track, which we can take to define

axis of an inertial reference frame (the Home Frame).
A train (the Other Frame) moves at a constant 25 m/s in
the *x direction relative to the track frame. Assume the
Galilean transformation equations are true.
(a) Do the firecrackers explode simultaneously in the train
the

frame?

(b) How far apart are the explosions as measured in the

train frame? (Hi: If x2 - xt : I25 m, what is x'r- x|,?)
(c) Suppose that instead of the explosions being simultaneous, the firecracker farther ahead in the *x direction
explodes 3.0 s before the other in the ground frame.
How far apart would the explosions be in the train
frame if this were true?

Chapter

20

R1

The Principle of Relativity

Initial:

Initial:
m

l,1:l,1
7J1

a2

m
a1

Final:

2m

a2

Final:
, =0

a3

2m

x'

Figure R1.11
A collision between two blobs, as observed in two different reference frames

R1M.4

Suppose that construction crews have set up a pair

of blinking lights 60 m apart to mark a construction zone
along a road. A car moves along the road with a speed of
30 m/s toward the lights. Let us take the road to be the
Home Frame and the car to be the Other Frame. In the
Home Frame, the light farther from the car (light 2) blinks
0.66 s before the other (light 1) blinks as the car approaches.
(a) What is the time interval t|- t't between the blink
events in the car's frame? (Hlnf; Think about the slg.)
(b) \ Ihat is the x displacement x'r- xibetween the events
in the car frame?

R1M.5 In a certain particle accelerator experiment, two

subatomic particles A and B are observed to fly away in
opposite directions from a particle decay. Particle A is
observed to travel with a speed of 0.6 c relative to the labotafoty, and particle B is observed to travel with a speed of
0.9 c, where c is the speed of light (3.0 x 108 m/s). For simplicity's sake, let's agree to take the direction of motion of
particle B to define the *x direction in both the laboratory
frame and particle A's frame. We would like to calculate
the relative velocity of particles A and B.
(a) What object should we choose to be the F{ome Frame

in this problem, according to the conventions established in this chapter? What object defines the Other
Frame? \tVhat object is the object whose velocity we
want to determine in both frames? Please explain.

(b) Use the Galilean velocity transformation equations to

calculate the relative velocity of the two particles as a
fraction or multiple of c. Please explain your work.

RLM.6 Ahighway patrol

officer on the ground uses a radar

gun to measure a suspect's car's speed to be 50 m/s. A

patrol car is traveling at 40 m/ s relative to the ground, but
is moving in the opposite direction as it approaches the car.
Let the direction in which the patrol car is traveling define
the *x direction for everyone. We would like to calculate
the relative velocity of the suspect and the patrol car.
(a) Suppose we choose the patrol car to be the Home
Frame. What is the sign ol B n this case, according
to the conventions established in this chapter? Please
explain your reasoning.
(b) Use the Galilean velocity transformation equations to
calculate the velocity of the suspect's car relative to the
patrol car. Please explain your work.

R1M.7 Figure R1.11 above shows an inelastic collision

between two blobs as observed in two different inertial reference frames. Assume the Galilean transformation equations are true.
(a) Which frame is the Home Frame, according to the convention established in this chapter? What is the sign
of B? Explain your reasoning.
(b) IVhat is zri, in terms ol ld' lz
(c) IrVhat is oi, in terms of ltlz
(d) \^Ihat is the system's total r-momentum in the Home
Frame, both initially and finally, in terms of mlrl Z
(e) \^[hat is the system's total x-momentum in the Other
frame, both initially and finally, in terms of mlrl Z
(f) Is momentum conserved in the Home Frame?
(g) Is momentum conserved in the Other Frame?
(hl Assuming that energy is conserved in the Home Frame,
show that it is also conserved in the Other Frame.
(Ignore gravity. Note that some kinel'tc energy is converted to thermal energy U in this collision, but also
that AU must be t1rre same in both frames, because aII
observers will agree about how much hotter the final
blob is compared to the initial blobs, and thus on how
much the system's internal energy has increased.)

R1M.8

Some people are playing a game of shuffleboard on

an ocean cruiser moving down the Hudson River at a condirection relative to the
stant speed of 17 m/s in the
shore. During one shot, a puck (which has a mass of 750 g

and is traveling at 10 m/s in the -x'direction in the boat

frame) hits a puck having the same mass at rest. After the
collision, the first puck comes to rest, and the other puck
travels at 10 m/s in the -x direction in the boat frame.
(Assume that the ground frame's r axis points in the same
direction as the boat frame's x'axis.)
(a) Show that the total x-momentum of the two-puck system is conserved in the boat frame. Explain carefully.
(b) Imagine that someone sitting on abridge under which

the boat is passing takes a video of this important

game. What r-velocity will each puck be measured to
have relative to the shore? Explain carefully.

(c) Show that in spite of the fact that the puck's r-velocities
have signs and magnitudes that are dffirent fromlhose
measured on the boat, the total momentum of the
two-puck system is still conserved in the shore frame.
Explain your work.

Answers

to

Exercises

21

The engines on a 4000-kg jet plane agcelerating for

off exert a constant thrust of magnitude I F* | ZO,OOO N
on the jet as it accelerates from rest a distance D = 1000 meters
down the runway before taking off. This take-off is observed
by someone riding in a trainwith a constantspeed of 30 m/s

alongside the runway. Assume that the train and jet move
in the same direction relative to the ground (which we will
take to be the *x direction), and assume that both the passenger and the plane are at x : 0 in the ground frame when
the plane starts its run at f : 0.
(a) ln unit C, we saw that if the net external force F.", on
an object is constant, the change in the object's kinetic
ene-rgy during a given interval of time should be equal
to F^", . Ala, where A76r is the displacement of the
object's center of mass during that time. Use this to
show that the jet's speed relative to the ground at takeoff is 100 m/s. Explain your reasoning.
(b) Assuming that the jet's acceleration is constant, show
that it takes it 20 s to reach this speed. Explain.
(c) \ /hat are the plane's initial and final -velocities in the
train frame? Explain.
(d) Assume that the passenger's position defines the origin x' : 0 in the train's frame. What is the jet's initial
x-position in the train frame?
(e) What is its r-position at take-off in the train frame?

(f)

Explain.
Show that the change in the jet's kinetic energy K' in the
train frame is equal to F'"", . A7u in that frame. (Thus,
this law of physics, the momentum requirement, is the
same in both frames, even though the numerical values
of K = Fn",. Al1 and K' :F',.r. A7'arenot.)

R1M.10 A person in an elevator drops a ball of mass m from

rest from a height h above the elevator floor. The elevator
is moving at a onstant speed l | downward with respect
to its enclosing building.
(a) How far will the ball fall in the building frame before it
hits the floor? (Hint: >h!)
(b) \^Ihat is the ball's initial vertical velocity in the building frame? (Hlnf: Not 0!)
(c) Use the law of conservation of energy in the building
frame to compute the ball's final speed (as measured
in that frame) just before it hits the elevator floor.
(d) Use the Galilean velocity transformation equations
and the result of part (c) to find the ball's final speed in
the elevator frame.
(e) Assume that conservation of energy applies in the
building's frame. Use the result of part (d) and the fact
that the ball's acceleration is l$ I to show that energy is
also conserved in the elevator frame.

Derivation
R1D.1

A totally symmetric way to orient a pair of reference

frames is so that their *x directions point in the direction
that the other frame is moving. How is this different from
the "standard" orientation (draw a picture)? How would the
Galilean position and velocity trasformation equations be
different if we were to use this convention?

R1D.2 Imagine two inertial

reference frames in standard

orientation, where the Other Frame moves in the * direction with x-velocity B relative to the Home Frame. Suppose
an observer in the Home Frame observes the following collision: an object with mass r?1 and velocity , hits an object
with mass m2travelingwith velocity r. AfTer the collision,
the objects move off with velocities d3 and a, respectively.
Do not assume that all or even any of these velocities are in
the direction. Assume, though, that total momentum is
measured to be conserved in the Home Frame, that is, that
mrt + rflzz = mr, +

mrn

(assume

this!)

(R1.7)

Using this equation and the Galilean transformation equations, show that if the Newtonian view of time is correct,
then the total momentum of the two objects will also be
conserved in the Other Frame
mt| + m2i= mt +

mr'a

(prove

thisl)

(R1.8)

even though the velocities measured in the two frames are

very different. Please show your work in detail.

Rich-Context
R1R.1

Design a first-law detector that does not use a floating ball as the basic active element. Your detector should
primarily test Newton's first law and not some other law
of physics (although it is fine if other laws of physics are
involved in addition to the first law). Preferably, your
detector should be reasonably practical and (if at all possible) usable in a gravitational field. (Nofe: There are many
possible solutions to this problem. Be creative!)

R1R.2

Consider the economy reference frame described in

problem R1M.2. Prove that an object that actually moves at

a constant velocity close to that of light will be observed in
the economy frame to move faster as it approaches the origin and slower as it departs. Also describe what happens if
the object moves faster than light.

ANSWERS TO EXERCISES
R1X.1

Imagine that the Home Frame is an inertial frame.

Consider a set of isolated objects arrayed around the Home
Frame that happen to be initially at rest in that frame. Since
their velocity with respect to the inertial Home Frame has
to be constant, these objects w111 remain at rest relative to
the Home Frame. Now, if the Home Frame moves at a constant velocity relative to the Other Frame (and the latter is

rigid and nonrotating

so that all parts of the Other Frame

move with a constant velocity relative to the Home Frame,
and vice versa), then our set of isolated objects must also
move with a constant velocity relative to the Other Frame.
Since these isolated objects are observed to move with a
constant velocity everywhere in the Other Frame, it must
be an inertial frame as well.

--F
aa'

Coordinate Tme
Chapter Overview
lntroduction
In this chapter, we take the first steps toward developing a conception of time that
solves the problem raised by Maxwell's equations. We also start developing crucial
tools for our exploration of relativity: the spcetime digrm and the geometic nalogy .

Section

R2.1

Relativistic Clock Synchronization

If Maxwell's equations are true laws of physics, then the speed of light c in a vacuum
must be the same in all inertial reference frames. This in turn implies, as Einstein
noted, that ny rlrrethod for synchronizing an inertial frame's clocks, if the method
is to be consistent with the principle of relativity, should be equivalent to using light
flashes to synchronize the clocks ssuming that the speed of those flashes is c.

Section R2.2: SR Units

The speed of light in this approach is a frame-independent universal constant that
connects time and space units. (Indeed, physicists currently define lhe meter to be the
distance that light moves in exactly 1/299,792,458 s.)
In Systme Relatiaistique (SR) units, we avoid the conversion factor entirely by
measuring distances in seconds, where 1 s of distance is the distance that light moves
irr 1 s of time. The speed of light in this unit system is 1 s/1 s = 1; all other speeds are
similarly unitless; and energy, momentum, and mass all have units of kilograms. To
convert a quantity from SI to SR units, you multiply it by whatever power of the unit
operator 1 : (3 x 108 m/1 s) makes the units of meters go away; to convert from SR to
SI units, you multiply by whatever power gives you the appropriate power of meters.

Section R2.3: Spacetime Diagrams

A

spacetime diagram is an important visual aid for displaying the relationship

between events. Such a diagram is simply a graph with a vertical time axis and horizontal spatial axes. Euents in spacetime correspond To points on such a diagram. The
connected sequence of events that describe a particie's motion in spacetime is represented by a curve on the diagram called a worldline, which displays how the particle's position varies with time. If the axes on the diagram have the same scale and
the object moves only along the x axis, then the slope A't / Ax of its diagram worldline
at any instant of time is 1/ a,, where o. is the particle's x-velocity at that instant. Note
that the slope of any light-flash worldline will be +1.

Section R2.4: Spacetime Diagrams as Movies

If you view a spacetime diagram through a horizontal slit moving upward at a constant rate, you see a one-dimensional movie that displays the events and moving
particles depicted on the diagram (see figure R2.5).

Section R2.5: The Radar Method

We can determine an event's spacetime coordinates using a method that is completely
equivalent to that described in section R2.1 but requires only one clock at the frame's

aaaaaaaoaoaaataaaaooaaaaaaaaaaaataaaaaaaaaaaaaa!aaaaaaaaaaaoaaaaaa

22

rr

a a a a a a a a a a a a a a a a a a a a a a a a a a o a a a a a a a a a a a a a a a a a o o a o o o o a a a a a a a a a a a a a a

origin. Suppose a light flash emitted by the master clock at time t is reflected by an
event and returns to the master clock at time f. Since light moves at speed 1 in all
inertial frames, event E's coordinates must be

tE:+(B+ tA) .rr: llru -

ro

(R2.4)

Purpose: This equation expresses an event E's spacetime coordinates f and x

in terms of the time f that a light flash leaves a given frame's origin and the
time fs when it returns after being reflected by the event. (Both times are measured by a master clock at the frame's origin.)
o Limitations: These equations assume that the event is located on the .v axis,
that the master clock is in an inertial reference frame, and that we are using
SR units.

This approach to determining coordinates is called the radar method, because radar
tracking of airplane positions is essentially done in this way.

Section R2.6: Coordinate Time ls Frame-Dependent

The coordinate time Af between two events is the time measured in the context of
an inertial reference frame. Usually, the two events will occur at different places in
a given frame, and in such a case the coordinate time is the difference between the
event times as recorded by two synchronized clocks in the frame, one present at each
event. If the events happen to occur at the same place in the frame, then a slngle clock
present at that position suffices to measure Af .
The coordinate time between a given pair of events is afrme-dependent number,
because although the method of clock synchronization described in section R2.1 is
self-consistent Llithin a given inertial frame, an observer in another inertial reference
frame will nof observe these clocks to be synchronized when comparing them to synchronized clocks in her or his own frame! This can cause, for example, an observer in
one frame to think that two events are simultaneous, while an observer in a different
frame concludes they are not.
This is not due to problems with the process of synchronizing multiple clocks.
One can show that the radar method of defining coordinates leads to exactly the same
results. The disagreement is a simple consequence of defining time so that the speed
of light is the same in all inertial reference frames.

Section R2.7: A Geometric Analogy

with analogous frame-dependent quantities in a related but more
familiar guise. Imagine superimposing two ry Cartesian coordinate systems on a town,
one rotated with respect to the other. We know that surveyors using properiy established but different coordinate systems will disagree about both the north-south displacement Ay and the east-west displacement Ar between two given locations in the
town. The way that different inertial reference frames define spacetime coordinates
for events is directly analogous to the way different coordinate axes define x and y
coordinates for points on the two-dimensional plane, so we shouLd expect observers
in different reference frames to disagree about the value of the coordinate Af between
events (as well as about their spatia separation lA l).
Note that we can use a tape measure to direcly measure the separation of two points
on a plane, either by measuring the pthlength along some arbifiary path between the
points or by measuring the distance between them along the unique straight iine connecting those points. Since we can measure these quantities directly without setting up a
coordinate system, any person who calculates these quantities using coordinate-based
quantities mllst get the same result as any other: the pthlength along a path and the
distance between two points are thus coordinate-independent, We will define analogous frame-independent time quanlittes in chapter R3.
We have no trouble

23

24

Chapter

R2.1

R2

Coordinate Time

RelativisticClockSynchronization

At the end of chapter R1, we saw that experiments show that light has the
same speed in all inertial reference frames. Einstein proposed that this was
because Maxwell's equations are true laws of physics and, since they predict
that electromagnetic waves must move at a certain speed c, that speed (by the
principle of relativity) must be the same in all reference frames. The problem
is that the concept of universal and absolute time (and the Galilean transfor-

Einstein's method of clock

synchronization

mation equations that follow from it) contradicts this proposal. But if time is
not universal and absolute, how can we even define what it means?
The solution, as Einstein was the first to see, is that we must define
what we mean by "tirr.;:e" opertionally within each inertial frame by specifying a concrete and specific procedure for synchronizing that frame's
clocks that is consistent with both the principle of relativity and tiiLe laws of
electromagnetism. But how can we synchronize clocks in such a manner?
Here is one method. Maxwell's equations imply that light moves through
a vacuum at a certain fixed speed c. The principle of relativity requires that
this speed be the same in every inertial reference frame. Therefore, any svnchronization method consistent with the principle of relativity will lead to
light being found to have speed c in any inertial reference frame.
Since the speed of light must be c in every inertial frame artryay,let us in
fact synchronize the clocks in our inertial reference frame by ssuming this
is true! How do we do this? Suppose we have a master clock at the spatial
origin of our reference frame. At exactly I : 0, we send a light flash from that
clock that ripples out to the other clocks in the frame. Since we are assuming that light travels at a speed of c : 299,792,458 m/s, this light flash will
reach a lattice clock exactly 1.0 meter from the master clock at exactly time
: (1.0 rn)/(299,792,458lnr-/s):3.33564095 x 10 e s:3.33564095ns. Therefore, if we set that clock to read 3.33564095 ns exactly as the flash passes, then
we know it is synchronized with the master clock. The process is similar for
all the other clocks in the lattice.
So here is a first draft of a description of Einstein's method for synchronizing the clocks in an inertial reference frame:

A light flash is emitted by clock A in an inertial frame at time fn (as

read on clock A) and received by clock B in the same frame at time
fs (as read on clock B). These clocks are defined to be synchronized
if c(ts - f) is equal to the distance between the clocks. That is, the
clocks are synchronized i they measure the speed of a light signal
traveling between them to be c.

Exercise R2X.1
Imagine we have a clock on the earth and a clock on the moon. FIow can we
tell if these clocks are synchronized according to this definition? Suppose we
send a flash of light from the earth's clock toward the moon at exactly noon,
as registered by the earth's clock. \Alhat time will the clock on the moon read
when it receives the flash if the two clocks are synchronized? (The distance
between the earth and the moon is 384,000,000 m.)
Note that we are ssuming tlnal
Maxwell's equations obey the

principle of relativity

"Now, wait just a minute!" I hear you cry. "Isn't all this a bit circular? You
claim that Maxwell's equations predict that light will be measured to have
the same speed c in every inertial reference frame. But then you go and set up
the clocks so this result is ensured.Is this fair?"

R2.2 SR Units

25

This ls fair. We are not trying here to proae that Maxwell's equations
obey the principle of relativity-we are ssuming they do, so we can determine the consequences of this assumption. To make this clear in his original
paper, Einstein actually stated the frame independence of the speed of light
as a separate postulate, emphsizing that it is an assumption. However, this is
not really a separnte assumption: it is a consequence of the assumption that
the principle of relativity applies to all laws of physics, including Maxwell's
equations. The point is that if the principle of relativity is true, the speed of
light will be measured to have speed c no matter what valid synchronization
method we use, so why not use a method based on that fact?
Moreover, t}l:erc re other valid ways of synchronizing the clocks in a
given inertial reference frame, ways that make no assumptions whatsoever
about the frame independence of the speed of light.* If such a method were
used to synchronize clocks in an inertial frame, such a frame could be used
to verify independently lhat the speed of light is indeed frame-independent.
These alternative methods yield the same consequences as one gets assuming the frame-independence of the speed of light. These methods are, however, also more complex and abstract: the definition of synchronization in
terms of light is much more vivid and easy to use in practice.

SR Units

In ordinary SI units, the speed of light c is equal to 299,792,458 m/s, a somewhat ungainly quantity. The definition of clock synchronization given in section R2.1 means that we will often need to calculate how long it would take
light to cover a certain distance or how far light would travel in a certain
time. You can perhaps see how messy such calculations will be.
For this reason (and many others), it will be convenient when we study relativity theory to measure distance not in the conventional unit of meters but in a
new unit called a light-second or just second for short. A light-second or second
of distance is defined to be the distance light travels in L second of time. Since
1983, the meter has in fact been officially defined by international agreement
as the distance light travels lr.I/299,792.458 s. Therefore, 299,792,458 meter
is equal to 1 light-secondby definition.
We can, of course, measure distance in any units we please: there is
nothing magical about the meter. Choosing to measure distance in lightseconds has some important advantages. First, light travels exactly L second
of distance in L second of time by definition. This allows us to talk about
clock synchronization much more easily. For example, if clock A and clock
B arc 7.3 light-seconds apart in an inertial reference frame and a light flash
leaves clock A when it reads t : 4.3 s, the flash should arrive at clock B at
tu : (4.3 + 7.3) s : 11,.6 s if the two clocks are correctly slmchronized, since
*One of the simplest is described by Alan Macdonald (Am. Phys., vol. 5L, no. 9,
I.
1983). Macdonald's method is as follows. Assume that clocks A and B emit flashes of
light toward each other at (as read on each clock's own face).If the readings of the two
clocks also agree when they receive the light signal from the other clock, they are
synchronized. This method only assumes that the light flashes take the same time to
travel between the clocks in each direction (that is, there is no preferred direction for
light travel): it does not assume light has any frame-independent speed. Achin Sen
(Am. l. Phys.,vol.62,no.2,1994) presents a particularly nice example of an approach
that sidesteps the synchronizalion issue, showing mathematically that the results of
relativity follow directly from the principle of relativity. Sen's article also contains an
excellent list of references.

Definition of the "second" of

distance

Advantages of using this unit

of distance

Chapter

26

R2

Coordinate Time

light travels exactly L light-second in 1 second of time by definition. So we avoid

performing ungainly unit conversions if we measure distance in this way.
Indeed, agreeing to measure distance in seconds allows us to state the
definition of clock synchronization in an inertial frame in a particularly nice
and concise manner (this will be our finnl draft of this description):
We define two clocks in an inertial reference frame to be synchronized if the time interval (in seconds) registered by those clocks for a
light flash to travel between them is equal to their spatial separation
(in light-seconds).

SR

units

In spite of the tangible advantages that measuring distance in seconds

yields when it comes to talking about synchronization, this is not the most
important reason for choosing to do so. In the course of working with relativity, we will uncover a deep relationship between time intervals and distance ntervals, akin to the relationship between distances measured north
and distances measured east on a plane. We would not think of measurng
northward distances in feet and eastward distances in meters: that would
obscure the fundamental similarity and relationship of these quantities. Similarly, measuring time intervals in seconds while measuring distance intervals
in meters obscures the fundamental similarity in these measurements that
will be illuminated by relativity theory. Choosing to measure time in the same
units as distance will make this beautiful symmetry of nature more apparent.
The standard unit system used by scientists studying ordinary phenomena comprises t}lte Systme Interntional, or SI, units. In this system, the
units for mechanical quantities (such as velocity, momentum, force, energy,
angular momentum, and pressure) are based on three fundamental units: the
meter, tlre secondt and the kilogrm.In this text, however, we will use a slightly
modified version of SI units (let's call it Systme Relatiaistiqu, or SR, units)
where distance is measured in seconds (that is, light-seconds) instead of in
meters (with the other basic r.rnits being the same).
As discussed in chapter C1, we can use unit operators to convett a quantity from one kind of unit to another. We first write down an equation stating the basic relationship between the units in question: 1 mi : 1.609 km,
for example. We then rewrite this in the form of a ratio equal to 1: either
1 : 1 mi/7.609 km or 1 : 7.609 kr/Imi. Since multiplying by 1 doesn't
change a quantity, you can multiply the original quantity by whichever ratio
leads you to the correct final units upon cancellation of any units that appear
in both the numerator and denominator. For example, to determine what a
distance of 25 km is in miles, you simply multiply 25 km by the first of the
two ratios described above, as follows:
2s km

25

km

zs[_
+-', : 1.60e
,27 mi = 16 mi
' \1.60ekl

(R2.1)

The factors used to convert from SI distance units to SR distance units are
based on the fundamental definition of the light-second: 299,792,458 m 1 s.
=
Thus, the basic conversion factors we need are 1 (1 s/299,792,458 m), or
t (299,792,458 m/1. s). For example, a distance of 25 km can be converted

to a distance in light-seconds

as

follows:

25km:25km.1.1: 2skltooory'\f
'

\1t

: 8.3 x 10-s s : 83 Us

ls

l\2.ee8x10o)
(R2.2)

meaning that 25 km is equivalent to the distance that light travels in

83 millionths of a second.

R2.3 Spacetime Diagrams

27

The light-second is a rather large unit of distance (the moon is only about
from the earth!). Amore appropriate unit on the human
scale is the light-nanosecond, where 1 light-nanosecond = 10-e light-second
(:0.2999 m = 1 ft). On the astronomical scale, the light-year (where 1 lightyear = 3.16 x 107 s = 0.95 x 1016 m) is an appropriate (and commonly used)
distance unit. The dimensions of the solar system are conveniently measured
in light-hours (it is about l0light-hours in diameter). All these units represent
extensions of the basic unit of the light-second.
In SR units, we consider the light-second to be equialenf to the second of
time, and both units are simply referred to as seconds. This means these units
1.3 light-seconds away

can be canceled if one appears in the numerator of an expression and the

other in the denominator. For example, in SI units, velocity has units of meters
per second, but in SR units it has units of seconds per second = unitless(!).
Thus, an object that travels 0.5 light-second in 1.0 s has a speed in SR units of
0.5 s/1.0 s : 0.5 (no units!). This bare number for a speed actually represents
a comparison of the object's speed to the speed of light, since light covers 1.0 s
of distance in 1.0 s of time by definition. Thus, an object traveling at a speed
of 0.5 (in SR units) is traveling at one-half the speed of light.
In a similar manner, one can find the natural units for any physical quantity in SR units. For example, a particle's kinetic energy has the same units as
mass times speed squared. In SI units, these units would be kg.rnz / s2. Thus,
the natural SI energy unit is the joule, where Il = 7kg.m2 / s2.ln SR units, mass
times speed squared has rmits okg.s2 /s2: kg. Thus, the natural SR rmit for
energy is the kilogram (the same as the unit of mass!). How much energy is
represented by an SR kilogram of energy? We can determine this by using the
standard conversion factor to convert the SR distance unit of seconds to the
SI distance unit of meters:
1 ks(energy )

Velocity is unitless in SR turits

Natural SR units for other

physical quantities

: lUr/r(2'ss} 1,108 v\\', (,. rt r r, rl : s.oas x 10'6 J (R2.3)

r \ rF I \rtq'r/r- /{ I

This unit is roughly equal to the yearly energy

ouut of 10 full-size electric

power plants!

In general, what we need to do to convert from SR units to SI units is

to multiply the SR quantity by whatever power of the conversion factor
7 : (299,792,458 m/Is) yields the correct power of meters in the units of
the final result. Similarly, to convert from SI units to SR units, we simply
multiply the SI quantity by whatever power of this factor causes the units of
meters to disappear. (See appendix RA for a complete discussion of how to
convert units and equations from one system to the other.)

R2.3

Spacetime Diagrams

The clock synchronization method described in section R2.1 provides the last
bit we needed to know to build and operate an inertial reference frame. We
now know how to assign spacetime coordinates to any event occurring within
that inertial frame in a manner consistent with the principle of relativity.
Problems in relativity theory often involve studying how physical events
relate to one another. We can conveniently depict the coordinates of events
by using a special kind of graph called a spacetime diagram. Throughout
this unit, we will find spacetime diagrams to be indispensable in helping us
express and visualize the relationships between events.
Consider an event ,4 whose spacetime coordinates are measured in
some inertial frame tobe t, xd, A, z. To simplify our discussion somewhat,

How to draw events

spacetime diagram

or-r a

Chapter

28

R2

Coordinate Time

6s

(a) How

to plot an event A on

A worldline's slope is the

inuerse of its particle's

r-velocity

4s

tA
a

spacetime diagram. ln this particular case, event A has a time

coordinate tt = 5.4 s and an x
coordinate xt = 4.5 s. (b) Plotting
an event A that has nonzero x
and y coordinates (but a zero z
coordinate).

Worldlines

A
I

Figure R2.1

LA

.,4

4s

2s

2s

2s

4s

r4

v
Aa

(a) o

2s

4s

6s

(b)

assume that y e : ZA : 0 (that is, the event occurs somewhere along the x axis
of the frame). Now imagine drawing a pair of coordinate axes on a sheet of
paper. Label the vertical axis with a I and the horizontal axis with an x (it is
conventional to take the f axis to be vertical in spacetime diagrams). Choose
an appropriate scale for each of these axes. Now you can represent when and
where event,4 occurs by plotting the event as a point on the diagram (in the
usual manner you would use in plotting a point on a graph), as shown in
figure R2.1a. We can plot any event that occurs along the r axis in space on
such a diagram in a similar manner.
Note that the point marked O in figure R2.La also rePresents an event.
This event occurs at time f : 0 and at position x : 0. We call this event the
diagram's origin event.
If we need to draw a spacetime diagram of an event A that occurs
in space somewhere nt}te xy plane (that is, which has za: 0 but rror:.Zto t,
x,Ud),we must add another axis to the spacetime diagram (see figure R2'1b).
The resulting diagram is less satisfactory and more difficult to draw because
we are trying to represent a three-dimensional graph on a two-dimensional
sheet of paper. We can't draw a spacetime diagram showing events with
three nonzero spatial components at all, as this would involve trying to represent afour-dimensional graph on a two-dimensional sheet of paper. A fourdimensional diagram is hard to visualize at all, much less draw! Fortunately,
one or two spatial dimensions will be sufficient for most purposes.
In chapter RL, we visualized describing an object's motion by imagining the object to carry a strobe light that blinks at regular intervals. If we can
specify when and where each of these blink events occurs y specifying its
spacetime coordinates), we can get a pretty good idea of how the object is
moving. If the time interval between these blink events is reduced, we get
an even clearer picture of the object's motion. In the limit that the interval
between blink events goes to zero, the object's motion can be described in
unlimited detail by a list of such events. Thus, we can describe any object's
motion in terms of a connected sequence of events. We call the set of all events
occurring along the path of an object that object's worldline.
On a spacetime diagram, an event is represented by a point. Therefore, a worldline is represented on a spacetime diagram by an infinite set of
infinitesimally separated points, which is a curae. This curve is nofhing more
than a graph of the object's position versus time (except that the time axis
is conventionally taken to be vertical on a spacetime diagram). Figure R2.2
illustrates worldlines for several example objects moving in the direction'
Note that because the time axis is taken to be vertical, the slope of the
curve on a spacetime diagram representing the worldline of an object traveling at a constant velocity in the r direction is not its x-velocity o" (as one
might expect) but rise/run : ;t/A.x:\/a,l Thus, the slope of the curve

R2.4 Spacetime Diagrams as Movies

29

Lx
a

'

^f

"Bliok
events"
(b)

(a)

A sequence of events
along an object's
path marks out its

motion.

As the time between

the events goes to
zero, the set of
events becomes a
curve.

(c)

The slope of this

worldline is equal to
the object's inverse
x-velocity (1./a,).

(d)
The vertical line is the
worldline of a particle
at rest. The other is the
worldline of a particle
accelerating from rest.

\l

v
x

Figure R2.2

Figure R2.3

A sequence of spacetime diagrams illustrating various important things to know

about worldlines.

The worldline of a particle traveling

in a circular path in the xy plane.

representing the worldline of an object at rest is infinity and decreses as 7)x

increases. The worldline of an object traveling at a constant r-velocity has a
constant slope.
Occasionally, we need to draw a spacetime diagram of an object moving
in two spatial dimensions. The spacetime diagram in such a case is necessarily three-dimensional, which is hard to draw on two-dimensional paper.
Figure R2.3 shows an example of such a spacetime diagram.
In drawing spacetime diagrams, we also conventionally use tlrre same-size
scale on both axes. If we do this, then the worldline of a flash (that is, a very
brief pulse) of light always has a slope of either 1 (if the flash is moving in
the *r direction) or -1 (if the flash is moving in the -r direction), since light
travels 1.0 second of distance in 1.0 second of time by definition in every
inertial'reference frame, We also conventionally draw the worldline of a flash
of light with a dashed line instead of a solid line (see figure R2.4).

Light-flash worldlines have

slopes of t1
t

6s

4s
ra

Exercise R2X.2
A

On figure R2.4, draw the worldline of a particle moving in the -r direction

through the origin event with a speed of 0 .2 (1 light-second per 5 s).

2s

2s

4s

Figure R2.4

,'-

82.4

Spacetime Diagrams as Movies

One can easily get confused about what a spacetime diagram really represents. For example, in the spacetime diagram shown in figure R2.4, one can
easily forget that the light flashes shown are moving in only one dimension
(along the r axis), not in two. Their velocity vectors therefore point opposite
to each other, not perpendicular to each other.
Here is a technique you can use to make the meaning of any spacetime
diagram clear and vivid: turn it into a movie! Here's how Take a 3 x 5 index
card and cut a slit about fr inch wide and about 4 inches long, using a knife
or a razo blade. This slit represents the spatial r axis at a given instant of
time. Now place the slit over the r axis of the spacetime diagram. What you
see through the slit is what is happening along the spatial x axis at time f : 0.
Now slowly move the slit up the diagram, keeping it horizontal. You will see

Light-flash worldlines on a
spacetime diagram. Worldline A represents a light flash

moving in the *x direction.

Worldline B represents a flash
traveling in the opposite
direction.

Making a movie viewer for a

spacetime diagram

30

Chapter

R2

Coordinate Time

READ THIS FIGURE FROM THE

BOTTOMUP!

Figure R2.5d
Now the light flashes have passed
each other and are moving away
from each other. You can also see
through the slit the momentary
flash representing the firecracker
explosion (event E).

Figure R2.5c
At this instant, the light flashes pass
through each other at a position of

aboutx:+Is.

Figure R2.5b
As time passes (and you move

the slit up the diagram), the dots

representing the light flashes
approach each other.

Figure R2.5a
The spacetime diagram is basically
the same as figure R2.4 with a
firecracker explosion (event E)
thrown in to make things more

interesting. At time

t:

0, the light

flashes are represented by black dots

that you can see through the slit at
x: -3 s and x: *4 s.

through the slit what is happening along the spatial axis at successively
later times. You can watch the objects whose worldlines afe shown on the
diagram move to the left or right as the slit exposes different parts of their
worldlines. Events drawn as dots on the diagram will show up as flashes as
you move the slit past them. what you see through the slit as you move it up
the diagram is essentia a movie of what happens along the axis as time
passes. Figure R2.5 illustrates the Process.
If you employ this technique, you cannot fail to interpret a spacetime
diagram correctly. After a bit of practice with the card, you willbe able to
convert diagrams to movies in your head.

R2.5 The Radar Method

31

The Radar Method

Reflected

If we are willing to confine our attention to events occurring only along the
r axis (and thus to objects moving only along that axis), we can determine the
spacetime coordinates of an event with a single master clock and some light
flashes: we don't need to construct a lattice at all! The method is analogous
to locating an airplane by using radar.
Suppose that at the spatial origin of our reference frame (that is, at x : 0),
we have a master clock that periodically sends flashes of light in the *x
directions. Imagine a certain flash emitted by the master clock at f happens
to illuminate an event of interest that occurs somewhere down the r axis.
The reflected light from the event travels back along the x axis to the master
clock, which registers the reception of the reflected flash at time fs (see the
spacetime diagram of figure R2.6).
The values of the emission and reception times ta and fs are sufficient
to determine both the location and the time that event E occurred! Consider
first how we can determine the location. The light flash's round-trip time is
t - t.Since in this time the light covered the distance from x : 0 to event
E and back, and since the light flash travels 1 second of distance in 1 second
of time by definition, the distance to event (in seconds) must be one-half
of the round-trip time (in seconds). The r coordinate of event is therefore
xr:*l(tu- fo). We can determine the sign of r by noting whether the
reflected flash comes from the -x or lx direction. (In this case, the reflected
flash comes from the *r direction, so we select the plus sign.)
We can determine the event's time coordinate as follows. Since the light
flash traveled the same distance to the event as back from the event, and
since the speed of light is a constant, the event must have occurred exactly
halfwaybetween times f and fs. The midpoint in time between times fa and
f3 can be found by computing the average, so t:
l1t" + to.
r summary, therefore, the spacetime coordinates of event E are

tE: l(t, +

tA) xr: |1tu - tn'

tB

light-flash
dline
Worldline
of master

clock

t.

_,

Event E

Outgoing
light-flash

worldline

:0

Figure R2.
At time t, the master clock at
rest at x = 0 in the Home Frame
sends out a flash of light, which
reflects from something at event
E and returns to the master clock
at time tr.

(R2.4)

o Purpose: This equation expresses an event E's spacetime coordinates

fr and xB in terms of the time f at which a light flash leaves a given

frame's origrn and the time fs when it returns after being refiected by
the event. (Both times are measured by a master clock at the frame's
origin.)
Limitatiors: This equation assurnes that the event is located on the
x axis, that the master clock is in an inertial reference frame, and that
we are using SR units.

Equations R2.4 represent a method of determining the spacetime coordinates of an event that does not require the use of a complete lattice of synchronized clocks. But you should be able to convince yourself thatthis method
produces exactly the sme coordinate aalues as you would get from a clocklattice.For
example, imagine you actually had a lattice clock at r (the location of event
). The distance between that clock and the master clock at the origin mustbe
equal to one-half the time it would take a flash of light to go from one clock
to the other and back, since light travels 1 second of distance in 1 second of
time by definition. The lattice clock at x, at the time of event E must read ta *
(the light travel time between the two clocks) : t -f T(t" - t^) :
l1t, + t^,
since we are assuming that the lattice clocks are s)mchronized, which means

This method yields the same

result as the clock lattice

method

32

Chapter

R2

Coordinate Time

(by definition) that they measure a light flash to travel between them at the
speed of light (1 second of distance in 1 second of time).
Using this method to determine spacetime coordinates is therefore equivalent to using a lattice of synchronized clocks. In some cases in this text, we
will find it clearer or more convenient to use one method, and in some cases
the other. The important thing to realize is that either ltre radar method or the
clock lattice method provides specific and well-defined method for assigning
time coordinates to events, and they are equivalent because both express the
assumption that the speed of tight is a frame-independent constant. These
methods essentially define what time means in special relativity, and thus
will provide the foundation for most of the arguments in the remainder of

Adapting the method for three

spatial directions

the text. Make sure you thoroughly understand both methods.

We call this section's method the radar method because radar installations actually use a three-dimensional generalization of this approach to
track the trajectories of aircraft (the impracticality of using a clock lattice to
do the same is obvious!). we can precisely locate an object in three spatial
dimensions if we record not only the time that the outgoing pulse was sent
and the time that the reflected pulse was received but also tlite direction fuorrt
which the reflected pulse was received. The analysis is a bit more complicated (see problems R2M.7 and R2M.8), but the basic idea is the same.

R2.

Coordinate Time ls Frame-DePendent

Once we have satisfactorily synchronized the clocks in an inertial reference frame, we can use them to measure the time coordinates of various
events that occur in that frame. In particular, we can measure the time
between two events A and B in our reference frame by subtracting the time
read by the clock nearest event,4. when it happened from the time read by
the clock nearest event B when it happened: Ltg: t" - t.Note that this
method of measuring the time difference between two events requires the
use of a pair of synchronized clocks in an established inertial reference
frame. Such a measurement therefore cannot be performed in the absence of
an inertial frame.
So define the coordinate time between two events as follows:
The definition of coordinate
timebetween events

The coordinate time between

events depends on your choice
of frame

The coordinate time Af between two events either by a pair of synchronized clocks at rest in a giaen inertil reference frame (one clock
present at each event) or by a single clock at rest in tht inertial frme
(if both events happen to occur at that clock in that frame).

Now, suppose the observer in some inertial reference frame (let's call
this frame the Other Frame: we'll talk about a Home Frame in a bit) sets
out to s)mchronize its clocks. In particular, let us focus on two clocks in that
frame that lie on the r axis an equal distance to the left and right of the
master clock at x' : 0. At t' :0, the observer causes the center clock to emit
two flashes of light, one traveling in the -lx'direction and the other in the
-r' direction. Let's call the emission of these flashes rorr. x' : 0 at t' : 0 the
origin event O.
As both of the other clocks are the same distance fromthe center clock and
since the speed of light is 1 (light)second/second in eaery inertial reference
frame,the left-hand clock will receive the left-going light flash (call the event
of reception event A) at the same time as the right-hand clock receives the
right-going flash (event B).By the definition of synchronization, both clocks
should therefore be set to read ttle snme time at events A and B (a time in
seconds equal to their common distance from the center clock).

R2. Coordinate Time ls Frame-Dependent

33

t'

Left-going
light flash

Right-going

light flash

Figure R2.7
A

Worldline of
center clock

Worldline
of left clock

Worldline
of right clock

The synchronization of two clocks

equally spaced from a center
clock, as observed in the Other
Frame. lf the right and left clocks
are set to agree at events A and
B, they will be synchronized with
each other.

The spacetime diagram in figure R2.7 illustrates this process. Note that
since all three clocks are at rest in this frame, their worldlines on the spacetime diagram are vertical. Moreover, since the speed of light is 1 s/s in this
(and every other inertial) frame, the worldlines of the light flashes will have
slopes of -|1 on the spacetime diagram (that is, they make a 45' angle with
each axis) as long as the axes have the same scale. On this diagram, it is clear
that events A and B really do occur at the same time in the Other Frame.
Now consider a different inertial reference frame (the Home Frame),

within which the Other Frame moves in the *r direction at an r-velocitr

B. How will an observer in this frame interpret these same er)ents? For con-

venience, let us take the event of the emission of the flashes to be the origin
event in this frame also (so event O occurs att : x: 0 in the Home Frame).
The observer in the Home Frame will agree that the right and left clocks
in the Other Frame are always equidistant from the center clock in the Other
Frame. Moreover, at t : 0, when the centet clock passes the point r : 0 in
the Home Frame as it emits its flashes, the right and left clocks are equidistant from the emission event. But as the light flashes are moving to the outer
clocks, the Home Frame observer observes the left clock to move up the
x axis towrd the flash coming toward it, and the right clock to move up the
x axis way from the flash coming toward it. The left-going light flash therefore has less distance to travel to meet the left clock than the right-going flash
does to meet the right clock. Since the speed of light is 1 in the Home Frame
as well as in the Other Frame, this means that the left clock receives its flash

first. Therere, the Home.Frme obseraer obseraes eaent Ato lccur before eaent B.
Figure R2.8 shows spacetime diagram of the process as observed in
the Home Frame. Note that the clocks ate not at rest in the Home Frame, so
their worldlines on a Home Frame spacetime diagram will be equally spaced
lines with slopes o11./ indicating that the clocks are moving to the right at
a speed B. The light flashes have a speed of 1 s/s in the Home Frame (as they
do in nny inertial frame), so we must draw their worldlines with a slope of
-l-1 on the spacetime diagram.

tB

Left-going
light flash

Worldline of
center clock

.
A

Figure R2.8

Worldline
of left clock

Worldline
of right clock

The same events as observed in

the Home Frame. ln this frame,
event B is measured to occur after
event A.

Chapter

34

R2

Coordinate Time

In summary, the coordinate time between events A and B as measured

in the Other Frame is Af ': 0 y construction here), but the coordinate time
between these events as measured in the Home Frame is Lt * 0. We see that
the coordinate times between the snme two eoents measured in different reference frames ae not generally equal. We say that the coordinate time differences
are relative (that is, they depend on one's choice of inertial reference frame).

Spatial coordinate differences

are relatirre also

VVhy? If each observer synchronizes the clocks in his or her own reference frame according to our definition, ech will conclude that the clocks in
the other's frnme re not synchronlzed. Notice that the Other Frame observer
has set the right and left clocks to read Ilrle same time at events A and B. Yet
these events do not occur at the same time in the Home Frame. Therefore, the
Home Frame observer will claim that the clocks in the Other Frame are not
s)mchronizecl. (Of course, the Other Frame observer will say the same thing
about the clocks in the Home Frame.) The definition of synchronization that
we are using makes perfect sense within any inertial reference frame, but it
does not allow us to slmchronize clocks rn difrent inertial frames. In fact, the
definition rcquires that observers in different inertial frames measure dffirent
time intervals between the same two events, as we have just seen.
In general, two observers in different frames will also disagree about the
spatial coordinate separation between the events. Consider events C and D
that both occur at the center clock in the Other Ftame, but at different times'

Since the center clock defines the location x' :0 in the Other Frame, the
events have the same / coordinate in that frame, so Ar' : 0. But in the Home
Frame, the center clock is measured to move in the time between the events,
and so the two events do not occur at the same place: A'x * 0 (see figure R2.9) '

Worldline of
the center
clock in the
Other Frame
C

Exercise R2X.3
o

rcxo

Figure R2.9
Events C and D both occur at the
same place in the Other Frame
(Ax': 0) but not in the Home
Frame (Ax * 0). (Ths diagrani
is drawn by an observer in the

Home Frame.)

The raclar rnethod yields the

same conclusion

Note that the frame dependence of ttle sptinl coordinate difference between
two events has nothing to do with clock synchronization or relativity: this
would be true even if time were universal and absolute. Show, using the
Galilean transformation equations, that if the separation between two events
in the Other Frame is Ax' : 0 but Lt' + 0, then the separation between these
events in the Home Frame isnot zero (^r # 0).

The reason why observers in different inertial frames disagree about

whether the clocks in a given frame are synchronize d is that synchronization is
defined so that light flnshes arc mesured to haae speed of f in eaery inertial f'rame:
the frame-dependence of coordinate time differences is a logical consequence
of this. This can be illustrated by considering the radar method of assigning
spacetime coordinates. Although the radar method does not involve the use
of slmchronized clocks, it does depend on the assumption that the speed of
light is the same in every inertial frame. Does the radar method also imply
that the coordinate time difference between two events is frame-dependent?
Figure R2.10 shows that it does. Figure R2.10a shows the observer in the
Other Frame using the radar method to determine the spacetime coordinates
of event C. The observer in that frame will conclude that event C and event D

occur at the same time, where D is the event of the master clock at r' : 0 registering tb: TG) i t[), thatis, at the instant of time halfway between the emission of the radar pulse at fj and the reflection's reception at f f . According to the
rdar method, then, the coordinate time between events C and D is Af ': 0.
[Radar and visible light are both electromagnetic waves (with different frequencies), so both will move at a speed of 1 light-second/second : 1.]

R2.7 A Geometric Analogy

35

t'

Figure R2.10

Reflected
radar puise

worldline

worldline

Worldline
of Other

Frame's
master

tc

clock

LI)

>C
D

'

Outgoing
radar pulse

(a) r'=

x'
0

outgoing
radar puise

worldline

Reflected
radar pulse

(b)

worldline

(a) ln the Other Frame, events C and D are

defined to be simultaneous if D occurs at
the master clock at a time halfway between
the emission event A and the reception
event B. The coordinate time difference
between events C and D in the Other Frame
is thus At'= 0. (b) ln the Home Frame,
the Other Frame's master clock moves to
the right as time passes, so its worldline
is slanted. On the other hand, radar pulse
worldlines still have slope +1, as shown.
This means that an observer in the Home
Frame will conclude that event C happens
after event D, so the coordinate time
difference between the events is
+ 0.

^t

When the same sequence of events is viewed from the Home Frame,
though, a different conclusion emerges (see figure R2.10b). According to
observers in the Home Frame, the Other Frame's master clock is moving
along the x axis with some speed , so in a spacetime diagram based on
Home-Frame measurements, that clock's worldline will appear as a slanted
line (with slope 1./ {3) instead of being vertical. Radar pulse worldlines, on
the other hand, still have slopes oI +1, just as they did in the Other Frame
spacetime diagram. The inevitable result (as you can see from the diagram) is
that observers in the Home Frame are forced to conclude that event C occurs
after event D does, and thus that the time difference between events C andD
in the Home Frame is At * 0.
The point is that the relativity of the coordinate time interval between
events. is a direct consequence of the fact that we arc defining coordinate time
by assuming that the speed of light is 1 in every inertial reference frame.
Remember, though, that we must make this assumption if the laws of electromagnetism are to be consistent with the principle of relativity!

R2.7

Frame dependence of
coordinate time follows from
principle of relativity

A Geometric Analogy

You may find it troubling that coordinate differences between events are not

absolute but are instead frame-dependent. This is particularly true of the

time coordinate separation: it is not easy to let go of the Newtonian notion of
absolute time! The fact is, though, we haue no trouble at all with these ides when
they apper in a related but more familiar guise.
Consider AskeW a hypothetical town somewhere in the western
United States (its embarrassed residents wish its precise location to remain
secret). Most towns in the rural United States have streets that run directly
north/south or east/west. The surveyor who laid out the streets of Askew in
1882, however, tried to calibrate his compass against the North Star the night
before, but in fact had forgotten exactly where it was (it was a long time since
he had this stuff in high school, after all) and ended up choosing a star that
turned out to be 28' east of the true North Star. Therefore, all the streets of
Askew are twisted 28'from the standard directions.
Now, if we would like to assign r and y coordinates to points of interest in
this town (or any town), we need to set up a Cartesian coordinate system. We

conventionally to orient coordinate axes on

plot of land so that the x andy

An illustration of alternative
coordinate systems in plane
geometry

Chapter

R2

Coordinate Time

t
I

x
I

tt

1. City Hall
Statue ofthe

2.

Unknown Physicist

(b)

(a)

Figure R2.11
(a)

A standard nodhward-oriented Cartesian coordinate system superimposed on the town of Askew.

(b) A more convenient coordinate system oriented 28' clockwise'

axes point north and eas! respectively (see figure R2.11a). This is usually also
convenien! since the streets in rnosf towns are parallel to those axes. There is no

reason why thismustbedone, though, and inAskew's case, it is actually more

convenient to use a coordinate system tilted 28' clockwise (figure R2.11b).

Note that City Hall is the origin of both coordinate systems.
We can (of course) use any coordinate system we like to quantify the
positions of points of interest in the town, since coordinate systems are
rbitrary human artifacts that we impose for our convenience on the physical
world. F{owever, the coordinates we actually obtnin 1 various points
certainly do dependon the coordinate system used. For example, the coordinate differences between City Hall and the Statue of the Unknown Physicist
in Memorial Park might b" Ly:0, A.x: 852'0 m in the standard coordinate system ,buf Ly' : 399'9 m, Lx' :752.3 m in the conveniently oriented
coordinate system.
We are not surprised by frame

dependence of coordinates in

this case!

Is it surprising that the results are different? Do the differences in the

results cause us to suspect that one or the other coordinate system has been set
up incorrectly? Hardly! We already know nd expect that differently oriented
coordinate systems on a plane will yield different coordinate measulements.
This causes no discomfort, because we undefstand this is the way things are.

In an entirely analogous way, we have carefully and unambiguously

defined a procedure for setting uP an inertial reference frame and synchronizing its clocks. This definition happens to imply that spacetime coordinate
measurements in different frames yield different results. This should be no
more troubling to us than the fact thatAskew residents who use different sets
of coordinate axes will assign different coordinates to various points in town.
Coordinates hue mening only
frame we use to obserae them.

in the context of the coordinte system or inertil

The only reason that the relativity of time coordinate differences is a difficult idea is because we don't have common experience with inertial reference
frames moving with high enough relative speeds to display the difference.
The frames that we typically experience in daily life have relative speeds
below 300 m/s, or about one-millionth of the speed of light. If for some reason w could only construct Cartesian coordinate systems on the surface
of the earth that differed in orientation by no more than one-millionth of
a radian, then we might also consider Cartesian coordinate differences as
being "universal and absolute" as well!

R2.7 A Geometric Analogy

37

So, to summarize, t}iLe coordinate differences between points on a plane

(or events in spacetime) are "relative" because coordinate systems (or inertial reference frames) are human artifacts that we impose on the land (or on
spacetime) to help us quantify that physical reality by assigning coordinate
numbers to points on the plane (or events in spacetime). Because we are free
to set up coordinate systems (or reference frames) in different ways, the coordinate differences between two points (or events) reflect not only something
about their real physical separation, but also something about the artificial

choice of coordinate system (or reference frame) we have made.

So, is it true, then, that euerything is relative? Is there nothing we can measure about the physical separation of the points (or events) that is absolute
(that is, independent of reference frame)?
T1lre distnnce between two points is such a quantity. For example, the distance between Askew's City HaIl and the Statue of the Unknown Physicist is

Itl: (Lx,'+ /ry')''':

m)'+ 0]'/' : 852.0 minthenorth-orientedcoorl(Lx' )2 + (/\y' )'l' /' : [(3g9.g m)2 + (782.3 rn)'l' /' :

[(852.0

dinate system an ' | :

852.0 m in the convenient coordinate system. It doesn't matter what coordinate
system one uses to calculate ll l, yo" always get the same answer.
The distance between two points on a plot of ground thus reflects something deeply real about the nature of the plot of ground itself, independent of
the human coordinate systems we impose on it. This distance is independent
of coordinate systems because we can, in fact, determine it directly without
using a coordinte system at all simply by laying a tape measure between the
points! Since this method yields an unambiguous distance, valid calculations
of this distance in any coordinate system should yield the same value.
Of course, there are may ut)nys that one could lay a tape measure between
City Hall and the Statue of the Unknown Physicist. One could lay the tape
measure along a straight path between the two points: this would measute
the distance "as the crow flies," which is what is usually meant by the phrase
"the distance between two points." But there are other possibilities. One
might, for example, lay the tape measure two blocks down Elm Street from
City Hall, then one block over along Grove Avenue, then up Maple Street,
and so on. This would measure a different kind of distance between the two
points that we might call apthlength.
Both the straight-line distance and the more general pathlengthbetween
two points can be measured directly with a tape measure, and thus are quantities independent of any coordinate system. But the distance and the pathlength between two points may not be the same. In general, the pathlength
between two points will depend on the path chosen, and will always be
greater than (or at best equal to) the straight-line distance.
To summarize, we can quantify the separation of two points on a plane
three totally different ways. We can measure tlrre coordinte separationsbetween
the points, using a coordinate system. (The results will depend on our choice
of coordinate system.) We can measure tlrle pathlengfh between them with a
tape measure laid along a specified path. (The result here will depend on the
path chosen, but is independent of coordinate system.) Or we can measure
t},':re distnce between the points with a tape measure laid along the unique
path that is the straight line between the points. Because in this last case the
tape's path is unique, the distance between two points is a unique number
that quantifies in a basic way the separation of those points in space.
Analogously, we can measure the time between two events in spacetime
in three different ways.The coordinate timebetween events is analogous to the
coordinate separation of points and so depends on one's choice of reference
frame. In the next chapter, we will see how we can define frame-independent
times analogous to pathlength and distnnce.
|

Distance and pathlength,

on the other hand, are
coordinate-l nd ep end ent

The three kinds of spatial

separation

Chapter

38

R2T.1

R2

Coordinate Time

Imagine that in the distant future you (on earth)

A

are watching a transmission from Pluto, which at the time

is 5.0 tight-hours from earth. You notice that a clock on

the wall behind the person speaking in the video reads
12:10 p.m. You note that your watch reads exactly the same
time. Is the station clock synchronized with your watch?

A.
B.
C.

3s
r

The problem doesn't give enough information to tell.

Suppose you receive a message from a starbase that

15,2127. \trhat year does your calendar indicate at the time

of reception if your calendar and the station's calendar are

correctly synchronized?

A.
B.

2127

D.

Other (specify)

2114
2140

R2T.3

The speed of a typical car on the freeway expressed

in SR units is most nearly
A. 107
10-10

c.

108
106

F.
T.

Other (specify)
None of these answers is right: we must state units!

D.
E.

it is.
No, it isn't.

is 13.0 light-years from earth. The message is dated July

B.

Yes,

R2T.2

c.

10-4

R2T.4

Sqppose you are sitting at the origin of an inertial

reference frame. You see (that is, you receive the light from)
an event E occurring near a clock at x : -30 ns at a time
f : 80 ns. When do youobserue that event to occur?

A. fr=g
B. fr : 39 t'tt
C. fr = 59 t'tt
D. tE: 80 ns, of course
E. fr : 119 t',t
F. Some other time (specify)
R2T.5

The spacetime diagram in figure R2.12 shows the

worldlines of various objects. Which object has the largest

speedattimef:1s?

R2T.6

The spacetime diagram in figure R2.12 shows the

worldlines of various objects. Which object has the largest

n^

Figure R2.12
Worldlines of various objects.

R2T.9 A light flash leaves a master clock at r : 0 at time

t = -12 s, is reflected from an object a certain distance in
the -x direction from the origin, and then returns to the

s. From this information, we can infer that

the spacetime coordinates of the reflection event are ft, x1:

origin at f = 18

A. [4 s,20 s]
B. l-4 s, -20 sl
C. [10 s, -2 s]
D. [2 s, -10 s]
E. [-2 s, -10 s]
F. Other (specify)
R2T.10 Coordinate time would be frame-independent if the
Newtonian concept of time were valid. T or

F?

R2T.11 Consider a Home Frame and an Other Frame that

moves in the *r direction with respect to the Home Frame.
(a) Observers in the Home Frame will conclude that the
clocks in an Other Frame will be out of synchronization, even if the observers in the Other Frame have
carefully synchronized clocks using the Einstein prescription. T or F?
(b) Specifically, Home Frame observers will see that for
events farther and farther up the common *r axis, the
times registered by Other Frame clocks at the events

A. Become further and further ahead.

B. Become further and further behind.
C. Remain the same.
D. Have no clear relationship to the values that Home
Frame clocks register for the same events.

speedattimef:4s?

R2T.7 The spacetime diagram in figure R2.12 shows the

worldlines of various objects. Which worldline cannot
possibly be correct? (Explain why.)
R2T.8

In figure R2.12 the object whose worldline is labeled B

is moving along the axis. T or F?

R2T.12 In the geometric analogy, the coordinate time difference Af between two events in spacetime corresponds to

A. The north-south separation between points on a plane.

B. The distance between points on a plane.
C. A certain pathlength between points on a plane.
D. The separation between the events in spacetime.
E. Something else (specify).

Homework Problems

39

HOMEWORK PROBLEMS
Basic Skills
C

R28.L

(Practice with SR units.)

(a) What is the earth's diameter in seconds?
(b) Ahighway sign reads "Speed Limit 6 x 10-8," meaning
speed in SR units. What is this in miles per hour?
(c) Argue that the SR unit of acceleration is s 1. \Ihat is
1 s-1 expressed as a multipte of l$ ll

4s

R28.2

(Practice with SR units.)

(a) A sign on a hiking trail reads "Viewpoint: 5.5 ps."
About how long would it take you to walk to this
viewpoint at a typical walking speed of I m/s?
(b) Section R1.2 mentions that the Voyager 2 spacecraft

achieved speeds in excess o172,000 km,/h. What is this

speed in SR units?
(c) In SI units, power is measured in watts (where 1 W
=
1J/s = 1 kg.m2ls3). Argue that the natural SR units
of power are kgls. Let's define 1 SR-watt 1 SRW =

lkg/s. Alarge

electric power plant produces energy at

a rate of about 10e W.

What is this in SRW?

R28.3

(Practice with SR units.)

(a) Argue that the SR units of mass, momentum, and

kinetic energy are simply kilograms.

(b) Imagine that a truck with a mass of 25 metric tons (that
is,25,000 kg) is barreling down a highway at a speed of
59 mi/h. \Vhat is the truck's momentum in kilograms?
(c) What is the truck's kinetic energy in kilograms?

R28.4.

(Practice with SR units.)

(a) Argue that force has SR units of kg,/s

R28.5

For each of the worldlines shown in figure R2.13,

describe in words what the particles are doing, giving
numerical values for velocities when possible. For example, you might say, "ParticIe A is traveling in the *x direction with a constant speed of I "
.

Suppose you send out a light flash at f : 3.0 s, as

registered by your clock, and you receive a return reflection showing your kid brother making a sitty face at
t : \I s, as registered by your clock.
(a) At what time did your brother actually make this face?
(b) How far is your brother away from you (express your
result in seconds and kilometers)?

R28,6

nuisance?
Suppose you send out a radar pulse at t : -22h, as
registered by your clock, and receive a reflection from an
alien spacecraft at t : *12 h as registered by your clock.
(a) Is the spaceship inside or outside the solar system?
(b) When did the spaceship reflect the radar pulse?

R28.7

Figure R2.13
Wordlines of various objects.

R28.8 A

firecracker explodes 30 km away from an

observer sitting next to a certain clock A. The light from the
firecracker explosion reaches the observer at exactly f : 0
according to clock A. Suppose the firecracker's flash illu-

minates the face of another clock B that is sitting next

to the firecracker. \Alhat time will clock B register at the
moment of illumination if it is correctly synchronized with
clock,4? Express your answer in milliseconds.

R28.9
space

Suppose you are in an inertial frame in empty

with a clock, a telescope, and a powerful strobe

light. A friend is sitting in the same frame at a very large

(unknown) distance from your clock. At precisely 12:00:00
noon according to your clock, you set off the strobe lamp.
Precisely 30.0 s later, you see in your telescope the face of
your friend's clock illuminated by your strobe flash.
(a) How lar away is your friend from you (in seconds)?
(b) What should you see on the face of your friend's clock
if that clock is synchronized with yours?
Describe your reasoning in a few short sentences.

(b) What is 1 kgls in newtons?

(c) What are the SR units of pressure?
(d) What are the SR units of density?

(c) Is this far enough away that he can't really be

4s

-4s

R28.L0 Draw a spacetime diagram that displays worldlines

for the following particles.
(a) Particle ,4. travels at a constant speed of in the +x
I
direction and passes the point : 0 at time f : 2 s.
(b) Particle B, which at time f : 0 is at position : *2 s is
traveling at a speed of ] in the -r direction, is slowing
down as time passes, and eventually comes to rest at

r:0attime:8s.

(c) Light flash C passes the position *

as it travels in the *r direction.

0 at time f

-2

R28.11 Draw a spacetime diagram that shows the following worldlines.

(a) Particle ,4 travels at a constant speed of in the
$
-r
direction and passes the point x : 0 at time t : -2 s.
(b) Particle B, which at time f : 0 is at rest at position
= 0, accelerates in the *x direction asymptotically
toward the speed of light as time passes.
(c) Light flash C passes the position x :0 at time : *3 s
as it travels in the - direction.

Chapter

40

R2

Coordinate Time

r': 0 and
ns, respectively, on a train that is moving in the
*x direction relative to the ground frame' According to
synchronized clocks on the train, both firecrackers explode
simultaneously. \trhich firecracker explodes first according
to slmchronized clocks on the ground? Explain carefully.
(Hint: Study figure R2.8 carefully.)

R28.12 Two firecrackers A and B are placed at

x'

L00

R28.13 Figure R2.8 implies that an observer in the Home

Frame concludes that clocks that have been properly synchronized in the Other Frame are nof synchronized in the
Home Frame. Would an observer in the Other Frame conclude that clocks that have been carefully synchronized in
the Home Frame are not synchronized in the Other Frame?
Draw a spacetime diagram from the point of view of the
Other Frame to justify your response.
R28.14 Redraw figure

R2.8, assuming that the Newtonian

concept of time is true. How does your redrawn diagram

differ from the original, and how is this difference related
to the behavior of light according to the Newtonian and
relativistic models?

Draw a quntittioely accurate spacetime diagram of this

situation (as observed in a frame attached to the earth,
with the earth at x : 0) that shows the worldlines of the
earth, the spaceship, the returning laser signal, and events

A,B,andC.
R2M.3 An alien spaceship traveling at a constant velocity
of J in the * direction passes the earth (call this event,4)
at time f = 0. Just as the spaceship passes, people on the
earth launch a probe, which accelerates from rest toward
the spaceship at such a rate that it catches up to and Passes
the alien spaceship (call this event B) whenboth are 10 min
of distance from the earth. As it passes the alien spaceship,
the probe takes a photo and sends it back to the earth as an
encoded radio message that travels at the speed of light.
The message reaches the earth at event C. Draw a quantitatiaely accurate spacetime diagram of this situation (as
observed in a frame attached to the earth, with the earth
at r : 0) that clearly shows the worldlines of the earth, the
alien spaceship, the probe, the returning radio message,
and events , B, and C. In particular, clearly indicate when
the people on earth receive the photo.

R2M.4 A rocket launched from the moon travels away

Modeling
R2M.1 The spacetime diagram below shows the worldline of a rocket as it leaves the earth, travels for a certain
amount of time, comes to rest, and then explodes.

I
600

a Explosion

from it at a speed of !. Call the event of the rocket's launching event.4. After 125 s, as measured in the reference frame
of the moon, the rocket explodes: call this event B. The
light from the explosion travels back to the moon: call its
reception event C. Let the moon be located at = 0 in its
oivn reference frame, and let event.A define f : 0. Assume
that the rocket moves along the *x direction.
(a) Draw a spacetime diagram of the situation, drawing
and labeling the worldlines of the moon, the rocket,

and the light flash emitted by the explosion and

Worldline
of rocket

received on the moon.

Worldline
of earth

(b) Draw and label events ,4, B, and C as points at the

appropriate places on the diagram. Write down the f
and

300 s

300 s

(a) The rocket leaves the earth; the rocket comes to rest in
deep space; the rocket explodes. What are the coordinates of each of these three events?
(b) \Alhat is the rocket's constant speed relative to the earth
before it comes to rest?
(c) A light signal from the earth reaches the rocket just as
it explodes. Indicate on the diagram exactly where and

when this light signal was emitted.

R2M.2

Suppose a spaceship leaves the earth at event A and

travels in the * direction, accelerating at a constant rate

from rest at time f : 0 to a final speed of f at time t :Lh.
It remains at that speed thereafter. Just as the spaceship
reaches that speed, it emits a laser signal back toward the
earth (call this event B), which reaches the earth at event C.

r coordinates of these

events.

A spaceship in deep space is approaching a space

n at a constant speed of I t I : t. tet the space station
define the position : 0 in its own reference frame. At time
f :0, the spaceship is 10.0 light-hours away. At that time
and place (call this event,4), the spaceship sends a laser pulse
of light toward the station, signaling its intention to dock. The
station receives the signal at its position (call this event B),
and after a pause of 100 min (everyone was at lunch), emits

another laser pulse signaling permission to dock (call this

event C). The spaceship receives this pulse (call this event
D) and immediately begins to decelerate at a constant rate.
It arrives at rest at the space station (call this event ) 4.0 h
after event D, according to station clocks.
(a) Carefully construct a spacetime diagram that shows
the worldlines of the space station, the approaching
spaceship, and the two light pulses. Also indicate the

time and place (in the station's frame) of events A

through E by labeling the corresponding points on the
diagram. Scale your axes using the hour as the basic
time and distance unit (you might subdivide each hour
into units of I h)'

Homework Problems

41

(b) In particular, exactly when and where does event D

occur? Event ? Write down the coordinates of these
events in the station frame, and explain lrow youclcuIated them (reading them from the diagram is a useful
chec but is not enough).
(c) Compute the magnitude of the spaceship's average
acceleration between events D and E in SR units (s-1)

and as a multiple of li | . Note that a shockproof watch

can typically tolerate an acceleration of about 50 | i l.

R2M.6

Suppose a spaceship is docked at a space station

floating in deep space. Assume the space station defines

the origin in its own frame. At f

: 0 (call

R2M.9

(S

eeing is

not fhe same

as obseroing!) Suppose at

time

f = 0 you (on earth) simultaneously receiae a message (sent

via a laser transmission) emitted (event A) by the outpost
on Venus (which is 180 Gm away at the time) and a message emitted (event B) by the outpost on Mars (which is
270 Gm away). Each message requires an urgent response

that must be received no more than 40 min after it was

sent. Which message do you respond to first?

R2M.10

(Seeing is not the same as obseroing!) Imagine that a

bullet-train running at auery high speed passes two trackside signs (,4 and B), as shown in the aerial view below.

this event ,4)

the spaceship starts accelerating in the * direction away

from the space station at a constant rate (as measured in
the station frame). The spaceship reaches a cruising speed
of 0.5 after 8 h as measured in the station frame (call this
event B). fHint:Youwill find it helpful to construct a spacetime diagram as you go through the parts of this problem:
such a diagram is required in part (e) anyway.J
(a) IMere does event B occur? Explain your reasoning.
(b) Find the magnitude of the ship's acceleration in SR
units (s-1) and as a multiple
Note that a shock"f l.g l.
proof watch can tolerate an acceleration
of =50 l | .
(c) At event B, the spaceship sends a radio signal to the
station, reporting it has reached cruising speed. This
signal reaches the station at event C. When and where
does this event occur? (Hint: Radio signals are electromagnetic waves that travel at the speed of light.)
(d) The technician responds to this message 0.5 h later
after returning from a coffee break: call this event D.
When and where does the ship receive this acknowledgment (event E)?

(e) Draw a careful spacetime diagram of this situation,

showing the worldlines of the space statiory the spaceship, the radio signals, and events A through E. Be sure
to label all these items appropriately.

R2M.7 An air traffic control radar installation

receives a

radar pulse reflected from a certain jet plane 280 ps after

the pulse was sent. The signal comes from a direction that
is 35' north of west and 5.5' up from horizontal. If the
sending of the outgoing pulse defines f = 0, in a frame
fixed to the earth's surface and oriented in the usual way
with the installation at the spatial origin of the frame, then
what are the spacetime coordinates [t, x, y, z] of the plane
at the instant it reflects the pulse? (Hint: Radar pulses are
electromagnetic waves, so they travel at the speed of light.)

R2M.8 Imagine you are in a spaceship prospecting for

asteroids. Your radar installation receives a pulse reflected

from a nice large asteroid 1..24 s alter it was sent, and the
returning signal comes from a direction 25' to the right
and 18' up from the direction your ship is facing. Assuming that the direction your ship is facing defines the *x
direction, and the up direction is the *z directiory and that
we define f : 0 to be when the pulse is sent, what are the
spacetime coordinates of the asteroid at the time the
pulse is reflected? (Hint: Radar pulses are electromagnetic
waves, so they travel at the speed of light.)

Sign B

80 ns

45 ns

Sign,4

60 ns

o
Let event 24, be the passing of the front end of the train by
srgnA, and let event B be the passing of the rear end of the
train by sign B. An observer is located at the cross marked
by an O in the diagram.

(a) This observer

sees (that is, receives light with her eyes)

event A to occur at time f : 0 and ses event B to occur
at time f : 25 ns. When does she obserue these events
to occur? That is, what would a clock present at sign A
read at event,4, and what would a clock present at sign
B read at event B if these clocks were correctly synchronized with the clock at O?
(b) In what way is the diagram misleading about the
implied time relationship between events,4 and B?
(Hinf: Remember that the clocks at signs A and B must be
synchronized with the clock at O in such a way that they
would read the speed of a light signal traveling between
them and O to be 1 s/s. Given this, the distance between O
and A, and the time that light from event reached O, can
you infer when event A must have happened?)

R2M.11 After reading this chapter, your roommate exclaims,

"Relativity cannot be right! This chapter claims that events
that are simultaneous in one inertial reference frame are not
simultaneous in another. Yet it is clear that two events are
really simultaneous or really not simultaneous! This is not
something that different observers could disagree about;
or if they do, one has to be right and the other wrong!"
Carefully and politely argue to your roommate that relatll|y could be right even so, and pinpoint the assumption
that your roommate makes that could be debated. (Hint:
You might be able to use the geometric analogy to good
effect. Two different surveyors set up differently oriented
coordinate systems on a plot of land. In one system, two
rocks both lie along the x axis; in the other, they do not. Is
this a problem?)

Chapter

42

R2

Coordinate Time

Rich-Context
R2R.L Imagine that an advanced alien race, bent on keeping humans from escaping into the galaxy, places an
opaque spherical force field around the solar system. The
force field is 6 light-hours in radius, is centered on the sun,
and is formed in a single instant of time as measured by
slmchronized clocks in an inertial frame attached to the
sun. That instant corresponds to 9 p.m. on a certain night
in your time zone. When does the opaque sphere appear
to start blocking light from the stars from your vantage
point on earth (8.33 light-minutes from the sun)? Does the
opaque sphere appear all at once? If not, how long does it
take for the sphere to appeal and what does it look like as
it appears? Describe things as completely as you can. (This
is inspired by the novel Qurantineby Greg Egan.)

R2R.2 Two radar pulses sent from the earth at 6:00 a.m.
and 8:00 a.m. one daybounce off an alien spaceship and are
detected on earth at 3:00 p.m. and 4:00 p.m. ut you aren't
sure which reflected pulse corresponds to which emitted
pulse). Is the spaceship moving toward earth or away? If
its speed is constant (but less than c), when will it (or did it)
pass by the earth? (Hint:Draw a spacetime diagram.)

R2R.3

(Seeing is not the same as obseraing!) Imagine you

are sitting immediately adjacent to a set of train tracks. A
certain bullet-train running at a speed of 0.5 (in SR units)
on these tracks has three blinking lights, one at each end,
and one in the middle. The end lights are 200 ns of distance

from the middle light according to measurements in your

ground frame. As the train's center passes you (a negligible distance away), you see all three lights blink simultaneously. Do you obseroe them to blik simultaneously?

Explain. If you do not observe them to blink simultaneously, which light blinks first, and how much in advance of
the center light does it blink? Explain carefully.

R2R.4 A train

is moving due east at a large constant speed

on a straight track. Suppose that Harry is riding on the

train exactly halfway between its ends. Sally is sitting by
the tracks only a few feet from the train. Let the event of
Harry passing Sally be the origin event O in both frames.
At this same instant, both Harry and Sally receive the light
from lightning flashes that have struck both ends of the
trairy leaving scorch marks on both the train and the track.
Harry concludes that since he is in the middle of the train
and he received the light from the strikes at the same time,
the lightning strikes must have occurred at the same time
in his reference frame. Is he right? If not, which strike (the
one at the front of the train or the one at the rear) really
happened first? Can Sally conclude from her seeing the
flashes at the same time that the strikes happened at the
same time in the ground frame? Why or why not? If not,
which strike happened first in her frame? (This problem
is adapted from one of Einstein's own illustrations of the
implications of the frame-independent speed of light.)

Advanced

R24.1 A meter stick moves

at a speed of 0.5 (in SR units)

along a line parallel to its length that passes within 1 m of
a camera. The camera shutter opens for an incredibly brief
instant just as the meter stick's center passes closest to the
camera. Explain why the marks on the meter stick do tof
look equally spaced in the resulting picture, and describe
what they look like. (Ignore length contraction.)

ANSWERS TO EXERCISES
R2X.1

The distance between the earth and the moon in

seconds is

zs+.oook"l[loooYll

\ ry

1' :Lza,

i\a.o x ro'4/

So the clock on the moon

6s

(R2.5)
S

should read 1.28 s after noon if

it

is synchronized with the earth clock.

R2.6

With the new worldline (shown in color), figure

A.

R2.4 becomes figure R2.14. Note that the particle worldline has a slope of -5 on the diagram, since it moves 1 s of

distance in the

direction per 5 s of time.

xb

- xL: (xo - Btp) - (x - Bt)

: (xo - x - (t" - tc)

(R2.6)

2s

4s

Figure R2.14

R2X.3

Consider events C and D with x coordinates xand

xp, respectively, in the Home Frame and x' and xb, respectively, in the Other Frame. According to the Galilean transformation equations (equations CA), we have

Figure R2.4 with the new worldline

If Lx' :

- xt = 0, then
0: (ro - xr) - (tr- tr) + xo- xc: (to- t (1U.7)
So if
' : Lt: to - tc* 0, then A)c: )co - xc* 0 as well.
^f

xb