Kuhn, Thomas - Copernican Revolution, The (Harvard, 1985) PDF

THE COPERNICAN REVOLUTION
Planetary Astronomy in the Development of Western Thought

THOMAS 5. KUHN
HARVARD UNIVERSITY PRESS CAMBRIDGE,
MASSACHUSETTS, AND LONDON, ENGLAND
Copyright 1957 by the President and Fellows of Harvard College
renewed 1985 by Thomas S. Kuhn
EIGHTEENTH PRINTING, 1995
Library of Congress Catalog Card Number 57-7612
ISBN 0-674-17103-9 (paper)
Printed in the United States of America
TO L. K. NASH
FOR A VEHEMENT COLLABORATION
PREFACE
The story of the Copernican Revolution has been told many
times before, but never, to my knowledge, with quite the scope and
object aimed at here. Though the Revolution's name is singular, the
event was plural. Its core was a transformation of mathematical as-
tronomy, but it embraced conceptual changes in cosmology, physics,
philosophy, and religion as well. These individual aspects of the Revo-
lution have been examined repeatedly, and without the resulting
studies this book could not have been written. The Revolution's
plurality transcends the competence of the individual scholar working
from primary sources. But both specialized studies and the elementary
works patterned on them necessarily miss one of the Revolution's most
essential and fascinating characteristics - a characteristic which arises
from the Revolution's plurality itself.
Because of its plurality, the Copernican Revolution offers an ideal
opportunity to discover how and with what effect the concepts of
many different fields are woven into a single fabric of thought.
Copernicus himself was a specialist, a mathematical astronomer con-
cerned to correct the esoteric techniques used in computing tables of
planetary position. But the direction of his research was often deter-
mined by developments quite foreign to astronomy. Among these were
medieval changes in the analysis of falling stones, the Renaissance
revival of an ancient mystical philosophy which saw the sun as the
image of God, and the Atlantic voyages which widened the terrestrial
horizons of Renaissance man. Even stronger filiations between distinct
fields of thought appear in the period after the publication of
Copernicus' work. Though his De Revolutionibus consists principally
of mathematical formulas, tables, and diagrams, it could only be
assimilated by men able to create a new physics, a new conception of
space, and a new idea of man's relation to God. Creative interdiscipli-
nary ties like these play many and varied roles in the Copernican Revo-
lution. Specialized accounts are inhibited both by aim and method
from examining the nature of these ties and their effects upon the
growth of human knowledge.
This account of the Copernican Revolution therefore aims to dis-
viii
PREFACE
play the significance of the Revolution's plurality, and that object is
probably the book's most important novelty. Pursuit of the object has,
however, necessitated a second innovation. This book repeatedly
violates the institutionalized boundaries which separate the audience
for "science" from the audience for "history" or "philosophy." Occa-
sionally it may seem to be two books, one dealing with science, the
other with intellectual history.
The combination of science and intellectual history is, however,
essential in approaching the plural structure of the Copernican Revo-
lution. The Revolution centered in astronomy. Neither its nature, its
timing, nor its causes can be understood without a firm grasp upon the
data and concepts that were the tools of planetary astronomers. As-
tronomical observations and theories therefore make up the essential
"scientific" component which dominates my first two chapters and
recurs throughout the remainder of this book. They do not, however,
make up the whole book. Planetary astronomy was never a totally
independent pursuit with its own immutable standards of accuracy,
adequacy, and proof. Astronomers were trained in other sciences as
well, and they were committed to various philosophical and religious
systems. of their nonastronomical beliefs were fundamental first
in postponing and then in shaping the Copernican Revolution. These
nonastronomical beliefs compose my "intellectual history" component,
which, after Chapter 2, parallels the scientific. Given the purpose of
this book, the two are equally fundamental.
Besides, I am not convinced that the two components are really
distinct. Except in occasional monographs the combination of science
and intellectual history is an unusual one. Initially it may therefore
seem incongruous. But there can be no intrinsic incongruity. Scientific
concepts are ideas, and as such they are the subject of intellectual
history. They have seldom been treated that way, but only because few
historians have had the technical training to deal with scientific source
materials. I am myself quite certain that the techniques developed by
historians of ideas can produce a kind of understanding that science
will receive in no other way. Though no elementary book can fully
document that thesis, this one should provide at least preliminary evi-
dence.
Indeed it has already provided some evidence. The book grows out
of a series of lectures delivered each year since 1949 in one of the
PREFACE ix
science General Education courses at Harvard College, and in that
application the combination of technical and intellectual-historical
materials has been quite successful. Since students in this General
Education course do not intend to continue the study of science, the
technical facts and theories that they learn function principally as
paradigms rather than as intrinsically useful bits of information.
Furthermore, though the technical scientific materials are essential,
they scarcely begin to function until placed in a historical or philo-
sophical framework where they illuminate the way in which science
develops, the nature of science's authority, and the manner in which
science affects human life. Once placed in that framework, however,
the Copernican system or any other scientific theory has relevance and
appeal for an audience far broader than either the scientific or the
undergraduate community. Though my first purpose in writing it was
to supply reading for the Harvard course and for others like it, this
book, which is not a text, is also addressed to the general reader.
friends and colleagues, by their advice and criticism, have
helped to shape this book, but none has left so large or significant a
mark as Ambassador James B. Conant. Work with him first persuaded
me that historical study could yield a new sort of understanding of the
structure and function of scientific research. Without my own Coperni-
can revolution, which he fathered, neither this book nor my other
essays in the history of science would have been written.
Mr. Conant also read the manuscript, and its early chapters show
many signs of his productive criticisms. Others who will recognize here
and there the effects of their useful suggestions include Marie Boas,
I. B. Cohen, P. Gilmore, Roger Hahn, G. J. Holton, E. C. Kemble,
P. E. LeCorbeiller, L. K. Nash, and F. G. Watson. Each has applied
critical talent to at least one chapter; several read the entire manu-
script in an earlier version; and all have rescued me from mistakes or
ambiguities. The advice of Mason Hammond and Mortimer Chambers
has given my occasional Latin translations an assurance that they
would not otherwise possess. Arnolfo Ferruolo first introduced me to
Ficino's De Sole and showed me that Copernicus' attitude toward the
sun is an integral part of a Renaissance tradition more striking even
in the arts and literature than in the sciences.
The illustrations display the skill, but scarcely the patience, with
which Polly Horan has translated and retranslated my vague
X PREFACE
directions into communicative symbols. J. D. Elder and the staff of the
Harvard University Press have given me constant and sympathetic
guidance in the arduous transmutation to type of a manuscript that
conforms neither to the rules for scientifl.c publication nor to those for
history. The index attests the industry and intelligence of W. J. Charles.
The joint generosity of Harvard University and the John Simon
Guggenheim Memorial Foundation provided the year's leave of
absence during which most of my manuscript was first prepared. I am
also grateful to the University of California for a small grant which
assisted in the final preparation of the manuscript and in seeing it
through the press.
My wife has been an active participant throughout the book's de-
velopment, but that participation is the least of her contributions to it.
Brain children, particularly someone else's, are the most obstreperous
members of any household. Without her continuing toleration and fore-
bearance this one would never have survived.
Berkeley, California
November 1956
T. S. K.
Note to the Seventh Printing. This printing contains a number of cor-
rections and textual changes inadvertently omitted from earlier Harvard
editions. With this and subsequent printings, all changes previously in-
troduced in the Random House and Vintage paperback editions are also
included in the Harvard Paperback together with a few minor correc-
tions after the earlier paperback editions were prepared.
CONTENTS
Foreword by James Bryant Conant . . . xiii
Chapter 1: The Ancient Two-Sphere Universe . 1
Chapter 2: The Problem of the Planets. . . . 45
Chapter 3: The Two-Sphere Universe in Aristotelian Thought . 78
Chapter 4: Recasting the Tradition:
Aristotle to the Copernicans 100
Chapter 5: Copernicus' Innovation . . 134
Chapter 6: The Assimilation of Copernican Astronomy . 185
Chapter 7: The New Universe 229
Technical Appendix.
266
References . . . .
279
Bibliographical Notes .
283
Index .....
293
FOREWORD
In Europe west of the Iron Curtain, the literary tradition in
education still prevails. An educated man or woman is a person who has
acquired a mastery of several tongues and retained a working knowl-
edge of the art and literature of Europe. By a working knowledge I
do not refer to a scholarly command of the ancient and modern clas-
sics or a sensitive critical judgment of style or form; rather, I have in
mind a knowledge which can be readily worked into a conversation at
a suitable social gathering. An education based on a carefully circum-
scribed literary tradition has several obvious advantages: the distinc-
tion between the 5 to 10 percent of the population who are thus edu-
cated and the otlwrs makes itself evident almost automatically when
ladies and gentlemen converse. For those who truly enjoy art, literature
and music, there is a comforting sense of solidarity. For others who feel
compelled to enter into a discussion of these subjects, the area of
maneuver is conveniently limited; not too much effort is required to
keep fresh a portion of the knowledge painfully acquired at school.
The price of admission to the cultural tradition of a European nation
is paid once and for all when one is young. Theoretically, this price is
eight or nine years of hard work in special schools whose curricula are
centered on the languages and literature of Greece and Rome. I say
theoretically, since in practice the study of modern languages has
in this century made inroads on the study of Greek and to some extent
on a knowledge of Latin as well. But even these changes have not
fundamentally altered the basic idea of education for the few as
being the consequence of long years of school work devoted to the
study of languages and the literature of Europe.
There have been attacks on this type of education off and on for a
century at least. The claims of the physical sciences for a greater share
of the curriculum have been pressed and such claims have usually
been associated with demands for the substitution of modern Ian-
xiv FOREWORD
guages for the ancient. The place of mathematics has hardly been at
issue, since a thorough study of mathematics, including the calculus,
has long been accepted as a matter of course in all the of
those special schools which prepare a student to enter the umvers1ty.
Several generations ago, as a clear-cut alternative to the classical cur-
riculum, a course of study was suggested which would be based on
physics, chemistry, mathematics, and modern languages. But the pro-
ponents of the classical course are still vigorous and effective. In Ger-
many, at least, a series of compromises seems to have been the result
of the argument. But because of the importance attached to a study
of languages, it is hardly too much to say that the literary tradition
still dominates. Even in those schools which devote the most time to
science, it would hardly be correct to say that the scientific tradition
had replaced the literary. Rather, one might say, in varying degrees
German students entering a university have acquired a considerable
amount of information about the physical sciences. But whether such
knowledge subsequently affects the attitude of those who do not pro-
ceed with a scientific education is at least an open question. There
seems little or no concern with changing educational methods so that
the nonscientist will acquire a better understanding of science. In-
deed, it would not be strange if those whose education was primarily
literary would question whether understanding science was a matter
of importance to anybody but scientists or engineers.
In the United States the European literary tradition as a basis for
education disappeared almost a hundred years ago, or rather was
transformed beyond recognition. But it has not been replaced by an
based on the physical sciences, mathematics, and modern
languages. Some would say there has simply been no replacement.
At all events, there have been repeated attempts to provide some broad
base for the cultural life of the nation - broad enough to include the
physical, biological, and social sciences as well as the Anglo-Saxon
literary tradition and concern with art forms from various civilizations.
Whether such attempts, directed toward producing a future citizen
of a democracy who will be an enthusiastic participant in the nation's
developing culture, have created a medium sufficiently nutrient for
the life of the spirit in America may be a question. But no one can
deny that those responsible for the attempts have, with few exceptions,
endeavored to find a suitable place for the scientific tradition.
FOREWORD
XV
Experience has shown, however, both in the United States and
in the modem schools of Europe, how difficult it is to place the study
of science on anything like the same footing as the study of literature
or art or music. A scientist or engineer may be able to participate in a
stimulating manner in a discussion of pictures or books or plays, but
it is very hard indeed to keep a conversation going about physical sci-
ence in which the majority of the participants are not themselves
scientists or engineers. (And while I should be the first to deny that
facility in conversation was a goal of education, nevertheless listening
in on a social gathering may be a permissible diagnostic method.)
It is quite clear that studying science and studying literature in
school or college do not leave the same sort of residue in the student's
mind. A knowledge of the chemistry of metals and a knowledge of
Shakespeare's plays are two entirely different kinds of knowledge as
far as the needs of a human being are concerned. Of course, it is not
necessary to pick an example from the natural sciences; for the "chem-
istry of metals" in the preceding sentence one could quite as well
substitute the words "Latin grammar." Expressed in very simple terms,
the difference lies in the fact that Shakespeare's plays have been and
still are the subjects of endless debates in which the style and the
characters have been criticized from every conceivable angle, and
strong words of admiration and condemnation are constantly to be
heard. No one either admires or condemns the metals or the behavior
of their salts.
No, something more than a study of science as a body of organized
knowledge, something more than an understanding of scientific the-
ories is required to make educated people ready to accept the scien-
tific tradition alongside that literary tradition which still underlies
even the culture of the United States. This is so because the difficulties
of assimilating science into Western culture have increased with the
centuries. When in the time of Louis XIV scientific academies were
formed, new discoveries and new theories in science were far more
accessible to educated persons than today; the situation was the same
as late as the Napoleonic Wars. Sir Humphrey Davy fascinated Lon-
don society at the beginning of the 19th century by his lectures on
chemistry illustrated by spectacular experiments. Fifty years later
Michael Faraday delighted audiences of young and old who came
to hear him in the auditorium of the Royal Institution in London; his
xvi FOREWORD
lectures on the chemistry of the candle are classic examples of popu-
larizing science. In our own times there has been no lack of attempts
along similar lines; but the obstacles to be overcome have grown with
the years. Spectacular lecture-table experiments no longer astonish and
please sophisticated audiences as they once did; large-scale engineer-
ing outplays them almost daily. The scientific novelties of the current
year are too numerous as well as too complicated to form a topic of
conversation among laymen. The advances are made so rapidly and
on so many fronts that the layman is bewildered by the news; further-
more, to have any comprehension of the significance of a scientific
break-through one needs to be well versed on the state of the science
in question before the successful attack was launched. Even those
trained in one branch of science find it difficult to understand what is
going on in a distant field. For example, physicists are hardly in a
position to read even summary papers written by geneticists for other
geneticists, and vice versa. For the large group of people with scien-
tific and engineering training who wish to keep abreast of the progress
of science in general, there are some excellent periodicals, and useful
books are published from time to time. But I doubt very much if this
effort to popularize science reaches those who are not directly con-
nected with the physical or biological sciences or their application. And
some attempts at popularization are so superficial and sensational as
to be of no value for the purposes of providing a basis for the under-
standing of science by nonscientists.
In the last ten or fifteen years there has been growing concern in
American colleges as to the place of the physical and biological sci-
ences in the curriculum. The orthodox first-year courses in physics,
chemistry, and biology have been felt by many to be unsatisfactory
for those students who do not intend to enter into an intensive study
of science, engineering, or medicine. Various proposals have been
made and various experiments tried involving new types of scientific
courses which would be part of a liberal arts or general education pro-
gram. In particular, more emphasis. on the history of science has been
recommended and in this recommendation I have heartily joined. Ac-
tually, experience with one type of historical approach in Harvard
College for several years has increased my conviction as to the possi-
bilities inherent in the study of the history of science, particularly if
combined with an analysis of the various methods by which science
FOREWORD xvii
has progressed. While recognizing the educational value of an over-all
survey of the history of science in the last 300 years, I believe more
benefit can be obtained by an intensive study of certain episodes in
the development of physics, chemistry, and biology. This conviction
has found expression in a series of pamphlets entitled "Harvard Case
Histories in Experimental Science."
The cases considered in the Harvard series are relatively narrowly
restricted in point of both chronology and subject matter. The aim has
been to develop in the student some understanding of the interrelation
between theory and experiment and some comprehension of the com-
plicated train of reasoning which connects the testing of a hypothesis
with the actual experimental results. To this end an original scientific
paper is reprinted and forms the basis of the case; the reader is guided
by comments of the editors to follow as far as possible the investigator's
own line of reasoning. It is left to the professors who use these pam-
phlets to fit the case in question into a larger framework of the advance
of science on a broad front.
The Harvard Case Histories are too limited in scope and too much
concerned with experimental details and analysis of methods for the
general reader. Furthermore, though the episodes chosen are all of
them important in the history of physics, chemistry, and biology, their
significance is not at once apparent to the uninitiated. The reader will
soon be aware that the present volume does not suffer from these de-
fects. Everyone knows of the impact on Western culture of the change
from an Aristotelian universe centered on the earth to the Copernican
universe. Professor Kuhn is concerned not with one event in the his-
tory of science but with a series of connected events influenced by
and in turn influencing the attitude of learned men far removed in
their interests from the field of astronomy itself. He has not set him-
self the relatively easy task of merely retelling the story of the develop-
ment of astronomy during the revolutionary period. Rather he has
succeeded in presenting an analysis of the relation between theory and
observation and belief, and he has boldly faced such embarrassing
questions as why brilliant, devoted, and completely sincere students of
nature should have delayed so long in accepting the heliocentric ar-
rangement of the planets. This book is no superficial account of the
work of scientists; rather it is a thorough exposition of one phase of
scientific work, from which the careful reader may learn about the
xviii FOREWORD
curious interplay of hypothesis and experiment (or astronomical ob-
servation) which is the essence of modem science but largely un-
known to the nonscientist.
It is not my purpose in this foreword, however, to attempt to present
in capsule form a summary of the lessons on understanding science to
be derived from reading what Professor Kuhn has written. Rather, I
wish to register my conviction that the approach to science presented
in this book is the approach needed to enable the scientific tradition
to take its place alongside the literary tradition in the culture of the
United States. Science has been an enterprise full of mistakes and
errors as well as brilliant triumphs; science has been an undertaking
carried out by very fallible and often high!y emotional human beings;
science is but one phase of the creative activities of the Western world
which have given us art, literature, and music. The changes in man's
views about the structure of the universe portrayed in the following
pages affect to some degree the outlook of every educated person of
our times; the subject matter is of deep significance in and by itself.
But over and above the importance of this particular astronomical
revolution, Professor Kuhn's handling of the subject merits attention,
for, unless I am much mistaken, he points the way to the road which
must be followed if science is to be assimilated into the culture of
our times.
JAMES B. CONANT
THE
COPERNICAN REVOLUTION
1
THE ANCIENT
TWO-SPHERE UNIVERSE
Copernicus and the Modern Mind
The Copernican Revolution was a revolution in ideas, a trans-
formation in man's conception of the universe and of his own relation
to it. Again and again this episode in the history of Renaissance thought
has been proclaimed an epochal turning point in the intellectual
development of Western man. Yet the Revolution turned upon the
most obscure and recondite minutiae of astronomical research. How
can it have had such significance? What does the phrase "Copernican
Revolution" mean?
In 1543, Nicholas Copernicus proposed to increase the accuracy and
simplicity of astronomical theory by transferring to the sun many
astronomical functions previously attributed to the earth. Before his
proposal the earth had been the fixed center about which astronomers
computed the motions of stars and planets. A century later the sun
had, at least in astronomy, replaced the earth as the center of planetary
motions, and the earth had lost its unique astronomical status, becom-
ing one of a number of moving planets. Many of modern astronomy's
principal achievements depend upon this transposition. A reform in
the fundamental concepts of astronomy is therefore the first of the
Copernican Revolution's meanings.
Astronomical reform is not, however, the Revolution's only mean-
ing. Other radical alterations in man's understanding of nature rapidly
followed the publication of Copernicus' De Revolutionibus in 1543.
Many of these innovations, which culminated a century and a half
later in the Newtonian conception of the universe, were unanticipated
by-products of Copernicus' astronomical theory. Copernicus suggested
the earth's motion in an effort to improve the techniques used in pre-
2 THE COPERNICAN REVOLUTION
dieting the astronomical positions of celestial bodies. For other sciences
his suggestion simply raised new problems, and until these were solved
the astronomer's concept of the universe was incompatible with that
of other scientists. During the seventeenth century, the reconciliation
of these other sciences with Copernican astronomy was an important
cause of the general intellectual ferment now known as the scientific
revolution. Through the scientific revolution science won the great
new role that it has since played in the development of Western society
and Western thought.
Even its consequences for science do not exhaust the Revolution's
meanings. Copernicus lived and worked during a period when rapid
changes in political, economic, and intellectual life were preparing the
bases of modern European and American civilization. His planetary
theory and his associated conception of a sun-centered universe were
instrumental in the transition from medieval to modern Western so-
ciety, because they seemed to affect man's relation to the universe and
to God. Initiated as a narrowly technical, highly mathematical revision
of classical astronomy, the Copernican theory became one focus for the
tremendous controversies in religion, in philosophy, and in social theory,
which, during the two centuries following the discovery of America,
set the tenor of the modern mind. Men who believed that their terres-
trial home was only a planet circulating blindly about one of an infinity
of stars evaluated their place in the cosmic scheme quite differently
than had their predecessors who saw the earth as the unique and focal
center of God's creation. The Copernican Revolution was therefore also
part of a transition in \Vestern man's sense of values.
This book is the story of the Copernican Revolution in all three of
these not quite separable meanings - astronomical, scientific, and phil-
osophical. The Revolution as an episode in the development of
tary astronomy will, of necessity, be our most developed theme.
the first two chapters, as we discover what the naked eye can see m the
heavens and how stargazers first reacted to what they saw there,
astronomy and astronomers will be very nearly our only concern.
once we have examined the main astronomical theories developed m
the ancient world, our viewpoint will shift. In analyzing the strengths
of the ancient astronomical hadition and in exploring the requisites
for a radical break with that tradition, we shall gradually discover how
difficult it is to restrict the scope of an established scientific concept
to a single science or even to the sciences as a group. Therefore, in
THE ANCIENT TWO-SPHERE UNIVERSE
3
Chapters 3 and 4 we shall be less concerned with astronomy itself
than with the intellectual and, more briefly, the social and economic
milieu within which astronomy was practiced. These chapters will
d.eal primarily with the extra-astronomical implications -for science,
for religion, and for daily life - of a time-honored astronomical con-
ceptual scheme. They will show how a change in the conceptions of
mathematical astronomy could have revolutionary consequences.
Finally, in the last three chapters, when we turn to Copernicus' work,
its reception, and its contribution to a new scientific conception of the
universe, we shall deal with all these strands at once. Only the battle
that established the concept of the planetary earth as a premise of
Western thought can adequately represent the full meaning of the
Copernican Revolution to the modern mind.
Because of its technical and historical outcome, the Copernican
Revolution is among the most fascinating episodes in the entire history
of science. But it has an additional significance which transcends its
specific subject: it illustrates a process that today we badly need to
understand. Contemporary Western civilization is more dependent,
both for its everyday philosophy and for its bread and butter, upon
scientific concepts than any past civilization has been. But the scientific
theories that bulk so large in our daily lives are unlikely to prove final.
The developed astronomical conception of a universe in which the
stars, including our sun, are scattered here and there through an in-
finite space is less than four centuries old, and it is already out of date.
Before that conception was developed by Copernicus and his succes-
sors, other notions about the structure of the universe were used to
explain the phenomena that man observed in the heavens. These older
astronomical theories differed radically from the ones we now hold, but
most of them received in their day the same resolute credence that we
now give our own. Furthermore, they were believed for the same
reasons: they provided plausible answers to the questions that seemed
important. Other sciences offer parallel examples of the transiency of
treasured scientific beliefs. The basic concepts of astronomy have, in
fact, been more stable than most.
The mutability of its fundamental concepts is not an argument for
rejecting science. Each new scientific theory preserves a hard core of
the knowledge provided by its predecessor and adds to it. Science
progresses by replacing old theories with new. But an age as dominated
by science as our own does need a perspective from which to examine
the scientific beliefs which it takes so much for granted, and history
provides one important source of such perspective. If we can discover
the origins of some modern scientific concepts and the way in which
they supplanted the concepts of an earlier age, we are more likely to
evaluate intelligently their chances for survival. This book deals pri-
marily with astronomical concepts, but they are much like those em-
ployed in many other sciences, and by scrutinizing their development
we can learn something of scientific theories in general. For example:
What is a scientific theory? On what should it be based to command
our respect? What is its function, its use? What is its staying power?
Historical analysis may not answer questions like these, but it can
illuminate them and give them meaning.
Because the Copernican theory is in many respects a typical
scientific theory, its history can illustrate some of the processes by
which scientific concepts evolve and replace their predecessors. In its
extrascientific consequences, however, the Copernican theory is not
typical: few scientific theories have played so large a role in non-
scientific thought. But neither is it unique. In the nineteenth century,
Darwin's theory of evolution raised similar extrascientific questions.
In our own century, Einstein's relativity theories and Freud's psycho-
analytic theories provide centers for controversies from which may
emerge further radical reorientations of Western thought. Freud him-
self emphasized the parallel effects of Copernicus' discovery that the
earth was merely a planet and his own discovery that the unconscious
controlled much of human behavior. Whether we have learned their
theories or not, we are the intellectual heirs of men like Copernicus
and Darwin. Our fundamental thought processes have been reshaped
by t h ~ m just as the thought of our children or grandchildren will have
been reshaped by the work of Einstein and Freud. We need more than
an understanding of the internal development of science. We must also
understand how a scientist's solution of an apparently petty, highly
technical problem can on occasion fundamentally alter men's attitudes
toward basic problems of everyday life.
The Heavens in Primitive Cosmologies
Much of this book will deal with the impact of astronomical
observations and theories upon ancient and early modern cosmological
thought, that is, upon a set of man's conceptions about the structure of
the universe. Today we take it for granted that astronomy should affect
THE ANCIENT TWO-SPHERE UNIVERSE 5
cosmology. If we want to know the shape of the universe, the earth's
position in it, or the relation of the earth to the sun and the sun to the
stars, we ask the astronomer or perhaps the physicist. They have made
detailed quantitative observations of the heavens and the earth; their
knowledge of the universe is guaranteed by the accuracy with which
they predict its behavior. Our everyday conception of the universe, our
popular cosmology, is one product of their painstaking researches.
But this close association of astronomy and cosmology is both tempo-
rally and geographically local. Every civilization and culture of which
we have records has had an answer for the question, 'What is the
structure of the universe?" But only the Western civilizations which
descend from Hellenic Greece have paid much attention to the ap-
pearance of the heavens in arriving at that answer. The drive to con-
struct cosmologies is far older and more primitive than the urge to
make systematic observations of the heavens. Furthermore, the primi-
tive form of the cosmological drive is particularly informative because
it highlights features obscured in the more technical and abstract
cosmologies that are familiar today.
Though primitive conceptions of the universe display considerable
substantive variation, all are shaped primarily by terrestrial events,
the events that impinge most immediately upon the designers of the
systems. In such cosmologies the heavens are merely sketched in to
provide an enclosure for the earth, and they are peopled with and
moved by mythical figures whose rank in the spiritual hierarchy
usually increases with their distance from the immediate terrestrial
environment. For example, in one principal form of Egyptian cos-
mology the earth was pictured as an elongated platter. The platter's
long dimension paralleled the Nile; its flat bottom was the alluvial
basin to which ancient Egyptian civilization was restricted; and its
curved and rippled rim was the mountains bounding the terrestrial
world. Above the platter-earth was air, itself a god, supporting an
inverted platter-dome which was the skies. The terrestrial platter in its
tum was supported by water, another god, and the water rested upon
a third platter which bounded the universe symmetrically from below.
Clearly several of the main structural features of this universe were
suggested by the world that the Egyptian knew: he did live in an
elongated platter bounded by water in the only direction in which he
had explored it; the sky, viewed on a clear day or night, did and does
look dome-shaped; a symmetric lower boundary for the universe was
6
the obvious choice in the absence of relevant observations. Astronomi-
cal appearances were not ignored, but they were treated with less
precision and more myth. The sun was Ra, the principal Egyptian god,
supplied with two boats, one for his daily journey through the air and
a second for his nocturnal trip through the water. The stars were
painted or studded in the vault of the heaven; they moved as miuor
gods; and in some versions of the cosmology they were reborn each
night. Sometimes more detailed observations of the heavens entered,
as when the circumpolar stars (stars that never dip below the horizon)
were recognized as "those that know no weariness" or "those that know
no destruction." From such observations the northern heavens were
identified as a region where there could be no death, the region of the
eternally blessed afterlife. But such traces of celestial observation
were rare.
Fragments of cosmologies similar to the Egyptian can be found in
all those ancient civilizations, like India and Babylonia, of which we
have records. Other crude cosmologies characterize the contemporary
primitive societies investigated by the modern anthropologist. Ap-
parently all such sketches of the structure of the universe fulfill a basic
psychological need: they provide a stage for man's daily activities and
the activities of his gods. By explaining the physical relation between
man's habitat and the rest of nature, they integrate the universe for
man and make him feel at home in it. Man does not exist for long
without inventing a cosmology, because a cosmology can provide him
with a world-view which permeates and gives meaning to his every
action, practical and spiritual.
Tbough the psychological needs satisfied by a cosmology seem
relatively uniform, the cosmologies capable of fulfilling these needs
have varied tremendously from one society or civilization to another.
None of the primitive cosmologies referred to above will now satisfy
our demand for a world-view, because we are members of a civilization
that has set additional standards which a cosmology must meet in order
to be believed. We will not, for example, credit a cosmology that em-
ploys gods to explain the everyday behavior of the physical world: in
recent centuries, at least, we have insisted upon more nearly mechamcal
explanations. Even more important, we now demand that a 3atisfactory
cosmology account for many of the observed details of nature's .be-
havior. Primitive cosmologies are only schematic sketches agamst
which the play of nature takes place; very little of the play is in-
corporated into the cosmology. The sun god, Ra, travels in his boat
across the heavens each day, but there is nothing in Egyptian cosmology
to explain either the regular recurrence of his journey or the seasonal
variation of his boat's route. Only in our own Western civilization has
the explanation of such details been considered a function of cosmology.
No other civilization, ancient or modern, has made a similar demand.
The requirement that a cosmology supply both a psychologically
satisfying world-view and an explanation of observed phenomena like
the daily change in the position of sunrise has vastly increased the
power of cosmologic thought. It has channeled the universal com-
pulsion for at-homeness in the universe into an unprecedented drive
for the discovery of scientific explanations. Many of the most charac-
teristic achievements of Western civilization depend upon this com-
bination of demands imposed upon cosmologic thought. But the com-
bination has not always been a congenial one. It has forced modern
man to delegate the construction of cosmologies to specialists, primarily
to astronomers, who know the multitude of detailed observations that
modern cosmologies must satisfy to be believed. And since observation
is a two-edged sword which may either confirm or conflict with a
cosmology, the consequences of this delegation can be devastating.
The astronomer may on occasions destroy, for reasons lying entirely
within his specialty, a world-view that had previously made the uni-
verse meaningful for the mem hers of a whole civilization, specialist
and nonspecialist alike.
Something very much like this happened during the Copernican
Revolution. To understand it we must therefore become something of
specialists ourselves. In particular, we must get to know the principal
observations, all of them accessible to the naked eye, upon which de-
pend the two main scientific cosmologies of the West, the Ptolemaic
and the Copernican. No single panoramic view of the heavens will
suffice. Seen on a clear night, the skies speak first to the poetic, not
to the scientific, imagination. No one who views the night sky can
challenge Shakespeare's vision of the stars as "night's candles" or
:\tilton's image of the Milky Way as "a broad and ample road, whose
dust is gold, and pavement stars." But these descriptions are the ones
embodied in primitive cosmologies. They provide no evidence relevant
to the astronomer's questions: How far away is the Milky Way, the
sun, the planet Jupiter? How do these points of light move? Is the
material of the moon like the earth's, or is it like the sun's, or like a
star's? Questions like these demand systematic, detailed, and quantita-
tive observations accumulated over a long period of time.
This chapter deals, then, with observations of the sun and stars
and with the role of these observations in establishing the first scien-
tific cosmologies of ancient Greece. The next chapter completes the
roster of naked-eye celestial observations by describing the planets,
the celestial bodies which posed the technical problem that led to the
Copernican Revolution.
The Apparent Motion of the Sun
Before the end of the second millennium B.C. (perhaps very
much before), the Babylonians and the Egyptians had begun systematic
observations of the motion of the sun. For this purpose they devel-
oped a primitive sundial consisting of a measured stick, the gnomon,
projecting vertically from a smooth flat section of ground. Since the
apparent position of the sun, the tip of the gnomon, and the tip of its
shadow lie along a straight line at each instant of a clear day, measure-
ments of the length and direction of the shadow completely determine
the direction of the sun. When the shadow is short, the sun is high in
the sky; when the shadow points, say, to the east, the sun must lie in
the west. Repeated observations of the gnomon's shadow can there-
fore systematize and quantify a vast amount of common but vague
knowledge about the daily and annual variation of the sun's position.
In antiquity such observations harnesserl the sun as a time reckoner
and C!llendar keeper, applications that provided one important motive
for continuing and refining the observational techniques.
Both the length and the direction of a gnomon's shadow vary
slowly and continuously during the course of any one day. The shadow
is longest at sunrise and sunset, at which times it points in roughly
opposite directions. During the daylight hours the shadow moves
gradually through a symmetric fan-shaped figure which, in most of
the locations accessible to ancient observers, is much like one of those
shown in Figure 1. As the diagram indicates, the shape of the fan is
different on different days, but it has one very significant fixed feature.
At the instant of each day when the gnomon's shadow is shortest, it
always points in the same direction. This simple regularity provides
9
two fundamental frames of reference for all further astronomical meas-
urements. The permanent direction assumed by the shortest shadow
each day defines due north, from which the other compass points fol-
low; the instant at which the shadow becomes shortest defines a refer-
ence point in time, local noon; and the interval between two successive
local noons defines a fundamental time unit, the apparent solar day.
During the first millennium B.c., the Babylonians, Egyptians, Greeks,
and Romans used primitive terrestrial timekeepers, particularly water
clocks, in order to subdivide the solar day into smaller intervals from
which our modern units of time- hour, minute, and second- descend.
0
WINTER SOLSTICE
NOON ____ T __ ...,..._
DAWN DUSK
VERNAL EQUINOX
ANO
AUTUMNAL EQUINOX
NOON
Figure 1. The daily motion of the gnomon's shadow at various seasons in
middle-northern latitudes. At sunrise and sunset the shadow stretches momen-
tarily to infinite distance where its end "joins" the broken line in the diagram.
Between sunrise and sunset the end of the shadow moves slowly along the broken
line; at noon the shadow always points due north.
The compass points and the time units defined by the sun's daily
motion provide a basis for describing the changes in that motion from
day to day. Sunrise always occurs somewhere in the east and sunset
in the west, but the position of sunrise, the length of the gnomon's
noon shadow, and the number of daylight hours vary from day to day
with the changing seasons (Figure 2). The winter solstice is the day
(December 22 on the modern calendar) when the sun rises and sets
farthest to the south of the due east and west points on the horizon.
On this day there are fewer hours of daylight and the gnomon's noon
shadow is longer than on any other. After the winter solstice the points
" For astronomical purposes the stars provide a more convenient time reckoner
than the sun. But, on a time scale determined by the stars, the length of the
apparent solar day varies by almost a minute at different seasons of the year.
Though ancient astronomers were aware of this slight but significant irregularity
of apparent solar time, we shall ignore it here. The cause of this variation and its
effect upon the definition of a time scale are discussed in Section I of the Tt>chnical
Appendix.
10
at which the sun rises and sets gradually move north together along
the horizon, and the noon shadows grow shorter. On the vernal equinox
(March 21) the sun rises and sets most nearly due east and west;
nights and days are then of equal length. As more days pass, the sun-
rise and sunset points continue to move northward and the number
of daylight hours increases until the summer solstice (June 22), when
the sun rises and sets farthest to the north. This is the time when dav-

light lasts longest and when the gnomon's noon shadow is shortest.
After the summer solstice, the sunrise point again moves south, and the
nights grow longer. At the autumnal equinox (September 23) the sun
once again rises and sets almost due east and west; then it continues
south until the winter solstice recurs.
w
SUNSET
Figure 2. Relation between the position of sunrise, the sun's noon elevation,
and the seasonal variation of the gnomon's shadow.
As the modern names of the solstices and equinoxes indicate, the
motion of sunrise back and forth along the horizon corresponds to
the cycle of the seasons. Most ancient peoples therefore believed that
the sun controlled the seasons. They simultaneously venerated the sun
as a god and observed it as a calendar keeper, a practical indicator
of the passage of the seasons upon which their agricultural activities
depended. Prehistoric remains, like the mysterious structure of giant
stones at Stonehenge, England, testify to the antiquity and the strength
of this double interest in the sun. Stonehenge was an important temple
11
laboriously constructed from huge stones, some almost thirty tons in
weight, by the people of an early Stone Age civilization. It was almost
certainly also a crude sort of observatory. The stones were so ar-
ranged that an observer at the center of the array saw the sun rise
over a specially placed stone, called the "Friar's Heel," on the ancient
midsummer day, the summer solstice.
The length of the cycle of the seasons - the interval between one
vernal equinox and the next- defines the basic calendar unit the year
just as the sun's daily motion defines the day. But the ye:r is a fa;
more difficult unit to measure than the day, and the demand for useful
long-term calendars has therefore presented astronomers with a con-
tinuing problem whose prominence during the sixteenth century played
a direct role in the Copernican Revolution. The earliest solar calendars
of antiquity were based upon a year of 360 days, a neat round number
that nicely fitted the sexagesimal number system of the Babylonians. But
the cycle of the seasons occupies more than 360 days, so that the "New
Year's Day" of these early solar calendars gradually crept around the
cycle of the seasons from winter, to fall, to summer, to spring. The
calendar was scarcely useful over long periods of time, because im-
portant seasonal events, like the flooding of the Nile in Egypt, occurred
at later and later dates in successive years. To keep the solar calendar
in step with the seasons, the Egyptians therefore added five extra days,
a holiday season, to their original year.
There is, however, no integral number of days in the cycle of
the seasons. The year of 365 days is also too short, and after 40 years
the Egyptian calendar was ten days out of step with the seasons.
Therefore, when Julius Caesar reformed the calendar with technical
assistance from Egyptian astronomers, he based his new calendar upon
a year 6 5 ~ days in length; three years of 365 days were followed by one
of 366. This calendar, the Julian, was used throughout Europe from
its introduction in 45 B.c. until after the death of Copernicus. But the
seasonal year is actually 11 minutes and 14 seconds shorter than 6 5 ~
days, so that by Copernicus' lifetime the date of the vernal equinox
had moved backward from March 21 to March 11. The resulting de-
mand for calendar reform (see Chapters 4 and 5) provided one im-
portant motive for the reform of astronomy itself, and the reform that
gave the Western world its modern calendar followed the publication
of the De Revolutionibus by only thirty-nine years. In the new calendar,
12
imposed upon large areas of Christian Europe by Pope Gregory XIII
in 1582, leap year is suppressed three times in every four hundred
years. The year 1600 was a leap year and the year 2000 will be, but
1700, 1800, and 1900, all leap years in the Julian calendar, had just
365 days in the Gregorian, and 2100 will again be a normal year of
365 days.
All the observations discussed above show the sun approximately
as it would appear to an astronomer in middle-northern latitudes, an
area that includes Greece, Mesopotamia, and northern Egypt, the re-
gions in which almost all ancient observations were made. But within
this area there is a considerable quantitative variation in certain aspects
of the sun's behavior, and in the southernmost parts of Egypt there is
a qualitative change as well. Knowledge of these changes also played
a part in the construction of ancient astronomical theories. No varia-
tions are observed as an observer moves east or west. But toward the
south the noon shadow of the gnomon is shorter and the noon sun
higher in the .sky than they would be on the same day in the north.
Similarly, though the length of the whole day remains constant, the
difference between the lengths of daytime and nighttime is smaller in
the southern portion of middle-northern latitudes. Also, in this region
the sun does not swing quite so far north and south along the horizon
WINTER SOLSTICE
NOON
---r---
DAwN
0
uSK
VERNAL EQUINOX
AND
AUTUMNAL EOUI NOX
SUMMER SOLSTICE
Figure 3. The daily motion of the gnomon's shadow at various seasons in the
northern torrid zone.
during the course of the year. None of these variations alters the
qualitative descriptions supplied above. But, if an observer has moved
far into southern Egypt during the summer, he will see the noon
shadow of the gnomon grow shorter day by day until at last it vanishes
entirely and then reappears pointing to the south. In the southernmost
parts of Egypt the annual behavior of the gnomon's shadow is that
shown in Figure 3. Journeys still farther south or much farther north
13
will produce other anomalies in the observed motions of the sun. But
these were not observed in antiquity. We shall not discuss them until
we deal with the astronomical theories that made it possible to pre-
dict them even before they were observed ( pp. 33 ff).
The Stars
The motions of the stars are much simpler and more regular
than the sun's. Their regularity is not, however, so easily recognized
because systematic examination of the night sky requires the ability
to select individual stars for repeated study wherever in the heavens
they appear. In the modern world this ability, which can be acquired
only by long practice, is quite rare. Few people now spend much
time out of doors at night, and, when they do, their view of the heavens
is frequently obscured by tall buildings and street lighting. Besides,
observation of the heavens no longer has a direct role in the life of the
average man. But in antiquity the stars were an immediate part of the
normal man's environment, and celestial bodies served a universal
function as time reckoners and calendar keepers. Under these circum-
4. constellation Ursa Major in the northern skies. Notice the
Big Dipper whose handle forms the bear's tail. The North Star is the
star directly over the bear's right ear in the picture. It lies almost on
a line JOining the last two stars in the Dipper's bowl.
14
t nces the abilitv to identify stars at a glance was relatively common.
sa ,
h . b
Long before the beginning of recorded history men w ose JO s gave
them a continuing view of the night sky had mentally arranged the
stars into constellations, groups of neighboring stars that could be
seen and recognized as a fixed pattern. To find an individual star
amidst the profusion of the heavens, an observer would first. locate
the familiar star pattern within which it occurred and then piCk the
individual star from the pattern.
Many of the constellations used by modem astronomers are named
after mythological figures of antiquity. Some can be traced to Baby-
lonian cuneiform tablets, a few as old as 3000 B.c. Though modern
h mod
ified their definitions the major constellations are
astronomy as
'
fi
among our oldest traceable inheritances. How these rst
picked out is, however, still uncertain. Few people can see a bear m_the
stars of the constellation Ursa Major (Figure 4); other
present similar problems in visualization; the stars may
nally have been grouped for convenience and named But, If
so, they were very strangely grouped. The ancient have
very irregular boundaries, and they occupy. areas o_f different
size in the sky. They are not convenient ch01ces, whiCh IS one reason
why modern astronomers have altered their boundaries. Probably the
ancient shepherd or navigator, staring at the heavens hour. after hour,
really did "see" his familiar mythological characters traced m the stars,
just as we sometimes "see" faces in clouds or the outlines of The
experiments of modem Gestalt psychology demonstrate a
need to discover familiar patterns in apparently random groupmgs, a
ed that underlies the well-known "ink-blot" or Rohrschach tests.
knew more about their historical origin, the constellations might
useful information about the mental characteristics of the
prehistoric societies that first traced them. . . . .
Learning the constellations is like gaining w_Ith a map
and has the same purpose: thP constellations make It easwr to find
one's way around the sky. Knowing the constellations, a man can read-
ily find a comet reported to be "in Cygnus" (the Swa_n); he .. :Vould
almost certainly miss the comet if he knew only that It was m the
sky." The map provided by the constellation_s is, an unusual
one because the constellations are always m motion. Smce they all
move together, preserving their patterns and their relative positions,
the motion does not destroy their usefulness. A star in Cygnus will
always be in Cygnus, and Cygnus will always be the same distance
from Ursa Major.
0
But neither Cygnus nor Ursa Major remains for
long at the same position in the sky. They behave like cities on a map
pasted to a rotating phonograph record.
Both the fixed relative positions and the motions of the stars are
illustrated in Figure 5, which shows the location and orientation of
the Big Dipper (part of Ursa Major) in the northern sky at three
times during a single night. The pattern of seven stars in the Dipper
is the same at each viewing. So is the relation of the Dipper to the
North Star, which always lies 29 to the open side of the Dipper's
bowl on a straight line through the last two stars in the bowl. Other
diagrams would show similar permanent geometric relations among
the other stars in the heavens.
Figure 5 displays another important characteristic of the stellar
motions. As the constellations and the stars composing them swing
through the skies together, the North Star remains very nearly station-
ary. Careful observation shows that it is not, in fact, quite stationary
during any night, but there is another point in the heavens, now less
than 1 away from the North Star, which has precisely the properties
attributed to that star in Figure 5. This point is known as the north
celestial pole. An observer at a given location in northern latitudes
can always find it, hour after hour and night after night, at the same
fixed distance above the due-north point on his horizon. A straight
stick clamped so that it points toward the pole will continue to point
toward the pole as the stars move. Simultaneously, however, the celes-
tial pole behaves as a star. That is, the pole retains its geometric rela-
tions to the stars over long periods of time. t Since the pole is a fixed
0
"Distance" here means "angular distance," that is, the number of degrees
between two pointing from the eye to the two celestial objects
whose separation IS to be measured. This IS the only sort of distance that astrono-
mers can measure directly, that is, without making calculations based upon some
theory about the structure of the universe.
t made years apart that the pole's position among
the stars .Is very changmg (about 1" m 180 years). We shall neglect this
moti?n, whiCh IS part an effect known as the precession of the equinoxes,
until SectiOn 2 of the Techmcal Appendix. Though the ancients were aware of it
by the end of the second century B.c., precession played a secondary role in the
construction of their astronomical theories, and it does not alter the short-term
observa_tions described above. There always been a north celestial pole at the
same distance above the due-north pomt on the horizon, but the same stars have
not always been near it.
16
point for each terrestrial observer and since the stars do not change
their distance from this point as they move, every star seems to travel
along the arc of a circle whose center is the celestial pole. Figure 5
shows a part of this circular motion for the stars in the Dipper.
The concentric circles traced by the circumpolar motions of the
stars are known as their diurnal or daily circles, and the stars revolve
in these circles at a rate just over 15 per hour. No star completes a
full circle between sunset and sunrise, but a man observing the
northern skies during a single clear night can follow stars near the
pole through approximately a semicircle, and on the next night he
can find them again moving along the same circles at the same rate.
Furthermore, he will find them at just the positions they would have
reached if they had continued their steady revolutions throughout the
intervening day. Since antiquity, most observers equipped to recog-
nize these regularities have naturally assumed that the stars exist and
move during the day as during the night, but that during the day the
strong light of the sun makes them invisible to the naked eye. On this
interpretation the stars swing steadily through full circles, each star
completing a circle once every 23 hours 56 minutes. A star that is
directly below the pole at 9:00 o'clock on the evening of October 23
will return to the same position at 8:56 on the evening of October 24
POLE
STAR
/
..
... .. .
.. *
9:00P.M.

Figure 5. Successive positions of the Big Dipper at four-hour intervals on a
night in late October.
and at 8:52 on October 25. By the end of the year it will be reaching
its position below the pole before sunset and will therefore not be
visible in that position at all.
In middle-northern latitudes the celestial pole is approximately
45 above the northernmost point on the horizon. (The elevation of
the pole is precisely equal to the observer's angle of latitude- that
is one way latitude is measured.) Therefore stars that lie within 45
of the pole, or whatever the elevation at the observer's location may be,
can never fall below the horizon and must be visible at any hour of a
clear night. These are the circumpolar stars, "those that know no de-
struction," in the words of the ancient Egyptian cosmologists. They are
also the only stars whose motion is easily recognized as circular.
Stars farther from the poles also travel along diurnal circles, but
part of each circle is hidden below the horizon (Figure 6). Therefore
such stars can sometimes be seen rising or setting, appearing above or
disappearing below the horizon; they are not always visible throughout
the night. The farther from the pole such a star is, the less of its
diurnal circle is above the horizon and the more difficult it is to recog-
nize the visible portion of its path as part of a circle. For example, a
star that rises due east is visible on only half of its diurnal circle. It
travels very nearly the same path that the sun takes near one of the
equinoxes, rising along a slant line up and to the south (Figure 7a),
reaching its maximum height at a point over the right shoulder of
an observer looking east, and finally setting due west along a line
slanting downward and to the north. Stars still farther from the pole
appear only briefly over the southern horizon. Near the due-south
point they set very soon after they rise, and they never get very far
above the horizon (Figure 7b). Since during almost half the year
they rise and set during daylight, there are many nights when they
do not appear at all.
These qualitative features of the night sky are common to the entire
area within which ancient astronomical observations were made, but
the description has glossed over significant quantitative differences.
As an observer travels south, the elevation of the pole above the north-
ern horizon decreases approximately 1 o for every 69 miles of southward
motion. The stars continue to move in diurnal circles about the pole,
but since the pole is closer to the horizon, some stars that were cir-
cumpolar in the north are seen rising and setting by an observer
farther south. Stars that rise and set due east and west continue to
appear and disappear at the same points on the horizon, but toward
the south they appear to move along a line more nearly perpendicular
to the horizon, and they reach their maximum elevation more nearly
Figure 6. A set of the short circular arcs described by typical stars in the
northern sky during a two-hour period. The heavy circle tangent to the horizon
separates the circumpolar stars from those that rise and set.
Star trails like these can actually be recorded by pointing a fixed camera at
the celestial pole and leaving the shutter open as the heavens tum. Each addi-
tional hour's exposure adds 15 to the length of every track. Notice, however,
that the elevated camera angle introduces a deceptive distortion. If the pole is
45 above the horizon (a typical elevation in middle-northern latitudes), then a
star that appears at the very top of the heavy circle is actually directly above the
observer's head. Recognizing the distortion due to camera angle makes it possible
to relate the star trails in this diagram to those shown more schematically in
Figures 7 a and b.
19
over the observer's head. The appearance of the southern sky changes
more strikingly. As the pole declines toward the northern horizon, stars
in the southern sky, because they remain at the same angular distance
from the pole, rise to greater heights over the southern horizon. A star
that barely rises above the horizon when seen from the north will
rise higher and be seen for longer when observed from farther south.
A southern observer will still see stars that barely peek above the
southernmost point on his horizon, but these will be stars that the
northern stargazer never sees at all. As an observer moves south, he
sees fewer and fewer circumpolar stars - stars that are visible through-
DUE EAST
(a)
DUE SOUTH
(b)
Figure 7. Star trails over (a) the eastern and (b) the southern horizon. Like
Figure 6, these diagrams show the motion of typical stars over a go section of
the horizon during a two-hour period. In these diagrams, however, the "camera" is
directed to the horizon, so that only the first 40 above the horizon is shown.
out the night. But in the south he will, at some time or other, observe
stars that an observer in the north can never see.
The Sun as a Moving Star
Because the stars and the celestial pole retain the same rela-
tive positions hour after hour and night after night, they can be per-
manently located upon a map of the heavens, a star map. One form of
star map is shown in Figure 8; others will be found in any atlas or
book on astronomy. The map of Figure 8 contains all the brighter
stars that can ever be seen by an observer in middle-northern latitudes,
but not all the stars on the map can be seen at once because they are
not all above the horizon simultaneously. At any instant of the night
approximately two-fifths of the stars on the map lie below the horizon.
The particular stars that are visible and the portion of sky in which
they appear depend upon the date and hour of the observation. For
example, the solid black line on the map broken by the four cardinal
points of the compass, N, E, S, W, encloses the portion of the sky
that is visible to an observer in middle-northern latitudes at 9:00
o'clock on the evening of October 23. It therefore represents his hori-
zon. If the observer holds the map over his head with the bottom
toward the north, the four compass points will be approximately
aligned with the corresponding points on his physical horizon. The
map then indicates that at this time of night and year the Big Dipper
appears just over the northern horizon and that, for example, the con-
stellation Cassiopeia lies at a position near the center of the horizon-
window, corresponding to a position nearly overhead in the sky. Since
the stars return to their positions in just 4 minutes less than 24 hours,
the same orientation of the map must indicate the position of the
stars at 8:56 on the evening of October 24, at 8:52 on October 25, at
8:32 on October 30, and so on.
Now imagine that the solid black horizon line which encloses the
observer's field of view is held in its present position on the page while
the entire disk of the map is rotated slowly behind it in a counterclock-
wise direction about the central pole. Rotating the disk 15 brings into
the horizon-window just those stars that are visible at 10:00 o'clock
on the evening of October 23, or at 9:56 on the evening of Octo-
ber 24, and so on. A rotation of 45 moves the stars visible at midnight
on October 23 inside of the horizon line. The positions of all bright
~ .
.
..

..

.
~

..
..
..
.+

.

.. 1
* !;s
{.

*

. .
D
~
0
N
Figure 8. A circumpolar star map containing all the major stars ever visible to
an observer at approximately 45 northern latitude. The cross at the geometric
center of the map indicates the position of the celestial pole.
If the map is held horizontally overhead with its face toward the ground
and with the bottom of the page pointing north, it will show the orientation of
the stars as they appear to an observer in middle-northern latitudes at 9:00
o'clock on the evening of October 23. The stars within the solid line bounding
the horizon-window are the ones that the observer can see; those outside the line
are below the horizon on this day at this hour. Stars that lie within the horizon-
window near the point N on the map will be seen just over the due-north point on
the physical horizon (notice the Dipper); those near the east point, E, will be
just rising in the east; and so on. To flnd the position of stars at a later hour on
October 23, the horizon-window should be imagined stationary and the circular
map should be rotated behind it, counterclockwise about the pole, 15o for each
hour after 9:00 P.M. This motion leaves the pole stationary but carries stars up
over the eastern horizon and down behind the western one. To flnd the positions
of stars at 9:00 P.M. on a later day the map should be rotated clockwise behind
the stationary horizon-window, I o for each day after October 23. Combining these
two procedures makes it possible to flnd the positions of stars at any hour of
any night of the year.
The broken line that encircles the pole in the diagram is the ecliptic, the
sun's apparent path through the stars (see p. 23). The box that encloses a portion
of the ecliptic in the upper right-hand quadrant of the map contains the region of
the sky shown in expanded form in Figures 9 and 15.
stars at any hour of any night can be found in this way. A movable
star map equipped with a fixed horizon-window, like that in Figure 8,
is frequently known as a "star finder."
Star maps have other applications, however, besides locating bodies
that, like the stars, remain in constant relative positions. They can
also be used to describe the behavior of celestial bodies that, like the
moon, comets, or planets, slowly change their positions among the
stars. For example, as the ancients knew, the sun's motion takes a
particularly simple form as soon as it is related to the stars. Since the
stars appear shortly after sunset, an observer who knows how to follow

Figure 9. Motion of the sun through the constellations Aries and Taurus. The
circles represent the sun's position among the stars at sunset on successive evenings
from the middle of April to late May.
their motion can record the time and horizon position of the sunset,
measure the time between sunset and the first appearance of the stars,
and then locate the sun on a star map by rotating the map backward
to determine which stars were at the appropriate horizon position when
the sun set. An observer who plots the position of the sun on a star
map for several consecutive evenings will find it in almost the same
position each time. Figure 9 shows the position of the sun on a star
map on successive evenings for one month. It is not in the same posi-
tion on the map for two successive observations, but it has not moved
far. Each evening finds it about 1 from its position the previous
evening, and 1 o is a relatively small distance, about twice the angular
diameter of the sun.
These observations suggest that both the daily motion of the sun
and its slower shift north and south along the horizon may conveniently
be analyzed by regarding the sun as a body that moves slowly among
the stars from day to day. If, for some particular day, the position
of the sun among the stars is specified, then, on that day, the sun's
motion will be almost exactly the diurnal motion of a star in the
corresponding position on the map. Both will move like points on the
rotating map, rising in the east along a line slanting upward and to
the south and later setting in the west. One month later the sun will
again have the diurnal motion of a star, but now it will move very
nearly like a star 30 away from the position of the star whose motion
it copied a month earlier. During the intervening month the sun
has moved slowly and steadily between these two positions, 30 apart
on the map. Each day its motion has been almost that of a star, part of
a circle about the pole of the heavens, but it has not behaved like
quite the same star on two successive days.
If the sun's position is plotted on a star map day after day and the
points marking its successive positions in the evening are connected
together, a smooth curve is produced which rejoins itself at the end
of a year. This is the curve, called the ecliptic, that is indicated by
the broken line on the star map of Figure 8. The sun is always to be
found somewhere on this line. As the ecliptic is carried rapidly through
the heavens by the common diurnal motion of the stars, the sun is
carried along with it, rising and setting like a star located at a point
somewhere on the line. But simultaneously the sun is moving slowly
around the ecliptic, occupying a slightly different position each day,
hour, or minute. Thus the complex helical motion of the sun can be
analyzed as the result of two much simpler motions. The total apparent
motion of the sun is composed of its diurnal motion (the westward
circle due to the counterclockwise motion of the whole map) and a
simultaneous slow eastward motion (clockwise about the pole on the
map) along the ecliptic.
Analyzed in this way, the sun's motion shows close parallels to
the motion of a toll collector on a merry-go-round. The collector is
carried around rapidly by the revolutions of the platform. But as he
walks slowly from horse to horse collecting tolls his motion is not
quite the same as that of the riders. If he walks in a direction opposite
to that of the platform's spin, his motion over the ground will be
slightly slower than that of the platform, and the riders will com-
plete one circle somewhat more rapidly than he. If his toll collections
carry him toward and away from the center of the platform, his total
motion with respect to the ground will not be circular at all, but a
complex curve which does not rejoin itself at the end of a single revo-
lution. Though it is theoretically possible to specify precisely the path
along which the toll collector moves over the stationary ground, it
is far simpler to divide his total motion into its two component parts:
a steady rapid rotation with the platform and a slower less regular
motion with respect to the platform. Since antiquity astronomers have
used a similar division in analyzing the apparent motion of the sun.
Each day the sun moves rapidly westward with the stars (its so-called
diurnal motion); simultaneously the sun moves slowly eastward along
the ecliptic through the stars or with respect to the stars (its annual
motion).
With the sun's total motion divided into two components, its be-
havior can be described simply and precisely merely by labeling
neighboring points on the ecliptic with the day and hour at which
the sun reaches each of them. The series of labeled points specifies
the annual component of the sun's motion; the remaining diurnal
component is specified by the daily rotation of the map as a whole.
For since the ecliptic appears in Figure 8 as a somewhat
distorted and considerably off-center circle, there must be one point,
SS, on the ecliptic that is nearer the central pole than any other. No
other point on the ecliptic rises and sets as far to the north as SS,
and no other joint stays within the horizon-window for as long
during the map's rotation. Therefore SS is the summer solstice, and
the sun's center must pass through it around June 22. Similarly the
points AE and VEin Figure 8 are the equinoctial points, the two points
on the ecliptic that rise and set due east and west and that remain
inside the horizon-window for exactly one-half of each map rotation.
The center of the sun must pass through them on September 23 and
March 21 respectively, just as it must pass through WS, the point on
the ecliptic farthest from the pole, on or near December 22. The sol-
stices and equinoxes, which first appeared as days of the year, have
now received a more precise and astronomically more useful defini-
tion. They are points on a star map or in the sky. Together with the
corresponding dates (or instants, since the sun's center passes instan-
taneously through each point), these labeled positions on the ecliptic
specify the direction and approximate rate of the sun's annual motion.
Given these labels and others like them, a man who knows how to
simulate the diurnal motion by rotating a star map can determine the
hours and positions of sunrise and sunset and the maximum height
of the sun on every day of the year.
The solstices and the equinoxes are not the only positions on the
ecliptic to receive standard labels. Drawn on a star map, the ecliptic
passes through a group of particularly prominent constellations, known
as the signs of the zodiac. By a convention dating from remote an-
tiquity these signs divide the ecliptic into twelve segments of equal
length. To say that the sun is "in" a particular constellation is to specify
approximately its position on the ecliptic, which, in turn, specifies the
season of the year. The annual journey of the sun through the twelve
signs seems to control the cycle of the seasons, an observation that is
one root of the science or pseudo science of astrology with which we
shall deal further in Chaper 3.
The Birth of Scientific Cosmology- The Two-Sphere Universe
The observations described in the last three sections are an
important part of the data used by ancient astronomers in analyzing
the structure of the universe. Yet, in themselves, these observations
provide no direct structural information. They tell nothing about the
composition of the heavenly bodies or their distance; they give no ex-
plicit information about the size, position, or shape of the earth.
Though the method of reporting the observations has disguised the
fact, they do not even indicate that the celestial bodies really move.
An observer can only be sure that the angular distance between a
celestial body and the horizon changes continually. The change might
as easily be caused by a motion of the horizon as by a motion of the
heavenly body. Terms like sunset, sunrise, and diurnal motion of a
star do not, strictly speaking, belong in a record of observation at all.
They are parts of an interpretation of the data, and though this inter-
26
pretation is so natural that it can scarcely be kept out of the vocabu-
lary with which the observations are discussed, it does go beyond
the content of the observations themselves. Two astronomers can agree
perfectly about the results of observation and yet disagree sharply
about questions like the reality of the motion of the stars.
Observations like those discussed above are therefore only clues
to a puzzle for which the theories invented by astronomers are tentative
solutions. The clues are in some sense objective, given by nature; the
numerical result of this sort of observation depends very little upon
the imagination or personality of the observer (though the way in
which the data are arranged may). But the theories or conceptual
schemes derived from these observations do depend upon the imagina-
tion of scientists. They are subjective through and through. Therefore,
observations like those discussed in the preceding sections could be
collected and put in systematic form by men whose beliefs about the
structure of the universe resembled those of the ancient Egyptians.
The observations in themselves have no direct cosmological conse-
quences; they need not be, and for many millennia were not, taken very
seriously in the construction of cosmologies. The tradition that detailed
astronomical observations supply the principal clues for cosmological
thought is, in its essentials, native to Western civilization. It seems to
be one of the most significant and characteristic novelties that we in-
herit from the civilization of ancient Greece.
A concern to explain observations of the stars and planets is appar-
ent in our oldest fragmentary records of Greek cosmological thought.
Early in the sixth century B.c., Anaximander of Miletus taught:
TJw stars are compressed portions of air, in the shape of [rotating]
wheels filled with fire, and they emit flames at some point from small open-
ings ....
The sun is a circle twenty-eight times the size of the earth; it is like a
chariot-wheel, the rim of which is hollow and full of fire, and lets the fire
shine out at a certain point in it through an opening like the nozzle of a
pair of bellows. . . .
The eclipses of the sun occur through the orifice by which the fire finds
vent being shut up.
The moon is a circle nineteen times as large as the earth; it is like a
chariot-wheel, the rim of which is hollow and full of fire, like the circle
of the sun, and it is placed obliquely, as that of the sun also is; it has one
vent like the nozzle of a pair of bellows; its eclipses depend on the turnings
of the wheel.l
Astronomically these conceptions are far in advance of the Egyp-
tians'. The gods have vanished in favor of mechanisms familiar on the
earth. The size and position of the stars and planets are discussed.
Though the answers given seem extremely rudimentary, the problems
had to be raised before they could receive mature and considered
solutions. In the fragment quoted the diurnal circles of the stars and
the sun are handled with some success by treating the celestial bodies
as orifices on the rims of rotating wheels. The mechanisms for eclipses
and for the annual wandering of the sun (the latter accounted for by
the oblique position of the sun's circle) are less successful, but they are
at least begun. Astronomy has started to play a major role in cosmologi-
cal thought.
Not all the Greek philosophers and astronomers agreed with
Anaximander. Some of his contemporaries and successors advanced
other theories, but they advanced them for the same problems and they
employed the same techniques in arriving at solutions. For us it is the
problems and techniques that are important. The competing theories
need not be traced; moreover, they cannot be traced completely, for
the historical records are too incomplete to permit more than conjec-
ture about the evolution of the earliest Greek conceptions of the uni-
verse. Only in the fourth century B.c. do the records become approxi-
mately reliable, and by that time, as the result of a long evolutionary
process, a large measure of agreement about cosmological essentials
had been reached. For most Greek astronomers and philosophers,
from the fourth century on, the earth was a tiny sphere susp{)nded
stationary at the geometric center of a much larger rotating sphere
which carried the stars. The sun moved in the vast space between the
earth and the sphere of the stars. Outside of the outer sphere there was
nothing at all- no space, no matter, nothing. This was not, in an-
tiquity, the only theory of the universe, but it is the one that gained
most adherents, and it is a developed version of this theory that the
medieval and modern world inherited from the ancients.
This is what I shall henceforth call the "two-sphere universe," con-
sisting of an interior sphere for man and an exterior sphere for the
stars. The phrase is, of course, an anachronism. As we shall see in
the next chapter, all those philosophers and astronomers who believed
in the terrestrial and celestial spheres also postulated some additional
cosmological device to carry the sun, moon, and planets around in the
space that lay between them. Therefore the two-sphere universe is not
really a cosmology at all, but only the structural framework for one.
Furthermore, that structural framework housed a great many different
and controversial astronomical and cosmological schemes during the
nineteen hundred years that separate the fourth century B.C. from
the age of Copernicus. There were many two-sphere universes. But,
after its first establishment, the two-sphere framework itself was almost
never questioned. For very nearly two millennia it guided the imagi-
nation of all astronomers and most philosophers. That is why we begin
our discussion of the main Western astronomical tradition by con-
sidering the two-sphere universe, framework though it is, in abstraction
from the various planetary devices advanced by one astronomer or
another to complete it.
The origin of the two-sphere framework is obscure, but the source of
its persuasiveness is not. The sphere of the heavens is only a short step
from the domed heaven of the Egyptians and Babylonians, and the
heavens do look domed. The elongation that the Egyptians gave to the
heavens vanishes in a society not based upon a river like the Nile and
leaves a hemispherical shell. Connecting the vault above the earth with
a symmetric vault below gives the universe an appropriate and satisfy-
ing closure. The rotation of the resulting sphere is indicated by the
stars themselves; as we shall shortly see, a steady rotation of the outer
sphere, once every 23 hours 56 minutes, will produce just the diurnal
circles that we have already described.
There is, in addition, an essentially aesthetic argument in favor of
the spherical universe. Since the stars seem as far away as anything we
can s ~ and since they all move together, it is natural to suppose that
they are simply markings on the outer surface of the universe and that
they move with it. Furthermore, since the stars move eternally with
perfect regularity, the surface on which they move ought itself be
perfectly regular and it should move in the same manner forever.
What figure better fulfills these conditions than the sphere, the only
completely symmetric surface and one of the few that can tum
eternally upon itself, occupying exadly the same space in each instant
of its motion? In what other form could an eternal and self-sufficient
universe be created? This is essentially the argument employed by the
Greek philosopher Plato (fourth century B.c.) in his Timaeus, an
allegorical story of the creation in which the universe appears as an
organism, an animal.
[The Creator's] intention was, in the first place, that the animal should
be as far as possible a perfect whole and of perfect parts: secondly, that
it should be one, leaving no remnants out of which another such world
might be created: and also that it should be free from old age [eternal] and
unaffected by disease [incorruptible J . . . . Wherefore he made the world
in the form of a globe, round as from a lathe, having its extremes in every
direction equidistant from the center, the most perfect and the most like
itself of all figures; for he considered that the like is infinitely fairer than the
unlike. This he finished off, making the surface smooth all round for many
reasons; in the first place, because the living being had no need of eyes
when there was nothing remaining outside him to be seen; nor of ears when
there was nothing to be heard; and there was no surrounding atmosphere to
be breathed; nor would there have been any use of organs by the help of
which he might receive his food or get rid of what he had already digested,
since there was nothing which went from him or came into him: for there
was nothing beside him. Of design he was created thus, his own waste
providing his own food, and all that he did or suffered taking place in and
by himself. For the Creator conceived that a being which was self-sufficient
would be far more excellent than one which lacked anything; and, as he had
no need to take anything or defend himself against any one, the Creator did
not think it necessary to bestow upon him hands: nor had he any need of
feet, nor of the whole apparatus of walking; but the movement suited to his
spherical form was assigned to him, . . . and he was made to move in
the samt: manner and on the same spot, within his own limits revolving in
a circle.
2
Some of the ancient arguments for the sphericity of the earth are of
the same sort: what is more fitting than that the earth, man's abode,
should display the same perfect figure with which the universe was
created? But many of the demonstrations are more concrete and
familiar. The hull of a ship sailing from shore disappears before the
top of the mast. More of the ship and of the sea is visible from a high
observation point than from a low one (Figure 10). The shadow of the
earth on the moon during a lunar eclipse always has a circular edge.
(This explanation of eclipses, current even before the fourth century
B.c., is discussed in Section 3 of the Technical Appendix.) These argu-
ments are still difficult to evade or refute, and in antiquity their
effectiveness extended by analogy from the earth to the heavens: a
celestial region that mirrored the shape of the earth seemed specially
appropriate. Other arguments derived from the similarity and sym-
metric arrangement of the two spheres. The earth's central position, for
example, kept it stationary i11 the spherical universe. In which direction
can a body fall from the center of a sphere? There is no "down" at the
Figure 10. An ancient (and modem) argument for the earth's sphericity. An
observer at the base of the mountain can see only the tip of the ship's mast over
the earth's bulge. From the mountain top the entire mast and part of the hull are
visible.
center, and every direction is equally "up." Therefore the earth must
hang at the center, eternally stable as the universe rotates about it.
Though these arguments from symmetry may seem strange today
(arguments for a discredited conclusion usually do seem strange),
they were very important in ancient, medieval, and early modem
thought. A discussion of symmetry, like Plato's, displays the appro-
priateness of the two-sphere cosmology; it explains why the universe
was created in the spherical form. Even more important, as we shall
discover in Chapters 3 and 4, the symmetry of the two spheres pro-
vided important links between astronomical, physical, and theological
thought, because it was essential to each. In Chapter 5 we shall find
Copen)icus struggling vainly to preserve the essential symmetry of
ancient cosmology in a universe constructed to contain a moving
planetary earth. But we are now most concerned with the astronomical
functions of the two-sphere universe, and here the case is entirely
clear. In astronomy the two-sphere cosmology works and works very
well. That is, it accounts precisely for the observations of the heavens
described in the earlier portions of this chapter.
Figure 11 shows a spherical earth, much exaggerated in size, at
the center of a larger sphere of the stars. An observer on the earth, at
a position indicated by the arrow, 0, can see just half of the sphere.
His horizon is bounded by a plane (shaded in the diagram) tangent
to the earth at the point where he is standing. If the earth is very small
N
Figure 11. Astronomical functions of the two-sphere universe. The outermost
circle is a cross section of the sphere of the stars which rotates steadily eastwad
about the axis NS. The observer at 0 can see all portions of this sphere that lie
above the shaded horizon plane SWNE. If the diagram were drawn to scale,
the earth would be much reduced and the horizon plane would be tangent to the
earth at the observation point. But since a scale drawing would reduce the earth
to minuscule dimensions, the plane is here drawn through the center of the sphere
of the stars, and its orientation with respect to the observer is preserved by keep-
ing it perpendicular to the line from the observer to the earth's center.
The horizontal circles in the diagram are the paths traced by selected points
on the sphere as the sphere goes through its daily rotation. They are therefore the
diurnal circles of selected stars, drawn solid where they are visible to the observer
and broken where they lie below the horizon. The central circle is traced by a
star on the celestial equator. It rises at E, due east of the observer, moves upward
along a line slanting south, and so on. The uppermost and lowermost circles are
traced by stars that meet the horizon in only one point. The upper circle, CC, is the
diurnal circle of the southernmost circumpolar star; the lower circle, II, is traced
by the northernmost of the stars that remain invisible to the observer at 0.
compared with the sphere of the stars, this tangent plane will divide
the outer sphere into two almost precisely equal parts, one visible to
the observer, and the other hidden from him by the surface of the
earth. Any objects permanently mounted on the outer sphere will,
like the stars, retain the same relative positions when seen from the
tiny central earth. If the sphere turns steadily about an axis through
the diametrically opposite points N and S, all the stars will move with
it unless they are actually located at N or S. Since S is invisible to the
observer in the diagram, N is the single stationary point in his heaven,
his celestial pole, and it is in fact located just about 45 over the due-
north point on his horizon, as it should be for an observer at 0, a point
in middle-northern latitudes.
Objects near the point N on the outer sphere appear to the observer
at 0 to rotate slowly in circles about the pole; if the sphere rotates once
in every 23 hours 56 minutes, these objects complete their circles in the
same period as the stars; they represent stars in the model. All stars
that lie near enough the pole to be inside the circle CC on the diagram
are circumpolar, for the sphere's rotation never carries them below the
horizon. Stars that lie farther from N, between the circles CC and II,
rise and set at an angle to the horizon once in each rotation of the
sphere, but stars nearest the circle II barely get above the southern
horizon and are visible only briefly. Finally, stars located within the
circle II, near S, never appear to the observer at 0; they are always
hidden by his horizon. They would, however, be visible to an observer
at other locations on the inner sphere; S is at least a potentially visible
fixed point in the heavens, a second pole. Call it the south celestial pole
and the visible point at N the north celestial pole.
If .. the observer in the diagram moves northward from 0 (that is,
toward a point on the inner sphere directly under the north celestial
pole), his horizon plane must move with him, becoming more and
more nearly perpendicular to the axis of the stellar sphere as he ap-
proaches the terrestrial pole. Therefure, as the observer moves north,
the celestial pole must appear farther and farther from the north point
on the horizon until at last it lies directly over his head. Simultaneously,
the circle CC, always drawn tangent to the northernmost point on the
horizon, must expand so that more and more stars become circumpolar.
Since the circle II also expands when the observer moves north, the
number of invisible stars must increase as well. If the observer moves
south, the effects are exactly reversed, the pole coming closer and
closer to the north point on the horizon and the circles CC and II
shrinking in size until they just enclose the north and south celestial
poles when the observer gets to the equator. Figure 12 shows the two
limiting cases, an observer at the north pole of the earth and an ob-
server at the terrestrial equator. In the first case the horizon is shown
horizontal; the north celestial pole is directly over the observer's head;
N N
{ol (b)
Figure 12. Stellar motions in the two-sphere universe as seen by an observer
(a) at the north terrestrial pole and (b) at the equator.
the stars in the upper half of the sphere swing continuously in circles
parallel to the horizon; and the stars in the lower hemisphere are never
seen at all. In the second diagram the horizon is vertical; the north
and south celestial poles are fixed at the north and south points on the
horizon; all stars can be seen at some time or other; but none can be
seen through more than a semicircle.
Except that these last extreme examples were not observed in an-
tiquity, the motion of the stars in the two-sphere model of the uni-
verse coincides precisely with the previously discussed observations
of real stars. There is no more convincing argument for the two-sphere
cosmology.
The Sun in the Two-Sphere Universe
A complete discussion of the sun's motion in the two-sphere
universe demands that the cosmology be elaborated to account for the
sun's intermediate position between the central earth and the periph-
eral rotating stellar sphere. This elaboration is part of the larger
problem of the planets; it will be considered in the next chapter. But
34
even the skeletal cosmology described above permits a great simplifica-
tion in the description of the sun's apparent motion. Seen from the
central earth against the spherical backdrop provided by the stellar
sphere, the sun's motion acquires a regularity that was not apparent
until the stars were localized on a rotating earth-centered sphere.
The new simplicity of the sun's apparent motion is described in
Figure 13, which is a reduced sketch of the stellar with _its
pole visible and with the direction of westward dmrnal rotatiOn mdi-
cated by an arrow about the pole. Halfway between the north and
south celestial poles is drawn the celestial equator, a great circle on
which lie all the stars (and all points on the sphere) that rise and set
exactly due east and west. A great circle is the simplest of all curves
that can be drawn on the surface of a sphere- the intersection of the
sphere's surface with that of a plane through the sphere's center -
and the new simplicity of the sun's motion results from the fact that on
a celestial sphere the ecliptic, too, is just a great circle, dividing the
sphere into two equal halves. In Figure 13 the ecliptic is the slanted
N
s
Figure 13. The equator and the ecliptic on the celestial sphere.
circle, intersecting the celestial equator in two diametrically opposite
points at an angle of
it contains all the points at which the center

of the sun is seen against the sphere of the stars by an observer on the
earth. At each instant the sun's center appears at a point on this great
circle, participating in the diurnal westward motion of the
sphere, but simultaneously the sun slips slowly. (arrows m
diagram), completing its journey around the ecliptiC once m a
Because, during any 24-hour period, the sun appears to remam very
near a single point on the ecliptic, it must move daily in a diurnal circle
very much like that of a star. But the sun moves slowly eastward with
respect to the sphere as the sphere itself turns rapidly westward.
Therefore the sun must complete its diurnal circle slightly more slowly
than the stars complete theirs, losing a little ground in its race with
the stars each day and being completely "lapped" by them once each
year. More precisely, since the sun must move through 360 to traverse
the ecliptic and since it completes this journey in just over 365 days,
its eastward motion along the ecliptic must cover just under 1 o per
day, and this is the figure derived earlier from observation ( p. 23).
It is the distance which the sun slips backward (or loses) with
respect to the stars each day. Furthermore, since the length of the day
is defined by the diurnal motion of the sun and since the stars (moving
15 per hour or 1 every 4 minutes) get 1 farther ahead of the sun
each day, a star that gets, say, overhead at midnight tonight will com-
plete its diurnal motion and return to the same position in the sky
just 4 minutes before midnight tomorrow night. Once again a detail
about the behavior of the heavens, initially introduced as one among
many assorted observations ( p. 16), has become part of a coherent
pattern in the two-sphere universe.
A similar order is apparent in the positions that the equinoxes and
solstices assume on the stellar sphere. The two equinoxes must be
the two diametrically opposite positions on the stellar sphere where
the ecliptic intersects the celestial equator. These are the only points
on the ecliptic that always rise and set due east and west. Similarly,
the two solstices must be the points on the ecliptic midway between
the two equinoxes, for these are the points on the ecliptic that lie
farthest north and south of the celestial equator. When the sun is at
one of these points it must rise farther north (or south) of the due-
east point than it does at any other time. Since the sun moves steadily
eastward from the summer solstice toward the autumnal equinox, the
individual equinoxes and solstices are readily identified on the sphere.
Each of them is labeled on the ecliptic in Figure 13, and once the
ecliptic has been drawn and labeled in this manner it is possible, by
constructing an appropriate horizon plane inside the stellar sphere, to
discover how the sun's behavior varies during the course of a year
when observed from any location on the earth's surface. Three particu-
larly significant examples of the sun's motion at various seasons of the
year are derived from the two-sphere conceptual scheme in the dia-
grams of Figure 14. In these diagrams the full force of the conceptual
scheme begins to appear.
The Functions of a Conceptual Scheme
Unlike the observations described in the early sections of this
chapter, the two-sphere universe is a product of the human imagina-
tion. It is a conceptual scheme, a theory, deriving from observations
but simultaneously transcending them. Because it will not yet account
N N
s
(a) (b)
N
s
(c)
Figure 14. The motion of the sun observed from different locations on the
earth.
(a) Observer in middle-northern latitudes: At the summer solstice the sun
rises along a slanting line far north of the due-east point; more than one-half of
its diurnal circle lies above the horizon, so that the days are longer than the nights.
At the equinoxes the sun rises due east, and only one-half of its diurnal circle is
visible. At the winter solstice the sun rises far to the south, and the days are
shorter than the nights. The sun's maximum daily elevation above the horizon is
greateit in summer, but at all seasons the sun's noon shadow must point due
north.
(b) Observer at the equator: Whatever the sun's location on the ecliptic, the
horizon plane divides the sun's diurnal circle into two equal parts. Days and
nights are always of equal length, and there is little seasonal variation of climate.
During half the year (vernal equinox to autumnal equinox) the sun rises north
of the due-east point, and its noon shadow points due south. During the other
half year the sun rises south of east, and its noon shadow points north.
(c) Observer at the north terrestrial pole: Half of the ecliptic is always below
the horizon, so that for half the year (autumnal to vernal equinox) the sun is
completely invisible. At the vernal equinox the sun begins to peek above the hori-
zon, swinging around it daily in a gradually rising spiral until the summer solstice.
Then the sun gradually spirals back to the horizon, disappearing slowly below it
at the autumnal equinox. Between the vernal and autumnal equinoxes the sun
does not set.
for the motions of all celestial bodies (the planets, in particular, have
been ignored to this point), the two-sphete cosmology is not complete.
But already it provides cogent illustrations of some logical and psycho-
logical functions that scientific theories can perform for the men who
develop or make use of them. The evolution of any scientific con-
ceptual scheme, astronomical or nonastronomical, depends upon the
way in which it performs these functions. By making some of them
explicit before the two-sphere universe is elaborated in the next two
chapters, we can highlight in advance a few of the most fundamental
problems that will emerge during this study of the Copernican Revo-
lution.
Perhaps the most striking characteristic of the two-sphere universe
is the assistance that it gives to the astronomer's memory. This char-
acteristic of a conceptual scheme is often called conceptual economy.
Though they were both carefully selected and systematically pre-
sented, the observations of the sun and stars discussed in earlier sec-
tions were, as a group, extremely complex. To a man not already
thoroughly acquainted with the heavens, one observation, like the
direction of the slant line along which the sun rises or the correspond-
ing behavior of the gnomon's shadow, seems unrelated to another
observation, like the location of the celestial pole or the brief appear-
ance of the stars in the southern sky. Each observation is a separate
item in a long list of bare facts about the heavens, and it is difficult to
retain the whole list in memory simultaneously.
The two-sphere universe presents no such problem: a gigantic
sphere bearing the stars rotates steadily westward on a fixed axis
once every 23 hours 56 minutes; the ecliptic is a great circle on this
sphere tilted 2 3 ~
0
to the celestial equator, and the sun moves steadily
eastward about the ecliptic once every 3 6 5 ~ days; the sun and stars
are observed from a tiny fixed sphere located at the center of the giant
stellar sphere. That much can be committed to memory once and for
all, and while it is remembered the list of observations may be for-
gotten. The model replaces the list, because, as we have already seen,
the observations can be derived from the model. Frequently they need
not even be derived. A man who observes the heavens with the two-
sphere universe firmly fixed in his mind will find that the conceptual
scheme discloses a pattern among otherwise unrelated observations,
that a list of the observations becomes a coherent whole for the first
time, and that the individual items on the list are therefore more easily
remembered. Without these ordered summaries which its theories
provide science would be unable to accumulate such immense stores
of detailed information about nature.
Because it provides a compact summary of a vast quantity of im-
portant observational materials, the two-sphere universe is actively
employed by many people today. The theory and practice of both navi-
gation and surveying can be developed with great simplicity and pre-
cision from models built to the specifications of Figure 11, and, since
the model demanded by modem astronomy is far more complex, the
two-sphere universe is normally used in preference to the Copernican
when teaching these subjects. Most handbooks of navigation or sur-
veying open with some sentence like this: "For present purposes we
shall assume that the earth is a small stationary sphere whose center
coincides with that of a much larger rotating stellar sphere." Evalu-
ated in terms of economy, the two-sphere universe therefore remains
what it has always been: an extremely successful theory.
In other respects, however, the two-sphere universe is no longer
at all successful and has not been since the Copernican Revolution. It
has remained economical only because economy is a purely logical
function. The celestial observations known to ancient astronomers and
used by modem navigators are logical consequences of the two-sphere
model whether or not the model is thought to represent reality. The
attitude of the scientist, his belief in the "truth" of the conceptual
scheme, does not affect the scheme's logical ability to provide an eco-
nomical summary. But conceptual schemes have psychological as well
as logical functions, and these do depend upon the scientist's belief or
incredulity. For example, the psychological craving for at-homeness,
discussed in the second section, can be satisfied by a conceptual scheme
only if that scheme is thought to be more than a convenient device for
summarizing what is already known. During antiquity and again in
the later Middle Ages the European world did have this additional
commitment to the conception of a two-sphere universe. Scientists and
nonscientists alike believed that the stars really were bright spots on
a gigantic sphere that symmetrically enclosed man's terrestrial abode.
As a result, two-sphere cosmology did for centuries provide many men
with a world view, defining their place in the created world and giving
physical meaning to their relation with the gods. As we shall see in
Chapters 3 and 4, a conceptual scheme that is believed and that there-
fore functions as part of a cosmology has more than scientific signifi-
cance.
Belief also affects the way in which conceptual schemes function
within the sciences. Economy as a purely logical function, and cosmo-
logical satisfaction as a purely psychological function, lie at opposite
ends of a spectrum. Many other significant functions lie within the
spectrum, between these limits, depending both upon the logical struc-
ture of the theory and upon its psychological appeal, its ability to
evoke belief. For example, an astronomer who believes in the validity
of the two-sphere universe will find that the theory not only provides
a convenient summary of the appearances, but that it also explains
them, enabling him to understand why they are what they are. Words
like "explain" and "understand" apparently refer simultaneously to the
logical and psychological aspects of conceptual schemes. Logically,
the two-sphere universe explains the motions of the stars because the
motions can be deduced from the far simpler model. Complexity is
reduced, and such logical reduction is one essential component of
explanation. But it is not the only one. Psychologically, the two-sphere
universe provides no explanation unless it is believed to be true. The
modem navigator uses the two-sphere universe on his job, but he does
not explain the stellar motions in terms of a rotation of the outer sphere.
He believes that the diurnal motion of the stars is only an apparent
motion, and he must therefore explain it as the result of a real rota-
tion of the earth.
A scientist's willingness to use a conceptual scheme in explana-
tions is an index of his commitment to the scheme, a token of his be-
lief that his model is the only valid one. Such commitment or belief
is always rash, because economy and cosmological satisfaction cannot
guarantee truth, whatever "truth" may mean. The history of science is
cluttered with the relics of conceptual schemes that were once fer-
vently believed and that have since been replaced by incompatible
theories. There is no way of proving that a conceptual scheme is final.
But, rash or not, this commitment to a conceptual scheme is a common
phenomenon in the sciences, and it seems an indispensable one, be-
cause it endows conceptual schemes with one new and all-important
function. Conceptual schemes are comprehensive; their consequences
are not limited to what is already known. Therefore, an astronomer
committed to, say, the two-sphere universe will expect nature to show
the additional, but as yet unobserved, properties that the conceptual
scheme predicts. For him the theory will transcend the known, be-
coming first and foremost a powerful tool for predicting and explor-
ing the unknown. It will affect the future of science as well as its past.
The two-sphere universe tells the scientist about the behavior of
the sun and stars in parts of the world (like the southern hemisphere
and the terrestrial poles) to which he has never traveled. In addition
it informs him of the motion of stars that he has never observed sys-
tematically. Since they are fastened to the stellar sphere, they must
revolve in diurnal circles as the other stars do. This is new knowledge,
derived initially not from observation but directly from the conceptual
scheme, and such new knowledge can be immensely consequential.
For example, two-sphere cosmology states that the earth has a cir-
cumference, and it suggests a set of observations (discussed in Sec-
tion 4 of the Technical Appendix) by which the astronomer can dis-
cover how large the earth's circumference is. One set of these obser-
vations (a bad one, as it happened, for the resulting value of the cir-
cumference was far too small) led Columbus to believe that the cir-
cumnavigation of the globe was a practical undertaking, and the results
of his voyages have been recorded. Those voyages and the subsequent
travels of Magellan and others provided observational evidence for
beliefs that had previously been derived solely from theory, and they
supplied science with many unanticipated observations besides. The
voyages would not have been undertaken, and the novel observations
would not have accrued to the sciences, if a conceptual scheme had
not ,pointed the way.
Columbus' voyages are one example of the fruitfulness of a con-
ceptual scheme. They show how theories can guide a scientist into
the unknown, telling him where to look and what he may expect to
find, and this is perhaps the single most important function of con-
ceptual schemes in science. But the guidance provided by conceptual
schemes is rarely so direct and unequivocal as that illustrated above.
Typically a conceptual scheme provides hints for the organization of
research rather than explicit directives, and the pursuit of these hints
usually requires extension or modification of the conceptual scheme
that provided them. For example, the two-sphere universe was initially
developed principally to account for the diurnal motions of the stars
and for the way in which those motions varied with the observer's
location on the earth. But once it had been developed, the new theory
was readily extended to give order and simplicity to observations of
the sun's motion as well. And, having disclosed the unsuspected regu-
larity that underlay the complexity of the sun's behavior, the con-
ceptual scheme provided a framework within which could be studied
the even more irregular motions of the planets. That problem had
seemed unmanageable until the over-all motion of the heavens was
reduced to order.
Much of this book will be concerned with the fruitfulness of par-
ticular conceptual schemes, that is, with their effectiveness as guides
for research and as frameworks for the organization of knowledge.
The next two chapters, in particular, will examine the role of the two-
sphere universe in the ancient solution, first, of the problem of the plan-
ets and, then, of some problems lying entirely outside astronomy. Later
we shall discover the rather different sort of guidance given to scien-
tific research by Copernicus' novel conception of a moving planetary
earth. The very best example of fruitfulness is, however, the story told
in the whole of this book. The Copernican universe is itself the product
of a series of investigations that the two-sphere universe made possible:
the conception of a planetary earth is the most forceful illustration of
the effective guidance given to science by the incompatible conception
of a unique central earth. That is why a discussion of the Copernican
Revolution must begin with a study of the two-sphere cosmology which
Copernicanism ultimately made obsolete. The two-sphere universe is
the parent of the Copernican; no conceptual scheme is born from
nothing.
Ancient Competitors of the Two-Sphere Universe
The two-sphere conception of the universe was not the only
cosmology suggested in ancient Greece. But it was the one taken most
seriously by the largest number of people, particularly by astronomers,
and it was the one that later Western civilization first inherited from
the Greeks. Yet many of the alternate cosmologies proposed and re-
jected in antiquity show far closer resemblances to modern cosmologi-
cal beliefs than does the two-sphere universe. Nothing more clearly
illustrates the strengths of the two-sphere cosmology and foreshadows
the difficulties to be encountered in overthrowing it than a compari-
42
son of the scheme with a few of its superficially more modern alterna-
tives.
As early as the fifth century B.C., the Greek atomists, Leucippus and
Democritus, visualized the universe as an infinite empty space, popu-
lated by an infinite number of minute indivisible particles or atoms
moving in all directions. In their universe the earth was but one of
many essentially similar heavenly bodies formed by the chance aggre-
gation of atoms. It was not unique, nor at rest, nor at the center. In
fact, an infinite universe has no center; each part of space is like every
other; therefore the infinite number of atoms, some of which aggre-
gated to form our earth and sun, must have formed numerous other
worlds in other portions of the empty space or void. For the atomists
there were other suns and other earths among the stars.
Later in the fifth century the followers of Pythagoras suggested a
second cosmology which set the earth in motion and partially deprived
it of its unique status. The Pythagoreans did place the stars on a
gigantic moving sphere, but at the center of this sphere they placed
'ln immense fire, the Altar of Zeus, invisible from the earth. The fire
could not be seen, because the earth's populated areas were always
turned away from the fire. For the Pythagoreans the earth was just
one of a number of celestial bodies, including the sun, all of which
moved in circles about the central fire. A century later Heraclides of
Pontus (fourth century B.C.) suggested that it was a daily rotation
of the central earth rather than a rotation of the peripheral sphere of
the stars that produced the apparent motion of the heavens. He also
obscured the symmetry of the two-sphere universe by suggesting that
the.planets Mercury and Venus revolved in circles about the moving
sun rather than in independent circular orbits about the central earth
(see Chapter 2). Still later, in the middle of the third century B.C.,
Aristarchus of Samos, whose ingenious and influential measurements
of astronomical dimensions are described in the Technical Appendix,
advanced the proposal that has earned for him the title of "the Coper-
nicus of antiquity." For Aristarchus the sun was at the center of an
immensely expanded sphere of the stars, and the earth moved in a
circle about the sun.
These alternative cosmologies, particularly the first and last, are
remarkably like our modern views. We do believe today that the earth
is but one of a number of planets, circulatin)Y .bout the sun, and that
43
the. sun is but one of a multitude of stars, some of which may have
theu But though some of these speculative suggestions
gave nse to s1gmficant minority traditions in antiquity, and though all
of them were a continuing source of intellectual stimulus to innovators
like Copernicus, they were not originally supported by the arguments
that now us believe them, and in the absence of these arguments
they were rejected by most philosophers and almost all astronomers in
the ancient world. In the Middle Ages they were ridiculed or ignored.
The reasons for the rejection were excellent. These alternative cosmolo-
gies violate the first and most fundamental suggestions provided by
the senses about the structure of the universe. Furthermore this viola-
tion of common sense is not compensated for by any in the
effectiveness with which they account for the appearances. At best they
are no more economical, fruitful, or precise than the two-sphere uni-
verse, and they are a great deal harder to believe. It was difficult to
take them seriously as explanations.
All of these alternative cosmologies take the motion of the earth
as a premise, and all (except Heraclides' system) make the earth
move as one of a number of heavenly bodies. But the first distinction
suggested by the senses is that separating the earth and the heavens.
The earth is not patt of the heavens; it is the platform from which we
view them. And the platform shares few or no apparent characteristics
the bodies seen from it. The heavenly bodies seem bright
pomts of hght, the earth an immense nonluminous sphere of mud and
rock. Little change is observed in the heavens: the stars are the same
night after night and apparently have remained so throughout the
many centuries covered by ancient records. In contrast the earth is
the home of birth and change and destruction. Vegetation and animals
alter from week to week; civilizations rise and fall from century to
century; legends attest the slower topographical changes produced on
earth by flood and storm. It seems absurd to make the earth like
celestial bodies whose most prominent characteristic is that immutable
regularity never to be achieved on the corruptible earth.
The idea that the earth moves seems initially equally absurd. Our
senses tell us all we know of motion, and they indicate no motion for
the earth. Until it is reeducated, common sense tells us that, if the
earth is in motion, then the air, clouds, birds, and other objects not
attached to the earth must be left behind. A man jumping would de-
scend to earth far from the point where his leap began, for the earth
would move beneath him while he was in the air. Rocks and trees,
cows and men must be hurled from a rotating earth as a stone Hies
from a rotating sling. Since none of these effects is seen, the earth is
at rest. Observation and reason have combined to prove it.
Today in the Western world only children argue this way, and
only children believe that the earth is at rest. At an early age the author-
ity of teachers, parents, and texts persuades them that the earth is
really a planet and in motion; their common sense is reeducated; and
the arguments born from everyday experience lose their force. But
reeducation is essential - in its absence these arguments are immensely
persuasive - and the pedagogic authorities that we and our children
accept were not available to the ancients. The Greeks could only
rely on observation and reason, and neither produced evidence for
the earth's motion. Without the aid of telescopes or of elaborate
mathematical arguments that have no apparent relation to astronomy,
no effective evidence for a moving planetary earth can be produced.
The observations available to the naked eye fit the two-sphere universe
very well (remember the universe of the practical navigator and sur-
veyor), and there is no more natural explanation of them. It is not
hard to realize why the ancients believed in the two-sphere universe.
The problem is to discover why the conception was given up.
2
THE PROBLEM
OF THE PLANETS
Apparent Planetary Motion
If the sun and stars were the only celestial bodies visible to
the naked eye, modern man might still accept the fundamental tenets
of the two-sphere universe. Certainly he would have accepted them
until the invention of the telescope, more than half a century after
Copernicus' death. There are, however, other prominent celestial
bodies, particularly the planets, and the astronomer's interest in these
bodies is the principal source of the Copernican Revolution. Once
again we consider observations before dealing with interpretive ex-
planations. And once again the discussion of interpretations will con-
front us with a new and fundamental problem about the anatomy of
scientific belief.
The term planet is derived from a Greek word meaning "wanderer,"
and it was employed until after Copernicus' lifetime to distinguish
those celestial bodies that moved or "wandered" among the st'lrs from
those whose relative positions were fixed. For the Greeks and their
successors the sun was one of the seven planets. The others were the
moon, Mercury, Venus, Mars, Jupiter, and Saturn. The stars and these
seven planets were the only bodies recognized as celestial in antiquity.
No additional planets were discovered untill781, long after the Coper-
nican theory had been accepted. Comets, which were well known in
the ancient world, were not considered celestial bodies before the
Copernican Revolution (Chapter 6).
All of the planets behave somewhat like the sun, though their mo-
tions are uniformly more complex. All have a westward diurnal motion
with the stars, and all move gradually eastward among the stars until
they return to approximately their original positions. Throughout their
46
motions the planets stay near the ecliptic, sometimes wamlering north
of it, sometimes south, but very seldom leaving the band of the zodiac,
an imaginary strip in the sky extending for 8 on either side of the
ecliptic. At this point the resemblance between planets ends, and the
study of planetary irregularities begins. .
The moon travels around the ecliptic faster and less steadily than
the sun. On the average it completes one journey through the zodiac
in days, but the time required for any single journey may differ
from the average by as much as 7 hours. In addition, the
of the moon's disk changes markedly as it moves. At new moon Its disk
is completely invisible or very dim; then a thin bright crescent
which gradually waxes until, about a week after new moon, a
circular sector is visible. About 2 weeks after new moon the full cu-
cular disk appears; then the cycle of phases is reversed, and the moon
gradually wanes, reaching new moon again about 1 after th,e
preceding new moon. The cycle of phases is recurrent, hke the moons
journey through the signs of the zodiac, but the two lunar cycles are
significantly out of step. New recurs after an interval
days (individual cycles may differ by as much as day from this
average), and, since this is 2 days longer than the period of an average
journey around the zodiac, the position of successive new moons must
gradually move eastward through the constellations. If new
occurs at the position of the vernal equinox one month, the moon Will
still be waning when it returns to the vernal equinox days later.
New moon does not recur for about 2 days more, by which time the
moon has moved almost 30 east from the equinox.
J?ecause they are easily visible and conveniently spaced, the moon's
phases provided the oldest of all calendar units. Primitive forms of
both the week and the month appear in a Babylonian calendar from
the third millennium B.C., a calendar in which each month began with
the first appearance of the crescent moon and was subdivided at the
7th, 14th, and 21st days by the recurrent "quarters" of the moon's cycle.
At the dawn of civilization men must have counted new moons and
quarters to measure time intervals, and as c;vilization progressed they
repeatedly attempted to organize these fundamental units into a co-
herent long-term calendar - one that would permit the compilation of
historical records and the preparation of contracts to be honored at a
specified future date.
THE PROBLEM OF THE PLANETS
47
But at this point the simple obvious lunar unit proved intractable.
Successive new moons may be separated by intervals of either 29 or
30 days, and only a complex mathematical theory, demanding genera-
tions of systematic observation and study, can determine the length
of a specified future month. Other difficulties derive from the incom-
mensurable lengths of the average lunar and solar cycles. Most soci-
eties (but not all, for pure lunar calendars are still used in parts of
the Middle East) must adjust their calendars to the sun-governed an-
nual climatic variation, and for this purpose some systematic method
for inserting an occasional thirteenth month into a basic year of 12
lunar months ( 354 days) must be devised. These seem to have been
the first difficult technical problems encountered by ancient astronomy.
More than any others, they are responsible for the birth of quantita-
tive planetary observation and theory. The Babylonian astronomers who
finally solved these difficulties between the eighth and third centuries
B.c., a period during most of which Greek science was still in its in-
fancy, accumulated much of the fundamental data subsequently incor-
porated into the developed structure of the two-sphere universe.
Unlike the moon and sun, the remaining five planets appear as
mere points of light in the heavens. The untrained naked-eye observer
can distinguish them from stars with assurance only by a series of
observations that discloses their gradual motion around the ecliptic.
Usually the planets move eastward through the constellations: this is
their so-called "normal motion." On the average, both Mercury and
Venus require 1 year for each complete circuir of the zodiac; the length
of Mars's cycle averages 687 days; Jupiter's average period is 12 years;
and Saturn's is 29 years. But in all cases the time required for any single
journey may be quite different from the average period. Even when
moving eastward through the stars, a planet does not continue at a
uniform rate.
Nor is its motion uniformly eastward. The normal motion of all
planets except the sun and moon is occasionally interrupted by brief
intervals of westward or "retrograde" motion. Compare Mars retro-
gressing in the constellation Taurus, shown in Figure 15, with the
normal motion of the sun through Taurus, shown in Figure 9 ( p. 22).
Mars enters the diagram in normal (eastward) motion, but as its
motion continues, the planet gradually slows until at last it reverses
its direction and begins to move westward, in retrograde. Other planets
behave in much the same way, each one repeating the interlude of
retrograde motion after a fixed length of time. Mercury briefly reverses
its motion through the stars once every 116 days, and Venus retro-
gresses every 584 days. Mars, Jupiter, and Saturn show retrograde
motion every 780, 399, and 378 days, respectively.

*

TAURUS

Figure 15. Mars retrogressing in Aries and Taurus. The section of sky is the
same as that shown in Figure 9 and in the box on the star map of Figure 8. The
broken iine is the ecliptic and the solid line the path of the planet. Note that
Mars does not stay on the ecliptic and that, though its over-all motion is eastward
among the stars, there is a period from the middle of June to early August during
which it moves to the west. The retrogressions of Mars are always of approxi-
mately this form and duration, but they do not always occur on the same date
or in the same part of the sky.
In their gradual eastward motions interrupted by periodic westward
retrogressions, the five wandering stars behave quite similarly. But
there is an additional characteristic of their motion which divides them
into two groups; this is the correlation between their position and the
sun's. Mercury and Venus, the two so-called inferior planets, never
get very far from the sun. Mercury is always found within 28 of the
sun's moving disk, and Venus's maximum "elongation" is 45. Both
planets move in a continuous slow shuttle, back and forth across the
THE PROBLEM OF THE PLANETS 49
moving sun; for a time they move eastward with the sun, then retro-
gress across its disk, and finally reverse themselves to overtake the
sun once more. When to the east of the sun, either of these inferior
planets appears as an "evening star," becoming visible shortly after
sunset and then rapidly following the sun below the horizon. Mter
retrogressing westward across the sun's disk, the planet becomes a
"morning star," rising shortly before dawn and disappearing in the
brilliant light of sunrise. But in between, when close to the sun,
neither Mercury nor Venus can be seen at all. Therefore, until their
motion was analyzed with respect to the sphere of the stars, neither
of the inferior planets was recognized as the same celestial body
when it appeared as a morning and as an evening star. For millenniums
Venus had one name when it rose in the east shortly before dawn and
another when, weeks later, it again became visible just over the west-
ern horizon shortly after sunset.
Unlike Mercury and Venus, the superior planets, Mars, Jupiter,
and Saturn, are not restricted to the same part of the sky as the sun.
Sometimes they are very close to or "in conjunction" with it; at other
times they are 180 across the sky or "in opposition" to the sun; be-
tween these times they assume all the intermediate positions. But
though their positions are unrestricted, their behavior does depend
upon their relation to the sun. Superior planets retrogress only when
they are in opposition. Also, when in retrograde motion across the sky
from the sun, superior planets appear brighter than at any other
time. This increased brilliance, which has usually been interpreted
(at least since the fourth century B.C.) as indicating a decrease in the
planet's distance from thP ~ a r t h is particularly striking in the case of
Mars. Normally a relatively inconspicuous planet, Mars in opposition
will frequently outshine every celestial body in the night sky except
the moon and Venus.
Interest in the five wandering stars is by no means so ancient as a
concern with the sun and moon, presumably because the wandering
stars had no obvious practical bearing upon the lives of ancient peoples.
Yet observations of the appearance and disappearance of Venus were
recorded in Mesopotamia as early as 1900 B.C., probably as omens,
portents of the future, like the signs to be read in the entrails of sacri-
ficial sheep. These scattered observations presage the much later de-
velopment of systematic astrology, a means of forecasting whose inti-
50
mate relation to the development of planetary astronomy is considered
in the next chapter. The same concern with omens clearly motivated
the more systematic and complete records of eclipses, retrograde mo-
tions, and other striking planetary phenomena compiled by Babylonian
observers after the middle of the eighth century B.C. Ptolemy, the
dean of ancient astronnmers, later complained that even these records
were fragmentary, but fragmentary or not they provided the first data
capable of specifying the full-scale problem of the planets as that prob-
lem was to develop in Greece after the fourth century B.C.
The problem of the planets is partially specified by the description
of the planetary motions sketched in the preceding pages. How are
the complex and variable planetary motions to be reduced to a simple
and recurrent order? Why do the planets retrogress, and how account
for the irregular rate of even their normal motions? These questions
indicate the direction of most astronomical research during the two
millenniums from the time of Plato to the time of Copernicus. But be-
cause it is almost entirely qualitative, the preceding description of the
planets does not specify the problem fully. It states a simplified prob-
lem and in some respects a misleading one. As we shall shortly see,
qualitatively adequate planetary theories are easily invented: the de-
scription above can be reduced to order in several ways. The astrono-
mer's problem, on the other hand, is by no means simple. He must
explain not merely the existence of an intermittent westward motion
superimposed upon an over-all eastward motion through the stars, but
also the precise position that each planet occupies among the stars
on different days, months, and years over a long period of time. The
real problem of the planets, the one that leads at last to the Copernican
Revolution, is the quantitative problem described in lengthy tables
which specify in degrees and minutes of arc the varying position of
every planet.
The location of the Planets
The two-sphere universe, as developed in the last chapter,
provided no explicit information about the positions or motions of the
seven planets. Even the sun's location was not discussed. To appear
"at" the vernal equinox (or any other point on the stellar sphere) the
sun need merely be somewhere on a line stretching from the observer's
eye to or through the appropriate point in the background of stars.
51
Like the other planets, it might be either inside, on, or perhaps even
outside the sphere of the stars. But though the two-sphere universe
fails to specify the shape or location of the planetary orbits, it does
make certain choices of position and orbit more plausible than others,
and it therefore at once guides and restricts the astronomer's approach
to the problem of the planets. That problem was set by the results of
observation, but, from the fourth century B.C., it was pursued in the
conceptual climate of two-sphere cosmology. Both observation and
theory made essential contributions to it.
Within a two-sphere cosmology, for example, the planetary orbits
should if possible preserve and extend the fundamental symmetry em-
bodied in the first two spheres. Ideally the orbits should therefore be
earth-centered circles, and the planets should revolve in these circles
with the same regularity that is exemplified in the rotation of the stellar
sphere. The ideal does not quite conform to observation. As we shall
see presently, an earth-centered circular orbit located in the plane of
the ecliptic provides a good account of the sun's annual motion, and a
similar circle can give an approximate account of the somewhat less
regular motion of the moon. But circular orbits do not even hint at an
explanation of the gross irregularities, like retrogression, observed in
the motions of the other five wandering "stars." Nevertheless, astrono-
mers who believed in the two-sphere universe could, and for centuries
did, think that earth-centered circles were the natural orbits for
planets. Such orbits at least accounted for the over-all average east-
ward motions. Observed deviations from the average motion- changes
in the rate or direction of a planet's motion- indicated that the planet
itself had deviated from its natural circular orbit, to which it would
again return. On this analysis the problem of the planets became
simply that of explaining the observed deviation from average motion
through the stars in terms of a corresponding deviation of each planet
from its single circular orbit.
We shall examine some of the ancient explanations of these devia-
tions in the next three sections, but first notice, as the ancients also did,
how far it is possible to proceed by neglecting the planetary irregulari-
ties and assuming simply that all orbits are at least approximately
circular. Almost certainly, in the two-sphere universe, the planets move
in the region between the earth and the stars. The stellar sphere itself
was often viewed as the outer boundary of the universe, so that the
planets could not be outside it; the difference between planetary and
stellar motions indicated that the planets were probably not located on
the sphere, but in some intermediate region where they were affected
by some influence that was inactive at the stellar sphere; the whole
argument gained force from the detail visible on the face of the moon,
presumptive evidence that one planet, at least, must be nearer than the
stars. Ancient astronomers, therefore, laid out the planetary orbits in
that vast and previously empty space between the earth and the sphere
of the stars. By the end of the fourth century B.C., the two-sphere uni-
verse was filling up. Later it was to become crowded.
Once the general location and shape of their orbits were known, it
proved possible to make a plausible and satisfying guess about the
order in which the planets were arranged. Planets like Saturn and
Jupiter, whose eastward motion was slow and whose total motion,
therefore, very nearly kept pace with the stars, were supposed to be
close to the stellar sphere and far from the earth. The moon, on the
other hand, which loses over 12 a day in its race with the stars, must
be closer to the stationary surface of the earth. Some ancient philos-
ophers seem to have justified this hypothetical arrangement by im-
agining that the planets floated in a gigantic aethereal vortex, whose
outer surface moved rapidly with the sphere of the stars and whose
interior was at rest at the earth's surface. Any planet caught in such a
vortex would lose more ground with respect to the stellar sphere if it
were closer to the earth. Other philosophers reached the same con-
clusion by a different sort of argument, later recorded, at least in its
essentials, by the Roman architect Vitruvius (first century B.C.). In
the differences between the intervals required by different
planets for trips about the ecliptic, Vitruvius suggested an illuminating
analogy:
Place seven ants on a wheel such as potters use, having made seven chan-
nels on the wheel about the center, increasing successively in circum-
ference; and suppose those ants obliged to make a circuit in these channels
while the wheel is turned in the opposite direction. In spite of having
to move in a direction contrary to that of the wheel, the ants must neces-
sarily complete their journeys in the opposite direction, and that ant which
is nearest the center must finish its circuit sooner, while the ant that is going
round at the outer edge of the disk of the wheel must, on account of the
size of its circuit, be much slower in completing its course, even though it is
moving just as quickly as the other. In the same way, these stars, which
53
struggle on against the course of the firmament, are accomplishing an orbit
on paths of their own; but, owing to the revolution of the heaven, they are
swept back as it goes round every day.
1
Before the end of the fourth century B.c., arguments like the above
had led to an image of the universe similar to the one sketched in
Figure 16; diagrams like this, or their verbal equivalents, remained
current in elementary books on astronomy or cosmology until the early
seventeenth century, long after Copernicus' death. The earth is at the
center of the stellar sphere, which bounds the universe; immediately
inside this outer sphere is the orbit of Saturn, the planet that takes
longest to move around the zodiac; next comes Jupiter and then Mars.
Figure 16. Approximate planetary orbits in the two-sphere universe. The
outermost circle is a cross section of the stellar sphere in the plane of the ecliptic.
To this point the order is unambiguous: the planets are arranged, from
the outside, in the order of decreasing orbital period; the same tech-
nique places the lunar orbit closest to the earth. But the remaining
three planets present a problem; the sun, Venus, and Mercury all com-
plete their journeys about the earth in the same average time, 1 year,
and their order therefore cannot be determined by the device applied
to the other planets. There was, in fact, much disagreement about their
order during antiquity. Until the second century B.c. most astronomers
placed the sun's orbit just outside the moon's, Venus's outside the
sun's, then Mercury's, and then Mars's. After that date, however, the
order shown in the diagram- moon, Mercury, Venus, sun, .Mars, etc.-
became increasingly popular. In particular, it was adopted by Ptolemy,
and his authority imposed it upon most of his successors. We shall
therefore take this order as standard throughout the early chapters
of this book.
As a structural diagram Figure 16 is still very crude. It gives no
meaningful indication of relative dimensions of the various orbits, and
it makes no attempt to provide for the observed planetary irregulari-
ties. But the conception of the universe embodied in the diagram had
two important functions in the subsequent development of astronomy
and cosmology. In the first place, the diagram contains most of the
structural information about the earth-centered universe that ever
became common knowledge among nonastronomers. The further
achievements of ancient astronomy, to which we shall turn in a mo-
ment, were too mathematical for most laymen to understand. As the
next two chapters indicate more fully, the most influential cosmologies
developed in antiquity and the Middle Ages did not follow ancient
astronomy very far beyond this point. Astronomy now becomes
esoteric; its further development tlucs not provide man with a home.
In addition, the structural diagram in Figure 16 is, despite its
crudity, an immensely powerful tool in astronomical research. For
many purposes it proved both economical and fruitful. For example,
during the fourth century B.C., the concepts embodied in the diagram
provided a complete qualitative explanation of both the phases of the
moon and lunar eclipses; during the fourth and third centuries these
same concepts led to a series of relatively accurate determinations of
the circumference of the earth; and during the second century B.C.,
they provided the basis for a brilliantly conceived estimate of the sizes
and distances of the sun and moon. These explanations and measure-
ments, particularly the last, typify the immense ingenuity and power
of the ancient astronomical tradition. They are, however, here relegated
to the Technical Appendix (Sections 3 and 4), because they were not
affected by the change in astronomical theory during the Copernican
Revolution. Nevertheless, they are relevant to the Revolution. The
ability of the developed two-sphere universe to explain and ultimately
to predict prominent celestial phenomena like eclipses, as well as its
ability to specify some linear dimensions of the celestial regions, im-
measurably increased the hold of the two-sphere conceptual scheme
upon the minds of both astronomers and laymen.
These achievements do not, however, touch upon the fundamental
problem posed by the continuous irregularity of planetary motion, and
this problem provides the pivot upon which the Copernican Revolution
ultimately turns. Like so many other problems of ancient astronomy, it
seems first to have emerged during the fourth century B.C., when the
two-sphere universe, by explaining diurnal motion, enabled Greek
astronomers to isolate the residual planetary irregularities for the first
time. During the following five centuries successive attempts to explain
these irregularities produced several planetary theories of unprece-
dented accuracy and power. But these attempts also constitute the
most abstruse and mathematical part of ancient astronomy, and they
are therefore usually omitted from books like this. Though a simpli-
fied precis of ancient planetary theory seems a minimal requisite for
an understanding of the Copernican Revolution, a few readers may
prefer to skim the next three sections (particularly the first, in which
the technical presentation is especially compact), picking up the
narrative again with the discussion of scientific belief that closes this
chapter.
The Theory of Homocentric Spheres
The philosopher Plato, whose searching questions dominated
so much of subsequent Greek thought, seems to have been the first to
enunciate the problem of the planets, too. Early in the fourth century
B.C. Plato is said to have asked: "What are the uniform and ordered
movements by the assumption of which the apparent movements of
the planets can be accounted for?"
2
and the first answer to this ques-
tion was provided by his onetime pupil Eudoxus ( c.408 - c.355 B.C.).
In Eudoxus' planetary system each planet was placed upon the inner
sphere of a group of two or more interconnected, concentric spheres
whose simultaneous rotation about different axes produced the ob-
served motion of the planet. Figure 17a shows a cross section of two
such interlocked spheres whose common center is the earth and whose
points of contact are the ends of the slanted axis of the inner sphere,
which serve as pivots. The outer sphere is the sphere of the stars, or
at least it has the same motion as that sphere; its axis passes through
the north and south celestial poles, and it rotates westward about this
axis once in 23 hours 56 minutes. The inner sphere's axis makes contact
with the outer sphere at two diametrically opposite points

away
from the north and south celestial poles; therefore the equator of the
inner sphere, viewed from the earth, always falls on the ecliptic of the
sphere of the stars, regardless of the rotation of the two spheres.
If the sun is now placed at a point on the equator of the inner
sphere and that sphere is turned slowly eastward about its axis once in
a year while the outer sphere turns about its axis once a day, the sum
of the two motions will reproduce the observed motion of the sun. The
outer sphere produces the observed westward diurnal motion of rising
and setting; the inner sphere produces the slower annual eastward
motion of the sun, around the ecliptic. Similarly, if one eastward rota-
tion of the inner sphere occurs every days and if the moon is
placed on the equator of this sphere, then the motion of this inner
sphere must produce the average motion of the moon around the
ecliptic. The north and south deviations of the moon from the ecliptic
and some of the irregularities in the time required by the moon for
successive journeys can be approximated by adding one more very
slowly moving sphere to the system. Eudoxus also used (though un-
necessarily) a third sphere to describe the motion of the sun, so that
six spheres were required to treat the moon and sun together.
The spheres shown in Figure 17 a were known as homocentric
spheres, because they have a common center, the earth. Two or three
such spheres can approximately represent the total motion of the sun
and of the moon, but they cannot account for the retrograde motions
of the planets, and Eudoxus' greatest genius as a geometer was dis-
played in the modification of the system that he introduced in treating
the behavior of the remaining five planets. For each of these
he used a total of four spheres, sketched in cross section in Figure 17b.
The two outer spheres move just like the spheres of Figure 17a:
the outer sphere has the diurnal motion of the sphere of the stars and
the second sphere (counting from the outside) turns eastward once
in the average time required by the planet to complete a journey
around the ecliptic. (Jupiter's second sphere, for example, turns once
in 12 years.) The third sphere is in contact with the second sphere at
two diametrically opposite points on the ecliptic (the equator of the
second sphere), and the axis of the fourth or innermost sphere is
fastened to the third sphere at an angle that depends upon the
characteristics of the motion to be described. The planet itself (Jupiter
in the example above) is located on the equator of the fourth sphere.
Suppose now that the two outer spheres are held stationary and
that the two innermost spheres rotate in opposite directions, each
completing one axial rotation in the interval that separates two sue-
23r N
s
(a )
N
s
(b)
Figure 17. Homocentric spheres. In the two-sphere system (a) the outermost
sphere produces the diurnal rotation, and the inner sphere moves the planet (sun
or moon) steadily eastward around the ecliptic. In the four-sphere system (b)
the planet Plies out of the plane of the paper, ahnost on a line from the earth E to
the reader's eye. The two innermost spheres then produce the looped motion shown
in Figure 18, and the two outer spheres produce both the diurnal motion and the
average eastward planetary drift.
cessive retrogressions of the planet ( 399 days for Jupiter). An observer
watching the motion of the planet against the temporarily stationary
second sphere will see it move slowly in a figure eight, both of whose
loops are bisected by the ecliptic. This is the motion sketched in Figure
18; the planet passes slowly around the loops from positions 1 to 2,
2 to 3, 3 to 4, . . . , spending equal times between each numbered
point and the next, and returning to its starting point after the interval
between retrogressions. During its motion from 1 to 3 to 5 the planet
moves eastward along the ecliptic; during the other half of the time,
while the planet moves from 5 to 7 and back to 1, it moves westward.
Now allow the second sphere to rotate eastward, carrying the two
rotating inner spheres with it, and suppose that the total motion of the
planet is observed against the background of stars on the first sphere,
again held temporarily stationary. At all times the planet is moved
eastward by the motion of the second sphere, and half of the time
(while it moves from 1 to 5 in Figure 18) the planet receives an addi-
EASTWARO
6 2
__

4 8
Figure 18. The looped motion generated by the two innermost homocentric
spheres. In the full four-sphere system this looped motion is combined with the
steady eastward motion of the second sphere, a motion that by itself would carry
the planet around the ecliptic at a uniform rate. When the looped motion is
added, the total motion of the planet varies in rate and is no longer confined to
the ecliptic. While the planet moves from 1 to 5 on the loop, its total motion is
rapid than the average eastward motion generated by the second sphere;
while the planet moves back from 5 to 1 on the loop, its eastward motion is slower
than that produced by the second sphere, and when it gets near 3, it may actually
move westward, in retrograde.
tional eastward motion from the two inner spheres so that its net
motion is eastward and even more rapid than that of the second sphere
alone. But during the other half of the time (as the planet moves from
5 to 1 in Figure 18) the eastward motion of the second sphere is op-
posed by a westward motion due to the two inner spheres, and, when
this westward motion is most rapid (near 7 in Figure 18), the net
motion of the planet against the sphere of the stars may actually be to
the west, in the retrograde direction. This is just the characteristic of
observed planetary motions that Eudoxus was striving to reproduce
in his model.
A system of four interlocked homocentric spheres approximates the
retrograde motion of Jupiter, and a second set of four spheres can ac-
count for the motion of Saturn. For each of the three remaining
planets, five spheres are needed (an extension provided by Eudoxus'
successor Callippus around 330 B.C.), and analysis of the resulting
motions becomes correspondingly more complex. Fortunately, we need
not pursue these complex combinations of rotating spheres further, be-
cause all homocentric systems have one severe drawback which in
antiquity led to their early demise. Since Eudoxus' theory places each
planet on a sphere concentric with the earth, the distance between a
planet and the earth cannot vary. But planets appear brighter, and
therefore seem closer to the earth, when they retrogress. During an-
tiquity the homocentric system was frequently criticized for its failure
to explain this variation in planetary brilliance, and the system was
abandoned by most astronomers almost as soon as a more adequate
explanation of the appearances was proposed.
But though short-lived as a significant astronomical device, homo-
centric spheres play a major role in the development of astronomical
and cosmological thought. By a historical accident the century during
which they seemed to provide the most promising explanation of
planetary motion embraced most of the lifetime of the Greek philoso-
pher Aristotle, who incorporated them in the most comprehensive, de-
tailed, and influential cosmology developed in the ancient world. No
comparably complete cosmology ever incorporated the mathematical
system of epicycles and deferents which, in the centuries after Aris-
totle's death, was employed to explain planetary motion. The concep-
tion that planets are set in rotating spherical shells concentric with the
earth remained an accepted portion of cosmological thought until early
in the seventeenth century. Even the writings of Copernicus show
important vestiges of this conception. In the title of Copernicus' great
work, De Revolutionibus Orbium Caelestium, the "orbs" or spheres are
not the planets themselves but rather the concentric spherical shells in
which the planets and the stars are set.
Epicycles and Deferents
The origin of the device that replaced homocentric spheres in
explaining the details of planetary motion is unknown, but its features
were early investigated and developed by two Greek astronomers and
mathematicians, Apollonius and Hipparchus, whose work spans the
period from the middle of the third century to the end of the second
century B.C. In its simplest form (Figure 19a) the new mathematical
mechanism for the planets consists of a small circle, the epicycle, which
rotates uniformly about a point on the circumference of a second rotat-
ing circle, the deferent. The planet, P, is located on the epicycle, and
the center of the deferent coincides with the center of the earth.
The epicycle-deferent system is intended to explain only motion
with respect to the sphere of the stars. Both the epicycle and the defer-
ent in Figure 19a are drawn on the plane of the ecliptic, so that the
rotation of the stellar sphere carries the entire diagram (except the
central earth) through one rotation per day and thus produces the
diurnal motion of the planet. If the epicycle and deferent of the figure
were stationary and did not have an additional motion of their own, the
planet would be fixed in the plane of the ecliptic and would therefore
have the motion of a zodiacal star, a westward circle executed once in
every 23 hours 56 minutes. From now on, whenever reference is made
to the motion of the deferent or the epicycle, it is the additional motion
of these circles in the plane of the ecliptic that is meant. The diurnal
rotation of the sphere and of the plane of the ecliptic will be taken for
granted.
Suppose, for example, that the deferent rotates eastward once in a
year and that the sun is placed on the deferent at the position now
occupied by the center of the epicycle, the epicycle itself being re-
moved. Then the rotation of the deferent carries the sun through its
annual journey around the ecliptic, and the sun's motion has been
analyzed, at least approximately, in terms of the motion of a single
deferent in the plane of the ecliptic. This is the technique taken for
granted in the explanation of average planetary motions in Figure 16.
Now imagine that the sun is removed and the epicycle is returned
to its position on the deferent. If the epicycle rotates just three times
around its moving center while the deferent rotates once and if the two
circles rotate in the same direction, then the total motion of the planet
withip the sphere of the stars produced by the combined motions of
the epicycle and the deferent is just the looped curve shown in Figure
19b. When the rotation of the epicycle carries the planet outside of the
deferent, the motions of both the epicycle and the deferent combine to
move the planet to the east. But when the motion of the epicycle places
the planet well inside the deferent, the epicycle carries the planet west-
ward, in opposition to the motion of the deferent. Therefore, when the
planet is closest to the earth, the two motions may combine to produce
a net westward or retrograde motion. In Figure 19b the planet retro-
gresses whenever it is on the interior part of one of the small loops;
everywhere else the planet moves normally toward the east, but at a
variable rate.
r
Figure 19c shows the motion of the planet through one of the
loops as viewed against the sphere of the stars by an observer on
earth. Since both observer and loop are in the same plane, that of the
ecliptic, the observer cannot see the open loop itself. What he sees is
merely the position of the planet against the background provided by
the ecliptic. Thus as the planet moves from position 1 to 2 in Figures
19b and 19c, the observer sees it move along the ecliptic toward the
east. As the planet approaches position 2, it appears to move more
slowly, stopping momentarily at 2 and then moving westward along the
ecliptic as it travels from 2 towards 3. Finally the westward journey of
)J
D
)}
D
(a)
(b)
EASTWARD
4 3
2
(c)
Figure 19. The basic epicycle-deferent system. A typical deferent and epicycle
are shown in (a); the looped motion that they generate in the plane of the ecliptic
is illustrated in (b); the third diagram (c) shows a portion (1-2-3-4) of the
motion in (b) as it is seen by an observer on the central earth, E.
the planet on the ecliptic halts, and the planet again moves eastward,
leaving position 3 on the loop and moving toward 4.
A system of one epicycle and one deferent therefore carries a planet
around the ecliptic in an interval that, on the average, exactly equals
the time required for one revolution of the deferent. The eastward
motion is, however, interrupted and the planet temporarily moves west-
ward at regular intervals equal to the time required for one revolution
of the epicycle. The rates of revolution of the epicycle and deferent
may be adjusted to fit the observations for any planet, yielding just
that intermittent eastward motion among the stars which planets are
observed to have. Furthermore, the epicycle-deferent system repro-
duces one other important qualitative feature of the appearances: a
planet can retrogress only when its motion brings it nearest to the earth
and that is the position in which the planet should and does appear
brightest. Its great simplicity plus this novel explanation of varying
planetary brilliance are the primary causes for the new system's victory
over the older system of homocentric spheres.
The epicycle-deferent system described by Figure 19 incorporates
one special simplification that is not characteristic of the motion of any
planet. The epicycle is made to revolve exactly three times for each
revolution of the deferent. Therefore, whenever the deferent completes
one revolution, the epicycle returns the planet to the same position it
occupied at the beginning of the revolution; the retrograde loops
J.lways occur at the same places; and the planet always requires the
same amount of time to complete its trip around the ecliptic. When
designed to fit the observations of real planets, however, epicycle-
deferent systems never perform in quite this manner. For example,
is observed to require an average of 1 year to complete a
JOUrney around the ecliptic, and it retrogresses once every 116 days.
Therefore Mercury's epicycle must revolve just over three times while
the deferent turns once; the epicycle completes three revolutions in 348
days which is less than the year required for a rotation of the deferent.
Figure 20a shows the path of a planet carried through one trip
around. the ecliptic by an epicycle that turns slightly more than three
times for each rotation of its deferent. The planet starts in the middle
of a retrograde loop and completes its third full loop before the defer-
completes its first full rotation; the planet therefore averages
shghtly more than three retrograde loops in each trip around the
If the motion of Figure 20a were continued through a second
tnp, the new set of retrograde loops would fall slightly to the west of
those generated during the first trip. Retrograde motion would not
occur at the same position in the zodiac on successive trips, and this is
characteristic of the observed progress of planets along the ecliptic.
Figure 20b indicates a second characteristic of the motion gen-
erated by an epicycle that does not revolve an integral number of times
in each revolution of the deferent. The planet at P in the figure is at
the position closest to the earth, the position from which the journey of
Figure 20a began. After one revolution of the deferent, the epicycle
will have turned slightly more than three times, and the planet will
have arrived at position P', so that it now appears to the west of its
p'
(a) (b) (c)
Figure 20. Motion generated by an epicycle and deferent when the epicycle
turns slightly more than three times for each revolution of the deferent. The
planet's path during a single complete journey through the stars is shown in (a).
This journey requires more than one revolution of the deferent, as indicated by
(b), which shows the planet's position at the beginning ( P) and the end ( P')
of the deferent's first full revolution. Diagram (c) shows the planet's position at
the beginning and end of a later revolution of the deferent, one that carries the
planet more than once around the ecliptic.
starting point. The deferent must turn eastward through more than a
single revolution to carry the planet fully around the ecliptic; the cor-
responding trip through the constellations therefore requires more time
than the average. Others, however, require less. After several more
revolutions of the deferent, each ending with the planet still farther
from the earth, the planet might start a new journey from the new
position P in Figure 20c. One more revolution of the deferent would
carry the planet to P', a point to the east of P Since this revolution of
the deferent carries the planet through more than one trip around the
ecliptic, this journey is a particularly rapid one. Figures 20b and 20c
represent very nearly the extreme values of the time required for a
journey around the ecliptic; intermediate trips consume intermediate
amounts of time; on the average, a journey around the ecliptic re-
quires the same time as a rotation of the deferent. But the epicycle-
deferent system allows for deviations from one trip to the next. Once
again it provides an economical explanation of an observed irregularity
of the planetary motions.
To describe the motions of all the planets a separate epicycle-
deferent system must be designed for each. The motion of the sun and
moon can be treated approximately by a deferent alone, for these
planets do not retrogress. The sun's deferent turns once a year; the
moon's revolves once in 7 3 ~ days. The epicycle-deferent system for
Mercury is much like the one discussed above; the deferent turns once
a year and the epicycle once in 116 days. By utilizing the observations
recorded early in this chapter, we could design similar systems for
other planets. Most of these would yield looped planetary paths like
the one shown in Figure 20a. If the epicycle is larger relative to the
deferent, the size of the loops is increased. If the epicycle turns
more quickly relative to the speed of the deferent, then there are more
loops included in one journey around the ecliptic. There are approxi-
mately eleven loops in each trip made by Jupiter, and approximately
twenty-eight in each by Saturn. In short, by appropriate variations in
the relative sizes and the speeds of the epicycle and the deferent, this
system of compounded circular motions can be adjusted to fit ap-
proximately an immense variety of planetary motions. A properly de-
signed combination of circles will even give a good qualitative account
of the immense irregularities in the motion of an atypical planet like
Venus (Figure 21).
Ptolemaic Astronomy
The discussion of the previous section illustrates the power
and versatility of the epicycle-deferent system as a method for ordering
and predicting the motions of the planets. But this is only the first step.
Once the system was available to account for the most striking irregu-
larities of planetary motion - retrogression and the irregular amounts
of time consumed in successive journeys around the ecliptic - it be-
came clear that there were still other, though very much smaller, ir-
regularities to be considered.
Just as the two-sphere model provided a precise mechanism for
the diurnal motions, thus permitting detailed study of the principal
planetary irregularities, so the epicycle-deferent system, by providing
I
L
an account of the main planetary motions, permitted the observational
isolation of smaller irregularities. This is the first example of the con-
p
(a) (b)
Figure 21. (a) A one-epicycle one-deferent system for Venus and (b) the
motion that it generates in the plane of the ecliptic.
In (a), notice the following characteristics of the design: The deferent rotates
once in a year, so that if the center of the epicycle is once aligned with the
earth, E, and the center of the sun, S, it will stay in alignment forever, and
Venus will never appear very far from the sun. The angles SEP' and SEP'" are the
largest angles that can appear between the sun and Venus, and the condition that
these angles of maximum elongation be 45 completely determines the relative
sizes of the epicycle and deferent. The epicycle rotates once in 584 days, so that
if Venus starts at P, close to the sun, it will arrive at P' (maximum elongation as
an evening star) after 219 days ( 3/8 revolution); at P" after 292 days ( l/2 revo-
lution); and at P"' (maximum elongation as a morning star) after 365 days ( 5/8
revolution).
The second diagram shows the path along which Venus is carried by the
moving circles sketched in (a). Here P is the starting point, as in the first dia-
gram; P' is Venus's position at maximum eastward elongation (219 days); P" is
the planet's location midway through a retrograde loop ( 292 days); and P'" is its
position at maximum westward elongation ( 365 days). Venus' s first journey around
the ecliptic ends at p after 406 days (note the great length) and includes one
retrogression and two maximum elongations. Its next trip ( p to p' to p") requires
only 295 days and includes none of these characteristic phenomena. At p' Venus
is again closest to the sun, a position reached after one complete revolution of the
epicycle ( 584 days). This is, at least qualitatively, the way Venus does behave!
66
cept's fruitfulness. When the motion predicted by a one-epicycle one-
deferent system is compared with the observed motion of an individ-
ual planet, it turns out that the planet is not always seen at quite the
position on the ecliptic where the geometry of the model says it
should be. Venus does not, if observed precisely, always attain its
maximum deviation of 45 from the sun; the intervals between suc-
cessive retrogressions of a single planet are not always quite the same;
and none of the planets, except the sun, stays on the ecliptic through-
out its motion. The one-epicycle one-deferent system was not, there-
fore, the final answer to the problem of the planets. It was only a very
promising start and one that lent itself to both immediate and long-
continued development. During the seventeen centuries that separate
Hipparchus from Copernicus all the most creative practitioners of
technical astronomy endeavored to invent some new set of minor
geometric modifications that would make the basic one-epicycle one-
deferent technique precisely fit the observed motion of the planets.
In antiquity the greatest of these attempts was made around A.D.
150 by the astronomer Ptolemy. Because his work displaced that of his
predecessors and because all of his successors, including Copernicus,
modelled their work upon his, the whole series of attempts for which
Ptolemy provides the archetype is now usually known as Ptolemaic
astronomy. That phrase, "Ptolemaic astronomy," refers to a traditional
approach to the problem of the planets rather than to any one of the
particular putative solutions suggested by Ptolemy himself, his prede-
cessors, or his successors. Each of the particular individual solutions,
and especially Ptolemy's, has an intense interest, both technical and
historical, but both the particular solutions and their historical inter-
relationships are too complex to be considered here. Instead of at-
tempting a general developmental account of the various Ptolemaic
planetary systems, we shall therefore simply survey the main sorts of
modification to which the basic epicycle-deferent system was subjected
at various times between its first invention three centuries before Christ
and its rejection by the followers of Copernicus.
Though their most important application is to the complex motions
of the planets, the principal ancient and medieval modifications of the
epicycle-deferent system are most simply described in their occasional
applications to the apparently simpler motions of the sun and moon.
The sun, for example, does not retrogress, so its motion does not require
67
a major epicycle of the sort described in the last section. But fixing the
sun on a deferent th<1t rotates uniformly about the earth as center does
not give a quantitatively precise account of the solar motion, for, as
shown by a reexamination of the dates of the solstices and equinoxes
listed in Chapter 1, the sun takes almost 6 days longer to move from the
vernal equinox to the autumnal equinox ( 180 along the ecliptic) than
it does to move back from the autumnal equinox to the vernal equinox
(again 180). The sun's motion along the ecliptic is slightly more rapid
during the winter than summer, and such a motion cannot be produced
by a fixed point on a uniformly rotating earth-centered circle. Examine
Figure 22a, in which the earth is shown at the center of a uniformly
rotating deferent circle and in which the positions of the vernal and
autumnal equinoxes on the sphere of the stars are indicated by the
dashes VE and AE. Uniform rotation of the deferent will carry the sun,
S, from VEto AE in the same time that it takes to carry it back from AE
to VE, and this corresponds only approximately with observation.
Suppose, however, that the sun is removed from the deferent and
placed on a small epicycle that rotates once westward while the defer-
ent rotates once eastward. Eight positions of the sun in such a system
are shown in Figure 22b. It is clear that the summer half of the defer-
ent's rotation does not carry the sun the entire distance from VE to AE
and that the winter half of the rotation carries the sun farther than the
distance from AE to VE. So the effect of the epicycle is to increase the
time spent by the sun in the 180 between VE and AE and to decrease
ss
I
I
WS
(a)
AE-
ss
I
I
ws
(b)
ss
I
I
ws
(c)
-V
Figure 22. Functions of a minor epicycle. In (a) the sun, moved by a single
earth-centered deferent, requires the same time to move from AE to VE that it
needs to move back. In (b) the joint motion of deferent and minor epicycle
carries the sun along the broken curve, so that more time is required for the trip
from VE to AE than for the return. Diagram (c) shows the curve generated when
the minor epicycle revolves at twice the rate used in constructing (b).
the time spent in the other half of the ecliptic between AE and VE.
If the radius of the little epicycle is 0.03 the radius of the deferent, the
difference in the time spent by the sun in the winter and the summer
halves of the ecliptic will be the required 6 days.
The epicycle employed in the preceding discussion to correct a
minor irregularity of the sun's motion is relatively small, and it pro-
duces no retrograde loops. Its function is therefore quite different from
that of the larger epicycles considered in the last section, and, though
Ptolemaic astronomers never did so, it will prove convenient to keep
these two functions apart. Henceforth we s!1all use the term "major
epicycle" for the large epicycles used to produce the qualitative ap-
pearance of retrograde motion and the term "minor epicycle" for the
additional circles used to eliminate small quantitative discrepancies
between theory and observation. All versions of the Ptolemaic system,
both before and after Ptolemy, had just five major epicycles, and it is
these with which Copernicus' reform did away. In contrast, the number
of minor epicycles and similar devices needed to account for small
quantitative discrepancies depended only on the precision of the avail-
able observations and on the accuracy of the predictions demanded
from the system. The number of minor epicycles employed in the
various versions of Ptolemaic astronomy therefore varied greatly from
one version to the next. Systems employing half a dozen to a dozen
minor epicycles were not uncommon in antiquity and the Renaissance,
for by an appropriate choice of the size and speed of minor epicycles
almost any sort of small irregularity could be explained away. That is
why, as we shall see, Copernicus' astronomical system was so nearly as
complex as Ptolemy's. Though his reform eliminated major epicycles,
Copernicus was as dependent upon minor epicycles as his predecessors.
One sort of irregularity was treated with the aid of a minor epicycle
in Figure 22b; another sort is shown in Figure 22c. There the minor
epicycle rotates twice westward while the deferent moves once east-
ward. Combining the two rotations results in a total motion (broken
line in the figure) along a flattened circle. A planet moving on this
curve moves faster and spends less time in the vicinity of the summer
and winter solstices than it does near the two equinoxes. If the epicycle
had turned slightly less than twice while the deferent rotated once,
then the positions on the ecliptic at which the planet's apparent speed
was greatest would have changed on successive trips around the
ecliptic. If it had appeared fastest near the summer solstice on one trip,
it would have passed the summer solstice before gaining its greatest
speed on the next trip. Other variations of this sort can be produced
at will.
Uses of the minor epicycle are not limited to the nonretrogressing
planets, the sun and moon. A minor epicycle can be placed upon a
major epicycle and used in the prediction of the more elaborate plane-
tary motions; in fact, planetary motions provided the minor epicycle's
main astronomical application. One such application, an epicycle on
an epicycle on a deferent, is shown in Figure 23a. If the major
epicycle turns eight times eastward and the minor epicycle once west-
ward during one rotation of the deferent, then the path within the
sphere of the stars described by the planet is that shown in Figure 23b.
It has eight normal retrograde loops, but these are somewhat more
densely clustered in the half of the ecliptic between the vernal equinox
and the autumnal equinox than in the half between the autumnal and
vernal equinox. If the rate of rotation of the minor epicycle is now
A E- -VE
(a) {b)
Figure 23. An epicycle on an epicycle on a deferent (a) and a typical path
through space (b) generated by this system of compounded circles. For simplicity,
the path has been shown rejoining itself smoothly, a situation that does not occur
in the motion of real planets.
doubled, the path described by the planet is flattened as in Figure 22c.
These diagrams begin to suggest the complexities of the paths that
minor epicycles can produce.
Nor is a minor epicycle the only device available for correcting
minor discrepancies between one-epicycle one-deferent systems and
70
the observed behavior of the planets. A glance at Figure 22b indicates
that the effect there produced by a minor epicycle that rotates west-
ward once as the deferent turns through a single eastward rotation can
equally well be achieved by a single deferent whose center is displa.ced
from the center of the earth. Such a displaced circle, known to ancwnt
astronomers as an eccentric, is shown in Figure 24a. If the distance
AE-
ss
I
I
ws
(a)
-vE
(b)
(c)
Figure 24. An eccentric (a), an eccentric on a deferent (b), and an eccentric
on an eccentric ( c) .
between the earth, E, and the center of the eccentric, 0, is about 0.03
the radius of the eccentric, this displaced circle will account for the 6
extra days that the sun spends between the vernal and autumnal
equinoxes. That is the particular device that Ptolemy used in own
account of the sun. Other values of the distance EO, employed m con-
junction with one or more epicycles, will account for other minor
planetary irregularities. Additional effects may be obtained by
the of the eccentric on a small deferent (Figure 24b), or on a
second smaller eccentric (Figure 24c). These two devices can be shown
to be geometrically fully equivalent to a minor epicycle on a
and a minor epicycle on an eccentric, respectively, and most Ptolema1c
astronomers used these small central circles in preference to minor
epicycles. In all cases one or more epicycles may be added, and any or
all of these circles mav be tilted into different planes to account for
the north and south of the planets from the ecliptic.
One more device, the equant, was developed in antiquity to aid in
the reconciliation of the theory of epicycles with the results of
observation. This device is of particular importance because Coperni-
cus' aesthetic objections to it (Chapter 5) provided one essential
71
motive for his rejection of the Ptolemaic system and his search for a
radically new method of computation. Copernicus used epicycles and
eccentrics like those employed by his ancient predecessors, but he did
not use equants, and he felt that their absence from his system was
one of its greatest advantages and one of the most forceful arguments
for its truth.
One form of equant, designed, for simplicity of illustration, to
account for the previously discussed irregularity in the sun's motion,
is shown in Figure 25. The center of the sun's deferent coincides as
before with the center of the earth, E, but the deferent's rate of rotation
is now required to be uniform not with respect to its geometric center
E, but with respect to an equant point, A, displaced in this case toward
the summer solstice. That is, the angle a subtended at the equant point
A by the sun and the summer solstice is required to change at a
constant rate. If the angle increases by 30 in one month, then it must
increase by 30 in every month of the same length. In the figure the
sun is shown at the vernal equinox, VE. To reach the autumnal equinox,
AE, it must complete a semicircle, which will change the angle a by
more than 180, and to return from AE to VE it must complete a second
semicircle, which will change a by less than 180. Since every 180
increase in a requires the same amount of time, the sun must take longer
to go from VE to AE than it requires for the return journey from AE
to VE. Therefore, viewed from the equant point A, the sun travels at
an irregular rate, fastest near the winter solstice and slowest near the
summer solstice.
That is the defining feature of the equant. The rate of rotation of a
deferent or some other planetary circle is required to be uniform, not
with respect to its own geometric center, but with respect to an equant
point displaced from that center. Observed from the geometric center
of its deferent, the planet seems to move at an irregular rate or to
wobble. Because of the wobble, Copernicus felt that the equant was
not a legitimate device for application to astronomy. For him the ap-
parent irregularities of the rotation were violations of the uniform
circular symmetry that made the system of epicycles, deferents, and
eccentrics so plausible and attractive. Since the equant was normally
applied to eccentrics and since similar devices occasionally made the
epicycle wobble as well, it is noi: hard to imagine how Copernicus
might have considered this aspect of Ptolemaic astronomy monstrous.
72
The mathematical devices sketched in the preceding pages were
not all developed at a stroke or by Ptolemy. Apollonius, in the thi:d
century B.C., knew both major epicycles (Figure 19a) and eccentrics
with moving centers (Figure 24b ) . During the following century
Hipparchus added minor epicycles and a more general of
eccentrics to the arsenal of astronomical weapons. In additiOn he
combined these devices to provide the first quantitatively adequate
account of the irregularities in the motions of the sun and moon.
Ptolemy himself added the equant, and during the thirteen centuries
between his time and that of Copernicus, first Moslem and then Euro-
pean astronomers employed still other combinations of circles -
cluding the epicycle on an epicycle (Figure 23a) and the eccentric
on an eccentric ( Figure 24c) - to account for additional planetary
irregularities.
AE-
ss
I
I
ws
Figure 25. The equant. The sun, S, moves on the earth-centered but at _an
irregular rate determined by the condition that the angle a vary uniformly w1th
time.
But Ptolemy's contribution is the outstanding one, and this entire
technique of resolving the problem of the planets is appropriately
known by his name, because it was Ptolemy who first put together a
particular set of compounded circles to account, not merely for the
motions of the sun and moon, but for the observed quantitative regu-
larities and irregularities in the apparent motions of all the seven
planets. His Almagest, the book that epitomizes the greatest achieve-
ments of ancient astronomy, was the first systematic mathematical
treatise to give a complete, detailed, and quantitative account of all the
celestial motions. Its results were so good and its methods so powerful
that after Ptolemy's death the problem of the planets took a new form.
To increase the accuracy or simplicity of planetary theory Ptolemy's
successors added epicycles to epicycles and eccentrics to eccentrics,
exploiting all the immense versatility of the fundamental Ptolemaic
technique. But they seldom or never sought fundamental modifications
of that technique. The problem of the planets had become simply a
problem of design, a problem to be attacked principally by the re-
arrangement of existing elements. What particular combination of de-
ferents, eccentrics, equants, and epicycles would account for the
planetary motions with the greatest simplicity and precision?
We cannot pursue further the individual quantitative solutions of
this problem proposed by Hipparchus, Ptolemy, and their successors.
The complete quantitative systems are mathematically too complex.
:\1uch of Ptolemy's Almagest consists of quantitative mathematical
tables, diagrams, formulas, and proofs, of long illustrative computations,
and of lists of numerous observations. Yet the problems that set Coper-
nicus to searching for a new approach to the problem of the planets and
the advantages that he claimed to derive from his new system all lie
within this abstruse body of quantitative theory. Copernicus did not
attack the two-sphere universe, though his work ultimately overthrew it,
and he did not abandon the use of epicycles and eccentrics, though these
too were abandoned by his successors. What Copernicus did attack
and what started the revolution in astronomy was certain of the ap-
parently trivial mathematical details, like equants, embodied in the
complex mathematical systems of Ptolemy and his successors. The
initial battle between Copernicus and the astronomers of antiquity was
fought over technical minutiae like those sketched in this section.
The Anatomy of Scientific Belief
For its subtlety, flexibility, complexity, and power the
epicycle-deferent technique sketched in the two preceding sections
has no parallel in the history of science until quite recent times. In its
most developed form the system of compounded circles was an as-
tounding achievement. But it never quite worked. Apollonius' initial
conception solved the primary planetary irregularities - retrograde
motion, variation of brightness, alteration in the time required for
successive journeys around the ecliptic - and it did so simply and at
a stroke. But it also disclosed some residual secondary irregularities.
Some of these were explained away by the more elaborate system of
compounded circles developed by Hipparchus, but still the theory did
not quite match the results of observation. Even Ptolemy's complex
combination of deferents, eccentrics, epicycles, and equants did not
precisely reconcile theory and observation, and Ptolemy's was neither
the most complex nor the last version of the system. Ptolemy's many
successors, first in the Moslem world and then in medieval Europe,
took up the problem where he had left it and sought in vain for the
solution that had evaded him. Copernicus was still grappling with the
same problem.
There are many variations of the Ptolemaic svstem besides the
one that Ptolemy himself embodied in the Almagest: and some of them
achieved considerable accuracy in predicting planetary positions. But
the accuracy was invariably achieved at the price of complexity -
the addition of new minor epicycles or equivalent devices - and in-
creased complexity gave only a better approximation to planetary mo-
tion, not finality. No version of the system ever quite withstood the
test of additional refined observations, and this failure, combined with
the total disappearance of the conceptual economy that had made
cruder versions of the two-sphere universe so convincing, ultimately
led to the Copernican Revolution.
But the Revolution was an incredibly long time coming. For almost
1800 years, from the time of Apollonius and Hipparchus until the birth
of Cqpernicus, the conception of compounded circular orbits within an
earth-centered universe dominated every technically developed attack
upon the problem of the planets, and there were a great many such
attacks before Copernicus'. Despite its slight but recognized inaccuracy
and its striking lack of economy (contrast the earlier two-sphere uni-
verse described in Chapter 1 ), the developed Ptolemaic system had
an immense life span, and the longevity of this magnificent but clearly
imperfect system poses a pair of closely related puzzles: How did the
two-sphere universe and the associated epicycle-deferent planetary
theory gain so tight a grip upon the imagination of the astronomers?
And, once gained, how was the psychological grip of this traditional
approach to a traditional problem released? Or to put the same ques-
tions more directly: Why was the Copernican Revolution so delayed?
And how did it come to pass at all?
These are questions about the history of a particular set of ideas,
and as history they will be considered at some length below. But they
are also more generally concerned with the nature and structure of
conceptual schemes and with the process by which one conceptual
scheme replaces another. It is therefore illuminating to approach them
first by returning briefly to the abstract logical and psychological cate-
gories introduced in the penultimate section of the first chapter. \Ve
there examined the functions of a conceptual scheme: we now ask
how a smoothly functioning scheme, like the early two-sphere uni-
verse, can be replaced. Examine the logic of the phenomenon first.
Logically there are always many alternative conceptual schemes
capable of bringing order to any prescribed list of observations, but
these alternatives differ in their predictions about phenomena not in-
cluded on the list. Both the Copernican and the Newtonian systems
will account for naked-eye stellar and solar observations just as ade-
quately as will the two-sphere system; Heraclides' system will do the
same and so will the system developed by Copernicus' successor, Tycho
Brahe; in theory there are an infinite number of other alternatives
besides. But these alternatives agree principally about observations
that have already been made. They do not give identical accounts of
all possible observations. The Copernican system, for example, differs
from the two-sphere universe in predicting an apparent annual motion
of the stars, in demanding a much larger diameter for the stellar sphere,
and in suggesting (though not to Copernicus) a new sort of solution
for the problem of the planets. It is because of differences like these
(and there are many others besides) that a scientist must believe in his
system before he will trust it as a guide to fruitful investigations of the
unknown. Only one of the different alternatives can conceivably repre-
sent reality, and the scientist exploring new territory must feel confi-
dent that he has chosen that one or the closest of the available approxi-
mations to it. But the scientist pays a price for this commitment to a
particular alternative: he may make mistakes. A single observation in-
compatible with his theory demonstrates that he has been employing
the wrong theory all along. His conceptual scheme must then be aban-
doned and replaced.
That, in outline, is the logical structure of a scientific revolution. A
76
conceptual scheme, believed because it is economical, fruitful, and
cosmologically satisfying, finally leads to results that are incompatible
with observation; belief must then bii surrendered and a new theory
adopted; after this the process starts again. It is a useful outline, be-
cause the incompatibility of theory and observation is the ultimate
source of every revolution in the sciences. But historically the process
of revolution is never, and could not possibly be, so simple as the logical
outline indicates. As we have already begun to discover, observation is
never absolutely incompatible with a conceptual scheme.
To Copernicus the behavior of the planets was incompatible with
the two-sphere universe; he felt that in adding more and more circles
his predecessors had simply been patching and stretching the Ptole-
maic system to force its conformity with observations; and he believed
that the very necessity for such patching and stretching was clear evi-
dence that a radically new approach was imperatively required. But
Copernicus' predecessors, to whom exactly the same sorts of instru-
ments and observations were available, had evaluated the same situa-
tion quite differently. What to Copernicus was stretching and patch-
ing was to them a natural process of adaptation and extension, much
like the process which at an earlier date had been employed to incor-
porate the motion of the sun into a two-sphere universe designed in-
itially for the earth and stars. Copernicus' predecessors had little doubt
that the system would ultimately be made to work.
In short, though scientists undoubtedly do abandon a conceptual
scheme when it seems in irreconcilable conflict with observation, the
emphasis on logical incompatibility disguises an essential problem.
What is it that transforms an apparently temporary discrepancy into
an inescapable conflict? How can a conceptual scheme that one gen-
eration admiringly describes as subtle, flexible, and complex become
for a later generation merely obscure, ambiguous, and cumbersome?
Why do scientists hold to theories despite discrepancies, and, having
held to them, why do they give them up? These are problems in the
anatomy of scientific belief. They are the primary concern of the next
two chapters, which set the stage for the Copernican Revolution proper.
Our immediate problem, however, is the analysis of the grip ex-
erted upon men's minds by the ancient tradition of astronomical re-
search. How could this tradition provide a set of mental grooves that
guided the astronomical imagination, limited the conceptions avail-
able in research, and made certain sorts of innovations difficult to
conceive and more difficult to accept? We have already dealt, at least
implicitly, with the strictly astronomical aspects of this problem. Both
the two-sphere universe and the associated epicycle-deferent technique
were initially highly economical and fruitful; their first successes
seemed to guarantee the fundamental soundness of the approach;
surely only minor modifications would be required to make the mathe-
matical predictions correspond with observation. A conviction of this
sort is difficult to break, particularly once it has been embodied in the
practice of a whole generation of astronomers who transmit it to their
successors through their teaching and writing. This is the band-wagon
effect in the realm of scientific ideas.
The band-wagon effect is not, however, the whole explanation of
the strength of the astronomical tradition, and in trying to complete
the explanation we shall be temporarily led away from astronomical
problems altogether. The two-sphere universe provided a fruitful guide
to the solution of problems outside as well as inside astronomy. By the
end of the fourth century B.C. it had been applied not only to the
problem of the planets, but also to terrestrial problems, like the fall of
a leaf and the flight of an arrow, and to spiritual problems, like the
relation of man to his gods. If the two-sphere universe, and particularly
the conception of a central and stable earth, then seemed the indubi-
table starting point of all astronomical research, this was primarily
because the astronomer could no longer upset the two-sphere universe
without overturning physics and religion as well. Fundamental astro-
nomical concepts had become strands in a far larger fabric of thought,
and the nonastronomical strands could be as important as the astro-
nomical in binding the imagination of astronomers. The story of the
Copernican Revolution is not, therefore, simply a story of astronomers
and the skies.
3
THE TWO-SPHERE UNIVERSE
IN ARISTOTELIAN THOUGHT
The Aristotelian Universe
In. order to examine the ancient world view, in which astro-
nomical and nonastronomical concepts were woven into a single co-
herent conceptual fabric, we must reverse chronological order and
return for a while to the middle of the fourth century B.C. At that time
the technically developed attack upon the problem of the planets had
scarcely begun, but the same two-sphere cosmology that was guiding
the mathematical researches of planetary astronomers had already
acquired essential nonastronomical functions. Many of these can be
discovered in the voluminous works of the great Greek philosopher
and scientist, Aristotle ( 384-322 B.C.), whose immensely influential
opinions later provided the starting point for most medieval and much
Renaissance cosmological thought.
Aristotle's writings, which have reached us only in imperfect and
highly edited form, deal with the scientific subjects now called physics,
che"listry, astronomy, biology, and medicine, as well as with such
nonscientific fields as logic, metaphysics, politics, rhetoric, and literary
criticism. To each of these, particularly biology, logic, and metaphysics,
he contributed new ideas that were uniquely his own, but even more
important than his many scattered substantive contributions was his
organization of all knowledge into a systematic and coherent whole.
He was not quite successful; it is not difficult to find inconsistencies
and occasional contradictions within the body of Aristotle's writings.
But there is a fundamental unity in his view of man and the universe
that has never since been achieved in a synthesis of comparable scope
and originality. That is one of the reasons why his writings had such
immense influence; others will be examined at the end of this chapter.
r
THE TWO-SPHERE UNIVERSE 79
We first, however, require a brief structural sketch of the Aristotelian
universe itself, followed by a more detailed discussion of the multiple
functions attributed to the terrestrial and celestial spheres in Aris-
totelian thought.
For Aristotle the entire universe was contained within the sphere
of the stars, or, more precisely, within the outer surface of the sphere.
At every point inside the sphere there was some sort of matter -
no holes or vacuums could exist in Aristotle's universe. Outside the
sphere there was nothing- no matter, no space, nothing at all. In
Aristotelian science matter and space go together; they are two aspects
of the same phenomenon; the very notion of a vacuum is absurd. That
is how Aristotle managed to explain the finite size and the uniqueness
of the universe. Matter and space must end together: one need not
build a wall to bound the universe and then wonder what bounds
the wall. As Aristotle put it in his book, On the Heavens:
It is plain, then, . . . that there is not, nor do the facts allow there to be,
any bodily mass beyond the heaven. The world in its entirety is made up of
the whole sum of available matter. . . , and we may conclude that there is
not now a plurality of worlds, nor has there been, nor could there be. This
world is one, solitary, and complete. It is clear in addition that there is neither
place nor void . . . beyond the heaven; for in all place there is a possi-
bility of the presence of body, [and] void is defined as that which, although
at present not containing body, can contain it. . . .1
Like Plato's universe, part of which was briefly described in Chap-
ter I, the Aristotelian universe is self-contained and self-sufficient, leav-
ing nothing outside itself. But Aristotle differentiates the interior of
the universe in far more detail than Plato. The largest part of the in-
terior is filled with a single element, the aether, which aggregates in
a homocentric set of nesting shells to form a gigantic hollow sphere
whose surfaces are the outside of the sphere of the stars and the inner
surface of the homocentric sphere carrying the lowest planet, the
moon. Aether is the celestial element- a crystalline solid in Aristotle's
writings, though its solidity was frequently questioned by his succes-
sors. Unlike the substances known on earth, it is pure and unalterable,
transparent and weightless. From it are made the planets and stars
as well as the nest of concentric spherical shells whose rotations ac-
count for the celestial motions.
Between the times of Aristotle and Copernicus a number of differ-
ent views prevailed about the form and the physical reality of these
celestial spheres that moved the heavens. Aristotle's view was the most
detailed and explicit of all. He believed that there were just fifty-five
real crystalline shells made of aether and that these shells embodied
in a physical mechanism the mathematical system of homocentric
spheres developed by Eudoxus and his successor, Callippus. Aristotle
almost doubled the number of spheres used by the earlier mathe-
maticians, but the spheres that he added were mathematically super-
fluous. Their function was to provide the mechanical linkages neces-
sary to keep the whole set of concentric shells turning; they trans-
formed the entire nest of spheres into a gigantic piece of celestial clock-
work, driven from the sphere of the stars. Since the universe was full,
all spheres were in contact, and the rubbing of sphere on sphere pro-
vided drive for the entire system. The sphere of the stars drove its
nearest interior neighbor, the outermost of the seven homocentric
shells that moved Saturn. That shell drove its next interior neighbor
in Saturn's set, and so on, until the motion was finally transmitted to
the lowest sphere in the set that carried the moon. This was the inner-
most of the aetherial shells, the lower boundary of the celestial or
superlunary region.
The set of epicycles and deferents, which replaced homocentric
spheres for purposes of mathematical astronomy, did not fit very well
into crystalline spheres like those proposed by Aristotle. As a result the
attempt to find a mechanical explanation of the epicyclic motions was
often neglected after the fourth century B.C., and the real existence of
crystalline spheres was occasionally questioned. It is not, for example,
clem- from the Almagest whether Ptolemy believed in them at all. But
in the period separating Ptolemy and Copernicus most educated
people, including astronomers, seem to have believed in at least a
bastard version of Aristotle's spheres. They allowed one spherical
shell for the stars and one for each planet, and they supposed that
each planetary shell was just thick enough for the planet to be at its
inner surface when closest to the central earth and at its outer surface
when farthest from the earth. These eight spheres were nested one
inside the other to fill the entire celestial region. The motion of the
sphere of the stars provided a precise explanation of the diurnal circles
of the stars. The continual rotation of the seven planetary spheres ex-
plained the average motion, but only the average motion, of the planets.
81
Men who did not know or care about the irregularities of planetary
motion could take the thick spheres quite literally: each planet was
fixed in and carried around by its sphere. Planetary astronomers used
epicycles, deferents, equants, and eccentrics to account for each planet's
motion within its own thick spherical shell. For them the shells usually
had at least metaphorical reality, but they rarely bothered with a physi-
cal explanation of a planet's motion within its sphere.
Five centuries or more after Aristotle's death this conception of
thick nested spherical shells added one important technical ingredient
to the astronomy of the post-Ptolemaic period. It enabled astronomers
to compute the actual size of the individual planetary spheres and
therefore of the universe as a whole. Observations of a planet's motion
among the stars enables an astronomer. to determine only the relative
sizes of its epicycle and deferent or the relative amount of its eccen-
tricity. Shrinking or expanding a planet's system of compounded circles
does not change the position at which the planet appears against the
ecliptic, provided that the relative dimensions of epicycle, deferent,
and eccentric are held constant. But if each spherical shell must be
just thick enough to contain the planet both when it is closest and
when it is farthest from the earth, then a knowledge of the relative
dimensions of epicycle, deferent, and so on, suffices to determine the
ratio of the inside and outside diameters of each sphere. Furthermore,
if the spheres must nest so that they fill the entire celestial region, the
outer diameter of one must be equal to the inner diameter of the
next, and the ratio of the distances to all the intershell boundaries can
be computed. Finally, these relative dimensions can be transformed
to absolute distances by utilizing the distance to the sphere of the
moon determined during the second century B.C. by the technique
discussed in the Technical Appendix, Section 4.
Estimates of size based on the conception of space-filling spheres
just large enough to house each planet's set of epicycles and other
circles do not appear in the astronomical literature until after Ptol-
emy's death, presumably because the first planetary astronomers were
skeptical about the reality of such spheres. But after the fifth century
A.D., estimates of this sort became quite common, and once again
they helped to make the entire cosmology seem real. One widely
known set of cosmological dimensions was provided by the Arab
astronomer AI Fargani, who lived in the ninth century A.D. Accord-
ing to his calculation the outer surface of the moon's sphere was 647{;
earth's radii from the center of the universe, the outer surface of
Mercury's sphere was 167 earth's radii from the center; of Venus's,
ll20 earth's radii; of the sun's, 1220; of Mars's, 8867; of Jupiter's, 14405;
and of Saturn's, 20ll0. Since AI Fargani gave the earth's radius as
3250 Roman miles, he placed the sphere of the stars more than 75
million miles from the earth. It is an immense distance but, by mod-
ern theory, almost a million times smaller than the distance to the
nearest star.
A glance at AI Fargani's measurements shows that the terrestrial
region, the space below the underside of the moon's sphere, is but a
minuscule portion of the universe. Most of space is the heavens; most
matter is the aether of the crystalline spheres. But the small size of
the sublunary region does not make it unimportant. Even in Aris-
totle's version, and to a far greater extent in the medieval Christian
revision of Aristotelian cosmology, the tiny central core of the uni-
verse is the kernel for which the rest was made. It is man's abode,
and its character is very different from that of the celestial region
above it.
The sublunary region is filled not with one element but with four
(or, in later writers, some other small number), and the distribution
of these four terrestrial elements, though simple in theory, is extremely
complex in fact. According to the Aristotelian laws of motion, to be
discussed below, the elements would, in the absence of any external
pushes and pulls upon them, settle into a series of four concentric
shells like the aetherial spheres of the fifth element surrounding them.
Eartb, the heavy element, would move naturally into a sphere at the
geometric center of the universe. Water, also heavy but not so heavy
as earth, would settle in a spherical shell about the central region of
earth. Fire, the lightest element, would rise spontaneously to form a
shell of its own immediately below the moon's sphere. And air, also
a light element, would complete the structure by filling the remaining
shell between water and fire. Having achieved these positions, the
elements would rest in place with their full elemental purity. Left to
itself, undisturbed by outside forces, the sublunary region would be
a static region, mirroring the heavenly spheres in its structure.
But ~ h terrestrial region is never undisturbed. It is bounded by
the movmg sphere of the moon, and the movement of this boundary
constantly pushes the layer of fire beneath it, setting up currents which
jostle and mix the elements together throughout the sublunary world.
Therefore, the elements can never be observed in their pure form.
The continual chain of pushes, deriving immediately from the sphere
of the moon and ultimately from the sphere of the stars, keeps them
mixed together in various and varying proportions. The structure of
shells is still approximated; the appropriate element predominates in
the appropriate region. But each element contains at least traces pf
the others, and these transform its character, giving rise, according
to the proportions of the mixture, to all of the varied substances that
can be discovered on the earth. The motions of the heavens are there-
fore responsible for all change and almost all variety observed in the
sublunary world.
It is within this Aristotelian universe, whose scope and adequacy
are scarcely represented by the preceding sketch, that we must search
for the strength of the pre-Copernican astronomical tradition. Why, de-
spite the real difficulties encountered by the Ptolemaic system, did
astronomers continue for so long to assume that the earth had to be
the stable center of the universe and of at least the average planetary
orbits? One familiar answer to this question is already apparent:
Aristotle, the greatest philosopher-scientist of antiquity, had declared
the earth to be immobile, and his word was taken with immense seri-
ousness by his successors, for many of whom he became "the Philoso-
pher," the first authority on all questions of science and cosmology.
But Aristotle's authority, though important, is only the beginning
of an answer, because Aristotle said a great many things which later
philosophers and scientists did not have the least difficulty in rejecting.
In the ancient world there were other schools of scientific and cos-
mological thought, apparently little influenced by Aristotelian opinion.
Even in the late centuries of the Middle Ages, when Aristotle did
become the dominant autlwrity on scientific matters, learned men did
not hesitate to make drastic changes in many isolated portions of his
doctrine. The list of alterations introduced by later Aristotelians into
Aristotle's original teachings is almost endless, and some of these altera-
tions were far from trivial. As we shall see in the next chapter, a few
of the criticisms directed at Aristotle by his successors play a direct
and causal role in the Copernican Revolution.
Yet no later Aristotelian suggested that the earth was a planet or
that it was located away from the center of the universe. That innova-
tion proved a peculiarly difficult one for an Aristotelian to conceive
or to accept, because the concept of a unique central earth was inter-
woven with so many other important concepts within the fabric of
Aristotelian thought. An Aristotelian universe can be built as well with
three or five terrestrial elements as with four and very nearly as well
with epicycles as with homocentric spheres, but it cannot and did not
survive the modification that made the earth a planet. Copernicus tried
to design an essentially Aristotelian universe around a moving earth,
but he failed. His followers saw the full consequeaces of his innova-
tion, and the entire Aristotelian structure cnunbled. The concept of a
central and stable earth was one of the few major constitutive concepts
in a closely knit and coherent world view.
The Aristotelian laws of Motion
A first example of the integration of astronomical and non-
astronomical thought is provided by Aristotle's explanation of terres-
trial motion. As we have already noted, Aristotle believed that, in the
absence of external pushes derived ultimately from the heavens, each
of the terrestrial elements would remain at rest in that part of the
terrestrial region natural to it. Earth rests naturally at the center, fire
at the periphery, and so on. In fact, the elements and the bodies com-
posed of them are constantly wrested from their natural positions.
But that requires the application of a force; an element resists displace-
ment; and, once displaced, it strives to regain its natural position by
the shortest possible path. Pick up a rock or some other earthy material
and f ~ l it tug away, attempting to reach its natural position at the
geometric center of the universe. Or watch the flames of a fire leap
upward on a clear night as they strive for their natural place at the
periphery of the terrestrial region.
We shall examine later the psychological sources and the strength
of this Aristotelian explanation of terrestrial motion. But first notice
t ~ bulwark that these doctrines, drawn from terrestrial physics, pro-
vide for the earth-centered universe of the astronomer. In an important
passage from On the Heavens Aristotle derives the sphericity, stability,
and central location of the earth from them. We have previously seen
them derived by astronomical arguments, but note how secondary
a role astronomical considerations play here.
The natural motion of the earth as a whole, like that of its parts, is
towards the center of the Universe: that is the reason why it is now lying
at the center. It might be asked, since the center of both is the same point, in
which capacity the natural motion of heavy bodies, or parts of the earth, is
directed towards it; whether as center of the Universe or of the earth. But
it must be towards the center of the Universe that they move. . . . It so
happens that the earth and the Universe have the same center, for the
heavy bodies do move also towards the center of the earth, yet only inci-
dentally, because it has its center at the center of the Universe ....
From these considerations it is clear that the earth does not move,
neither does it lie anywhere but at the center. In addition the reason for its
immobility is clear from our discussions. If it is inherent in the nature of
earth to move from all sides to the center (as observation shows), and of
fire to move away from the center towards the extremity, it is impossible
for any portion of earth to move from the center except under con-
straint. . . . If then any particular portion is incapable of moving from the
center, it is clear that the earth itself as a whole is still more incapable, since
it is natural for the whole to be in the place towards which the part has a
natural motion. . . .
Its shape must be spherical. . . . To grasp what is meant we must
imagine the earth as in process of generation. . . . It is plain, first, that if
particles are moving from all sides alike towards one point, the center, the
resulting mass must be similar on all sides; for if an equal quantity is added
all round, the extremity must be at a constant distance from the center.
Such a shape is a sphere. But it will make no difference to the argument
even if the portions of the earth did not travel uniformly from all sides
towards the center. A greater mass must always drive on a smaller mass in
front of it, if the inclination of both is to go as far as the center, and the
impulsion of the less heavy by the heavier persists to that point. . . .
Further proof is obtained from the evidence of the senses. (i) If the
earth were not spherical, eclipses of the moon would not exhibit segments
of the shape they do .... (ii) Observation of the stars also shows not only
that the earth is spherical but that it is of no great size, since a small change
of position on our part southward or northward visibly alters the circle of
the horizon, so that the stars above our heads change their position con-
siderably, and we do not see the same stars as we move to the North or
South. Certain stars are seen in Egypt and the neighborhood of Cyprus,
which are invisible in more northerly lands, and stars which are continuously
visible in the northern countries are observed to set in the others. This
proves both that the earth is spherical and that its periphery is not large, for
otherwise such a small change of position could not have had such an im-
mediate effect. For this reason those who imagine that the region around
the Pillars of Heracles joins on to the regions of India, and that in this way
the ocean is one, are not, it would seem, suggesting anything utterly in-
crPdible.2
Passages like this indicate that astronomy and terrestrial physics
are not independent sciences. Observ'ations and theories developed for
one become intimately entangled with those drawn from another.
Therefore, though difficulty in solving the problem of the planets might
have provided an astronomer with a motive for experimentir. g in
astronomy with the conception of a moving earth, he could not do so
without upsetting the accepted basis of terrestrial physics in the
process. The very notion of a moving earth would be unlikely to occur
to him, because, for reasons drawn from his nonastronomical knowl-
edge, the conception seemed so implausible. That seems to be what
Ptolemy and his successors meant when they later described the astro-
nomical hypotheses of Aristarchus, Heraclides, and the Pythagoreans
as "ridiculous" even though astronomically satisfactory.
Examine, for example, the following passage from the Almagest,
in which Ptolemy rejects Heraclides' theory that the sphere of the
stars is stationary and that its apparent westward diurnal motion is
due. to a .real eastward diurnal rotation of the central earth. Ptolemy
begms with arguments for the sphericity and central position of the
earth much like those given by Aristotle in the passage quoted above.
Then he continues:
Certain thinkers, though they have nothing to oppose to the above
arguments, ?ave a scheme which they consider more acceptable,
and they thmk that no evidence can be brought against them if they suggest
for the sake of argument that the heaven is motionless, but that the earth
rotates about one and the same axis from west to east, completing one
revolution approximately every day. . . .
These persons forget however that, while, so far as appearances in the
stellar .world are concerned, there might, perhaps, be no objection to this
theory. . , yet, t? judge by the [terrestrial] conditions affecting ourselves
those m the air about us, such a hypothesis must be seen to be quite
ndiCulous .... [If the earth] made in such a short time such a colossal
turn back to the same position again, . . . everything not actually standing
on the earth must have seemed to make one and the same movement always
in the contrary sense to the earth, and clouds and any of the things that fly
or can be thrown could never be seen travelling towards the east, because
the . earth would always be anticipating them all and forestalling their
motion towards the east, insomuch that everything else would seem to
recede towards the west and the parts which the earth would be leaving
behind it. a
The gist of Ptolemy's argument is the same as Aristotle's, and many
other arguments were derived from the same principles during the
Middle Ages and the Renaissance. Unless it is pushed, a body will
move straight toward its natural position and then rest there. These
natural positions and the lines by which bodies move to them are
determined entirely by the intrinsic geometry of an absolute space, a
space in which each position and direction is permanently labeled
whether or not the position is occupied. Therefore, as Aristotle says
elsewhere in On tlze Heavens, "If the earth were removed to where
the moon is now, separate parts of it would not move towards the
whole, but towards the place [the center] where the whole is now."
4
The natural motion of a stone is governed by space alone, not by the
stone's relation to other bodies. Therefore, a stone thrown vertically
upward moves away and returns along a straight line fixed once and
for all in space, and if the earth moves while the stone is in the air,
the stone will not rejoin the earth at the point from which it departed.
By the same token, clouds which already occupy their natural posi-
tions would be left behind as the earth rotates. Only if the moving earth
carries the air with it, could a cloud or stone follow the earth at all, and
even the motion of the air would not push a stone hard enough to
keep it in step with the earth's rotation.
There are, of course, difficulties in this Aristotelian theory of mo-
tion, and some of them will later play a significant role in the -Coperni-
can Revolution. But, like the two-sphere universe itself, Aristotle's
theory of motion is an excellent first step toward an understanding of
motion, and it does necessitate a central stationary earth. Advocates
of a planetary earth will therefore require a new theory of motion, and
until such a theory is invented, as it was during the Middle Ages, a
knowledge of terrestrial physics will inhibit the astronomical imagina-
tion.
The Aristotelian Plenum
A second illustration of the blinders fitted to the astronomer
by the coherent interrelations between his astronomical and nonastro-
nomical knowledge is provided by the Aristotelian conception of a full
universe or plenum. This example is more typical than the last, for
the ties between the various strands ofknowledge are here both more
numerous and less binding than those illustrated above. The complex
pattern of Aristotelian thought now begins to emerge.
88
The ancient conception of the fullness of the universe is often re-
ferred to as the horror vacui, nature's abhorrence of a vacuum. As an
explanatory principle it can be paraphrased to read: Nature will al-
ways act to prevent the formation of a vacuum. In this form the
Greeks derived it from and used it to explain a large variety of natural
phenomena. Water will not flow from an open bottle with a small neck
unless a second hole is made in the bottle, because without a second
hole at which air can enter the emerging water would leave a vacuum
behind it. Siphons, water clocks, and pumps were economically ex-
plained on the same basis. Some ancient thinkers applied the horror
vacui to the explanation of adhesion and to the design of hot-air and
steam engines. The experimental basis of the principle could not be
challenged. Convincing approximations to vacuums cannot be pro-
duced on earth without apparatus of which the Greeks knew nothing.
There were no pneumatic phenomena to challenge the principle until,
with the large-scale development of deep mining during the sixteenth
century, it was discovered that lift pumps will not raise water more
than 30 feet. Rejecting the horror vacui necessarily meant destroying
a thoroughly satisfactory scientific explanation of a host of terrestrial
phenomena.
For Aristotle and most of his successors, however, the horror vacui
was more than a successful experimental principle applicable on and
near the surface of the earth. Aristotle held not only that there are in
fact no vacuums in the terrestrial world, but that there can in prin-
ciple be no vacuums anywhere in the universe. The very concept of
a vacuum was to him a contradiction in terms, like the concept "square
circle." Today, when everybody has seen a "vacuum" tube and heard
of a "\'acuum" pump, Aristotle's logical proofs of the impossibility of
a void convince almost no one, though it is frequently difficult to dis-
cover the faults in his arguments. But in the absence of the experi-
mental counterevidence, which we now possess, they seemed convinc-
ing, for they arose from a genuine difficulty inherent in the words
with which we discuss problems of matter and space. Apparently space
can be defined only in terms of the volume occupied by body. In the
absence of material body there is nothing in terms of which to define
space; it cannot apparently exist by itself at all. Matter and space
are inseparable, two sides of the same coin. There can be no space
without matter. In Aristotle's more cumbersome words, "there is no
such thing as a dimensional entity, other than that of material sub-
stances."
5
The theory of a full universe therefore entered ancient science with
the combined authority of logic and experiment, and it immediately
became an essential ingredient of cosmological and astronomical the-
ories. It is, for example, involved in the Aristotelian explanation of the
endurance of motion within the sphere of the stars. If any one of
the celestial or terrestrial shells were replaced by a void, all motion
within that shell would cease. The rubbing of shell on shell produces
all motion, except return to natural position, and a void would break
the chain of pushes. Again, as we have already noted, the impossibility
of a void is the basis of the universe's finitude. Beyond the sphere of
the stars is neither space nor matter - nothing at all. Without a con-
cept that indissolubly united matter and space, the Aristotelian would
be forced to admit the infinity of the universe. Matter could be bounded
by void, and void could, in turn, be bounded by matter, but there
could be no terminus, no last boundary at which the universe ended
once and for all.
But an infinite universe could scarcely remain an Aristotelian uni-
verse for two reasons. An infinite space has no center: every point is
equally distant from all points on the periphery. And if there is no
center, there is no preferred point at which the heavy element earth
can aggregate, and there is no intrinsic "up" and "down" to determine
the natural motion of an element returning to its proper place. In fact
there is no "natural place" in an infinite universe, for each place is like
every other. The whole Aristotelian theory of motion is, as we shall
see more fully later, inextricably bound to the conception of a finite
and fully occupied space. The two stand or fall together.
Nor are these the only difficulties presented to an Aristotelian by
the infinitude of space. If space is infinite and there is no special center
point, it is scarcely plausible that all the earth, water, air, and fire in
the universe should have aggregated at one and only one point. In
an infinite universe, it is natural to presume that there are other worlds
scattered here and there through all of space. Perhaps there are plants,
men, and animals on these other worlds. Thus the earth's uniqueness
vanishes; the peripheral force that drives the whole disappears with
it; man and the earth cease to be at the focus of the universe. Both in
antiquity and in the Middle Ages, most of those philosophers who,
like the atomists, believed that the universe was infinite felt them-
selves impelled to accept the reality of the void and a plurality of
worlds as well. And until the seventeenth century no one who embraced
this set of concepts produced a cosmology able to compete with the
Aristotelian in the explanation of everyday terrestrial and celestial
phenomena. The infinite universe may be a common-sense universe
today, but today common sense has been reeducated.
The multifarious roles of the conception of a full universe in Aris-
totelian thought is our one full-dress example of the coherence of a
cosmology or a world view. The plenum is implicated in pneumatics,
the endurance of motion, the finitude of space, the laws of motion,
the uniqueness of the earth. The list could be extended. Note that the
plenum does not logically necessitate either the uniqueness, or the
central position, or the immobility of the earth. It simply fits into a
coherent pattern in which the unique, central, and immobile earth is
a second essential strand. Conversely, the earth's motion does not
necessitate either the existence of a vacuum or the infinity of the uni-
verse. But it is no accident that both these views won acceptance
shortly after the victory of the Copernican theory.
Copernicus himself believed in neither. As we shall see, he tried to
preserve most of the central features of Aristotelian and Ptolemaic
cosmology. But by giving the earth an axial motion, he made the sphere
of the stars immobile, depriving it of physical function. And by giving
the earth an orbital motion, he made necessary a vast increase in the
size of the sphere. Copernicus' cosmology thus took away from inter-
planetary matter many of its essential Aristotelian functions and simul-
taneously demanded that there be vastly more of it. His successors soon
fractured the now functionless sphere, scattered .the stars through all
of space, admitted a vacuum or something very like it between them,
and dreamed of other worlds inhabited by other men in the vast ex-
panses beyond our solar system. Even the terrestrial principle of the
horror vacui did not survive for long. In the new universe it was very
much easier for scientists to recognize that practical miners had for
a century been producing a terrestrial vacuum at the top of overlong
water pumps. Air pressure soon replaced the vacuum in the pneu-
matic conceptions of the seventeenth century. Many other forces
played an essential role in the modification of pneumatics- the story
is complex- but the new astronomy of Copernicus is a necessary in-
91
gredient of its plot. Once again astronomical displays its inti-
mate entanglement with the theories of other sciences, and those other
sciences condition the astronomical imagination.
The Majesty of the Heavens
The extra-astronomical entanglements of astronomical theory
are not, however, exclusively ties to other sciences. As our previous
discussions of the motives for celestial observation have repeatedly
hinted, the ancient astronomical tradition is partially indebted for its
very existence to a widespread primitive perception of
between the power and stability of the heavens and the Impotent m-
security of terrestrial life. This same perception is incorporated into
Aristotle's cosmology by the absolute distinction between the super-
lunary and sublunary regions. But in Aristotle's highly articulated ver-
sion the distinction comes to depend explicitly upon both the central
position of the earth and the perfect symmetry of the spheres that gen-
erate both the stellar and planetary motions.
According to Aristotle, the underside of the sphere of the moon
divides the universe into two totally disparate regions, filled with dif-
ferent sorts of matter and subject to different laws. The terrestrial re-
gion in which man lives is the region of variety and change, birth and
death, generation and corruption. The celestial region is, in contrast,
eternal and changeless. Only aether, of all the elements, is pure and
incorruptible. Only the interlocked celestial spheres move naturally
and eternally in circles, never varying their rate, always occupying
exactly the same region of space, forever turning back upon them-
selves. The substance and the motion of the celestial spheres are the
only ones compatible with the immutability and majesty of the heavens,
and it is the heavens that produce and control all variety and change
on earth. In Aristotle's physical description of the universe, as much
as in any primitive religion, the encircling heavens are the locus of
the perfection and the power upon which terrestrial life depends. On
the Heavens puts the point unequivocally:
From what has been said it is clear why . . . the primary body of all
[that is, celestial matter] is eternal, suffers neither growth nor diminution,
but is ageless, unalterable and impassive. I think too that the argument bears
out experience and is borne out by it. All men have a conception of gods, and
all assign the highest place to the divine, both barbarians and Hellenes, as
many as believe in gods, supposing, obviously, that immortal is closely linked
with immortal. It could not, they think, be otherwise. If then - and it is
true - there is something divine, what we have said about the primary
bodily substance [namely that it is weightless, indestructible, unalterable,
and so on] is well said. The truth of it is also clear from the evidence of the
senses, enough at least to warrant the assent of human faith; for throughout
all past time, according to the records handed down from generation to
generation, we find no trace of change either in the whole of the outermost
heaven or in any one of its proper parts. It seems too that the name of this
first body has been passed down to the present time by the ancients. . . .
Thus thev, believing that the primary body was something different from
earth and fire and air and water, gave the name aether to the uppermost
region, choosing its title from the fact that it "runs always" and eternally.
6
Aristotle himself carried the conception of the majesty and divin-
ity of the celestial regions little further. Both the matter and the motions
of the heavens are perfect; all terrestrial change is caused and governed
by a succession of pushes initiated by the uniform motions of the
celestial spheres which symmetrically enclose the earth. Already a
significant nonscientific argument for the earth's unique central loca-
tion is apparent, and, in the centuries after Aristotle's death, this argu-
ment was reinforced by elaborating the conception of the perfect
heavens and integrating it with two other important sets of beliefs,
both of which had originated independently. One of these develop-
ments - the detailed integration of Aristotelian cosmology with Chris-
tian theology- we must postpone until its proper chronological posi-
tion in the next chapter. It resulted in a universe each of whose struc-
tural details carried religious as well as physical significance: Hell was
at geometric center; God's throne was beyond the stellar sphere;
each planetary sphere and epicycle was turned by an angel. But an-
other significant application of the concept of celestial majesty- the
science of astrology - is older than the Christian-Aristotelian cosmol-
ogy, and it had an even more immediate impact on practitioners of
astronomy. Because it involved them professionally, astrology may
well have been the most important of the forces binding astronomers to
the conception of the earth's uniqueness.
We have already noted the roots of astrological belief
and their relation to the Aristotelian conception of the power of the
heavens. Distance and immutability make the heavens a plausible locus
for the gods who can intervene at will in men's affairs. Disruptions of
celestial regularity- particularly comets and eclipses- were regarded
from an early date as portents of unusual felicity or disaster. In addi-
tion, there is good observational evidence for celestial control of at
least some terrestrial events. It is hot when the sun is in the constel-
lation Cancer and cold when it is in Capricorn; the varying height of
the tide follows the variation of the moon's phases; the menstrual
cycle of women throughout the earth recurs at intervals coincident
with the length of the lunar month. In an era when man's need to
understand and control his fate immeasurably transcended his physical
and intellectual tools, this apparent evidence of celestial power was
naturally extended to the other celestial wanderers. Particularly after
Aristotle supplied a physical mechanism - the frictional drive -
through which heavenly bodies could produce terrestrial change, there
was a plausible basis for the belief that an ability to predict the future
configurations of the heavens would enable men to foretell the future
of men and nations.
Before the second century B.C., ancient records show few signs of
a fully developed attempt to predict the details of terrestrial affairs
from the observed and computed positions of the stars and planets.
But after this relatively late start, astrology was inseparably linked
to astronomy for 1800 years; together they constituted a single pro-
fessional pursuit. The astrology that predicted the future of men from
the stars was known as judicial astrology; the astronomy that pre-
dicted the future of stars from their present and past was known as
natural astrology; those who gained fame in one branch were usually
well known in the other as well. Ptolemy, whose Almagest exhibited
ancient astronomy in its most developed form, was equally famous for
his Tetrabiblos, antiquity's classic contribution to judicial astrology.
European astronomers like Brahe and Kepler, who late in the Renais-
sance put Copernicus' system into something very like its modern
form, were supported financially and intellectually because they were
thought to cast the best horoscopes.
During most of the period with which the rest of this book is
concerned, astrology exercised an immense influence upon the minds
of the most educated and cultured men of Europe. Early in the Middle
Ages it was partially suppressed by the Church, whose doctrinal in-
sistence that men are free to choose the Christian good was incom-
patible with astrology's strict determinism. But during the five cen-
turies centering on the birth of Christ, and again during the late
Middle Ages and the Renaissance, astrology was the guide of kings and
of their people, and it is no accident that these are just the periods
during which earth-centered astronomy made most rapid progress. The
elaborate tables of planetary position and the complex computational
techniques developed by planetary astronomers from antiquity to the
Renaissance were the main prerequisite for astrological prediction.
Until after Copernicus' death these major products of astronomical
research had little other socially significant application. Astrology
therefore provided the principal motive for wrestling with the prob-
lem of the planets, so that astrology became a particularly important
determinant of the astronomical imagination.
Astrology, however, and the perception of celestial power that
underlies it lose much plausibility if the earth is a planet. A planetary
earth will act as forcefully on Saturn as Saturn can act on it; the
same argument applies to the other planets; and the terrestrial-celestial
dichotomy breaks down. If the earth is a celestial body, it must share
the immutability of the heavens, and the heavens in tum must partici-
pate in the corruptibility of earth. It cannot be coincidence that astrol-
ogy's stranglehold upon the human mind finally relaxed during just
the period in which the Copernican theory first gained acceptance.
It may even be significant that Copernicus, the author of the theory
that deprived the heavens of special power, belonged to
the mmonty group of Renaissance astronomers who did not cast horo-
scopes.
Astrology and the majesty of the heavens therefore provide one
of the indirect consequences of the earth's stability and
umqueness, consequences that have been repeatedly illustrated but by
n.o means exhausted in this extended discussion of the multiple func-
tiOns of a central stable earth in the Aristotelian world view. It is, of
course, precisely these consequences and others like them that make
the Copernican Revolution a revolution. To describe the innovation
initiated by Copernicus as the simple interchange of the position of
the earth and sun is to make a molehill out of a promontory in the
development of human thought. If Copernicus' proposal had had no
consequences outside astronomy, it would have been neither so long
delayed nor so strenuously resisted.
The Aristotelian World View in Perspective
The Aristotelian world view was the single most important
source and support for the pre-Copernican tradition of astronomical
practice. But Aristotle's day is not our day, and a real mental trans-
position is therefore necessary in approaching his writings, particu-
larly those dealing with physics and cosmology. Failure to make this
transposition has resulted in some strained and distorted explanations
of the endurance of Aristotelian physics in antiquity and during the
Middle Ages.
We are, for example, often told that it is only because medieval
scientists preferred the authority of the written word, preferably an-
cient, to the authority of their own eyes that they could continue to
accept Aristotle's absurd dictum that heavy bodies fall faster than
light ones. Modem science, on this prevalent interpretation, began
when Galileo rejected texts in favor of experiments and observed that
two bodies of unequal weight, released from the top of the tower of
Pisa, struck the ground simultaneously. Today every schoolboy knows
that heavy bodies and light bodies fall together. But the schoolboy is
wrong and so is this story. In the everyday world, as Aristotle saw,
heavy bodies do fall faster than light ones. That is the primitive per-
ception. Galileo's law is more useful to science than Aristotle's, not
because it represents experience more perfectly, but because it goes
behind the superficial regularity disclosed by the senses to a more
essentiaL but hidden, aspect of motion. To verify Galileo's law by
observation demands special equipment; the unaided senses will not
yield or confirm it. Galileo himself got the law not from observation,
at least not from new observation, but by a chain of logical arguments
like those we shall examine in the next chapter. Probably he did not
perform the experiment at the tower of Pisa. That was performed by
one of his critics, and the result supported Aristotle. The heavy body
did hit the ground first.
The popular story of Galileo's refutation of Aristotle is largely a
myth, motivated by a failure of historical perspective. We like to
forget that many of the concepts in which we believe were painfully
drummed into us in our youth. We too easily take them as natural and
indubitable products of our own unaided perceptions, dismissing con-
cepts different from our own as errors, rooted in ignorance or stupid-
ity and perpetuated by blind obedience to authority. Our own edu-
cation stands between us and the past. Particularly it stands between
us and Aristotelian physics, frequently leading us to misinterpret the
nature and source of Aristotle's immense influence on subsequent
generations.
Part of the authority of Aristotle's writings derives from the bril-
liance of his own original ideas, and part derives from their immense
range and logical coherence, which are as impressive today as ever.
But the primary source of Aristotle's authority lies, I believe, in a third
aspect of his thought, one which it is more difficult for the modern
mind to recapture. Aristotle was able to express in an abstract and
consistent manner many spontaneous perceptions of the universe that
had existed for centuries before he gave them a logical verbal rationale.
In many cases these are just the perceptions that, since the seventeenth
century, elementary scientific education has increasingly banished from
the adult Western mind. Today the view of nature held by most sophis-
ticated adults shows few important parallels to Aristotle's, but the
opinions of children, of the members of primitive tribes, and of many
non-Western peoples do parallel his with surprising frequency. Some-
times the parallels are difficult to discover, because they are hidden by
Aristotle's abstract vocabulary and by his elaborately logical method.
These are the elements of Aristotelian dialectic, and only their rudi-
ments can be found in primitives and children. But Aristotle's sub-
stantive ideas about nature, in contrast to the way he expresses and
documents them, do show important residues of earlier and more
elementary perceptions of the universe. Unless alert to these residues
we may miss the meaning and will surely miss the force of important
segments of Aristotelian doctrine.
The nature of the primitive residues and the manner in which
they are transformed by the impact of Aristotelian dialectic are clearly
illustrated in Aristotle's discussions of space and motion. The world
views of primitive societies and of children tend to be animistic. That
is, children and many primitive peoples do not draw the same hard and
fast distinction that we do between organic and inorganic nature, be-
tween living and lifeless things. The organic realm has a conceptual
priority, and the behavior of clouds, fire, and stones tends to be ex-
plained in terms of the internal drives and desires that move men and,
presumably, animals. Asked why balloons go up, one child of four an-
r
I
I
97
swers, "Because they want to fly away." Another, age six, explains that
balloons go up because "they like the air. So when you let go they
go up in the sky." Asked why a box falls to the ground, Hans, age five,
answers, "Because it wants to go there." Why? "Because it is a good
thing [for it to be there]."
7
Primitives frequently give similar ex-
planations, though they are often harder to unravel because expressed
in myths which cannot be taken quite literally. We have already ex-
amined the Egyptian explanation of the sun's motion as that of a god
sailing in his boat across the heavens.
Aristotle's stones are not alive, though his universe frequently seems
to be, at least metaphorically. (There are passages in Aristotle reminis-
cent of the passage from Plato's Timaeus quoted in the first chapter.)
But his perception of the stone leaping from the hand to achieve its
natural place at the center of the universe is not so very different from
the child's perception of the balloon that likes the air or of the box
that falls to the ground because it is good for it to be there. The vocab-
ulary has changed; the concepts are manipulated with adult logic;
animism has been transmuted. But much of the appeal of the Aris-
totelian doctrine must lie in the naturalness of the perception that
underlies the doctrine.
Animism is not, however, the entire psychological base of Aris-
totle's explanation of motion. A subtler and, I think, more important
element derives from the Aristotelian transmutation of a primitive
perception of space. To the members of prehistoric civilizations and
primitive tribes, space seems very different from the Newtonian space
in which we were all brought up, usually without knowing it. The
latter is physically neutral. A body must be located in space and move
through space, but the particular part of space and the particular
direction of motion exert no influence on the body. Space is an inert
substratum for all bodies. Each position and each direction is like
every other. In modern terminology, space is homogeneous and iso-
tropic; it has no "top" or "bottom," "east" or "west."
The space of the primitive, in contrast, is often more nearly a life
space: the space in a room, or in a house, or in a community. It has a
"top" and "bottom," and "east" and "west" (or "front" and "back" - in
many primitive societies words for direction derive from words for
parts of the body and reflect the intrinsic differences of these parts).
Each position is a position "for" some object or "where" some char-
acteristic activity occurs. Each region and direction of space is charac-
teristically different from every other, and the differences partially
determine the behavior of bodies in each region. Usually the primitive's
space is the active dynamic space of everyday life; distinct regions
have distinct characteristics.
Egyptian cosmology provided an example: the region of the cir-
cumpolar stars became the region of eternal life, of those who never
die. A similar perception of space provides one important source of
astrological thought. The nature and power of planets depends upon
their position in space. One old Babylonian text states: "When the
star Marduk [the planet Jupiter J is in the ascendant [that is, low on
the eastern horizon], it is then Nebo [the god Mercury]. When it has
risen . . . [number omitted] double hours, it is Marduk [the god
Jupiter]. When it stands in the middle of the heavens, it is Nibiru [the
highest one, the omnipotent god]. Each planet becomes this at its
zenith."
8
The primitive residues inherent in the Aristotelian conception of
space are seldom so clear. But examine the following discussion of
motion from Aristotle's Physics:
The typical locomotions of the elementary natural bodies - namely fire,
earth, and the like- show not only that place is something, but also that it
exerts a certain influence. Each is carried to its own place, if it is not
hindered, the one up, the other down. . . . It is not every chance direction
which is "up," but where fire and what is light are carried; similarly, too,
"down" is not any chance direction but where what has weight and what
is made of earth are carried - the implication being that these places do not
differ merely in relative position, but also as possessing distinct potencies.
9
This passage is an almost perfect summary of the conception of
space that underlies the Aristotelian explanation of motion: "place
... exerts a certain influence"; "places do not differ merely in rela-
tive position, but also as possessing distinct potencies." These are
places in a space that has an active and dynamic role in the motion of
bodies. Space itself supplies the push that drives flre and stones to
their natural resting places at the periphery and the center. The in-
teractions of matter and space determine the motion and rest of bodies.
To us this is an unfamiliar notion, because we are the heirs of the
Copernican Revolution, which made it necessary to discard and re-
place the Aristotelian conception of space. But the concept is not im-
plausible. Perhaps by mere coincidence, the spatial concepts embodied
in Einstein's general theory of relativity are, in important respects,
closer to Aristotle's than to Newton's. And Einstein's universe may,
like Aristotle's and unlike Newton's, be finite.
Aristotle's world view was not the only one created in antiquity,
nor was it the only one that gained adherents. But Aristotle's was far
nearer to many primitive conceptions of the world than its ancient com-
petitors, and it corresponded more closely with t ~ e e v i d e ~ e of un-
aided sense perception. That is another reason why It was so lffiffiensely
influential, particularly during the late Middle Ages. Having isolated
at least part of its appeal, we can better appreciate the strength that
Aristotelian cosmology contributed to the ancient astronomical tradi-
tion. Now we must discover what happened to that tradition to pre-
pare the way for Copernicus.
4
RECASTING THE TRADITION:
ARISTOTLE TO THE COPERNICANS
European Science and Learning to the Thirteenth Century
Aristotle was the last great cosmologist of antiquity, and
Ptolemy, who lived almost five centuries after Aristotle, was its last
great astronomer. Until after the death of Copernicus in 1543, the
writings of these two men dominated the astronomical and cosmologi-
cal thought of the West. Copernicus seems their immediate heir, for
in the thirteen centuries that separate Ptolemy's death from Copernicus'
birth no large and enduring modification had been imposed upon their
work. Because Copernicus began where Ptolemy had stopped many
people conclude that there was no science during the intervening cen-
turies. In fact, there was much intense though spasmodic scientific
activity, and it played an essential role in preparing the ground for
the inception and success of the Copernican Revolution.
If there is a paradox here, it is only apparent. Thirteen centuries
of spasmodic research did not materially modify the substantive be-
liefs of the researchers. Copernicus' teachers still believed that the
structure of the universe was about as described by Aristotle and
Ptolemy, and their beliefs place them in an ancient tradition. But their
attitude toward those beliefs was not ancient. Conceptual schemes age
with the succession of the generations that behold them. At the begin-
ning of the sixteenth century men still believed in the ancient descrip-
tion of the universe, but they evaluated it differently. Their concepts
were the same, but they saw new strengths and weaknesses in those
concepts. Just as we have explored the sources and the strengths of
the ancient astronomical tradition, so we must discover what hap-
pened to it as it aged. We shall have to begin by finding out how the
RECASTING THE TRADITION 101
ancient tradition was lost and then rediscovered, for the first changes
in European attitudes toward the tradition arose from the necessity
of recovering it.
The West's loss of ancient science occurred in two stages, the
first a slow decline in the quality and quantity of scientific activity,
the second a genuine disappearance of traditional learning. After the
second century B.C., Mediterranean civilization was increasingly domi-
nated by Rome, and it declined with the decline of Roman hegemony
during the first few centuries of the Christian era. Ptolemy, the astrono-
mer, and Galen, the physician, were the last great figures in ancient
science, and they both lived in the second century A.D. After their time
the major scientific works of the West were commentaries and en-
cyclopedias. When the Moslems invaded the Mediterranean basin
during the seventh century, they found only the documents and the
tradition of ancient learning. The activity had largely ceased.
The Islamic invasions shifted the geographic center of Europear.
Christendom northward from the Mediterranean and thus enforced
the continued decline of Western learning. During the seventh century
Europeans were deprived even of the documents in which the ancient
learned tradition was embodied. Euclid was known only in the in-
complete Latin translation prepared by Boethius early in the sixth
century; that version stated only the more important theorems and
included no proofs. Ptolemy was apparently totally unknown, and Aris-
totle was represented only by a few works on logic. Encyclopedic
collections by men like Boethius and Isidore of Seville preserved frag-
ments of ancient science, but even these fragments were too often
inaccurate, intellectually debased, and heavily interlarded with fable.
There was little learned activity of any sort. The economic level of
European Christendom barely provided subsistence. Science was par-
ticularly neglected because, as we shall see in the next section, the
Catholic Church was initially hostile to it.
During the centuries when European learning reached its nadir,
there was a great renaissance of science in Islam. After the middle of
the seventh century the Moslem world rapidly expanded from an
Arabic.,l oasis to a Mediterranean empire, and this new empire in-
herited the manuscripts and the tradition that Christendom had lost.
Moslem scholars first reconstituted ancient science by translating Syriac
102
THE COPERNICAN R'EVOLUTION
versions of original Greek texts into Arabic; then they added contribu-
tions all their own. In mathematics, chemistry, and optics they made
original and fundamental advances. To astronomy they contributed
both new observations and new techniques for the computation of
planetary position. Yet the Moslems were seldom radical innovators
in scientific theory. Their astronomy, in particular, developed almost
exclusively within the technical and the cosmological tradition estab-
lished in classical antiquity. Therefore, from our present restricted
viewpoint, Islamic civilization is important primarily because it pre-
served and proliferated the records of ancient Greek science for later
European scholars. Christendom recovered ancient learning first from
the Arabs and usually in Arabic translation. The title Almagest by
which we know Ptolemy's major work is not its Greek name at all, but
a contraction of the Arabic title which it received from a ninth-
century Moslem translator.
Europeans rediscovered ancient learning in Islam during the period
of general European recovery which makes the tenor of the later
Middle Ages so different from that of the Dark Ages. Beginning slowly
in the tenth century and culminating in what is now known as the
Twelfth Century Renaissance, there was a gradual increase in the
tempo of all aspects of European life. For the first time Christendom
achieved relative political security; with it came a great increase of
population and trade, including trade with the Moslem world. In-
tellectual contacts with Islam multiplied as commerce grew. New-found
wealth and security provided leisure to explore the newly opened
horizons of learning. The first Latin translations from the Arabic were
in the tenth century and multiplied rapidly in those that fol-
lowed. Late in the eleventh century students from all over Europe began
to assemble informally but in steadily increasing numbers to hear some
master read and comment upon a new translation of an ancient text.
In the twelfth and thirteenth centuries these initially informal gather-
ings became so large that they required the regulations and charters
that transformed them officially to universities, a new sort of learned
institution indigenous to Europe. Starting as centers for the oral propa-
gation of ancient learning, these universities rapidly became the home
of an original and creative tradition of European scholarship, the
critical and combative philosophical tradition known as scholasticism.
RECASTING THE TRADITION
103
The rediscovery of ancient astronomy was a part of the larger
reclamation of the science and philosophy of the ancient world. The
first astronomical tables to be widely exploited by Europeans were
imported from Toledo in the eleventh century.
and most of Aristotle's astronomical and physical wntmgs were latm-
ized during the twelfth, and in the following century they were
steadily, though selectively, integrated into the curriculum of the
medieval university. Copernicus studied them there at the end of the
fifteenth century, his return to these classics of ancient science
makes him the heir of Aristotle and Ptolemy. But they would scarcely
have recognized as their own work the inheritance Copernicus
received. Old problems, though still unsolved, had disappeared; ne':"
ones though sometimes merely pseudo problems, had taken theu
In addition, the purposes and methods of the
tradition differed significantly from the ones that had guided ancient
scholars.
Some of the new problems were purely textual in origin. Ancient
texts were recovered piecemeal, in an order governed more by chance
than by logic. Arabic manuscripts were faithful to
their Greek or Syriac sources; the medieval Latin mto which they were
retranslated did not at first possess a vocabulary adequate to their
technical and abstract subject matter; even good translations inevitably
deteriorated during successive transcriptions by men who did not fully
understand them. Discovering how Aristotle or Ptolemy had answered
a particular question was often difficult and Yet
medieval scholars repeatedly insisted on reconstructmg ancient thought
before venturing a judgment of their own. The brilliance, scope, and
coherence of their unexpected legacy dazzled men emerging from the
Dark Ages; they naturally felt that their first task .to assimilate
their heritage. Problems of interpretation and reumficatwn therefore
bulked large in scholastic thought. . . .
The task of the medieval scholar was further and artificially compli-
cated by a foreshortened historical perspective. He expected to
reestablish a broad and coherent system of knowledge modeled
Aristotle's and he did not always recognize that the "antiquity" from
whom system was to derive had had a number of
about a great many questions of detail. Though the scholastics found It
104
hard to recognize (often pleading errors of transmission or
Aristotle himself had not always been consistent. Nor had his con-
temporaries accepted all his views. Occasional equivocations and
tradictions had characterized the ancient tradition from the start. The1r
range had been immensely broadened by the Hellenistic and .Islamic
commentators, whose works, written during the fifteen centunes sep-
arating Aristotle from his European disciples, were recovered with, and
sometimes before, those of the master. To us these inconsistencies in the
tradition seem natural products of its evolution and transmission, but
to the medieval scholar they often appeared as internal contradictions
in a single body of knowledge, the hypothetical unit "ancient
In part because of this confusion, the comparison and reconc1hat10n
conflicting authorities became a distinctive characteristic of scholastic
thought. As we shall see more concretely later in this chapter, the re-
vived tradition of learning was less empirical, more verbal, logical, and
rational, than its ancient counterpart had been.
One of the inconsistencies embedded in the tradition played a
particularly significant role in the development of astronomy: the ap-
parent conflict between the spheres of Aristotelian cosmology and the
epicycles and deferents of Ptolemaic astronomy. Though we have not
previously noted it, these were really characteristic of t"':o
distinct ancient civilizations, the Hellenic and the Hellemstlc. Hellemc
civilization centered on the Greek mainland during the period when
Greece dominated the Mediterranean basin. The science to which it
gave rise was predominantly qualitative in and cosmo_Iogical
i'fl orientation. Aristotle was its greatest representative and also Its last.
Just before his death the evolution of Hellenic science came to a prema-
ture halt at the time when Alexander the Great conquered Greece and
joined it to a great empire embracing all of Asia Minor, Egypt, and
Persia to the Indus river. The Hellenistic civilization that emerged
after Alexander's conquests centered in commercial and cosmopolitan
metropolises like Alexandria. There scholars of many nations and races
merged elements of their diverse cultures to produce a science that
less philosophical, more mathematical and numerical, than its Hellemc
predecessor had been. Astronomy perfectly the contrast.
The cosmological framework of ancient astronomy IS largely a
of the Hellenic tradition which culminated in the works of Anstotle.
The mathematical astronomy of Hipparchus and Ptolemy belongs to
105
the tradition which, in astronomy, flourished only two
centunes and more after Aristotle's death.
The Hellenistic astronomers who measured the universe, cata-
logued the stars, and grappled with the problem of the planets were
clearly not indifferent to the cosmology developed by their Hellenic
But were they much concerned with cosmological
mm.utlae. They ndiCuled the authors of cosmologies that differed
radically the norm, and they occasionally wrote short
cosmological treatises of their own. Ptolemy himself composed a
thoroughly cosmological work, the Hypotheses of the Planets which
a rather unsatisfactory physical mechanism for
mohons. But when designing mathematical systems to predict
planetary position Hellenistic astronomers seldom worried about the
possibility of Constructing mechanical counterparts for their geometric
constructs. To them the physical reality of the spherical shells and the
mechanisms which kept the planets moving within them had become
at. most secondary In short, Hellenistic scientists acquiesced
Without discomfort 1ll a partial bifurcation of astronomy and
a mathematical technique for predicting plan-
positiOn did not have to conform entirely to the psychological
reqmrement of cosmological reasonableness.
In the sixteenth century this bifurcation offered an important prece-
dent to he too saw astronomy as essentially
ma.themahcal, the physical mcongruity of a moving epicycle in a
.of spheres c.ould provide a dim anticipation of the physical
mcongrmty of a movmg earth. But this was not the bifurcation's first
contribution. Four centuries earlier, at the time when Aristotle and
Ptolemy were first recovered by Europeans, it had also helped pave
road to revolution, though in a very different way. Because their
of the preceding centuries had telescoped their sense of
history, the scholastics viewed Aristotle and Ptolemy very nearly as
:hey appeared as exponents of a single tradition,
th: differences between their systems became
very hke mconsistencws m a single body of doctrine. Changes that
had as natural evolution of knowledge over the five
centunes separatmg him from Aristotle, often seemed contradictions to
the scholastics, and contradictions raised novel problems of reconcil' _
t' s. h
la
wn. mce t e passage of time proved reconciliation to be both difficult
106
and inconclusive, these apparent contradictions, like other conflicts in
medieval thought, finally helped to cast doubt upon the entire tra-
dition.
As in the Middle Ages, the ancient tradition of learning
had acqmred a new look, and the preceding pages indicate that some
of the important novelties derived from the mere necessity of revival.
But there were also more substantive changes in the revived tradition,
changes by indigenous characteristics of the Middle Ages
the Renaissance. For example, though science played a large part
m the thought of the later Middle Ages, the dominant intellectual
forces were theological, and the practice of science in a theological
milieu shifted both the strengths and the weaknesses of the scientific
tradition. Besides, medieval science was not itself static. Aristotle's
scholastic critics developed important alternatives for some of his doc-
trines, and a few of these alternatives played a major role in preparing
the way for Copernicus. By the sixteenth century still other forces -
intellectual, economic, and social- were at work, and among these
are some.with bearing upon the problems of astronomy and of
the earths motion. These changes demand an independent treatment
to which we now turn.
'
Astronomy and the Church
Throughout the Middle Ages and much of the Renaissance
the Catholic Church was the dominant intellectual authority of all
Eu.rope: European scholars were members of the clergy; the
umvers1ties m which ancient learning was assembled and studied were
Chur&h schools. From the fourth century to the seventeenth the
Church's attitude toward science and about the structure of the uni-
verse was a determining factor in the progress or stagnation of astron-
omy. But the Church's attitude and its practice were not uniform
during these centuries. After the Dark Ages the Church began to
support a learned tradition as abstract, subtle, and rigorous as any
t?e world has known . But before the tenth century and again after the
sixteenth the Churchs influence was, on balance, antiscientific. The
Copernican theory evolved within a learned tradition sponsored and
supported by the Church; Copernicus himself was the nephew of a
bishop and a canon of the cathedral at Frauenburg; yet in 1616 the
Church banned all books advocating the reality of the earth's motion.
107
No single generalization will describe the Church's overwhelming in-
fluence upon science, for the influence changed with the changing
situation of the Church.
In the early centuries of the Christian era the Church fathers were
crusaders and proselytizers for a new faith, fighting for its very exist-
ence. Their calling itself demanded that they deprecate the pagan
learning of their predecessors and maximize the attention given to the
problems of Christian theology by the rapidly contracting learned
world. In addition, they deeply believed that Scripture and Catholic
exegesis contained the sum of knowledge necessary for salvation. To
them science was secular learning. Except when essential for daily
life it was useless at best, dangerously distracting at worst. Therefore,
in his Enchiridion or handbook for Christians, St. Augustine, the most
influential of the early Church fathers, counseled the faithful as
follows:
When, then, the question is asked what we are to believe in regard to
religion, it is not necessary to probe into the nature of things, as was done
by those whom the Greeks call physici; nor need we be in alarm lest the
Christian should be ignorant of the force and number of the elements, - the
mvtion, and order, and eclipses of the heavenly bodies; the form of the
heavens; the species and the natures of animals, plants, stones, fountains,
rivers, mountains; about and distances; the signs of coming
storms; and a thousand other things which those philosophers either have
found out, or think they have found out. . . . It is enough for the
Christian to believe that the only cause of all created things, whether
heavenly or earthly, whether visible or invisible, is the goodness of the
Creator, the one true God; and that nothing exists but Himself that does
not derive its existence from Him.l
This attitude was not incompatible with an admiring knowledge of
ancient learning, at least not before the Moslem invasions. Augustine
himself had read Greek science attentively, and his writings testify to
his admiration of its accuracy and scope. But his attitude was incom-
patible with the active pursuit of scientific problems, and it readily
lent itself to further negative elaboration. In the writings of Augustine's
less liberal contemporaries and successors, his spiritual depreciation of
pagan science was usually coupled with an outright rejection of its con-
tent. Astronomy, because of its ties to astrology, was particularly
scorned, for the explicit determinism of astrology made it seem incom-
patible with Christian doctrine.
At the beginning of the fourth century, for example, Lactantius,
tutor to the son of the Emperor Constantine, devoted the third book
of his Divine Institutions to "the false wisdom of the philosophers" and
allowed himself one chapter to ridicule the concept of the spherical
earth. For him it was sufficient to point to the absurdity of a region in
which men hung head down and to the impossibility of the heavens'
being below the earth. Later in the century the Bishop of Cabala
achieved the same effect with Biblical evidence. The heavens are not
a sphere, but a tent or tabernacle, for "it is He . . . that stretcheth out
the heavens as a curtain, and spreadeth them out as a tent to dwell in"
(Isaiah 40:22). There are "waters . . . above the firmament" (Genesis
1:7). The earth is flat, for "the sun was risen upon the earth when Lot
entered into Zoar" (Genesis 19:23). By the middle of the sixth century
Kosmas, an Alexandrian monk, could replace the pagan system with
a detailed Christian cosmology derived primarily from the Bible. His
universe is shaped like the tabernacle that the Lord instructed Moses to
build in the wilderness. It has a flat bottom, perpendicular sides, and a
semicylindrical roof, like an old-fashioned traveling trunk. The earth,
the footstool of the Lord, is a rectangular plane, twice as long as it is
broad, and resting on the flat bottom of the universe. The sun does not
travel below the earth at night, but is hidden behind its northernmost
portions, which are higher than the regions to the south.
The cosmologies of men like Lactantius and Kosmas never became
official Church doctrine. Nor did they entirely supplant the ancient
universe of spheres which survived in fragmentary descriptions within
the more erudite medieval encyclopedias. There was no Christian
about cosmology during the first half of the Middle Ages.
Science and cosmology were not significant enough to demand it. But
though these cosmologies, compounded from the nai:vest sense per-
ceptions plus a smattering of Scripture, were never official, they are
representative. They illustrate the decadence of secular learning that
characterized the Dark Ages, and they therefore prepare us for the
surprise and awe with which later Christian scholars greeted the re-
discovery of ancient knowledge during the eleventh and twelfth
centuries.
By the time that Christian Europe reestablished commercial and
cultural ties with the Eastern Church in Byzantium and with the
Moslems of Spain, Syria, and Africa, the Church's attitude toward
pagan wisdom had changed. The main areas of continental Europe had
been converted; the Church's intellectual and spiritual authority was
complete; the hierarchy of ecclesiastical administration was fixed.
Pagan and secular learning were no longer a threat, provided that
the Church could maintain intellectual leadership by absorbing them.
Churchmen therefore devoted some of the leisure provided by new-
found prosperity to a vigorous pursuit of the rediscovered learning.
By broadening the range of knowledge acceptable for Christian
scholarship, they preserved for five more centuries the Catholic
monopoly of learning. In the twelfth century "the nature of things,"
including the heavens and the earth, again b0came a suitable topic for
intensive study. By the thirteenth, if not earlier, the main outlines of
the two-sphere universe were once more taken for granted in the dis-
cussions of educated Christians. During the last centuries of the
Middle Ages the setting of Christian life, both terrestrial and celestial,
was a fully Aristotelian universe.
We have been calling the process by which Christians discovered
that they lived in an Aristotelian universe a recovery of ancient learn-
ing, but "recovery" is clearly an inadequate word. vVhat occurred was
far more nearly a revolution in both Christian thought and the ancient
scientific tradition. From the fourth century on, Aristotle, Ptolemy, and
other Greek writers had been attacked by Churchmen because of the
conflict between their cosmological opinions and Scripture. Those con-
flicts still existed in the twelfth and thirteenth centuries, and they were
recognized. In 1210 a provincial council at Paris prohibited thE teach-
ing of Aristotelian physics and metaphysics. In 1215 the Fourth
Lateran Council issued a similar, though more restricted, anti-Aris-
totelian edict. Other interdictions issued from the papacy throughout
the century. They were unsuccessful, winning lip service alone, but
they are not insignificant. The edicts testify to the impossibility of
simply adding ancient secular learning to the existing body of medieval
theology. Both ancient texts and Scripture required modification in the
creation of a new fabric of coherent Christian doctrine. When the
new fabric was completed, theology had become an important bulwark
for the ancient concept of a central stationary earth.
The physical and cosmological structure of the new Christian uni-
verse was predominantly Aristotelian. St. Thomas Aquinas ( 1225-
1274), the scholastic who contributed most to the final pattern of the
110
fabric, describes the perfection and appropriateness of the celestial
motions in words that, except for their clarity, might have been written
by Aristotle himself:
It is therefore clear that the material of the heavens is, by its intrinsic
nature, not susceptible to generation and corruption, since it is the primary
sort of alterable body and closest in its nature to those bodies which are
intrinsically changeless. [The only truly changeless body in the Christian
universe is God, from whom all change on earth and in the heavens derives.]
That is why the heavens experience only the absolute minimum of altera-
tion. Motion is the only sort of change they experience, and this sort of
alteration [unlike change of size, weight, color, and so on] does not modify
their intrinsic nature in the least. Furthermore, among the sorts of motion
which they might experience, theirs is circular, and circular motion is the
one which produces the very minimum of alteration because the sphere as a
whole does not change place.
2
Aristotle could not always be embraced quite so literally. Many
scholastics felt forced, for example, to abandon his proof of the abso-
lute impossibility of a void, because it seemed arbitrarily to limit God's
infinite power. No Christian could accept Aristotle's view that the
universe had always existed. The first words of the Bible are, "In the
beginning God created the heaven and the earth." Besides, the Crea-
tion was an essential ingredient in the Catholic explanation of the
existence of evil. On a matter of this significance Aristotle had to give
way; the universe had been created at a determinate first instant of
time. But more often the Bible gave way, usually to metaphorical
interpretation. For example, in discussing the scriptural text, "Let
there be a firmament made amidst the waters; and let it divide the
from the waters" (Genesis 1:6), Aquinas first outlined a cosmo-
logical theory that would preserve the literal sense of the passage and
then said:
As, however, this theory can be shown to be false by solid reasons, it
cannot be held to be the sense of Holy Scripture. It should rather be con-
sidered that Moses was speaking to ignorant people, and that out of con-
descension to their weakness he put before them only such things as are
apparent to sense. Now even the most uneducated can perceive by their
senses that earth and water are corporeal, whereas it is not evident to all that
air also is corporeal. . . . Moses, then, while he expressly mentions water
and earth, makes no express mention of air by name, to avoid setting before
ignorant persons something beyond their knowledge.a
By reading "water" as "air" or "transparent substance" the integrity
of Scripture is preserved. But in the process the Bible becomes, in some
sense, a propaganda instrument, composed for an ignorant audience.
The device is typical, the scholastics employed it again and again.
The painstaking thoroughness with which Aquinas and his con-
temporaries attacked the task of reconciliation is illustrated by the diffi-
culties they discovered in the Biblical account of the Ascension. Ac-
cording to Scripture Christ "ascended up far beyond all heavens, that
he might fill all things" (Ephesians 4:10). Aquinas succeeded in fitting
this bit of Christian history into a universe of spheres, but to do so he
had to resolve many varied problems, among which were the follow-
ing:
It seems that it was not fitting for Christ to ascend into heaven. For the
Philosopher [Aristotle] says (On the Heavens, Book II) that things which
are in a state of perfection possess their good without movement. But Christ
was in a state of perfection .... Consequently, He has His good without
movement. But ascension is movement. Therefore it was not fitting for Christ
to ascend ....
Further, there is no place above the heavens, as is proved in On the
Heavens, I. But every body must occupy a place. Therefore Christ's body
did not ascend above all the heavens. . . .
Further, two bodies cannot occupy same place. Since, then, there is
no passing from place to place except through the middle space, it seems
that Christ could not have ascended above all the heavens unless [the
crystal spheres of] heaven were divided; which is impossible.
4
Aquinas' answers need not concern us. It is the objections them-
selves that are astounding, particularly since the Ascension is only one
of the aspects of Christian history to present problems and since
Aquinas is only the greatest of the many Churchmen concerned with
them. Aquinas' Summa Theologica, from which most of the preceding
quotations were taken, is a compendium of Christian knowledge, often
printed in twelve fat volumes. In all of them Aristotle's name (or his
more revealing designation as "The Philosopher") occurs again and
again. Only through a multiplicity of works like these could ancient
learning, particularly Aristotelian learning, have again become a
foundation for Western thought.
Aquinas and his thirteenth-century contemporaries certified the
compatibility of Christian belief with much of ancient learning. By
making Aristotle orthodox they licensed his cosmology to become a
112
creative element in Christian thought. But the very detail and erudi-
tion of their works obscured the over-all structure of the new Christian
universe that was emerging late in the Middle Ages. If we are to
understand the pregnant meaning which gave that universe, including
the central, stable earth, its hold upon the medieval and Renaissance
mind, we require a more comprehensive view. That larger view can
scarcely be discovered in the thirteenth century at all. It evolved only
after Aristotle had been licensed, appearing perhaps first and certainly
most forcefully in the works of the Italian poet Dante, particularly in
his great epic, the Divine Comedy.
Taken literally, Dante's epic is a description of the poet's journey
through the universe as conceived by the fourteenth-century Christian.
The journey begins on the surface of the spherical earth; descends
gradually into the earth via the nine circles of Hell which symmetri-
cally mirror the nine celestial spheres above; o and arrives at the vilest
and most corrupt of all regions, the center of the universe, the ap-
propriate locus of the Devil and his legions. Dante then returns to the
surface of the earth at a point diametrically opposite the one where
he had entered, and there he finds the mount of purgatory with its
base on the earth and its top extending into the aerial regions above.
Passing through purgatory, the poet travels through the terrestrial
spheres of air and fire to the celestial region above. At the last he
journeys through each of the celestial spheres in turn, conversing with
the spirits that inhabit them, until finally he contemplates God's Throne
in the last, the Empyrean, sphere. The setting of the Divine Comedy
is a literal Aristotelian universe adapted to the epicycles of Hipparchus
a'hd the God of the Holy Church.
For the Christian, however, the new universe had symbolic as well
as literal meaning, and it was this Christian symbolism that Dante
wished most of all to display. Through allegory his Divine Comedy
made it appear that the medieval universe could have had no other
structure than the Aristotelian-Ptolemaic. As he portrays it, the uni-
verse of spheres mirrors both man's hope and his fate. Both physically
0
The ninth sphere, which appears throughout medieval astronomy, was added
to the eight spheres of ancient cosmology by Moslem astronomers in order to
account for the precession of the equinoxes, and the motion of the pole of the
heavens (see the Technical Appendix, Section 2). In the Moslem system it is the
ninth sphere that rotates every 24 hours, as the sphere of the stars had done in the
older system.
113
and spiritually man occupies a crucial intermediate position in this
universe filled, as it is, by a hierarchical chain of substances that
stretches from the inert clay of the center to the pure spirit of the
Empyrean. Man is compounded of a material body and a spiritual soul:
all other substances are either matter or spirit. Man's location, too, is
intermediate: the earth's surface is close to its debased and corporeal
center but within sight of the celestial periphery which surrounds it
symmetrically. Man lives in squalor and uncertainty, and he is very
close to Hell. But his central location is strategic, for he is everywhere
under the eye of God. Both man's double nature and his intermediate
position enforce the choice from which the drama of Christianity is
compounded. He may follow his corporeal, earthy nature down to its
natural place at the corrupt center, or he may follow his soul upward
through the successively more spiritual spheres until he reaches God.
As one critic of Dante has put it, in the Divine Comedy the "vastest
of all themes, the theme of human sin and salvation, is adjusted to the
great plan of the universe."
5
Once this adjustment had been achieved,
any change in the plan of the universe would inevitably affect the
drama of Christian life and Christian death. To move the earth was to
break the continuous chain of created being.
No aspect of medieval thought is more difficult to recapture than
the symbolism that mirrored the nature and fate of man, the micro-
cosm, in the structure of the universe, which was the macrocosm. Per-
haps we can no longer grasp the full religious significance with which
this symbolism clothed the Aristotelian spheres. But we can at least
avoid dismissing it as mere metaphor or supposing that it had no active
role in the Christian's nonastronomical thought. One of Dante's prose
works, written in part as a technical handbook to aid his contempor-
aries in interpreting his verse, concludes a literal physical description
of the spheres and epicycles employed in medieval astronomy with the
following words:
However, beyond all these [crystalline spheres], the Catholics place the
Empyrean Heaven . . . ; and they hold it to be immovable, because it has
within itself, in every part, that which its matter demands. And this is the
reason that the Primum Mobile [or ninth sphere] moves with immense
velocity; because the fervent longing of all its parts to be united with those
of this most quiet heaven, makes it revolve with so much desire that its
velocity is almost incomprehensible. And this quiet and peaceful heaven is
the abode of that Supreme Deity who alone doth perfectly behold himself.6
114
In this passage the astronomer charts the position (and elsewhere
the dimensions) of God's abode. He has become theologian, and in the
fourteenth and fifteenth centuries the astronomer's theological func-
tions did not always end with the measurement of heaven. Dante and
some of his contemporaries also turned to astronomy to discover the
kinds and occasionally even the numbers of the angelic inhabitants of
God's spiritual realm.
Dante himself outlines one typical medieval theory of the relation
between the spiritual hierarchy and the spheres in a passage from The
Banquet immediately following the description of the spheres referred
to above:
Since it has been demonstrated in the preceding chapter what this
heaven is, and how it is ordered within itself, it remains to show who they
are who move it. Therefore be it known, in the first place, that these are
Substances separate from matter, that is, Intelligences, whom the common
people call Angels .... The numbers, the orders, the hierarchies [of these
angelic beings], are recounted by the movable heavens, which are nine;
and the tenth announces the unity and stability of God. And therefore the
psalmist says, "The heavens recount the glory of God, and the firmament
announceth the work of His hands."
\Vherefore it is reasonable to believe that the motive powers [that is, the
beings who move the spheres] of the Heaven of the Moon are of the order
of Angels; and those of Mercury, Archangels; and those of Venus are the
Thrones. . . . And these Thrones, which are allotted to the government of
this heaven [of Venus], are not many in number, and the astrologers [or
astronomers] differ about their number, according to their differences about
the revolutions [of this heaven], although all are agreed in this, that their
nmnber is equal to that of these revolutions; which, according to the Book
of the Aggregation of the Stars . . . are three: one by which the star
revolves within its epicycle, the second by which the epicycle and the whole
heaven revolves equally with that of the Sun, and the third by which all that
heaven revolves, following the [precessional] motion of the stellar sphere
from west to east, one degree in a hundred years. So that for these three
motions are three motive powers [which are three members of the angelic
order of Thrones].
7
When angels become the motive force of epicycles and deferents,
the variety of spiritual creatures in God's legion may increase with the
complexity of astronomical theory. Astronomy is no longer quite sep-
arate from theology. Moving the earth may necessitate moving God's
Throne.
115
Aristotle's Scholastic Critics
The effects of medieval scholarship were not all so conserva-
tive as the integration that made theology a bulwark for the universe
of spheres. Aristotle and his commentators were the invariable starting
point for scholastic research, but they were often no more than that.
The very intensity with which Aristotelian texts were studied
gua.ranteed that inconsistencies of doctrine or proof would be quicklY.
noticed, and these inconsistencies were often the seeds of important
creative achievements. l\ledieval scholars caught scarcely a glimpse of
the new astronomy and cosmology produced by their sixteenth- and
seventeenth-century successors. But they extended Aristotle's logic,
discovered fallacies in his proofs, and rejected many of his explanations
because they failed the test of experience. In the process they forged
a number of the concepts and tools that proved essential to the ac-
complishments of men like Copernicus and Galileo.
Important anticipations of Copernican thought can be found, for
example, in the critical commentary on Aristotle's On the Heavens
written during the fourteenth century by Nicole Oresme, a member of
the important Parisian nominalist school. Oresme' s method is typically
scholastic. In his long manuscript Aristotle's text is broken into frag-
ments, each a few sentences in length, and these fragments are inter-
spersed with long critical or explanatory comments. On finishing the
work a reader discovers that Oresme agrees with Aristotle on ahnost
every substantive point except the Creation. But the reasons for his
agreement are far from clear; Oresme's brilliant critique has destroyed
many of Aristotle's proofs and suggested important alternatives for a
number of Aristotelian positions. These alternatives were seldom
adopted by the scholastics themselves. But medieval scholars con-
tinued to discuss them, and that discussion helped to create a climate
of opinion in which astronomers could imagine experimenting with
the idea of a moving earth.
Oresme was, for example, quite critical of Aristotle's principal argu-
ment for the earth's uniqueness.
8
Aristotle had said that if there were
two earths in space (and when the earth becomes a planet, there will
be six "earths"), then they would both fall to the center of the universe
and coalesce, because earth moves naturally to the center. This proof,
says Oresme, is invalid, because it presupposes a theory of motion that
is itself unproved. Perhaps earth does not move naturally to the center,
but simply to other nearby bits of earth. Our earth has a center, and it
may be to that center, wherever in the universe it is located, that loose
stones return. On this alternative theory the natural motion of a body
is governed, not by its position in an absolute Aristotelian space, but
by its position relative to other portions of matter. Some such theory
was prerequisite to the new cosmologies of the sixteenth and seven-
teenth centuries, cosmologies in which the earth was neither unique
nor at the center. Similar theories, in vari011s disguises, are common to
the work of Copernicus, Galileo, Descartes, and Newton.
Even more important anticipations of Copernican arguments
emerge when Oresme criticizes Aristotle's refutation of Heraclides, the
Pythagorean who had explained the diurnal motion of the stars by
positing an eastward axial rotation of the central earth. Oresme does
not believe that the earth rotates, at least he says he does not. But he
is concerned to show that the choice between a stable and a rotating
earth must be made on faith. No argument, he says, whether logical,
physical, or even scriptural, can disprove the possibility of the earth's
diurnal rotation. For example, nothing can be concluded from the ap-
parent motion of the stars, because, says Oresme:
I suppose that local motion can be perceived only when one body alters
its position relative to another. Therefore if a man in a smoothly riding boat,
a, which is moved either slowly or rapidly, can see nothing but a second
boat, b, which moves in just the same way as a, . . . then I say that it will
seem to him that neither boat is moved. And if a is at rest and b moves, it
will seem to him that b moves; and if it is a that moves and b is at rest, it will
still seem to him that a is at rest and that b is moved. . . . Therefore I say
that if the higher [or celestial] of the two parts of the universe mentioned
above were today moved with a diurnal motion, as it is, while the lower
[or terrestrial] part remained at rest, and if tomorrow on the contrary the
lower part were moved diurnally while the other part, i.e., the heavens were
at rest, we would be unable to see any change, but everything would seem
the same today and tomorrow. It would seem to us throughout that our loca-
tion was at rest while the other part of the universe moved, just as it seems
to a man in a moving boat that the trees outside the boat are in motion.9
This is the argument from optical relativity that plays such a large
part in the writings of Copernicus and Galileo. Oresme does not stop
with it, however. His treatise proceeds immediately to the demolition
of an even more important Aristotelian argument, the one that con-
117
eludes for immobility because an object thrown vertically upward
always returns to the point on earth from which it departed:
[In response to Aristotle's and Ptolemy's argument] one may say that an
arrow shot straight into the air is [also] moved rapidly eastward with the air
through which it passes and with the whole mass of the bottommost [or
terrestrial] portions of the universe described above, the whole [earth and
air and arrow] being moved with a daily rotation. Therefore the arrow
returns to the spot on the earth from which it was shot. This appears possi-
ble by analogy: if a man were on a ship moving rapidly eastward without his
being aware of its motion, and if he drew his hand rapidly downward, des-
cribing a straight line against the mast of the ship, it would seem to him
that his hand had only a vertical motion; and the same argument shows
why the arrow seems to us to go straight up or down,lO
Galileo's famous defense of the Copernican system, the Dialogue
on the Two Principal Systems of the World, is filled with arguments of
exactly this sort, and Galileo may well have been elaborating hints
derived from Copernicus' scholastic predecessors, including Oresme.
But that does not make Oresme a Copernicus. He does not conclude
for even the diurnal rotation of the earth; he does not dream of an
orbital motion about the center of the universe; and he has no notion
of the benefits that astronomers might derive by positing a mobile
earth. He does not, for that matter, even share Copernicus' motive,
and its absence makes his work all the more astounding. When
Oresme's arguments recur in the writings of Copernicus and Galileo,
they have a different and a more creative function. The later scientists
wished to show that the earth could move in order to exploit the ad-
vantages that would accrue to astronomy if, in fact, it did move.
Oresme wished to show only that the earth could move; he was in-
vestigating only Aristotle's proof. Like many of the other most fruitful
contributions of scholastic science, his Copernican arguments are
products of the preeminence that the late medieval mind accorded
Aristotle. Men who agreed with Aristotle's conclusions investigated
his proofs only because they were proofs executed by the master.
Nevertheless their investigations often helped to ensure the master's
ultimate overthrow.
We cannot, of course, be certain that Copernicus or Galileo read
Oresme. The tradition that requires the scholar and scientist to name
his sources was not established until long after the scientific revolution
118
of the sixteenth and seventeenth centuries. But Aristotle had many
scholastic critics, and they wrote numerous manuscripts which were
repeatedly copied in the years after their deaths. Five and a half
centuries after the composition of Oresme's commentary there are still
six extant medieval manuscript copies, several dating from the fifteenth
century, after Oresme's death. At the time of Copernicus' birth there
must have been many more. Furthermore, the tradition of scholastic
criticism was a continuous tradition. Key concepts which originated
at Paris in the fourteenth century can be traced to Oxford in the same
century and to Padua in the fifteenth and sixteenth. Copernicus studied
at Padua and Galileo taught there. Though we cannot be sure that
Copernicus derived any particular argument in the De Revolutionibus
from any particular scholastic critic, we cannot doubt that the critics
as a group facilitated the production of those arguments. At the very
least they created a climate of opinion in which topics like the earth's
motion were legitimate subjects for university discussion. Quite prob-
ably a few of Copernicus' key arguments were borrowed from earlier
and unacknowledged sources.
The preceding discussion of Oresme has illustrated the most typical
sort of scholastic criticism: the testing of Aristotelian proofs and the
investigation of possible alternative doctrines, usually discarded once
their logical possibility was demonstrated. But not all medieval science
was of this limited critical and perhaps evanescent variety. The scholas-
tics also introduced a few new areas of investigation and a few perman-
ent doctrinal modifications into the Aristotelian scientific tradition.
Among these the most significant were in the fields of kinematics and
dynamics whose subject is the motion of heavy bodies on the earth and
(after the Middle Ages) in the heavens. Some of Galileo's most sig-
nificant contributions, particularly his work on falling bodies, can ap-
propriately be viewed as a creative reordering of previously scattered
physical and mathematical insights gained with difficulty by medieval
scholars. But even before the seventeenth century, when Galileo wove
them into a new dynamics, one of these insights, the impetus theory
of motion, had had an important, if indirect, bearing on astronomical
thought.
The impetus theory was erected on the rubble of one of the weakest
explanations in the body of Aristotle's physics, the explanation of pro-
jectile motion. Aristotle had believed that, unless it was moved by an
119
external push, a stone would either remain at rest or move in a straight
line toward the center of the earth. It was a natural explanation for a
large number of phenomena, but it could not easily be fitted to the
observed behavior of a projectile. When released from the hand or
from a sling, a stone does not drop straight to earth. Rather it continues
to mo.ve in the direction toward which it was initially impelled even
aft:r Its contact with the initial projector (hand or sling) is broken.
Anstotle, who a shrewd observer, knew how a projectile moves,
and he patched his theory by conceiving the disturbed air as the source
of a push which prolongs the projectile's motion after contact with the
fir.st is broken. He does not seem to have been too pleased
this solutwn, for he provided at least two incompatible versions
It, and he was always a bit argumentative on this point. But for him
It was never his fundamental interests lay elsewhere; he
the projectile only as an aside and apparently only because it
might create difficulty for his theory.
.It did create difficulties, apparently almost immediately. John
Philoponus, the sixth-century Christian commentator who records the
:arliest extant of Aristotle's theory, attributes his own partial
nnpetus-theory solution to the Hellenistic astronomer Hipparchus.
commentators were at least troubled by this aspect of
Aristotle s thought. Perhaps no one, including its author ever took th
.
, e
arr as pusher quite seriously. But it was not until the fourteenth
c:ntury, when difficulties in Aristotle's text were problems in their own
nght, the is.sue the projectile was fully faced and resolved by a
modificatiOn of Aristotle's theory. Though its source was a
terrestrial problem, that modification proved to have immediate im-
plications for astronomy.
Both the problem and its medieval solution can be recovered in
brilliant detail from the Questions on the Eight Books of Aristotle's
Physics (a typical title for scholastic science) by Oresme' s teacher
Jean Buridan:
'
. is sought a projectile after leaving the hand of the projector
1s moved the arr or by what it is moved. . . . This question I judge to
be very difficult because Aristotle, as it seems to me, has not solved it well.
h: : [at one point] holds that the projectile swiftly leaves the place
m ':hlCh 1t was, and nature, not pennitting a vacuum, rapidly sends air in
behmd to fill up the vacuum. The air moved in this way impinges upon
the projectile and impels it along. This is repeated continually for a certain
distance .... [But] it seems to me that many experiences show this method
of proceeding to be valueless. . . .
[For example, one among many that Buridan gives,] a lance having a
conical posterior as sharp as its anterior would be moved after projection
just as swiftly as it would be without a sharp conical posterior. But surely
the air following could not push on a sharp end in this way since the air
would be easily divided by the sharpness [whereas it could push on a
blunt end, thus moving the lance with the blunt end farther]. . . .
Thus we can and ought to say that in the stone or other projectile
there is impressed something which is the motive force of that projectile.
This is evidently better than falling back on the statement that the air
would [continue to] move the projectile, for the air appears to resist. . . .
[The projector] impresses a certain impetus or motive force into the
moving body, which impetus acts in the direction toward which the mover
moved the moving body, either up or down, or laterally or circularly. And
by the amount the motor moves that moving body more swiftly, by the
same amount it will impress in it a stronger impetus. It is by that impetus
that the stone is moved after the projector ceases to move. But that impetus
is continually decreased by the resisting air and by the gravity of the stone
which inclines it in a direction contrary to that in which the impetus was
naturally predisposed to move it. Thus the movement of the stone continually
becomes slower until the impetus is so diminished or corrupted that the
gravity of the stone wins out over it and moves the stone down to its natural
place.ll
This is only a fraction of Buridan's elaborate discussion, and count-
less parallel treatments can be found in the works of his successors.
By the end of the fourteenth century impetus dynamics, in one of a
number of forms very like Buridan's, had replaced Aristotelian dynam-
ics in the work of the principal medieval scientists. The tradition en-
dured: it was taught at Padua around the time Copernicus studied
there; Galileo learned it from his master Bonamico at Pisa. Both of
them used it, explicitly or implicitly, as did their contemporaries and
successors. On a number of occasions and in a variety of ways the
impetus theory played an essential role in the Copernican Revolution.
One of these roles we have already seen, though without recogniz-
ing it. Oresme's refutation of Aristotle's central argument for the earth's
immobility takes the impetus theory, or something quite like it, for
granted. On the Aristotelian theory of motion a vertically thrown
stone must move along a radius fixed in space. If the earth moves while
the stone is in the air, the stone (or arrow) cannot accompany it and
will therefore not return to its point of departure. But It the earth's
eastward motion endows the stone with an eastward impetus while the
stone is still in contact with the projector, that impetus will endure and
will cause the stone to pursue the moving earth even after contact is
broken. The impetus theory enables the moving earth to endow terres-
trial bodies with an internal propellant, and that propellant enables
them to follow the earth afterward. Like his master Buridan, Oresme
believed in the impetus theory, and though his refutation of Aristotle
does not mention the theory explicitly, the refutation makes no sense
without it. In one way or another the impetus theory is implicated in
most of the arguments, both medieval and Renaissance, that make it
possible to move the earth without leaving terrestrial bodies behind.
Some adherents of the impetus theory immediately extended it
from the earth to the heavens. In the process they took a second long
step toward the Copernicanism that was to come. Buridan himself
said, in a passage that closely follows the preceding excerpt from his
Questions:
Also, since the Bible does not state that appropriate [angelic] intelli-
gences move the celestial bodies, it could be said that it does not appear
necessary to posit intelligences of this kind. For it could [equally well]
be answered that God, when He created the world, moved each of the
celestial orbs as He pleased, and in moving them He impressed in them
impetus which moved them without His having to move them any more
except by the method of general influence whereby he concurs as a coagent
in all things which take place. Thus on the seventh day He rested from all
work which he had executed by committing to others the actions and the
passions in tum. And these impetuses which He impressed in the celestial
bodies were not decreased nor corrupted afterwards, because there was no
inclination of the celestial bodies for other movements. Nor was there
resistance which would be corruptive or repressive of that

In Buridan' s writings, perhaps for the first time, the heavens and
the earth were at least tentatively subjected to a single set of laws, and
the same suggestion was carried further by Buridan's pupil, Oresme.
He suggested that, "when God created [the heavens] . . . , He im-
pressed them with a certain quality and force of motion, just as He
impressed terrestrial things with weight ... ; it is just the same as
a man building a clock and leaving it to run itself. Thus God left the
heavens to be moved continually . . . according to the order [He
had] established."
13
But to conceive the heavens as a terrestrial
122
mechanism, a piece of clockwork, is to break the absolute dichotomy
between the superlunary and sublunary regions. Though the impetus
theorists followed the suggestion no further, at least during the Middle
Ages, it was just this dichotomy, drawn from both Aristotle and the-
ology, that had to be broken if the earth was to be made a planet.
The possibility of the earth's motion and the partial unification of
terrestrial and celestial law were the impetus theory's two most direct
contributions to the Copernican Revolution. Its most important contri-
bution, however, was an indirect one, to which we shall revert briefly in
the last chapter. Through its role in the evolution of Newtonian
dynamics, the impetus theory helped to bring the Revolution to a suc-
cessful close more than a century after Copernicus' death. Copernicus
in the sixteenth century provided only a new mathematical description
of the way the planets move; he was not successful in explaining why
the planets moved as he said they did. Initially his mathematical as-
tronomy made no physical sense, and it therefore posed new sorts of
problems for his successors. Those problems were only resolved by
Newton, whose dynamics supplied the missing keystone to Copernicus'
mathematical system, and Newton's dynamics, even more than Coper-
nicus' astronomy, depended on the prior scholastic analyses of motion.
Impetus dynamics is not Newtonian dynamics, but by pointing to
new problems, new variables, and new abstractions impetus dynamics
helped to pave the way for Newton's work. Before the impetus theory
both Aristotle and experiment had testified that only rest endures.
Buridan and some other impetus theorists declared that, unless re-
sisted, motion too would endure forever, and they thus took a long
step toward what we now know as Newton's First Law of Motion.
A g a i ~ in a passage omitted from the descriptive quotation above,
Buridan equated the quantity of impetus in a moving body with the
product of the body's speed and its quantity of matter. The concept
of impetus became very like, though not identical with, the modern
concept of momentum; in Galileo's work the words "impetus" and
"momentum" are often used interchangeably. Elsewhere, to give one
final example, Buridan's discussion came very close to saying that the
gravity (or weight) of a freely falling body impresses equal incre-
ments of impetus (and therefore of velocity) upon the body in equal
intervals of time. Galileo was not the first of Buridan's successors to
say precisely this and to derive from it, with the aid of other analytical
devices supplied by the scholastics, the modern quantitative relation
between time of fall and distance. Contributions like these give scholas-
tic science an important role in the evolution of Newtonian dynamics,
and Newtonian dynamics was the keystone in the structure of the new
universe created by Copernicus and his successors.
During the seventeenth century, just when its full utility was being
demonstrated for the first time, scholastic science was bitterly attacked
by men trying to weave a radically new fabric of thought. The scholas-
tics proved easy to ridicule, and the image has stuck. Medieval scien-
tists more often found their problems in texts than in nature; many
of those problems do not now seem problems at all; by modern stand-
ards the practice of science during the Middle Ages was incredibly
inefficient. But how else could science have been reborn in the West?
The centuries of scholasticism are the centuries in which the tradition
of ancient science and philosophy was simultaneously reconstituted,
assimilated, and tested for adequacy. As weak spots were discovered,
they immediately became foci for the first effective research in the
modern world. The great new scientific theories of the sixteenth and
seventeenth centuries all originate from rents torn by scholastic criti-
cism in the fabric of Aristotelian thought. Most of those theories also
embody key concepts created by scholastic science. And more im-
portant even than these is the attitude that modern scientists inherited
from their medieval predecessors: an unbounded faith in the power
of human reason to solve the problems of nature. As the late Professor
Whitehead remarked, "Faith in the possibility of science, generated
antecedently to the development of modern scientific theory, is an
unconscious derivative from medieval theology."
14
Astronomy in the Age of Copernicus
In discussing late-medieval modifications of the Aristotelian-
Ptolemaic tradition, we have said almost nothing about developed
planetary astronomy. In fact there was almost none in Europe during
the Middle Ages, partly because of the intrinsic difficulty of the mathe-
matical texts and partly because the problem of the planets seemed
so esoteric. Aristotle's On the Heavens described the entire universe
in relatively simple terms; Ptolemy's elaborate Almagest dealt, for the
most part, only with the computation of planetary position. Therefore,
though the works of both Aristotle and Ptolemy were translated by
124
the end of the twelfth century, Aristotelian logic, philosophy, and
cosmology were assimilated far more rapidly than developed Ptolemaic
astronomy. Thirteenth-century metaphysics rivals Aristotle's in pro-
fundity; fourteenth-century physics and cosmology exceed Aristotle's
in depth and logical coherence. But Europeans produced no indige-
nous astronomical tradition to rival Ptolemy's until the middle of the
fifteenth century, if then. The first widely known European astronomi-
cal treatise, written around 1233 by John of Holywood, slavishly copied
an elementary Arabic treatise and devoted only one chapter to the
planets compared with Ptolemy's nine. The next two centuries pro-
duced only commentaries on Holywood's book and a few unsuccessful
rivals. Until two decades before Copernicus' birth in 1473 there was
little concrete evidence of technically proficient planetary astronomy.
Then it appeared in works like those of the German Georg Peuerbach
( 1423-1461) and his pupil Johannes Muller ( 1436-1476).
To Europeans of Copernicus' generation planetary astronomy was,
therefore, almost a new field, and it was practiced in an intellectual
and social environment quite different from any in which astronomy
had been practiced before. In part that difference arose from the theo-
logical accretions to the astronomical tradition that we have examined
in the work of Aquinas and Dante. Even more essential changes were
produced by the logical and cosmological criticism of men like Buridan
and Oresme. But these were medieval contributions, and Copernicus
did not live during the Middle Ages. His lifetime, 1473--1543, occu-
pied the central decades of the Renaissance and Reformation, and
novelties characteristic of this later age were also effective in inaugu-
rating and shaping his work.
Since stereotypes are most readily discarded during periods of
general ferment, the turbulence of Europe during the Renaissance
and Reformation itself facilitated Copernicus' astronomical innova-
tion. Change in one field decreases the hold of stereotypes in others.
Radical innovations in science have repeatedly occurred during periods
of national or international convulsion, and Copernicus' lifetime was
such a period. The Moslems again threatened to absorb vast areas of
a Europe now convulsed by the dynastic rivalries through which the
nation-state replaced the feudal monarchy. A new commercial aris-
tocracy, accompanied by rapid changes in economic institutions and
in technology, began to rival the older aristocracies of Church and
landed nobility. Luther and Calvin led the first successful revolts
:1gainst Catholic religious hegemony. In an age marked by such obvi-
ous upheavals in political, social, and religious life, an innovation in
rt,netary astronomy could at first seem no innovation at all.
Specific characteristics of the age had more concrete effects on
astronomy. The Renaissance was, for example, a period of voyages
and explorations. Fifty years before Copernicus' birth Portuguese voy-
ages along the African coast had begun to excite the imagination and
avarice of Europeans. Columbus' first landfall in America, made when
Copernicus was nineteen, only capped this earlier series of explora-
tions and provided the basis for a new series. Successful voyages de-
manded improved maps and navigational techniques, and these de-
pended in part on increased knowledge of the heavens. Prince Henry
the Navigator, the organizer and director of the early Portuguese voy-
ages, constructed one of the earliest European observatories. Explora-
tion therefore helped to create a demand for expert European astrono-
mers, and having done so, it partially changed their attitude toward
their field. Each new voyage disclosed new territory, new products,
and new people. Men rapidly learned how wrong ancient descriptions
of the earth could be. In particular, they learned how wrong Ptolemy
could be, for Ptolemy had been the greatest geographer as well as
the greatest astronomer and astrologer of antiquity. The astronomer's
awareness- an awareness which we shall shortly discover in Coper-
nicus himself- that Renaissance man could at last correct Ptolemy's
geography prepared him for changes in his own closely related field.
Agitation for calendar reform had an even more direct and dra-
matic effect on the practice of Renaissance astronomy, for the study
of calendars brought the astronomer face to face with the inadequacy
of existing computational techniques. The cumulative errors of the
Julian calendar had been recognized much earlier, and proposals for
calendar reform date from the thirteenth century or before. But these
proposals were ineffective until the sixteenth century, when the in-
creasing size of political, economic, and administrative units placed a
new premium upon an efficient and uniform means of dating. Then
reform became an official Church project, with results for astronomy
that are well illustrated in the biography of Copernicus himself.
Early in the sixteenth century Copernicus was asked to advise the
papacy about calendar reform. He declined and urged that reform be
126
postponed, because he felt that existing astronomical observations
and theories did not yet permit the design of a truly adequate calendar.
When Copernicus listed the aspects of contemporary astronomy that
had led him to consider his radical theory, he began, "For, first, the
mathematicians are so unsure of the movements of the Sun and Moon
that they cannot even explain or observe the constant length of the
seasonal year" (see p. 137 below). Reform of the calendar demanded,
said Copernicus, reform in astronomy. The preface of his De Revolu-
tionibus closed with the suggestion that his new theory might make
a new calendar possible. The Gregorian calendar, first adopted in
1582, was in fact based upon computations that made use of Coper-
nicus' work.
Recognition of the inadequacies in existing techniques for astro-
nomical computation was heightened by still another aspect of Renais-
sance life. During the fifteenth century, Europe experienced a second
great intellectual revival associated with a second rediscovery of classi-
cal models. Unlike its twelfth-century predecessor, the Renaissance re-
vival of learning was not primarily a scientific revival. Most of the
newly recovered documents exemplified ancient literature, art, and
architecture, subjects whose great tradition was then little known
in the West principally because Islamic culture had been indiffer-
ent to it. The manuscripts discovered in the fifteenth century did,
however, include a few important works of Hellenistic mathematics
and, more important, a great many authentic Greek versions of scien-
tific classics previously known only in Arabic. As a result, the Ptolemaic
system's recognized failure correctly to predict celestial motions could
no lovger be blamed upon errors accumulated in transmission and
translation. Astronomers could no longer believe that astronomy had
declined since Ptolemy.
Peuerbach, for example, began his career in astronomy by work-
ing from second-hand translations of the Almagest transmitted via
Islam. From them he was able to reconstruct a more adequate and
complete account of Ptolemy's system than any known before. But
his work only convinced him that a truly adequate astronomy could
not be derived from Arabic sources. Astronomers, he felt, would have
to work from Greek originals, and he was about to depart for Italy to
examine manuscripts available there when he died in 1461. His suc-
cessors, particularly Johannes Muller, did work from the Greek ver-
sions, and they then discovered that even Ptolemy's original formula-
tion was inadequate. By making available sound texts of ancient
authors, fifteenth-century scholars helped Copernicus' immediate pre-
decessors to recognize that it was time for a change.
Developments like those discussed above can help us understand
why the Copernican Revolution occurred when it did. They are essen-
tial parts of the climate for astronomical upheaval. But there are other,
more intellectual, aspects of the Renaissance which played a somewhat
different role in the Revolution. They are associated with humanism,
the dominant learned movement of the age, and their effect was less
upon the Revolution's timing than upon its shape. Humanism was not
principally a scientific movement. The humanists themselves were
often bitterly opposed to Aristotle, the scholastics, and the entire tra-
dition of university learning. Their sources were the newly recovered
literary classics, and, like literary men in other ages, many of them
rejected the scientific enterprise as a whole. The early humanist poet
Petrarch sounds a typical note, strangely and significantly reminiscent
of Augustine's earlier depreciation of science. "Even if all these things
were true, they help in no way toward a happy life, for what does
it advantage us to be familiar with the nature of animals birds fishes
and reptiles, while we are ignorant of the nature of the r ~ e of 'man t ~
which we belong, and do not know or care whence we come or whither
we go."
15
If humanism had been the only intellectual movement of
the Renaissance, the Copernican Revolution might have been long
postponed. The work of Copernicus and his astronomical contempo-
raries belongs squarely in that university tradition which the human-
ists most ridiculed.
The humanists did not, however, succeed in stopping science.
During the Renaissance a dominant humanistic tradition outside the
universities existed side by side with a continuing scientific tradition
within university walls. In consequence the first scientific effect of the
htL'llanists' dogmatic anti-Aristotelianism was to facilitate for others a
break with the root concepts of Aristotle's science. A second but more
important effect was the surprising fertilization of science by the strong
otherworldly strain that characterized humanist thought. From this
aspect of humanism, a first hint of which is contained in the preceding
quotation from Petrarch, some Renaissance scientists, like Copernicus,
Galileo, and Kepler, seem to have drawn two decidedly un-Aristotelian
128
ideas: a new belief in the possibility and importance of discovering
simple arithmetic and geometric regularities in nature, and a new
view of the sun as the source of all vital principles and forces in the
universe.
The otherworldliness of humanism derived from a well-defined
philosophical tradition which had greatly influenced Augustine and
other early Church fathers but which had been temporarily eclipsed by
the twelfth-century reco\ery of Aristotle's writings. That tradition, un-
like the Aristotelian, discovered reality in a changeless world of spirit
rather than in the transient affairs of everday life. Plato, who is the tra-
dition's ultimate source, often seems to dismiss the objects of this world
as mere impedect shadows of an eternal world of ideal objects or
"forms" existing outside of space and time. His followers, the so-called
Neoplatonists emphasized this tendency in their master's thought to
the exclusion of all others. Their mystical philosophy, which many
humanists took as a model, recognized only a transcendent reality.
Yet for all its mysticism, Neoplatonism contained elements that gave
a significant new direction to the science of the Renaissance.
The Neoplatonist leaped at once from the changeable and cor-
ruptible world of everyday life to the eternal world of pure spirit,
and mathematics showed him how to make the leap. For him mathe-
matics exemplified the eternal and real amid the imperfect and fluc-
tuating appearances of the terrestrial world. The triangles and circles
of plane geometry were the archetypes of all Platonic forms. They
existed nowhere- no line or point drawn on paper satisfies Euclid's
postulates -but they were endowed with certain eternal and neces-
sary properties which the mind alone could discover and which, once
discov;red, could be observed dimly mirrored in the objects of the
real world. The Pythagoreans, who had also envisioned the real
world as a shadow of the eternal world of mathematics, exemplified
the ideal of terrestrial science in their discovery that uniform strings
whose lengths are in the simple numerical ratios 1: :14 : 73 : produce
harmonious sounds. The mathematical strain in Neoplatonic thought
is often attributed to Pythagoras and identified as Neopythagoreanism.
Plato himself emphasized the necessity of mathematics as training
for the mind in pursuit of forms; over the door of his Academy he is
said to have inscribed, "Let no one destitute of geometry enter my
doors."
1
6 The Neoplatonists went further. They found in mathematics
the key to the essential nature of God, the soul, and the world soul
which was the universe. A typical passage from the fifth-century Neo-
platonist Proclus pedectly displays a part of this mystical vision of
mathematics:
The soul [of the world], therefore, is by no means to be compared to a
smooth tablet, void of all reasons; but she is an ever-written tablet, herself
inscribing the characters in herself, of which she derives an eternal pleni-
tude from intellect. . . . All mathematical species, therefore, have a pri-
mary subsistence in the soul: so that, before sensible numbers, there are to
be found in her inmost recesses, self-moving numbers; vital figures, prior to
the apparent; ideal proportions of harmony previous to concordant sounds;
and invisible orbs, prior to the bodies which revolve in a circle .... [We]
must conceive all these as subsisting ever vitally, and intellectually, as the
exemplars of apparent numbers, figures, reasons and motions. And here we
must follow the doctrine of Timaeus, who derives the origin, and con-
summates the fabric of the soul, from mathematical forms, and reposes in
her nature the causes of everything which existsP
Proclus and the humanists who espoused his cause are a long way
from the physical sciences. But they occasionally influenced their more
scientifically inclined contemporaries, and the result was a new search
by many late Renaissance scientists for simple geometric and arithme-
tic regularities in nature. Copernicus' friend and teacher at Bologna,
Domenico Maria de Novara, was a close associate of the Florentine
Neoplatonists who translated Proclus and other authors of his school.
Novara himself was among the first to criticize the Ptolemaic planetary
theory on Neoplatonic grounds, believing that no system so complex
and cumbersome could represent the true mathematical order of
nature. When Novara's pupil Copernicus complained that the Ptole-
maic astronomers "seem to violate the first principle of uniformity
in motion" and that they have been unable to "deduce the principal
thing- namely the shape of the Universe and the unchangeable sym-
metry of its parts" (seep. 138 below), he was participating in the same
Neoplatonic tradition. The Neoplatonic strain shows even more strongly
in Copernicus' great successor, Kepler. As we shall see, the search for
simple numerical relations runs through and motivates most of Kepler's
work.
The origin of the liaison between Neoplatonism and sunworship is
more obscure, but a hint about the ties that bound them together can
130
be found in the preceding quotation from Proclus. Neoplatonic thought
could never quite dispense with the real \Vorld. The "vital figures" and
"invisible orbs" that Proclus found in the soul of the world, or in
God, might be the primary philosophical entities, the only things that
have complete reality and existence. But the Neoplatonist could not
avoid granting some sort of existence to the impedect bodies attested
by his senses. He therefore regarded them as second-rate copies gen-
erated by the "vital figures" themselves. As Proclus said, the mathe-
matical forms that determine the nature of the world soul were also
the "causes of everything which exists." They generated countless de-
based and materialized copies of their own purely intellectual sub-
stance. The Neoplatonist's God was a self-duplicating procreative prin
ciple whose immense potency was demonstrated by the very multi-
plicity of the forms that emanated from Him. In the material universe
this fecund Deity was suitably represented by the sun whose visible
and invisible emanations gave light, warmth, and fertility to the uni-
verse.
This symbolic identification of the sun and God is found repeatedly
in Renaissance literature and art. Marsilio Ficino, a central figure in
the fifteenth-century humanist and Neoplatonic academy of Florence,
gave it typical expression in his treatise On the Sun:
Nothing reveals the nature of the Good [which is God] more fully
than the light [of the sun]. First, light is the most brilliant and clearest of
sensible objects. Second, there is nothing which spreads out so easily,
broadly, or rapidly as light. Third, like a caress, it penetrates all things
harmlessly and most gently. Fourth, the heat which accompanies it fosters
and nourishes all things and is the universal generator and mover. . . .
Similarty the Good is itself spread everywhere, and it soothes and entices
all things. It does not work by compulsion, but through the love which
accompanies it, like heat [which accompanies light]. This love allures
all objects so that they freely embrace the Good. . . . Perhaps light is itself
the celestial spirit's sense of sight, or its act of seeing, operating from a dis-
tance, linking all things to heaven, yet never leaving heaven nor mixing
with external things. . . . Just look at the skies, I pray you, citizens of
heavenly fatherland. . . . The sun can signify God hmself to you, and who
shall dare to say the sun is false.
18
With Ficino as with Proclus, we are a long way from science. Ficino
does not seem to understand astronomy. He certainly made no attempt
to reconstruct it. Though the sun has a new significance in Ficino's
universe, it retains its old position. Yet that position was no longer
appropriate. Ficino wrote, for example, that the sun was created first
and in the center of the heavens. Surely no lesser position in space
or in time could be compatible with the sun's dignity and creative
function. But the position was not compatible with Ptolemaic astron-
omy, and the resulting difficulties for Neoplatonism may have helped
Copernicus to conceive a new system constructed about a central
sun. In any case they gave him an argument for the new system. As
soon as he had discussed the new position of the sun, Copernicus ad-
verted to the fitness of his new cosmology (see p. 179 below). His
authorities are immediately Neoplatonic:
In the middle of all sits Sun enthroned. In this most beautiful temple
could we place this luminary in any better position from which he can illumi-
nate the whole at once? He is rightly called the Lamp, the Mind, the Ruler
of the Universe; Hermes Trismegistus names him the Visible God, Sophocles'
Electra calls him the All-seeing. So the sun Sits as upon a royal throne ruling
his children the planets which circle round him.
Neoplatonism is explicit in Copernicus' attitude toward both the
sun and mathematical simplicity. It is an essential element in the in-
tellectual climate that gave birth to his vision of the universe. But it
is often hard to tell whether any given Neoplatonic attitude is posterior
or antecedent to the invention of his new astronomy in Copernicus'
thought. No similar ambiguity exists with respect to the later Coper-
nicans. Kepler, for example, the man who made the Copernican sys-
tem work, is quite explicit about his reasons for preferring Copernicus'
proposal, and among them is the following:
[The sun] is a fountain of light, rich in fruitful heat, most fair, limpid,
and pure to the sight, the source of vision, portrayer of all colors, though
himself empty of color, called king of the planets for his motion, heart
of the world for his power, its eye for his beauty, and which alone we should
judge worthy of the Most High God, should he be pleased with a material
domicile and choose a place in which to dwell with the blessed angels. . . .
For if the Germans elect him as Caesar who has most power in the whole
empire, who would hesitate to confer the votes of the celestial motions on
him who already has been administering all other movements and changes
by the benefit of the light which is entirely his possession? ... [Hence]
by the highest right we return to the sun, who alone appears, by virtue of
his dignity and power, suited for this motive duty and worthy to become
the home of God himself, not to say the first mover.
1
9
Until after Copernicus' death the mathematical magic and the sun
that are so marked in Kepler's research remained the prin-
cipal points of explicit contact between Renaissance Neoplatonism
and the new astronomy. But late in the sixteenth century a third aspect
of Neoplatonism merged with Copernicanism and helped to reshape
the structure of Copernicus' universe. Unlike the Deity of the Neo-
platonists, whose perfection was measured by his immense fecundity,
the God of Aquinas and Aristotle had been conceived as an architect
who displayed His perfection in the neatne:;s and order of His creation.
Aquinas' God was well suited to the finite Aristotelian cosmos, but
the God of the Neoplatonists was not so easily bounded. If God's
perfection is measured by the extent and multiplicity of his procrea-
tion, a larger and more populous universe must connote a more per-
fect Deity. To many Neoplatonists the finitude of Aristotle's universe
was therefore incompatible with God's perfection. His infinite good-
ness would, they felt, be satisfied only by an infinite act of creation.
Even before Copernicus the resulting vision of a multipopulated uni-
verse, infinite in extent, had been the source of important divergences
from Aristotelian doctrine. During the Renaissance the revived em-
phasis on God's infinite creativity may have been a significant element
in the climate of opinion that bred Copernicus' innovation. Certainly,
as we shall see later, it was a major factor in the post-Renaissance
transition from Copernicus' finite universe to the infinite space of the
Newtonian world machine.
Neoplatonism completes the conceptual stage setting for the Coper-
nican Revolution, at least as we shall examine it here. For an astro-
revolution it is a puzzling stage, because it is set with so few
astronomical properties. Their absence, however, is just what makes
the setting important. Innovations in a science need not be responses
to novelties within that science at all. No fundamental astronomical
discovery, no new sort of astronomical observation, persuaded Coper-
nicus of ancient astronomy's inadequacy or of the necessity for change.
Until half a century after Copernicus' death no potentially revolu-
tionary changes occurred in the data available to astronomers. Any
possible understanding of the Revolution's timing and of the factors
that called it forth must, therefore, be sought principally outside of
astronomy, within the larger intellectual milieu inhabited by astron-
omy's practitioners. As we suggested at the beginning of this chapter,
133
Copernicus began his cosmological and astronomical researches ver
nearly where Aristotle and Ptolemy had stopped In that h '!
the imm d h . . . sense e IS
e mte eir of the ancient scientific tradition But hi. . h .
t t k 1m s m en-
ance oo a ost t.wo milleniums to reach him. In the interim the
very process of the medieval integration of science and
theology, the centunes of scholastic criticism and the n f
R . ' ew currents o
enaissance life and thought all had combin d t h
d . . ' e o c ange men's atti-
tu e towards the scientific heritage that they learned in school Just
great, and yet how strangely small, this essential change
e we shall discover in the next chapter as we tak C . '
innovation. e up opermcus
I s
COPERNICUS' INNOVATION
Copernicus and the Revolution
The publication of Copernicus' De Revolutioni.bus Orbium
Caelestium in 1543 inaugurates the upheaval in astronomiCal and
mological thought that we call the Copernican Revolution .. To this
point we have dealt only with the background of that RevolutiOn, set-
ting the stage upon which the Revolution occurred. N.ow we
the Revolution itself, dealing first, in this chapter, w1th Coperni:Uf>
contributions to it. So far as possible we shall discover those contnbu-
tions in Copernicus' own words, drawn from the De
the book that presented the new astronomy to the Almost Im-
mediately we shall encounter difficulties and upon
resolution depends our understanding of the Copernican Revolution
or since that Revolution is in many respects typical, of any other
conceptual upheaval in the sciences.
.
The De Revolutionibus is for us a problem text. Some of Its prob-
lems derive simply from the intrinsic difficulties of its. subject matter.
All put the introductory First Book is too to be read
with understanding by anyone except a techmcally
mer. We must deal with its essential technical contnbutwns m rela-
tively nonmathematical paraphrase, much like that employed. in treat-
ing the Almagest, and we shall by-pass process certam .of
essential problems that the De Revolutwmbus presented to Its six-
teenth-century readers. Had Copernicus propounded the new
omy in the simplified form to which we shall frequently resort. this
chapter, its reception might have been quite different. OppositiOn to
a more comprehensible work might, for example, have been marshaled
sooner. Our first probkm is therefore the barrier which a lack of
technical proficiency places between us and the central books of the
work that inaugurated the Revolution.
135
But the technical obscurity of the De Revolutionibus, though it
must be recognized at the start, is neither the most difficult nor the
most important sort of problem inherent in Copernicus' work. The
principal difficulties of the De Revolutionibus and the ones that we
may not evade arise rather from the apparent incompatibility between
that text and its role in the development of astronomy. In its conse-
quences the De Revolutionibus is undoubtedly a revolutionary work.
From it derive a fundamentally new approach to planetary astronomy,
the first accurate and simple solution of the problem of the planets,
and ultimately, with other fibers added to the pattern, a new cos-
mology. But, to any reader aware of this outcome, the De Revolutioni-
bus itself must be a constant puzzle and paradox, for, measured in
terms of its consequences, it is a relatively staid, sober, and unrevolu-
tionary work. Most of the essential elements by which we know the
Copernican Revolution - easy and accurate computations of planetary
position, the abolition of epicycles and eccentrics, the dissolution of
the spheres, the sun a star, the infinite expansion of the universe -
these and many others are not to be found anywhere in Copernicus'
work. In every respect except the earth's motion the De Revolutioni-
bus seems more closely akin to the works of ancient and medieval
astronor.ners and cosmologists than to the writings of the succeeding
generations who based their work upon Copernicus' and who made
explicit the radical consequences that even its author had not seen in
his work.
The significance of the De Revolutionibus lies, then, less in what
it says itself than in what it caused others to say. The book gave rise
to a revolution that it had scarcely enunciated. It is a revolution-making
rather than a revolutionary text. Such texts are a relatively frequent
and extremely significant phenomenon in the development of scientific
thought. They may be described as texts that shift the direction in
which scientific thought develops; a revolution-making work is at once
the culmination of a past tradition and the source of a novel future
tradition. As a whole the De Revolutionibus stands almost entirely
within an ancient astronomical and cosmological tradition; yet within
its generally classical framework are to be found a few novelties ,vhich
shifted the direction of scientific thought in ways unforeseen bv its
author and which gave rise to a rapid and complete break with the
ancient tradition. Viewed in a perspective provided by the history of
136 THE COPERNH.:AN REVOLUTION
astronomy, the De Revolutionibus has a dual nature. It is at once
ancient and modem, conservative and radical. Therefore its signifi-
cance can be discovered only by looking simultaneously to its past
and to its future, to the tradition from which it derived and to the
tradition which derives from it.
That double view of a single work is the principal problem of this
chapter. What is the relation of Copernicus to the ancient astronomi-
cal tradition within which he was educated? More precisely, what
aspects of that tradition led him to believe that some astronomical
innovation was essential, that certain aspects of ancient cosmology
and astronomy must be rejected? And, having resolved to break with
an old tradition, to what extent was he still necessarily bound by it
as the only source of those intellectual and observational tools re-
quired for the practice of astronomy? Again, what is Copernicus' rela-
tion to the tradition of modem planetary astronomy and cosmology?
Given the limitations imposed by the training and tools of classical
astronomy, what creative innovations could his work contain? How
could those innovations, which ultimately produced a radically new
astronomy and cosmology, be embedded initially in a predominantly
classical frame? And how could those novelties be recognized and
adopted by his successors? These problems and their corollaries are
symptomatic of the real difficulties of the De Revolutionibus or of any
scientific work which, though born within one tradition of scientific
thought, is the source of a new tradition that ultimately destroys its
parent.
Motives for Innovation- Copernicus' Preface
Copernicus is among that small group of Europeans who
first revived the full Hellenistic tradition of technical mathematical
astronomy which in antiquity had culminated in the work of Ptolemy.
The De Revolutionibus was modeled on the Almagest, and it was
directed almost exclusively to that small group of contemporary
astronomers equipped to read PtolemY's treatise. With Copernicus we
return for the first time to the sort of technical astronomical problem
with which we last dealt in Chapter 3 when examining the developed
Ptolemaic system. In fact we return to the same problem. The De
Revolutionibus was written to solve the problem of the planets, which,
Copernicus felt, Ptolemy and his successors had left unsolved. In
COPERNICUS' INNOVATION 137
Copernicus' work the revolutionary conception of the earth's motion
is initially an anomalous by-product of a proficient and devoted astron-
omer's attempt to reform the techniques employed in computing
planetary position. That is the first significant incongruity of the De
Revolutionibus, the disproportion between the objective that moti-
vated Copernicus' innovation and the innovation itself. It can be dis-
covered almost at the start of the prefatory letter that Copernicus
prefixed to the De Revolutionibus in order to sketch the motive, the
source, and the nature of his scientific achievement.I
TO THE MOST HOLY LORD, POPE PAUL III
The Preface of Nicholas Copernicus to the
Books of the Revolutions
I may well presume, most Holy Father, that certain people, as soon as
they hear that in this book about the Revolutions of the Spheres of the Uni-
verse I ascribe movement to the earthly globe, will cry out that, holding
such views, I should at once be hissed off the stage. For I am not so pleased
with my own work that I should fail duly to weigh the judgment which
others may pass thereon; and though I know that the speculations of a
philosopher are far removed from the judgment of the multitude - for his
aim is to seek truth in all things as far as God has permitted human reason
so to do - yet I hold that opinions which are quite erroneous should be
avoided.
Thinking therefore within myself that to ascribe movement to the Earth
must indeed seem an absurd performance on my part to those who know
that many centuries have consented to the establishment of the contrary
judgment, namely that the Earth is placed immovably as the central point
in the middle of the Universe, I hesitated long whether, on the one hand,
I should give to the light these my Commentaries written to prove the
Earth's motion, or whether, on the other hand, it were better to follow
the example of the Pythagoreans and others who were wont to impart their
philosophic mysteries only to intimates and friends, and then not in writing
but by word of mouth, as the letter of Lysis to Hipparchus witnesses. 1 This
letter, which Copernicus had at one time intended to include in the De
Revolutionibus, describes the Pythagorean and Neoplatonic injunction against
revealing nature's secrets to those who are not initiates of a mystical cult.
Reference to it here exemplifies Copernicus' participation in the Renaissance
revival of Neoplatonism discussed in the last chapter.] In my judgment
they did so not, as some would have it, through jealousy of sharing their
doctrines, but as fearing lest these so noble and hardly won discoveries of
the learned should be despised by such as either care not to study aught
save for gain, or- if by the encouragement and example of others they are
stimulated to philosophic liberal pursuits - yet by reason of the dullness of
their wits arc in the company of philosophers as drones among bees. Re-
flecting thus, the thought of the scorn which I had to fear on account of the
novelty and incongruity of my theory, well-nigh induced me to abandon
my projeet.
Thcsc misgivings and actual protests have been overcome by my friends
... [one of whom J often urged and even importuned me to publish this
work which I had kept in store not for nine years only, but to a fourth period
of nine years .... They urged that I should not, on account of my fears,
refuse any longer to contribute the fruits of my labors to the common ad-
vantage of those interested in mathematics. They insisted that, though my
theory of the Earth's movement might at first seem strange, yet it would
appear admirable and acceptable wlwn the publication of my elucidatory
comments should dispel the mists of paradox. Yielding then to their per-
suasion I at last permitted my friends to publish that work which they have
so long demanded.
That I allow the publication of these my studies may surprise your
Holiness the less in that, having been at such travail to attain them, I had
already not scrupled to commit to writing my thoughts upon the motion of
the Earth. [Some years before the publication of the De Revolutionibus
Copernicus had circulated among his friends a short manuscript called the
Commentariolus, describing an earlier version of his sun-centered astron-
omy. A second advance report of Copernicus' major work, the Narratio
Prima by Copernicus' student, Rheticus, had appeared in 1540 and again
in 1541.] How I came to dare to conceive such motion of the Earth, con-
trary to the received opinion of the Mathematicians and indeed contrary
to the impression of the senses, is what your Holiness will rather expect to
hear. So I should like your Holiness to know that I was induced to think
of a method of computing the motions of the spheres by nothing else than
the knowledge that the Mathematicians are inconsistent in these investi-
gations.
, For, hrst, t11e mathematicians are so unsure of the movements of the
~ and Moon that they cannot even explain or observe the constant length
of the seasonal year. Secondly, in determining the motions of these and
of the other five planets, they use neither the same principles and hypotheses
nor the same demonstrations of the apparent motions and revolutions. So
some use only homocentric circles [the Aristotelian system, derived by
Aristotle from Eudoxus and Callippus, and revived in Europe shortly before
Copernicus' death by the Italian astronomers Fracastoro and Amici], while
others [employ] eccentrics and epicycles. Yet even by these means they do
not completely attain their ends. Those who have relied on homocentrics,
though they have proven that some different motions can be compounded
therehom, have not thereby been able fully to establish a system which
agrees with the phenomena. Those again who have devised eccentric sys-
tems, though they appear to have well-nigh established the seeming motions
by calculations agreeable to their assumptions, have yet made many admis-
139
sions .[like .the .use of. the equant] which seem to violate the first principle
of uniformity m mohon. Nor have they been able thereby to discern or
deduce the principal thing- namely the shape of the Universe and the
unchangeable symmetry of its parts. With them it is as though an artist
were to gather the hands, feet, head and other members for his images
from diverse models, each part excellently drawn, but not related to a
single body, and since they in no way match each other, the result would be
monster rather than man. So in the course of their exposition, which the
mathematicians call their system, . . . we find that they have either omitted
some indispensable detail or introduced something foreign and wholly
irrelevant. This would of a surety not have been so had they followed fixed
principles; for if their hypotheses were not misleading, all inferences based
thereon might be surely verified. Though my present assertions are obscure,
they will be made clear in due course.
An honest appraisal of contemporary astronomy, says Copernicus,
shows that the earth-centered approach to the problem of the planets
is hopeless. The traditional techniques of Ptolemaic astronomy have
not and will not solve that problem; instead they have produced a
monster; there must, he concludes, be a fundamental error in the basic
concepts of traditional planetary astronomy. For the first time a tech-
nically competent astronomer had rejected the time-honored scientific
tradition for reasons internal to his science, and this professional aware-
ness of technical fallacy inaugurated the Copernican Revolution. A felt
necessity was the mother of Copernicus' invention. But the feeling of
necessity was a new one. The astronomical tradition had not previ-
ously seemed monstrous. By Copernicus' time a metamorphosis had
occurred, and Copernicus' preface brilliantly describes the felt causes
of that transformation.
Copernicus and his contemporaries inherited not only the Alma-
gest but also the astronomies of many Islamic and a few European
astronomers who had criticized and modified Ptolemy's system. These
are the men to whom Copernicus refers as "the mathematicians." One
had added or subtracted a few small circles; another had employed
an epicycle to account for a planetary irregularity that Ptolemy had
originally treated with an eccentric; still another had invented a means
unknown to Ptolemy of accounting for small deviations from the
motion predicted by a one-epicycle one-deferent system; others had,
With new measurements, altered the rates at which the compounded
circles of Ptolemy's system rotated. There was no longer one Ptolemaic
140
system, but a dozen or more, and the number was multiplying rapidly
with the multiplication of technically proficient astronomers. All these
systems were modeled on the system of the Almagest, and all were
therefore "Ptolemaic." But because there were so many variant sys-
tems, the adjective "Ptolemaic" had lost much of its meaning. The
astronomical tradition had become diffuse; it no longer fully specified
the techniques that an astronomer might employ in computing plane-
tary position, and it could not therefore specify the results that he
would obtain from his computations. Equivocations like these deprived
the astronomical tradition of its principal source of internal strength.
Copernicus' monster has other faces. None of the "Ptolemaic" sys-
tems which Copernicus knew gave results that quite coincided with
good naked-eye observations. They were no worse than Ptolemy's re-
sults, but they were also no better. After thirteen centuries of fruitless
research a perceptive astronomer might well wonder, as Ptolemy
could not have, whether further attempts within the same tradition
could conceivably be successful. Besides, the centuries that had inter-
vened between Ptolemy and Copernicus had magnified the errors of
the traditional approach, thus providing an additional source of dis-
content. The motions of a system of epicycles and deferents are not
unlike those of the hands of a clock, and the apparent error of a clock
increases with the passage of time. If a clock loses, say, 1 second per
decade, its error may not be apparent at the end of a year or the end
of ten. But the error can scarcely be evaded after a millenium, when
it will have increased to almost 2 minutes. Since Copernicus and his
contemporaries possessed astronomical data extending over a time span
thirteen centuries longer than that covered by Ptolemy's data, they
could impose a far more sensitive check upon their systems. They
were necessarily more aware of the errors inherent in the ancient
approach.
The passage of time also presented the sixteenth-century astrono-
mer with a countedeit problem which ironically was even more effec-
tive than the real motion of the planets in fostering recognition of
the error$ in the Ptolemaic method. Many of the data inherited by
Copernicus and his colleagues were bad data which placed the planets
and stars in positions that they had never occupied. Some of the
erroneous records had been collected by poor observers; others had
once been based upon good observations but had been miscopied or
141
misconstrued during the process of transmission. No simple planetary
system- Ptolemy's, Copernicus', Kepler's, or Newton's- could have
reduced to order the data that Renaissance astronomers thought they
had to explain. The complexity of the problem presented by Renais-
sance data transcended that of the heavens themselves. Copernicus
a victim of the data that had originally aided him in
rejectmg the system. His own system would have given far
better results if he had been as skeptical about his predecessors' ob-
servations as he was about their mathematical systems.
. Diffuseness . continued inaccuracy - these are the two prin-
cipal charactenshcs of the monster described by Copernicus. In so
fa: the Copernican Revolution depended upon explicit changes
withm the astronomical tradition itself, these are its major sources.
But they are not the only ones. We may also ask why Copernicus was
able to recognize the monster. Some of the tradition's apparent meta-
morphosis must have been in the eye of the beholder, for the tradition
been diffuse and inaccurate before. In fact we have already con-
Sidered this question. Copernicus' awareness of monstrosity depended
upon .that larger climate of philosophical and scientific opinion whose
genesis and nature were described in the last chapter. From the state
of contemporary astronomy a man without Copernicus' Neoplatonic
bias might have concluded merely that the problem of the planets
have no solution that was simultaneously simple and precise.
Similarly, an astronomer unacquainted with the tradition of scholastic
might have been unable to develop parallel criticisms for
his field. These and other novelties developed in the last chapter
are mam currents of Copernicus' time. Though he seems unaware of
them, Copernicus was carried by these philosophical currents, as his
contemporaries were unwittingly carried by the motion of the earth.
Copernicus' work remains incomprehensible unless viewed in its rela-
tion to both the internal state of astronomy and the larger intellectual
climate of the age. Both together produced the monster.
Discontent with a recognized monster was, however, only the first
step toward the Copernican Revolution. Next came a search whose
beginnings are described in the remaining portions of Copernicus'
prefatory letter:
li pondered long upon this uncertainty of mathematical tradition in estab-
shmg the motions of the system of the spheres. At last I began to chafe
142
a ree on any one certain theory of
that philosophers could no means for us by a supremely good and
the mechanism of the wrougt thev investigated with meticulous
orderly Creator, though m er s I 'LNote how Copernicus equates
I t t oints rehtmg to Its cnc es. 1
f
care t 1C mmu es P . ' , pect of his N eoplatonism rom
"orderly" with ::hemently dissented. There are
which any good Anstote wn k ains to read again the works of
other sorts. of orderliness.] I Phand to seek out whether any
all the philosophers on t the of the spheres were other than
of them had ever suppose a .
1
h
1
I found first in Cicero that
d d d b the mathemahca sc oo s.
d
those eman e y
had realized that the Earth move .
Hicetas [of Syracuse, fifth centu? B.c.] . tl rs had held the like opinion.
Afterwards I found in Plutarch ht ,lat certam leto make them accessible to
I think fit here to add Plutarc s own wor ,
all: "The rest hold the Earth to be stationary, but Philolaus
] that she moves around the [centra re
[fifth century B.C. says M H lides of Pontus and Ecphantus
1. . 1 l"k the Sun and oon. erac
ob 1que circ e I e
]
1
ake the Earth to move, not
the Pythagorean [fourth cbentury own center as a wheel on
indeed through space but y rotatmg roun
an axle from West to East." . b an to think of the mobility of the
Taking advantage of this I too eg t knowin now that others
Earth; and though the such as they chose
before had been grante r \ stars I considered that I also might
to explam the phenomena t eb some motion of the Earth,
easily be allowed to try whhe . er, f y th olution of the celestial spheres
sounder explanations than t eirs or e rev
might so be discovered. . l . work I ascribe to the Earth, by
Thus assuming motions, whic 1 m myl d" d that if the motions
b f I have at ast Jscovere ,
long and frequent o serva ght into relation with the circulation of
of the rest of the planets tion to the circles of each planet, not
tJ:te Earth and be reckoned m but the orders and magnitudes
only do their phenomena become so bound to-
of all stars and spheres, nay t h f ld be moved from its place with-
h part t ereo cou
gether that _not mg m_ any f ll the other parts and of the Universe as a
out producmg confus_IOn oh a . t to the single most striking difference
h le [Copermcus ere pom s
. .
1
w o . . . . d p 1 , I the Copernican system It IS no onger
between his an it of an one planet at will, holding
possible to shnnk or expa fi t t' f can determine the order and
the orbits without resort to the
the spheres. We shall discuss the point more fully
hypot esis o space-
1
, ]
C ystem with Pto emy s.
whet learned will
. . to com rehend and appreciate, not super Cia y u
if they alre t th! demands of this science, such reasoning as I
thorough y, accor mg o
143
bring to bear in support of my judgment. But that learned and unlearned
alike may see that I shrink not from any man's criticism, it is to your Holi-
ness rather than anyone else that I have chosen to dedicate these studies
of mine, since in this remote corner of Earth in which I live you are re-
garded as the most eminent by virtue alike of the dignity of your Office and
of your love of letters and science. You by your influence and judgment
can readily hold the slanderers from biting, though the proverb hath it that
there is no remedy against a sycophant's tooth. It may fall out, too, that
idle babblers, ignorant of mathematics, may claim a right to pronounce a
judgment on my work, by reason of a certain passage of Scripture basely
twisted to suit their purpose. Should any such venture to criticize and carp
at my project, I make no account of them; I consider their judgment .ash,
and utterly despise it. I well know that even Lactantius, a writer in other
ways distinguished but in no sense a mathematician, discourses in a most
childish fashion touching the shape of the Earth, ridiculing even those who
have stated the Earth to be a sphere. Thus my supporters need not be
amazed if some people of like sort ridicule me too.
Mathematics are for mathematicians, and they, if I be not wholly de-
ceived, will hold that these my labors contribute somewhat even to the
Commonwealth of the Church, of which your Holiness is now Prince. For
not long since, under Leo X, the question of correcting the ecclesiastical
calendar was debated in the Council of the Lateran. It was left undecided
for the sole cause that the lengths of the years and months and the motions
of the Sun and Moon were not held to have been yet determined with
sufficient exactness. From that time on I have given thought to their more
accurate observation, by the advice of that eminent man Paul, Lord Bishop
of Sempronia, sometime in charge of that business of the calendar. What
results I have achieved therein, I leave to the judgment of learned mathe-
maticians and of your Holiness in particular. And now, not to seem to promise
your Holiness more than I can perform with regard to the usefulness of the
work, I pass to my appointed task.
"Mathematics are for mathematicians." There is the first essential
incongruity of the De Revolutionibus. Though few aspects of West-
ern thought were long unaffected by the consequences of Copernicus'
work, that work itself was narrowly technical and professional. It was
mathematical planetary astronomy, not cosmology or philosophy,
that Copernicus found monstrous, and it was the reform of mathe-
matical astronomy that alone compelled him to move the earth. If his
contemporaries were to follow him, they would have to learn to under-
stand his detailed mathematical arguments about planetary position,
and they would have to take these abstruse arguments more seriously
than the first evidence of their senses. The Copernican Revolution was
144
not primarily a revolution in the mathematical techniques employed
to compute planetary position, but it began as one. In recognizing the
need for and in developing these new techniques, Copernicus made
his single original contribution to the Revolution that bears his name.
Copernicus was not the first to suggest the earth's motion, and he
did not claim to have rediscovered the idea for himself. In his preface
he cites most of the ancient authorities who had argued that the earth
was in motion. In an earlier manuscript he even refers to Aristarchus,
whose sun-centered universe very. closely resembles his own. Although
he fails, as was customary during the Renaissance, to mention his
more immediate predecessors who had believed that the earth was
or could be in motion, he must have known some of their work. He may
not, for example, have known of Oresme's contributions, but he had
probably at least heard of the very influential treatise in which the
fifteenth-century Cardinal, Nicholas of Cusa, derived the motion of
the earth from the plurality of worlds in an unbounded Neoplatonic
universe. The earth's motion had never been a popular concept, but
by the sixteenth century it was scarcely unprecedented. What was un-
precedented was the mathematical system that Copernicus built upon
the earth's motion. With the possible exception of Aristarchus, Coper-
nicus was the first to realize that the earth's motion might solve an
existing astronomical problem or indeed a scientific problem of any
sort. Even including Aristarchus, he was the first to develop a detailed
account of the astronomical consequences of the earth's motion. Co-
pernicus' mathematics distinguish him from his predecessors, and it
.. was in part because of the mathematics that his work inaugurated a
revolution as theirs had not.
Copernicus' Physics and Cosmology
For Copernicus the motion of the earth was a by-product of
the problem of the planets. He learned of the earth's motion by ex-
amining the celestial motions, and, because the celestial motions had
to him a transcendent importance, he was little concerned about the
difficulties that his innovation would present to normal men whose
concerns were predominantly terrestrial. But Copernicus could not
quite ignore the problems that the earth's motion raised for those
whose sense of values was less exclusively astronomical than his own.
He had at least to make it possible fo h. 14 5
the earth's motion he had t h r Is contemporaries to conceive
' 0
s ow that th
motiOn were not so devastat' h e consequences of this
b Th f mg as t ey were co I
e. ere ore Copernicus o d h . mmon y supposed to
technical sketch of th . pene t e De Revolutionibus with a non
e umverse that he had -
moving earth. His introductor F' B constructed to house a
d
. . y ust ook was d. d
an It mcluded all th Irecte to laymen
e arguments that he th h '
accessible to those witho t . oug t he could make
Th u astronomical training
ose arguments are profound! un . . .
derive from mathematical I y convmcmg. Except when they
I
ana yses that Co .
exp icit in the First Book th permcus failed to make
f
' ey were not new and th d d
con orm to the details of th . ' ey i not quite
t d I e astronomical system th C
o eve op in the later book 0 I at opernicus was
h
s. n y a man h l'k C
ot er reasons for supposing th t h w o, I e opernicus, had
th F' a t e earth m d I e Irst Book of the D R l . ove cou d have taken
e evo utwnibus t' I
But the First Book . . en Ire Y seriously.
h
Is not ummportant It
s adow the incredulity and 'd' I . . s very weaknesses ore-
n ICU e With h' h C
would be greeted by those who ld w IC opernicus' system
matical discussion of th b cou not follow the detailed mathe-
. e su sequent books It
upon Anstotelian and schola t' . s repeated dependence
C
s IC concepts and I h
even opernicus was abl t tr . aws s ow how little
t
. h' e o anscend his train' d h' .
cep m IS own narrow field f . I' . mg an Is times ex-
d
o speCia Izatwn Fin II h
ness an incongruities of the F' t B k . . a y, t e incomplete-
t d' Irs oo Illustrate h
o ra ItiOnal cosmology and t d' . agam t e coherence
I d
ra ItiOnal astrono C
was e to revolution by astrono . I . my. opernicus, who
t . d miCa motives I d
ne to restrict his innovation t on y an who inevitably
th d o astronomy co ld
e estructive cosmolog I ' u not evade entirely
Ica consequences of the earth's motion
BOOK ONE
1. That the Universe is Spherical
. the first place we must observe . .
Is either because that figure is th that the Umverse is spherical This
but whole and complete in itself :rm:st as not being
therefore best suited for that wh' h . ecause It IS the most capacious and
[of all solids with a give f Ic h IS to contain and preserve all th.
. b n sur ace t e h h mgs
agam ecause all the perfect part ere as the greatest volume]. or
hso formed; or because all thingss Idt, tnamely, Sun, Moon and S;ars
m t e case of d f n o assume thi h ' N rops o water and liquid b d. . s s ape, as is seen
o one doubts that such a shape h b o Ie.s m general if freely formed.
as een assigned to the heavenly bod' IeS.
146
2. That the Earth alw is Spherical.
. . n all sides it inclines [or falls J toward
The Earth also is sphencal, smce o ' . t o thward the North Pole
A
we pass from any pom n r '
d the center. s . l .
1
tll" other pole sinks correspon - l lly 'IS >S \V 11 e -' of the daily rotJ.twn grac u.l .
1
e
1
p
1
. . t set while certain stars , the N ort I o e o ,
1 ingly and more stars near F tl e. the change in altitude of the po e
in the South do not nse. ur
1
.I' d the Earth which could
. l t the dishnce traverse on ' d is always proportwna o ' H tl1e Earth must be finite an h 1 fi 'ure ence ' not be save on a sp g, 1 .d .. tl chapter with a few more argu-
h
. . . l [ Copermcus cone u es 1e
h
sp enca . . . . . . . . .. 1 f th classical sources that we ave ments for the earths sphencity typiC,! o e
already examined.]
3. II ow Earth, with theW ater on it, forms one Sphere.
E- th form the seas and fill the lower
The waters spread around the , . .'lr st be less than that of the Earth,
declivities. The volume of the b th by their weight, press
else they would swallow up the an (smfce of 1.' . g things stretches of
t ) Tl us for the sa ety o Jvm ' toward the same cen er . l ' d 1 e OtiS islands widely scattered.
l
l ft overed an a so num r the Eart 1 are e unc ' . d
1
hole of the :Mainland, but a Nay, what is a continc,It, and mdee t le w
vast island? . . t h w both that the terrestrial
[Copernicus wishes, in this chapter, do sho t and earth together
d . tl ade of earth an t at wa er
d globe is pre amman y m l b 1 Presumably he is looking ahea .
are required to mr.ke the go e a sp d motion of a solid globe
Earth breaks up less easily w will finally say
is more plausible than of a hqtu bgam, t . a sphere (see Chap-
h
t rally in c1rc es ecause 1 1s
that the eart moves na u h f d to show that both earth
h F
. t B k below) He t ere ore nee s
h ter 8 of is 1rs
00
. . f the sphere in order that bot
and water are essential to the composlt,lon o l t' , The passage is of . . h th sphere s natura mo 1011.
"'Will participate toget cr m . e d t. g his view of the structure of
l
. t t b cause m ocumen m
P
articu ar m eres , e . . . 'th the recent voyages . d' lavs h
1s acc1uamtance WI the earth Copernicus ISP '- '
tly be made in of discovery and with the corrections that must
Ptolemy's geographical writings. the depth of Ocean
If the terrestrial globe were pre omman y d ' d so neither island
1
. from the shore outwar s, an
would constant Y mcrease f l d ld be met by sailors, how h. f the nture o an wou
nor rock nor anyt mg o u k that between the Egyptian Sea
f
th ventured Yet we now
ar soever .11 . ,h in the middle of the great land-mass, Is a
and the Arabian Gulf, we On the other hand, in his Cosmography
passage barely 15 wJde. habitable land extends to the middle circle
Ptolemy would have
1
t that the . d' 180o eastward from
h h
. th gh a hem1sphere exten mg [of the eart , t at IS, rou . . b d where modern discovery
the Canary Islands] with a terra zncogmta f 60" of longitude.
d tensl
,e remon as ar as has added Cathay an a very ex o-
147
Thus we know now that the Earth is inhabited to a greater longitude than
is left for Ocean.
This will more evidently appear if we add the islands found in our own
time under the Princes of Spain and Portugal, particularly America, a
land named after the Captain who discovered it and, on account of its
unexplored size, reckoned as another Mainland - besides many other islands
hitherto unknown. We thus wonder the less at the so-called Antipodes or
Antichthones [the inhabitants of the other hemisphere]. For geometrical
argument demands that the Mainland of America on account of its position
be diametrically opposite to the Ganges basin in India. . . .
4. That the Motion of the Heavenly Bodies is Uniform, Circular,
and Perpetual, or Composed of Circular Motions.
We now note that the motion of heavenly bodies is circular. Rotation
is natural to a sphere and by that very act is its shape expressed. For here
we deal with the sin1plest kind of body, wherein neither beginning nor end
may be discerned nor, if it rotate ever in the same place, may the one be
distinguished from the other.
Because there are a multitude of spheres, many motions occur. Most
evident to sense is the diurnal rotation . . . marking day and night. By
this motion the whole Universe, save Earth alone, is thought to glide from
East to West. This is the common measure of all motions, since Time itself
is numbered in days. Next we see other revolutions in contest, as it were,
with this daily motion and opposing it from West to East. Such opposing
motions are those of Sun and Moon and the five planets. . . .
But these bodies exhibit various differences in their motion. First their
axes are not that of the diurnal rotation, but of the Zodiac, which is oblique
thereto. Secondly, they do not move uniformly even in their own orbits;
for are not Sun and Moon found now slower, now swifter in their courses?
Further, at tin1es the five planets become stationary at one point and an-
other and even go backward .... Furthermore, sometimes they approach
Earth, being then in Perigee, while at other times receding they are in
Apogee.
Nevertheless, despite these irregularities, we must conclude that the
motions of these bodies are ever circular or compounded of circles. For the
irregularities themselves are subject to a definite law and recur at stated
times, and this could not happen if the motions were not circular, for a
circle alone can thus restore the place of a body as it was. So with the Sun
which, by a compounding of circular motions, brings ever again the ooang-
ing days and nights and the four seasons of the year. Now therein it must
be that divers motions are conjoined, since a sin1ple celestial body cannot
move irregularly in a single circle. For. such irregularity must come of un-
evenness either in the moving force (whether inherent or acquired) or in
the form of the revolving body. Both these alike the mind abhors regarding
the most perfectly disposed bodies.
148
It is then generally agreed that the motions of Sun, Moon, and Planets
do but seem irregular either by reason of the divers directions of their
axes of revolution, or else by reason that Earth is not the center of the
circles in which they revolve, so that to us on Earth the displacements of
these bodies [along their orbits] seem greater when they are near [the
earth] than when they are more remote (as is demonstrated in optics [or in
everyday observation - boats or carriages always seem to move by more
quickly when they are closer]). Thus, equal [angular] motions of a sphere,
viewed from different distances, will seem to cover different distances in
equal times. It is therefore above all needful to observe carefully the rela-
tion of the Earth toward the Heavens, lest, searching out the things on
high, we should pass by those nearer at hand, and mistakenly ascribe earthly
qualities to heavenly bodies.
Copernicus here provides the fullest and most forceful version that
we have yet examined of the traditional argument for restricting the
motions of celestial bodies to circles. Only a uniform circular motion,
or a combination of such motions, can, he thinks, account for the regu-
lar recurrence of all celestial phenomena at fixed intervals of time. So
far every one of Copernicus' arguments is Aristotelian or scholastic,
and his universe is indistinguishable from that of traditional cosmology.
In some respects he is even more Aristotelian than many of his prede-
cessors and contemporaries. He will not, for example, consent to the
violation of the uniform and symmetric motion of a sphere that is im-
plicit in the use of an equant.
The radical Copernicus has so far shown himself a thoroughgoing
conservative. But he cannot postpone the introduction of the earth's
motion any longer. He must now take account of his break with tradi-
.. tion. And strangely enough, it is in the break that Copernicus shows
his dependence on the tradition most clearly. In dissent he still re-
mains as nearly as possible an Aristotelian. Beginning in the fifth
chapter, below, and culminating in the general discussion of motion
in the eighth and ninth chapters, Copernicus suggests that because
the earth is a sphere, like the celestial bodies, it too must participate
in the compounded circular motions which, he says, are natural to a
sphere.
5. Whether Circular Motion belongs to the Earth;
and concerning its position.
Since it has been shown that Earth is spherical, we now consider
whether her motion is conformable to her shape and her position in the
.
149
Umverse. Without these we cannot construct a
phenomena. Now authorities agree that theory of the heavenly
center of the Universe and th
d h
ds firm her place at the
b d
'
ey regar t e contrary
hink b
a sur . Yet if we examine mor l l . .
as unt a le, nay as
not so settled, and needs wider ceo c y will be seen that this question is
ns1 erahon
A seeming change of place rna com . f
observer, or again of unequal y e o movement either of object or
and parallel motions no movemmo;e:nents of _the two (for between equal
which the rotation of the H en. IS perceptible). Now it is Earth from
eavens 1s seen If the
.
assumed it will be reproduced in external b . n of Earth be
the opposite direction.
odws, wh1eh w1ll seem to move in
Consider first the diurnal rotation. B it h
.
alone and its contents ap
t
y t e whole Umverse, save Earth
I
' pears o move very swiftl y
revo ves from West to East and
"ll fi . y. et grant that Earth
elusion is right It is the It rc;: Wl nd, If you ponder it, that my con-
should not mo;ion be that contains all things, and why
tainer, to the located than the 1 r; ?er to the than to the con-
Heraclides and Ecphantus th The latter VIew was certainly that of
cording to Cicero). All of th e y and Hicetas of Syracuse ( ac-
Universe, believing that th eSmt rna e t e Earth rotate in the midst of the
e ars set owing to th E h
way, and rise again when it ha
d
e art coming in the
If .
s passe on.
this [possibility of the earth's motio . d .
less grave arises about the Ea th' . _n] IS a mitted, then a problem no
has hitherto held that the . s even though almost everyone
the earth can move at all it ISh at t e center of the Universe. [Indeed if
h
' may ave more than a 1 1
'
t e center of the Universe It
simp e axm motion about
d h
. may move away from th
l
an t ere are some good astronomical r
e a together,
For grant that Earth is not at th
easons for supposmg that it does.]
hi h h l
e exact center but at a d" t f
w c , w i e small compared [with th d.
Is ance rom it
considerable compared with [th d t e Istance] to the starry sphere, is yet
the other planets Then calculate ances to] the spheres of the Sun and
motions, assumin.g these [m t" e ] conbsequent variations in their seeming
o Ions to e really unif
d b
center other than the Earth' 0
h
orm an a out some
cause for the irregularity of s;h ne en perhaps adduce a reasonable
Planets are seen at varying d. et se varfla e motions. And indeed since the
. l
Is ances rom the Earth th
Is sure y not the center of their eire] N . .
. , e center of Earth
move toward and away from E th es. Eor IS It certam whether the Planets
is therefore justifiable to hold a;ha't and away from them.
bon to the diurnal rotation That th E art has another motion in addi-
several motions and is a Pl e arth, rotating, wanders with
the Pythagorean no mean mathe IS a VIew attributed to Philolaus
ha
'
' rna 1c1an and one wh Pl . .
ve sought out in Italy.
'
om ato 1s smd to
Copernicus is here pointing t th
.
astronomers of the con t f o most Immediate advantage for
cep o a movmg earth. If the earth moves in
150
an orbital circle around the center as well as spinning on its axis, then,
at least qualitatively, the retrograde motions and the different times
required for a planet's successive journeys around the ecliptic can be
explained without the use of epicycles. In Copernicus' system the
major irregularities of the planetary motions are only apparent. Viewed
from a moving earth a planet that in fact moved regularly would appear
to move irregularly. For this reason, Copernicus feels, we should be-
lieve in the orbital motion of the earth. But, strangely enough, in the
parts of his work accessible to the lay reader, Copernicus never
demonstrates this point any more clearly than he has above. Nor does
he demonstrate the other astronomical advantages that he cites else-
where. He asks the nonmathematical reader to take them for granted,
though they are not difficult to demonstrate qualitatively. Only in
the later books of the De Revolutionibus does he let the real advan-
tages of his system show, and since he there deals, not with retrograde
motions in general, but with the abstruse quantitative details of the
retrograde motions of each individual planet, only the astronomically
initiate were able to discover what the earlier references to astro-
nomical advantages meant. Copernicus' obscurity may have been de-
liberate, for he had previously referred with some approval to the
Pythagorean tradition which dictated withholding nature's secrets
from those not previously purified by the study of mathematics (and
by other more mystical rites). In any case, the obscurity helps ex-
plain the way in which his work was received.
In the next two sections we shall consider the astronomical con-
sequences of the earth's motion in detail, but we must first complete
"Copernicus' general sketch of physics and cosmology. Omitting for the
moment Chapter 6, Of the Vastness of the Heavens compared with
the Size of the Earth, we proceed to the central chapters in which
Copernicus, having asked indulgent readers to assume that astronomi-
cal arguments necessitate the earth's motion around the center, at-
tempts to make that motion physically reasonable.
7. Why the Ancients believed that the Earth is at rest,
like a Center, in the Middle of the Universe.
The ancient Philosophers tried by divers . . . methods to prove Earth
fixed in the midst of the Universe. The most powerful argument was drawn
from the doctrine of the heavy and the light. For, they argue, Earth is the
heaviest element, and all things of weight move towards it, tending to its
center. Hence since the Earth is spherical, and heavy things move vertically
151
it, they all rush together to the center if not stopped at the f
o:V those thmgs which move towards the center mu . su: ace.
mam at rest. Much more then will the who! E th st, reachmg It, re- t f h U .
e ar remam at rest at the
cen er o t e mverse. Receiving all falling bodies "t "ll .
able by its own weight.
'
1
WI rcmam immov-
Another argument is based on the sup osed nat r f . .
says that motion of a single and simple bo/y . . lu eAo . molhon.
b e"th t h 15
srmp e. s1mp emotion may
e 1 er s rmg t, or circular. Again a straight t" b .
do S . 1
mo Ion may e either up or
wn. o every simp e motion must be either toward th
downward, or away from the center name! u e center, namely
namely circular [That i d" ' . Y pward, or round the center,
natural motions: the onl; scholastic physics,
are caused by the nature of the body th t . :VI hout an external push,
each of the simple bodie (th fi la ISm motiOn. The natural motion of
h
s e ve e ements - earth water fi d
aet er) must itself be simple, because it is a con , 'tlr, . re, an
elementary nature And finall the sequence o a srmple or
motions within h '. 1 Y: re are only three (geometrically) simple
Now it is a up, circularly about the center.]
d d h
e eavy e ements earth and water to mo
ownwar ' t at is, to seek the center. But the light element . d fir ve
upward away from the center There . s arr an e move
to these four elements. The ascnbe motion
So far Aristotle.
Ies owever have crrcular motion.
If then, says Ptolemy, Earth moves at least with a diurnal .
result must be the reverse of that described ab the
of excessive rapidity, since in 24 h "t For the motiOn must be
to the Earth. Now thin s rota. ours I rm_part a complete rotation
are apt to disperse, unl!ss coh:ionf or, if united,
Earth would have been diss a d 1 . emy ere ore says that
absurdity) would have th oz;r ago, (which is the height of
living creatures and other heavy t emselves; and certainly all
mained on its surface but must h ee to move could not have re-
o_bjects reach their a;pointed offh could falling
time the Earth would ha d . y eneat ' smce m the mean-
1 d
ve move swiftly from under them Mo
c ou s and everything in th . ld
. reover that C . . e air wou continually move westward. [Note
opermcm has considerably elaborated Ptolemy's . . I quoted 85 1 b
ongma argument
far.] on P . t IS y no means clear that Ptolemy would have gone
8. The Insufficiency of these Arguments, and their Refutation.
of and like they say that Earth surely rests at the center
e mverse. Now 1 one should say that the Earth moves that ;
_as to that th_e motion is natural, not violent [or due to, an
happen to nature produce the opposite
imp t 7 Ic occur by vwlence. Things subjected to any force or
But ua or must be disintegrated, and cannot long exist.
1 processes bemg adapted to their purpose work smoothly. [That
152
is, if the earth moves at all, it does so because it is of the nature of earth
to move, and a natural motion cannot be disruptive.]
Idle therefore is the fear of Ptolemy that Earth and all thereon would
be disintegrated by a natural rotation, a thing far different from an artificial
act. Should he not fear even more for the Universe, whose motion must be
as much more rapid as the Heavens are greater than the Earth? Have the
Heavens become so vast because of their vehement motion, and would they
collapse if they stood still? If this were so the Heavens must be of infinite
size. For the more they expand by the force of their motion, the more rapid
will become the motion because of the ever increasing distance to be tra-
versed in 24 hours. And in turn, as the motion waxes, must the immensity
of the Heavens wax. Thus velocity and size would increase each the other
to infinity. . . .
They say too that outside the Heavens is no body, no space, nay not
even void, in fact absolutely nothing, and therefore no room for the Heavens
to expand [as we have suggested above that they would]. Yet surely it is
strange that something can be held by nothing. Perhaps indeed it will be
easier to understand this nothingness outside the Heavens if we assume
them to be infinite, and bounded internally only by their concavity, so that
everything, however great, is contained in them, while the Heavens remain
immovable ....
Let us then leave to Natural Philosophers the question whether the
Universe be finite or no, holding only to this that Earth is finite and spheri-
cal. Why then hesitate to grant Earth that power of motion natural to its
[spherical] shape, rather than suppose a gliding round of the whole uni-
verse, whose limits are unknown and unknowable? And why not grant that
the diurnal rotation is only apparent in the Heavens but real in the Earth?
It is but as the saying of Aeneas in Virgil- "We sail forth from the harbor,
and lands and cities retire." As the ship floats along in the calm, all external
things seem to have the motion that is really that of the ship, while those
within the ship feel that they and all its contents are at rest.
.. It may be asked what of the clouds and other objects suspended in the
air, or sinking and rising in it? Surely not only the Earth, with the water on
it, moves thus, but also a quantity of air and all things so associated with
the Earth. Perhaps the contiguous air contains an admixture of earthy or
watery matter and so follows the same natural law as the Earth, or perhaps
the air acquires motion from the perpetually rotating Earth by propinquity
and absence of resistance. . . .
We must admit the possibility of a double motion of objects which fall
and rise in the Universe, namely the resultant of rectilinear and circular
motion. [This is the analysis advocated earlier by Oresme.] Thus heavy
falling objects, being specially earthy, must doubtless retain the nature of
the whole to which they belong. . . . [Therefore a stone, for example, when
removed from the earth will continue to move circularly with the earth
and will simultaneously fall rectilinearly toward the earth's surface. Its net
153
will be some sort of spiral, like the motion of a bug that crawls
straight toward the center of a rotating potter's wheel.]
the of a simple body must be simple is true then primarily
of Circular motiOn, and only so long as the simple body rests in its own
natural pl.ace state. In that state no motion save circular is possible, for
motiOn IS wholly self-contained and similar to being at rest. But if
obJects o: mo:ed from their natural place rectilinear motion super-
Now It IS Inconsistent with the whole order and form of the Universe
that. It should be outside its own place. Therefore there is no rectilinear
motion sav.e of objects out of their right place, nor is such motion natural to
perfect obJects, smce [by such a motion] they would be separated from
the t,o which they belong and thus would destroy its unity. . . .
[CoperniCus argument shows how quickly the traditional distinction be-
tween the terrestrial and the celestial regions must disappear when the
earth becomes a he is here simply applying a traditional argu-
ment about cel.estial bodies to the earth. Circular motion, whether simple
com?ound, IS the nearest thing to rest. It can be natural to the earth
JUSt as It has been to the heavens, because it cannot disrupt
the observed umty and regulanty of the universe. Linear motion, on the
other hand, cannot be natural to any object that has achieved its
0
1
f 1' t' . d' .
wn pace,
or mo IOn IS Isruphve and a natural motion that destroys the uni-
verse IS absurd.]
Further, conceive immobility to be nobler and more divine than
change and mconstancy, which latter is thus more appropriate to Earth
than to the Universe. Would it not then seem absurd to ascribe motion to
that which contains or locates, and not rather to that contained and located
namely the Earth?
'
since the planets approach and recede from the Earth, both
therr motiOn round .the which is held [by Aristotelians] to be the
Earth, and also their motion outward and inward are the motion f
body. .this violates the very laws from which Aristotelians
central position of the earth, for according to these laws the planets should
have a single motion.] Therefore we must accept this motion round the
In a more general sense, and must be satisfied provided that every
has a proper center. From all these considerations it is more probable
the moves .than that it remains at rest. This is especially the case
th the dmrnal rotation, as being particularly a property of the Earth.
9. Whether more than one Motion can be attributed to the
Earth, and of the center of the Universe.
Since there is no reason why the Earth should not possess the
of motwn, we must consider whether in fact it has more motions
an one, so as to be reckoned as a Planet.
e Earth is not the center of all revolutions is proved by the appar-
ntly Irregular motions of the planets and the variations in their distances
154
from the Earth. These would be unintelligible if they moved in circles con-
centric with Earth. Since, therefore, there are more centers than one [that is,
a center for all the orbital motions, a center of the earth itself, and perhaps
others besides], we may discuss whether the center of the Universe is or
is not the Earth's center of gravity.
Now it seems to me gravity is but a natural inclination, bestowed on the
parts of bodies by the Creator so as .to co.mbine in the form of
a sphere and thus contribute to their umty and mtegnty. And we may
believe this property present even in the Sun, Moon, and Planets, so that
thereby they retain their spherical form notwithstanding their various
If therefore the Earth also has other motions, these must necessanly
r:semble the,many outside [planetary] motions having a yearly period
the earth now seems like a planet in so many other respects]. For If we
transfer the motion of the Sun to the Earth, taking the Sun to be at rest,
then morning and evening risings and settings of Stars will be unaffected,
while the stationary points, retrogressions, and progressions of
are due not to their own motions, but to that of the Earth, which theu
appearances reflect. Finally we shall place the Sun at .the center of
the Universe. All this is suggested by the systematic processiOn of events
and the harmony of the whole Universe, if only we face the facts, as they
"with both eyes open."
In these last three chapters we have Copernicus' theory of motion,
a conceptual scheme that he designed to permit his transposing the
earth and sun without tearing apart an essentially Aristotelian uni-
verse in the process. According to Copernicus' physics all matter,
celestial and terrestrial, aggregates naturally into spheres, and the
spheres then rotate of their own nature. A bit of matter separated from
its natural position will continue to rotate with its sphere, simultane-
returning to its natural place by a rectilinear motion. It is a
singularly incongruous theory (as Chapter 6 will demonstrate in more
detail), and, in all but its most incongruous portions, it is a relatively
unoriginal one. Copernicus may possibly have reinvented it for him-
self, but most of the essential elements in both his criticism of Aristotle
and his theory of motion can be found in earlier scholastic writers, par-
ticularly in Oresrne. Only when applied to Oresme's more limited
problem, they are less implausible.
Failure to provide an adequate physical basis for the earth's motion
does not discredit Copernicus. He did not conceive or accept the
earth's motion for reasons drawn from physics. The physical anc.l
cosmological problem treated so crudely in the First Book are of his
making, but they are not really his problems; he might have avoided
them altogether if he could. But the inadequacies of Copernicus'
physics do illustrate the way in which the consequences of his astro-
nomical innovation transcend the astronomical problem from which
the innovation was derived, and they do show how little the author
of the innovation was himself able to assimilate the Revolution born
from his work. The moving earth is an anomaly in a classical Aris-
totelian universe, but the universe of the De Revolutionibus is classical
in every respect that Copernicus can make seem compatible with the
motion of the earth. As he says himself, the motion of the sun has sim-
ply been transferred to the earth. The sun is not yet a star but the
unique central body about which the universe is constructed; it in-
herits the old functions of the earth and some new ones besides. As
we shall soon discover, Copernicus' universe is still finite, and con-
centric nesting spheres still move all planets, even though they can
no longer be driven by the outer sphere, which is now at rest. All
motions must be compounded of circles; moving the earth does not
even enable Copernicus to dispense with epicycles. The Copernican
Revolution, as we know it, is scarcely to be found in the De Revolu-
tionibus and that is the second essential incongruity of the text.
Copernican Astronomy- The Two Spheres
We have not quite finished with Copernicus' First Book.
But Chapters 10 and 11, which immediately follow the last section
quoted above, deal with more nearly astronomical matters, and we
shall consider them in the context of an astronomical discussion which
goes beyond the arguments that Copernicus made accessible for lay
readers. We shall again tum briefly to Copernicus' text in a later
section, but first we shall try to discover why astronomers might have
been more impressed than laymen with Copernicus' proposal. That
can scarcely be discovered anywhere in the First Book.
Copernicus endowed the earth with three simultaneous circular
motions: a diurnal axial rotation, an annual orbital motion and an
annual conical motion of the axis. The eastward diurnal is
the one that accounts for the apparent diurnal circles traced by the
stars, sun, moon, and planets. If the earth is situated at the center of
sphere of the stars, and rotates eastward daily about an axis through
Its own north and south poles, then all objects that are stationary or
nearly stationary with respect to the sphere of the stars will seem to
travel westward in circular arcs above the horizon, arcs just like those
in which the celestial bodies are observed to move in any short period
of time.
If Copernicus' or Oresme' s arguments to this effect are obscure,
refer again to the star trails shown in Figures 6 and 7 ( pp. 18 and 19).
Those tracks could be produced either by a circular motion of the
stars in front of a fixed observer (Ptolemy's explanation) or by a
rotation of the observer in front of fixed stars (Copernicus' explana-
tion). Or examine the new two-sphere universe shown in Figure 26, a
Figure 26. A rotating earth at the center of a fixed stellar sphere. In com-
paring this diagram with Figure 11, notice that here the horizon plane must be
turned with the earth, so that its geometric relation to the moving observer 0
stays fixed.
simpiified copy of the drawing that we first used in discussing the
motions of stars in the two-sphere universe (Figure 11, p. 31) except
that in the new version the poles are shown for the earth, not for
the celestial sphere, and the direction of rotation has been reversed.
When we first used a diagram like this, we held the earth, the ob-
server, and the horizon plane fixed, and we turned the sphere of the
stars westward. Now we must hold the outer sphere fixed and spin
the earth, observer, and horizon plane together eastward. An observer
sitting at the center of the horizon plane and moving with it will not
be able to tell, at least from anything he can see in the skies, any
difference between the two cases. In both he will see stars and planets
157
emerge along the eastern rim of the horizon and travel overhead to
the western horizon in the same circular paths.
To this point we have kept the spinning earth at the center of
the stationary sphere of the stars; we have, that is, considered the
model of the universe suggested by Heraclides and developed by
Oresme. This is only the first step toward a Copernican universe, how-
ever, and the next one is both more radical and more difficult. As
Copernicus points out in the portion of Chapter 5 already quoted, if
we are prepared to admit the possibility of the earth's motion at all,
we must be prepared to consider not only a motion at the center, but
also a motion of the earth away from the center. In fact, says Co-
pernicus, a moving earth need not be at the center. It need only be
relatively near the center, and as long as it stays close enough to the
center it may move about at will without affecting the apparent mo-
tions of the stars. This was a difficult conclusion for his astronomically
trained colleagues to accept because, in contrast to the conception of
the earth's immobility, which derives only from common sense and
from terrestrial physics, the notion of the earth's central position can
apparently be derived directly from astronomical observation. Co-
pernicus' conception of a noncentral earth therefore seemed initially
to conflict with the immediate consequences of pure astronomical ob-
servation, and it was to avoid this conflict, or a closely related one
which we shall consider at the end of the next section, that Copernicus
was forced to increase vastly the size of the sphere of the stars and
to take a first step toward the conception of an infinite universe elab-
orated by his successors. Copernicus' discussion of the earth's position
occurs in Chapter 6 of his First Book. Here we shall need a clearer
and more comprehensive version.
The earth's central position within the sphere of the stars can appar-
ently be derived from the observation that the horizon of any ter-
restrial observer bisects the stellar sphere. The vernal equinox and
the autumnal equinox are, for example, two diametrically opposite
points on the sphere of the stars, for they are defined as the intersections
of two great circles on the sphere, the equator and the ecliptic. Observa-
tion shows that whenever one of these points is just rising over the
horizon on the east, the other is just setting in the west. The same is
true of any other pair of diametrically opposite points on the sphere:
whenever one rises, the other sets. Apparently these observations can
158
be explained only if, as shown in Figure 26 or the earlier Figure 11,
the horizon plane is drawn through the center of the sphere of the
stars so that it, too, will intersect the sphere in a great circle. If and
only if the horizon plane intersects the sphere of the stars in a great
circle will diametrically opposite points on the sphere always rise and
set at the same moment.
But all horizon planes must also be drawn tangent to the spheri-
cal earth. (We have avoided this construction in Figures 26 and 11
only because we have there shown the earth immensely exaggerated in
size.) Therefore the observer must himself be at, or very uParl) at,
the center of the sphere of the sfars. The entire surface ot the terres-
trial sphere itself must be at or very nearly at the center; the earth
must be very small, almost a point, and it must be centrally located.
If, as in Figure 27, the earth (represented by the inner concentric
'Figure 27. If the earth's diameter is appreciable compared with that of the
sphere of the stars or if the earth is appreciably displaced from the center, the
horizon plane does not bisect the stellar sphere.
circle) were quite large with respect to the sphere of the stars or if
the earth (now represented by the black dot) were small but displaced
from the center, then the horizon plane would not seem to bisect the
sphere of the stars, and diametrically opposite points on the sphere
would not rise and set together.
As developed here the argument itself makes clear the weakness
exploited by Copernicus. Observation does not show that the earth
must be a point (if it did, even the Aristotelian and Ptolemaic uni-
verse would conflict with observation) or that it must be precisely
at the center, because observation can never say that, for example,
the vernal equinox rises exactly as the autumnal equinox sets. Crude
naked-eye observations will show that when the vernal equinox is
just setting, the autumnal equinox is within a degree or so of the hori-
zon. Refined naked-eye observation (appropriately corrected for
atmospheric refraction and for the irregularities Qf any actual hori-
zon) might show that when the winter solstice has just reached the
western horizon, the summer solstice is within 6' (or 0.1) of the
eastern horizon. But no naked-eye observation will do much better.
It can show only that the horizon very nearly bisects the sphere and
that all terrestrial observers must therefore be very close to the center
of the universe. Just how nearly the horizon bisects the sphere and
just how close to the center terrestrial observers must be depends upon
the accuracy of observation.
For example, if we know from observation that whenever one
solstice lies on the horizon, the other is no more than 0.1 o away from
the horizon, then no terrestrial observer may ever be farther from the
center of the sphere of the stars than a distance which is 0.001 the radius
of that sphere. Or if observation tells us (and few naked-eye observa-
tions are even approximately this good) that with one solstice on the
horizon the other is no more than 0.01 away from the horizon, then
the inner sphere of Figure 27 may have a radius no larger than 0.0001
the radius of the outer sphere, and the entire earth must again lie
somewhere within the inner circle at all times. If the earth moved
outside the inner circle, then the horizon plane would fail to bisect
the sphere of the stars by more than O.OP, and our hypothetical ob-
servations would discover the discrepancy, but with the earth any-
where inside of the inner circle, the horizon plane will seem, within
the limits of observation, to bisect the sphere.
That is Copernicus' argument. Observation only forces us to keep
the earth somewhere inside of a small sphere concentric with the
sphere of the stars. Within that inner sphere the earth may move
freely without violating the appearances. In particular, the earth may
have an orbital motion about the center or about the central sun, pro-
vided that its orbit never carries it too far away from the center. And
"too far" means only "too far relative to the radius of the outer
sphere." If the radius of the outer sphere is known, then observations
160
of known accuracy place a limit upon the maximum radius of the
earth's orbit. If the size of the earth's orbit is known (and it can, in
theory, be determined by Aristarchus' technique for measuring the
earth-sun distance), then observations of known accuracy place a limit
upon the minimum size of the sphere of the stars. For example, if the
distance between the earth and sun is, as indicated by Aristarchus'
measurement described in the Technical Appendix, equal to 764 earth
diameters ( 1528 earth radii) and if observations are known to be
accurate within 0.1 , then the radius of the sphere of the stars must
be at least 1000 times the radius of the earth's orbit or at least 1,528,000
earth radii.
Our example is a useful one, because, though Copernicus' obser-
vations were not quite this accurate, those made by his immediate
successor, Brahe, were if anything slightly more accurate than 0.1 ,
Ours is a representative estimate of the minimum size of the sphere
of the stars by a sixteenth-century Copernican. In principle, there is
nothing absurd about the result, for in the sixteenth and seventeenth
centuries there was no direct way of determining the distance to the
sphere of the stars. Its radius might have been more than 1,500,000
earth radii. But if it were that large - and Copernicanism demanded
that it should be - then a real break with traditional cosmology must
be admitted. AI Fargani, for example, had estimated the radius of the
sphere as 20,110 earth radii, more than seventy-five times smaller than
the Copernican estimate. The Copernican universe must be vastly
larger than that of traditional cosmology. Its volume is at least 400,000
times as great. There is an immense amount of space between the
sphere of Saturn and the sphere of the stars. The neat functional co-
h e r ~ n c e of the nesting spheres of the traditional universe has been
violated, though Copernicus seems to remain sublimely unaware of
the break.
Copernican Astronomy- The Sun
Copernicus' argument permits an orbital motion of the earth
in a vastly expanded universe, but the point is academic unless the
orbital motion can be shown to be compatible with the observed
motions of the sun and other planets. It is to those motions that
Copernicus turns in Chapters 10 and 11 of his First Book. We may
best begin with an expanded paraphrase of Chapter 11, in which
Copernicus describes the orbital motion of the earth and considers its
effect upon the apparent position of the sun. For the moment assume,
as shown in Figure 28, that the centers of the universe, the sun, and
the earth's orbit all coincide. In the diagram the plane of the ecliptic
is viewed from a position near the north celestial pole; the sphere of the
stars is stationary; the earth travels regularly eastward in its orbit
once in a year; and it simultaneously spins eastward on its axis once
in every 23 hours 56 minutes. Provided that the earth's orbit is much
S1 8 ---- ---rs"::- E2
_ ...... ~ E
_...,..- I
EARTH'S
ORBIT
STATIONARY
STELLAR
SPHERE
Figure 28. As the earth moves in its Copernican orbit from E, to E
2
, the
apparent position of the central sun, S, seen against the sphere of the stars shifts
from S, to s .
smaller than the sphere of the stars, the axial rotation of the earth
will account precisely for the diurnal circles of the sun, moon, and
planets, as well as for those of the stars, because from any position
in the earth's orbit all of these bodies must be seen against the sphere
of the stars and must seem to move with it as the earth rotates.
In the diagram the earth is shown in two positions which it occupies
thirty days apart. In each position the sun is viewed against the sphere
of the stars, and both apparent positions of the sun must lie on the
ecliptic, which is now defined as the line in which the plane of the
earth's motion (a plane that includes the sun) intersects the sphere.
But as the earth has moved eastward from position E
1
to position
E2 in the diagram, the sun has apparently moved eastward along the
ecliptic from position S1 to position S
2
Copernicus' theory therefore
predicts just the same eastward annual motion of the sun along the
ecliptic as the Ptolemaic theory. It also predicts, as we shall discover
162
immediately, the same seasonal variation of the height of the sun in
the sky.
Figure 29 shows the earth's orbit viewed from a point in the celes-
tial sphere slightly north of the autumnal equinox. The earth is
drawn at the four positions occupied successively at the vernal equinox,
the summer solstice, the autumnal equinox, and the winter solstice. In
all four of these positions, as throughout its motion, the earth's axis
remains parallel to an imaginary line passing through the sun and
tilted 2 3 ~

from a perpendicular to the plane of the ecliptic. Two
SUMMER
SOLSTICE
VERNAL
EQUINOX
AUTUMNAL
EQUINOX
WINTER
SOLSTICE
Figure 29. The earth's annual motion around its Copernican orbit. At all times
the earth's axis stays parallel to itself or to the stationary line drawn through the
sun. As a result an observer 0 at noon in middle-northern latitudes finds the sun
much more nearly overhead at the summer than at the winter solstice .
..
little arrows in the diagram show the position of a terrestrial observer
in middle-northern latitudes at local noon on June 22 and December 22,
the two solstices. Lines from the sun to the earth (not shown in the
diagram) indicate the direction of the rays of the noon sun, which
is clearly more nearly over the observer's head during the summer
solstice than during the winter solstice. A similar construction deter-
mines the sun's elevation at the equinoxes and at intermediate seasons.
The seasonal variation of the sun's elevation can therefore be
completely diagnosed from Figure 29. In practice, however, it is
simpler to revert to the Ptolemaic explanation. Since in every season
the sun appears to occupy the same position among the stars in the
Copernican as in the Ptolemaic system, it must rise and set with the
same stars in both systems. The correlation of the seasons with the
apparent position of the sun along the ecliptic cannot be affected by
the transition. With respect to the apparent motions of the sun and
stars the two systems are equivalent, and the Ptolemaic is simpler.
The last diagram also reveals two other interesting features of Co-
pernicus' system. Since it is the rotation of the earth that produces the
diurnal circles of the stars, the earth's axis must point to the center of
those circles in the celestial sphere. But, as the diagram indicates, the
earth's axis never does point to quite the same positions on the celes-
tial sphere from one year's end to the next. According to the Coperni-
can theory the extension of the earth's axis traces, during the course of
a year, two small circles on the sphere of the stars, one around the
north celestial pole and one around the south. To an observer on the
earth the center of the diurnal circles of the stars should itself seem
to move in a small circle about the celestial pole once each year. Or,
to put the same point in a way more closely related to observation, each
of the stars should seem slightly to change its position on the sphere of
the stars (or with respect to the observed pole of the sphere) during
the course of a year.
This apparent motion, which cannot be seen with the naked eye
and which was not even seen with telescopes until 1838, is known as
Figure 30. The annual parallax of a star. Because the line between a terrestrial
observer and a fixed star does not stay quite parallel to itself as the earth moves
in its orbit, the star's apparent position on the stellar sphere should shift by an
angle p during an interval of six months.
164
the parallactic motion. Because two lines drawn to a star from dia-
metrically opposite points on the earth's orbit are not quite parallel
(Figure 30), the apparent angular position of the star viewed from
the earth should be different at different seasons. But if the distance to
the star is very much greater than the distance across the earth's
orbit, then the angle of parallax, p in Figure 30, will be very, very
small, and the change in the apparent position of the star will not
be appreciable. The parallactic motion is not apparent only because
the stars are so very far away relative to the dimensions of the earth's
orbit. The situation is precisely equivalent to the one we discussed
above when considering why the earth's motion did not seem to
change the intersection of the horizon plane and the sphere of the
stars. In fact, we are dealing with the same problem. But the present
version of the problem is a more important one, because near the
horizon it is very difficult to make the precise measurements of stellar
position required to discover whether the horizon bisects the stellar
sphere. Unlike the rising and setting of the equinoxes, discussed above,
the search for parallactic motions need not be restricted to the horizon.
Parallax therefore provides a much more sensitive observational check
upon the minimum size of the sphere of the stars relative to the size
of the earth's orbit than is provided by the position of the horizon,
and the Copernican estimates of the sphere's size given above ought
really to have been derived from a discussion of parallax.
The second point illuminated by considering Figure 29 is not
about the skies at all but about Copernicus. We described the orbital
motion illustrated in the diagram as a single motion by which the
center is carried in a circle about the sun while its axis re-
mains always parallel to a fixed line through the sun. Copernicus de-
scribes the same physical motion as consisting of two simultaneous
mathematical motions. That is why he gives the earth a total of three
circular motions. And the reasons for his description give another sig-
nificant illustration of the extent to which his thought was bound to
the traditional patterns of Aristotelian thought. For him the earth
is a planet which is carried about the central sun by a sphere just
like the one that used to carry the sun about the central earth. If the
earth were firmly fixed in a sphere, its axis would not always stay
parallel to the same line through the sun; it would instead be carried
about by the sphere's rotation and would occupy the positions shown
165
in Figure 3la. After the earth had revolved 180 about the sun, the
earth's axis would still be tilted

away from the perpendicular
but in a direction opposite to that in which it had begun. To undo
this change in the direction of the axis, caused by the rotation of the
(a) (b)
MOTION OF NORTH
END OF AXIS
Figure 31. Copernicus' "second" and "third" motions. The second motion, that
of a planet fixed in a rotating sun-centered sphere, is shown in (a). This motion
does not keep the earth's axis parallel to itself, so that the conical third motion
shown in (b) is required to bring the axis back into line.
sphere that carries the earth, Copernicus requires a third circular
motion, this one applied to the axis of the earth only and shown in
Figure 3lb. It is a conical motion, which carries the north end of the
axis once westward each year, and thus just compensates for the effect
on the earth's axis of the orbital motion.
Copernican Astronomy- The Planets
So far the conceptual scheme developed by Copernicus is
just as effective as Ptolemy's, but it is surely no more so, and it seems
a good deal more cumbersome. It is only when the planets are added
to Copernicus' universe that any real basis for his innovation becomes
apparent. Consider, for example, the explanation of retrograde motion
to which Copernicus alluded without discussion at the end of Chap-
ter 5 in his introductory First Book. In the Ptolemaic system the retro-
grade motion of each planet is accounted for by placing the planet
on a rr.djor epicycle whose center is, in tum, carried about the earth
by the planet's deferent. The combined motion of these two circles
produces the characteristic looped patterns discussed in Chapter 3.
166
In Copernicus' system no major epicycles are required. The retrograde
or westward motion of a planet among the stars is only an apparent
motion, produced, like the apparent motion of the sun around the
ecliptic, by tht; orbital motion of the earth. According to C0pernicus
the motion tl1at Ptolemy had explained with major epicycles was
really the motion of the earth, attributed to the planets by a terrestrial
observer who thought himself stationary.
(a)
(b)
Figure 32. The Copernican explanation of retrograde motion for (a). superi?r
planets and (b) inferior planets. In each diagram the earth moves steadily on
orbit from E, to E, and the planet moves from P, to P,. Simultaneously the planets
apparent position the stellar sphere shifts from 1 to 7, but as the
two planets pass there is a brief westward retrogressiOn from 3 to 5.
The basis of Copernicus' contention is illustrated and clarified by
Figures 32a and 32b. Successive apparent positions of a moving su-
perior planet viewed from a moving earth against the fixed back-
ground provided by the stellar sphere are shown in diagram;
the second shows successive apparent positions of an Infenor planet.
Only the orbital motions are indicated; the earth's diurnal rotation,
which produces the rapid apparent westward motion of the.sun,
and stars together, is omitted. In both diagrams successive positiOns
167
of the earth in its sun-centered circular orbit are indicated by the
points Et. E2, ... , E
1; the corresponding consecutive positions of the
planets are marked P
1, P2, ... , P
7
; and the corresponding apparent
positions of the planet, discovered by extending a line from the earth
through the planet until it intersects the stellar sphere, are labeled
1, 2, ... , 7. In each case the more central planet moves more rapidly
in its orbit. Inspection of the diagram indicates that the apparent
motion of the planet among the stars is normal (eastward) from 1
to 2 and from 2 to 3; then the planet appears to retrogress (move west-
ward) from 3 to 4 and from 4 to 5; and finally it reverses its motion
again and moves normally from 5 to 6 and from 6 to 7. As the earth
completes the balance of its orbit, the planet continues in normal
motion, moving eastward most rapidly when it lies diametrically
across the sun from the earth.
Therefore, in Copernicus' system, planets viewed from the earth
should appear to move eastward most of the time; they retrogress
only when the earth, in its more rapid orbital motion, overtakes them
(superior planets) or when they overtake the earth (inferior planets).
Retrograde motion can occur only when the earth is nearest to the
planet whose motion is observed, and this is in accord with observa-
tions. Superior planets, at least, are most brilliant when they move
westward. The first major irregularity of planetary motion has been
explained qualitatively without the use of epicycles.
Figure 33 indicates how Copernicus' proposal accounts for a sec-
ond major irregularity of the planetary motions - the discrepancy
between the times required for successive trips of a planet around the
ecliptic. In the diagram it is assumed that the earth completes 1% east-
ward trips about its orbit while the planet, in this case a superior planet,
travels eastward through its orbit once. Suppose that at the start of the
series of observations the earth is at E
1 and the planet at P. The planet
is then in the middle of a retrogression and appears silhouetted against
the stationary stellar sphere at 1. When the planet has completed one
revolution in its orbit and returned to P, the earth has made U4 trips
around its orbit and reached E2. The planet therefore is seen at 2,
west of position 1 at which it started. It has not yet completed a full
journey around the ecliptic, and its first full trip will therefore con-
sume more time than the planet required to revolve once in its orbit.
As the planet makes its second trip about its orbit, the earth again
makes more than one orbital revolution and reaches E3 when the
planet has returned to P again. This time the planet is seen silhouetted
at 3, to the east of position 2. It has completed more than one journey
around the ecliptic while moving only once through its orbit, and its
second journey around the ecliptic was therefore a very rapid one.
Mter a third revolution the planet is again at P, but it appears at
position 4, east of 3, and its journey around the ecliptic was therefore
~ ~ ~ H E R E
~ \ ; / ~ S T R S
\ I I
\ I I
\I I
/p
Figure 33. The Copernican explanation of variations in the time required for a
superior planet to complete successive journeys around the ecliptic. While the
planet moves once eastward around its orbit from P to P, the earth makes 1 east-
ward revolutions from E1 to E1 and on to E2. During this interval the apparent posi-
tion of the planet among the stars moves eastward from l to 2, slightly less than a
full trip. During the planet's next revolution the earth moves from E
2
to E
2
and on
!{l E3, so that its apparent position among the stars shifts from 2 to 1 and on to 1
again, slightly more than one full trip around the ecliptic.
again a fast one. After a fourth revolution in its orbit the planet again
appears at 1, west of 4, and its final trip was therefore slow. The planet
has completed four trips about its orbit and four trips around the
ecliptic at the same instant. The average time required by a superior
planet to circle the ecliptic is therefore identical with the planet's
orbital period. But the time required for an individual trip may be
considerably greater or considerably less than the average. A similar
argument will account for the similar irregularities of an inferior
planet's motion.
Retrograde motion and the variation of the time required to circle
the ecliptic are the two gross planetary irregularities which in an-
tiquity had led astronomers to employ epicycles and deferents in
treating the problem of the planets. Copernicus' system explains these
same gross irregularities, and it does so without resorting to epicycles,
or at least to major epicycles. To gain even an approximate and
qualitative account of the planetary motions Hipparchus and Ptolemy
had required twelve circles - one each for the sun and moon, and
two each for the five remaining "wanderers." Copernicus achieved
the same qualitative account of the apparent planetary motions with
only seven circles. He needed only one sun-centered circle for each
of the six known planets- Mercury, Venus, Earth, Mars, Jupiter,
and Saturn - and one additional earth-centered circle for the moon.
To an astronomer concerned only with a qualitative account of the
planetary motions, Copernicus' system must seem the more eco-
nomical.
But this apparent economy of the Copernican system, though it
is a propaganda victory that the proponents of the new astronomy
rarely failed to emphasize, is largely an illusion. We have not yet
begun to deal with the full complexity of Copernicus' planetary
astronomy. The seven-circle system presented in the First Book of
the De Revolutionibus, and in many modern elementary accounts of
the Copernican system, is a wonderfully economical system, but it
does not work. It will not predict the position of planets with an
accuracy comparable to that supplied by Ptolemy's system. Its ac-
curacy is comparable to that of a simplified twelve-circle version of
Ptolemy's system- Copernicus can give a more economical qualita-
tive account of the planetary motions than Ptolemy. But to gain a
reasonably good quantitative account of the alteration of planetary
position Ptolemy had been compelled to complicate the fundamental
twelve-circle system with minor epicycles, eccentrics, and equants,
and to get comparable results from his basic seven-circle system
Copernicus, too, was forced to use minor epicycles and eccentrics.
His full system was little if any less cumbersome than Ptolemy's had
been. Both employed over thirty circles; there was little to choose
between them in economy. Nor could the two systems be distin-
guished by their accuracy. When Copernicus had finished adding
circles, his cumbersome sun-centered system gave results as accurate
as Ptolemy's, but it did not give more accurate results. Copernicus
did not solve the problem of the planets.
The full Copernican system is described in the latter books of the
De Revolutionibus. Fortunately we need only illustrate the sorts of
complexities there developed. Copernicus' system was not, for ex-
ample, really a sun-centered system at all. To account for the in-
creased rate at which the sun travels through the signs of the zodiac
during the winter, Copernicus made the earth's circular orbit eccen-
tric, displacing its center from the sun's. To account for other irregu-
larities, indicated by ancient and contemporary observations of the
sun's motion, he kept this displaced center in motion. The center of
the earth's eccentric was placed upon a second circle whose motion
continually varied the extent and direction of the earth's eccentricity.
The final system employed to compute the earth's motion is repre-
sented approximately in Figure 34a. In the diagram, S is the sun,
fixed in space; the point 0, which itself moves slowly about the sun,
is the center of a slowly rotating circle that carries the moving center
OE of the earth's eccentric; E is the earth itself.
Similar complexities were necessitated by the observed motions of
the other heavenly bodies. For the moon Copernicus used a total of
three circles, the first centered on the moving earth, the second cen-
tered on the moving circumference of the first, and the third on the
(a) (b)
Figure 34. Copernicus' account of the motion of (a) the earth and (b) Mars.
In (a) the sun is at 5, and the earth, E, revolves on a circle whose center, OE, re-
volves slowly ahout a point 0, which in turn revolves on a sun-centered circle. In
(b) Mars is placed on an epicycle revolving on a deferent whose center, O,, main-
tains a fixed geometric relation to the moving center OE of the earth's orbit.
circumference of the second. For Mars and most of the other planets
he employed a system much like that illustrated in Figure 34b. The
center of Mars's orbit, OM, is displaced from the center of the earth's
orbit, OE, and is moved with it; the planet itself is placed at M, not
on the eccentric but on an epicycle, which rotates eastward in the
same direction and with the same period as the eccentric. Nor do the
complexities end here. Still other devices, fully equivalent to Ptolemy's,
were required to account for the north and south deviations of each
planet from the ecliptic.
Even this brief sketch of the complex system of interlocking circles
employed by Copernicus to compute planetary position indicates the
third great incongruity of the De Revolutionibus and the immense
irony of Copernicus' lifework. The preface to the De Revolutionibus
opens with a forceful indictment of Ptolemaic astronomy for its in-
accuracy, complexity, and inconsistency, yet before Copernicus' text
closes, it has convicted itself of exactly the same shortcomings.
Copernicus' system is neither simpler nor more accurate than Ptolemy's.
And the methods that Copernicus employed in constructing it seem
just as little likely as the methods of Ptolemy to produce a single
consistent solution of the problem of the planets. The De Revolu-
tionibus itself is not consistent with the single surviving early version
of the system, described by Copernicus in the early manuscript Com-
mentariolus. Even Copernicus could not derive from his hypothesis
a single and unique combination of interlocking circles, and his
successors did not do so. Those features of the ancient tradition which
had led Copernicus to attempt a radical innovation were not elimi-
nated by that innovation. Copernicus had rejected the Ptolemaic
tradition because of his discovery that "the Mathematicians are in-
consistent in these [astronomical] investigations" and because "if
their hypotheses were not misleading, all inferences based thereon
might surely be verified." A new Copernicus could have turned the
identical arguments against him.
The Harmony of the Copernican System
Judged on purely practical grounds, Copernicus' new plane-
tary system was a failure; it was neither more accurate nor significantly
simpler than its Ptolemaic predecessors. But historically the new sys-
172
tern was a great success; the De Revolutionibus did convince a few
of Copernicus' successors that sun-centered astronomy held the key
to the problem of the planets, and these men finally provided the
simple and accurate solution that Copernicus had sought. We shall
examine their work in the next chapter, but first we must try to dis-
cover why they became Copernicans - in the absence of increased
economy or precision, what reasons were there for transposing the
earth and the sun? The answer to this question is not easily dis-
entangled from the technical details that fill the De Revolutionibus,
because, as Copernicus himself recognized, the real appeal of sun-
centered astronomy was aesthetic rather than pragmatic. To astrono-
mers the initial choice between Copernicus' system and Ptolemy's
could only be a matter of taste, and matters of taste are the most
difficult of all to define or debate. Yet, as the Copernican Revolution
itself indicates, matters of taste are not negligible. The ear equipped
to discern geometric harmony could detect a new neatness and co-
herence in the sun-centered astronomy of Copernicus, and if that
neatness and coherence had not been recognized, there might have
been no Revolution.
We have already examined one of the aesthetic advantages of
Copernicus' system. It explains the principal qualitative features of
the planetary motions without using epicycles. Retrograde motion,
in particular, is transformed to a natural and immediate consequence
of the geometry of sun-centered orbits. But only astronomers who
valued qualitative neatness far more than quantitative accuracy (and
there were a few - Galileo among them) could consider this a con-
l'incing argument in the face of the complex system of epicycles and
eccentrics elaborated in the De Revolutionibus. Fortunately there were
other, less ephemeral, arguments for the new system. For example,
it gives a simpler and far more natural account than Ptolemy's of
the motions of the inferior planets. Mercury and Venus never get very
far from the sun, and Ptolemaic astronomy accounts for this observa-
tion by tying the deferents of Mercury, Venus, and the sun together
so that the center of the epicycle of each inferior planet always lies
on a straight line between the earth and the sun (Figure 35a). This
alignment of the centers of the epicycles is an "extra" device, an ad
hoc addition to the geometry of earth-centered astronomy, and there
is no need for such an assumption in Copernicus' system. When, as in
173
Figure 35b, the orbit of a planet lies entirely within the earth's orbit,
there is no way in which the planet can appear far from the sun.
Maximum elongation will occur when, as in the diagram, the line
from the earth to the planet is tangent to the planet's orbit and the
SPE is a right angle. Therefore the angle of elongation, SEP,
IS the largest angle by which the inferior planet can deviate from the
sun. The basic geometry of the system fully accounts for the way in
which Mercury and Venus are bound to the sun.
s
(a )
(b)
.Figure 35. Limited elongation of inferior planets explained in (a) the Ptole-
maiC and (b) the Copernican systems. In the Ptolemaic system the angle be-
tween _the sun, S, and the planet, P, must be restricted by keeping the center of
epiCycle on the earth and the sun. In the Copernican system,
w1th the planets orbit entirely contained by the earth's, no such restriction is
necessary.
Copernican geometry illuminates another even more important
of. the behavior of the inferior planets, namely, the order of
therr orbits. In Ptolemaic system the planets were arranged in
earth-centered so that the average distance between a planet
and the earth mcreased with the time required for the planet to
traverse the ecliptic. The device worked well for the superior planets
and for the but Mercury, Venus, and the sun all require 1 year
for an average JOurney around the ecliptic, and the order of their
orbits had always been a source of debate. In the Copernican
system there Is no place for similar debate; no two planets have the
sam: orbital period. The moon is no longer involved in the problem,
for It travels about the earth rather than about the central sun. The
174
iter and Saturn, preserve their old order
superior planets, Mars, Jup , th . b'tal periods are the same as
h nt r because eu or I
,
about t e new ce e ,
d t . le the ecliptic. The earths
h f ti they nee o cue
the average lengt s o . the earth's orbital period, 1 year, is
orbit lies inside of Mars s, smce . t lace Mercury and Venus
th M ' 687 days It only remams o P
less an ars s
. . . f r the first time, uniquely deter-
in the system, and theu order IS, o
mined.
V . known to retrogress every
This can be seen as follows. IS be observed only when
d . trograde motion can
584 days, an smce re
b the time Venus requires
h th 584 days must e
.
Venus passes t e ear , .
. 't of the sun. Now m
rth . their common circm
to lap the ea once m d . b't
584
( = 12I9) times. Since Ve-
584 days the earth has traverse must have circled
d th th once durmg IS m erv ,
nus has lappe e ear . . 84 d B t a planet that circles its
. b' 2 219( _ 949) times m JUSt 5 ays. u
. I
Its or It
365
- 365
.
584
X 365 ( = 225) days to crrc e
orbit 949 times in 584 days must reqmre . 949 d . less than
365
f r since Venus's penod, 225 ays, IS
its orbit once. There o e,
th' d there is no am-
b 'd th ear s an
th's Venus's orbit must e msi e e '
' d
ear ,
.
M 's orbit inside Venus san
b. 'ty A similar calculation places ercury
1 th
Igm .
1 retrogresses and therefore aps e
closest to the sun. Since :tv ercury . : . 116 ( = 48I) times
th 116 days it must complete Its orbit JUSt 1 365 . 365 365
ear , every Th f ' 't will complete its orbit just once m 116 Xru
in 116 days. I . d of 88 days is the shortest of all, and it is
( = 88) days. Its or Ita peno
therefore the planet closest to the sun.
b "ts with the
So far we have ordered the. sun-centered o:a:th-centered
d . d by Ptolemaic astronomers
svme eviCe use
th t of the universe take longer to
b I t farther from e cen er
or its: p ane s
. th th . e of the orbit increases
Th assumptiOn at e SIZ
circle the center. e
. d f II n the Copernican than
with orbital period can be apphe more u y I. . . "ti lly arbitrary.
b . both systems It IS Im a
in the Ptolemaic system, ut m b h th" way like Vitruvius'
al th t planets should e ave Is ,
It seems natur a .
'ty that they do so. Perhaps the
h I but there IS no necessi
h
ants on a w ee ,
.
d tl lanets excepting t e sun
. . r ly gratmtous, an le p ,
h
assumption IS en Ire
b d' tl determined have anot er
h distances can e Irec Y
'
and moon, w ose
order.
d dering constitutes another very
The response to this suggeste reor . d the Ptolemaic sys-
ff
b tween the Copermcan an
important di erence e
. d . his preface Copernicus
d e which as we discovere m
,
terns, an on
'
175
himself particularly emphasizes. In the Ptolemaic system the deferent
and epicycle of any one planet can be shrunk or expanded at will
without affecting either the sizes of the other planetary orbits or the
position at which the planet, viewed from a central earth, appears
against the stars. The order of the orbits may be determined by assum-
ing a relation between size of orbit and orbital period. In addition, the
relative dimensions of the orbits mny be worked out with the aid of
the further assumption, discussed in Chapter 3, that the minimum dis-
tance of one planet from the earth is just equal to the maximum dis-
tance between the earth and the next interior planet. But though both
of these seem natirral assumptions, neither is necessary. The Ptolemaic
system could predict the same apparent positions for the planets with-
out making use of either. In the Ptolemaic system the appearances are
not dependent upon the order or the sizes of the planetary orbits.
There is no similar freedom in the Copernican system. If all the
planets revolve in approximately circular orbits about the sun, then
both the order and the relative sizes of the orbits can be determined
directly from observation without additional assumptions. Any change
in order or even in relative size of the orbits will upset the whole
system. For example, Figure 36a shows. an inferior planet, P, viewed
from the earth at the time when it reaches its maximum elongation
from the sun. The orbit is assumed circular, and the angle SPE must
therefore be a right angle when the angle of elongation, SEP, reaches
its maximum value. The planet, the sun, and the earth form a right
triangle one of whose acute angles, SEP, can be directly measured.
But knowledge of one acute angle of a right triangle determines the
ratio of the lengths of the sides of that triangle. Therefore the ratio
of the radius of the inferior planet's orbit, SP, to the radius of the earth's
orbit, SE, can be computed from the measured value of the angle SEP.
The relative sizes of the earth's orbit and the orbits of both inferior
planets can be discovered from observation.
An equivalent determination can be made for a superior planet,
though the techniques are more complex. One possible technique is
illustrated in Figure 36b. Suppose that at some determined instant
of time the sun, ilie earth, and the planet all lie on the straight line
SEP; this is the orientation in which the planet lies diametrically
across the ecliptic from the sun and is in ilie middle of a retrograde
motion. Since the earth traverses its orbit more rapidly than any su-
176
{a )
{b)
relative dimensions of orbits in the Copernican
(b) for a superior planet.
Figure 36. Determining the
system: (a) for an inferior planet;
. 1 t there must be some later instant of time when the earth
penor pane,
'P' 'th th
at E' and the planet at P' will form a right angle SE WI e sun,
d
. SE'P' I's the angle between the sun and the supenor planet
an smce
. d h
. d f om the earth it can be directly determmed an t e time
v1ewe r '
'E' be
re uired to achieve it can be measured. The angle :s can
d
qt ined for it must bear the same ratio to 360 as the time re-
e erm '
' h 365 d s that
uired by the earth to move from E to E bears to t e ' ay
le earth requires to complete its orbit. The angle PSP can be de-
. h t' ired by the planet
. d . J'ust the same way smce t e tme requ
termme m ' d b th
to complete its orbit is already known, and the time occupie y
lanet in oing from p to P' is the same as that needed by
P f g E to E' With PSP' and ESE' known, the angle p SE can
to go rom
. 1 SE'P'
found by subtraction. Then we again have a nght tnang_ e, '
1 P
'SE' known and the ratio of the radms of the
with one acute ang e, '
b
1 t
, bt't SP' to that of the earth's orbit, SE', can therefore e
pane s or , ,
determined just as for an inferior planet. b
B techniques like this the distances to all the planets can e
det:rmined in terms of the distance between the earth the sun,
. terms of any unit, like the stade, in which the radms of . the
or m
h fi Copermcus
rth's orbit has been measured. Now, fort e rst time, as
:: s in his prefatory letter, "the orders and magnitudes ?f all stars
yd h s become so bound together that nothing m any part
an sp ere .
. f
thereof could be moved from its place without produ:.mg con usion
of all the other parts and of the universe as a whole. Because the
177
relative dimensions of the planetary orbits are a direct consequence
of the first geometric premises of sun-centered astronomy, the new
astronomy has for Copernicus a naturalness and coherence that were
lacking in the older earth-centered version. The structure of the
heavens can be derived from Copernicus' system with fewer extra-
neous or ad hoc assumptions like plenitude. That is the new and
aesthetic harmony which Copernicus emphasizes and illustrates so
fully in the tenth chapter of his introductory First Book, to which
we now turn, having first learned enough about the new system (as
Copernicus' lay readers had not) to understand what he is talking
about.
10. Of the Order of the Heavenly Bodies.
No one doubts that the Sphere of the Fixed Stars is the most distant of
visible things. As for the order of the planets, the early Philosophers wished
to determine it from the magnitude of their revolutions. They adduce the
fact that of objects moving with equal speed, those farther distant seem
to move more slowly (as is proved in Euclid's Optics). They think that
the Moon describes her path in the shortest time because, being nearest to
the Earth, she revolves in the smallest circle. Farthest they place Saturn,
who in the longest time describes the greatest circuit. Nearer than he is
Jupiter, and then Mars.
Opinions differ as to Venus and Mercury which, unlike the others, do
not altogether leave the Sun. Some place them beyond the Sun, as Plato in
Timaeus; others nearer than the Sun, as Ptolemy and many of the moderns.
Alpetragius [a twelfth-century Moslem astronomer] makes Venus nearer
and Mercury farther than the Sun. If we agree with Plato in thinking that
the planets are themselves dark bodies that do but reflect light from the
Sun, it must follow, that if "'earer than the Sun, on account of their prox-
imity to him they would appear as half or partial circles; for they would
generally reflect such light as they receive upwards, that is toward the Sun,
as with the waxing or waning Moon. [See the discussion of the phases of
Venus in the next chapter. Neither this effect nor the following is dis-
tinctly visible without the telescope.] Some think that since no eclipse
even proportional to their size is ever caused by these planets, they can
never be between us and the Sun. . . . [Copernicus proceeds to note
many difficulties in the arguments usually used to determine the relative
order of the sun and the inferior planets. Then he continues:]
Uncondncing too is Ptolemy's proof that the Sun moves between those
bodies that do and those that do not recede from him completely [that is,
between the superior planets which can assume any angle of elongation
and the inferior planets whose maximum elongation is limited]. Con-
178
sideration of the case of the Moon, which does so recede, exposes its false-
ness. Again, what cause can be alleged, by those who place Venus nearer
than the Sun, and Mercury next, or in some other order? Why should not
these planets also follow separate paths, distinct from that of the Sun, as
do the other planets [whose deferents are not tied to the sun's.]? And
might be said even if their relative swiftness and slowness d1d not
their alleged order. Either then the Earth cannot be the center .to
the order of the planets and their Spheres is related, or
relative order is not observed, nor does it appear why a higher pos1t10n
should be assigned to Saturn than to Jupiter, or any o.ther ..
Therefore I think we must seriously consider the mgemous v1ew held
by Martianus Capella [a Roman encyclopedist of the fifth century who
recorded a theory of the inferior planets probably first suggested by
Heraclides] ... and certain other Latins, that Venus and Mercury do
not go round the Earth like the other planets run their courses with
the Sun as center, and so do not depart from h1m farther than the con-
vexity of their Spheres allows. . . . What else can . they than that
the center of these Spheres is near the Sun? So certamly the crrcle of
cury must be within that of Venus, which, it is agreed, is more than tw1ce
as great. . .
We may now extend this hypothesis to bnng Saturn, Jupiter and Mars
also into rel.\tion with this center, making their Spheres great enough to
contain those of Venus and Mercury and the Earth .... These outer
planets are always nearer to the Earth about the time of their evening
rising, that is, when they are in opposition to the Sun, and the Earth be-
tween them and the Sun. They are more distant from the Earth at the
time of their evening setting, when they are in conjunction with the Sun
and the Sun between them and the Earth. These indications prove that
their center pertains rather to the Sun than to the Earth, and that this is
the same center as that to which the revolutions of Venus and Mercury
.
' [Copernicus' remarks do not actually "prove" a thing. Ptolemaic
system explains these phenomena as completely as Copernican, .but the
Copernican explanation is again more natural, for, hke the Copernican ex-
planation of the limited elongation of the planets, it depends
on the geometry of a sun-centered astronomical syste'?, on the pa.rtJCu-
lar orbital periods assigned to the planets. Copermcus remarks will be
clarified by reference to Figure 32a. A superior planet retrogresses
the earth overtakes it, and under these circumstances it must be simul-
taneously closest to the earth and across the ecliptic from the sun. In the
Ptolemaic system a retrogressing superior planet must be closer to the
earth than at any other time, and it is in fact also across the sky from
the sun. But it is only across the sky from the sun because the rates of
rotation of its deferent and epicycle have particular values that happen to
put the planet back in opposition to the sun whenever the epicycle brings
l
179
the. planet close to the central earth. If, in the Ptolemaic system, the
penod or deferent were quantitatively slightly different, then
the quahtatlve regularity that puts a retrogressing superior planet across
the sky from the sun would not occur. In the Copernican system it must
occur regardless of the particular rates at which the planets revolve in
their orbits.]
But since all these [Spheres] have one center it is necessary that the
space between the convex side of Venus's Sphere and the concave side
of Mars's must also be viewed as a Sphere concentric with the others
capable of receiving the Earth with her satellite the Moon and whateve;
is contained within the Sphere of the Moon - for we must not separate
the Moon from the Earth, the former being beyond all doubt nearest to
the latter, especially as in that space we find suitable and ample room
for the Moon.
We therefore assert that the center of the Earth, carrying the Moon's
passes in a great circuit among the other planets in an annual revolu-
tion round the Sun; that near the Sun is the center of the Universe; and
that whereas the Sun is at rest, any apparent motion of the Sun can be
better explained by motion of the Earth. Yet so great is the Universe that
distance of the Earth from the Sun is not insignificant compared
w1th the s1ze of any other planetary path, in accordance with the ratios of
their sizes, it is insignificant compared with the distances of the Sphere
of the Fixed Stars.
I think it easier to believe this than to confuse the issue by assuming
a vast number of Spheres, which those who keep Earth at the center must
do. We thus rather follow Nature, who producing nothing vain or super-
often yrefers to endow one cause with many effects. Though these
VIews are difficult, contrary to expectation, and certainly unusual, yet in
the sequel we shall, God willing, make them abundantly clear at least to
mathematicians.
Given the above view - and there is none more reasonable - that the
periodic times are proportional to the sizes of the Spheres, then the order
of the Spheres, beginning from the most distant is as follows. Most distant
of all is of the Fixed Stars, containing all things, and being
therefore Itself Immovable. It represents that to which the motion and posi-
tion of. all t?e other bodies must be referred . . . . Next is the planet Saturn,
revolvmg m 30 years. Next comes Jupiter, moving in a 12-year circuit;
then Mars, goes round in 2 years. The fourth place is held by the
revolutwn [of the Sphere] in which the Earth is contained, together
With the of the Moon as on an epicycle. Venus, whose period is
9 months, IS m the fifth place, and sixth is Mercury, who goes round in
the space of 80 days.
In the middle of all sits Sun enthroned. In this most beautiful temple
we place this luminary in any better position from which he can
illummate the whole at once? He is rightly called the Lamp, the Mind, the
Ruler of the Universe; Hermes Trismegistus names him the Visible God,
Sophocles' Electra calls him the All-seeing. So the Sun sits as upon a royal
throne ruling his children the planets which circle round him. The Earth
has the Moon at her service. As Aristotle says, in his On [the Generation
of] Animals, the Moon has the closest relationship with the Earth. Mean-
while the Earth conceives by the Sun, and becomes pregnant with an
annual rebirth.
So we find underlying this ordination an admirable symmetry in the
Universe, and a clear bond of harmony in the motion and magnitude of
the Spheres such as can be discovered in no other wise. For here we may
observe why the progression and retrogression appear greater for Jupiter
than Saturn, and less than for Mars, but again greater for Venus than for
Mercury [a glance at Figure 32 will show that the closer the orbit of a
planet is to the orbit of the earth, the larger the apparent retrograde
motion of that planet must be- an additional harmony of Copernicus'
system]; and why such oscillation appears more frequently in Saturn than
in Jupiter, but less frequently in Mars and Venus than in Mercury [the
earth will lap a slowly moving superior planet more frequently than it laps
a rapid one, and conversely for an inferior planet]; moreover why Saturn,
Jupiter and Mars are nearer to the Earth at opposition to the Sun than
when they are lost in or emerge from the Sun's rays. Particularly Mars,
when he shines all night [and is therefore in opposition], appears to rival
Jupiter in magnitude, being only distinguishable by his ruddy color; other-
wise he is scarce equal to a star of the second magnitude, and can be
recognized only when his movements are carefully followed. All these
phenomena proceed from the same cause, namely Earth's motion.
That there are no such phenomena for the fixed stars proves their
immeasurable distance, because of which the outer sphere's [apparent]
annual motion or its [parallactic] image is invisible to the eyes. For every
visible object has a certain distance beyond which it can no more be seen,
as is proved in optics. The twinkling of the stars, also, shows that there is
still a vast distance between the farthest of the planets, Saturn, and the
Sphere of the Fixed Stars [for if the stars were very near Saturn, they
should shine as he does], and it is chiefly by this indication that they are
distinguished from the planets. Further, there must necessarily be a great
difference between moving and non-moving bodies. So great is this divine
work of the Great and Noble Creator!
Throughout this crucially important tenth chapter Copernicus'
emphasis is upon the "admirable symmetry" and the "clear bond of
harmony in the motion and magnitude of the Spheres" that a sun-
centered geometry imparts to the appearances of the heavens. If the
sun is the center, then an inferior planet cannot possibly appear far
from the sun; if the sun is the center, then a superior planet must be
in opposition to the sun when it is closest to the earth; and so on and
on. It is through arguments like these that Copernicus seeks to per-
suade his contemporaries of the validity of his new approach. Each
argument cites an aspect of the appearances that can be explained by
either the Ptolemaic or the Copernican system, and each then pro-
ceeds to point out how much more harmonious, coherent, and natural
the Copernican explanation is. There are a great many such argu-
ments. The sum of the evidence drawn from harmony is nothing if
not impressive.
But it may well be nothing. "Harmony" seems a strange basis on
which to argue for the earth's motion, particularly since the harmony
is so obscured by the complex multitude of circles that make up the
full Copernican system. Copernicus' arguments are not pragmatic.
They appeal, if at all, not to the utilitarian sense of the practicing
astronomer but to his aesthetic sense and to that alone. They had
no appeal to laymen, who, even when they understood the arguments,
were unwilling to substitute minor celestial harmonies for major ter-
restrial discord. They did not necessarily appeal to astronomers, for
the harmonies to which Copernicus' arguments pointed did not enable
the astronomer to perform his job better. New harmonies did not
increase accuracy or simplicity. Therefore they could and did appeal
primarily to that limited and perhaps irrational subgroup of mathe-
matical astronomers whose Neoplatonic ear for mathematical har-
monies could not be obstructed by page after page of complex mathe-
matics leading finally to numerical predictions scarcely better than
those they had known before. Fortunately, as we shall discover in
the next chapter, there were a few such astronomers. Their work is
also an essential ingredient of the Copernican Revolution.
Revolution by Degrees
Because he was the first fully to develop an astronomical
system based upon the motion of the earth, Copernicus is frequently
called the first modern astronomer. But, as the text of the De Revolu-
tionibus indicates, an equally persuasive case might be made for
calling him the last great Ptolemaic astronomer. Ptolemaic astronomy
meant far more than astronomy predicated on a stationary earth, and
it is only with respect to the position and motion of the earth that
Copernicus broke with the Ptolemaic tradition. The cosmological
frame in which his astronomy was embedded, his physics, terrestrial
and celestial, and even the mathematical devices that he employed
to make his system give adequate predictions are all in the tradition
established by ancient and medieval scientists.
Though historians have occasionally grown livid arguing whether
Copernicus is really the last of the ancient or the first of the modern
astronomers, the debate is in principle absurd. Copernicus is neither
an ancient nor a modern but rather a Renaissance astronomer in whose
work the two traditions merge. To ask whether his work is really
ancient or modern is rather like asking whether the bend in an other-
wise straight road belongs to the section of road that precedes the
bend or to the portion that comes after it. From the bend both sections
of the road are visible, and its continuity is apparent. But viewed
from a point before the bend, the road seems to run straight to the
bend and then to disappear; the bend seems the last point in a straight
road. And viewed from a point in the next section, after the bend,
the road appears to begin at the bend from which it runs straight on.
The bend belongs equally to both sections, or it belongs to neither.
It marks a turning point in the direction of the road's progress, just
as the De Revolutionibus marks a shift in the direction in which astro-
nomical thought developed.
To this point in this chapter we have emphasized primarily the
ties between the De Revolutionibus and the earlier astronomical and
cosmological tradition. We have minimized, as Copernicus himself
does, the extent of the Copernican innovation, because we have been
concerned to discover how a potentially destructive innovation could
be produced by the tradition that it was ultimately to destroy. But, as
we shall soon discover, this is not the only legitimate way to view the
De Revolutionibus, and it is not the view taken by most later Coper-
nicans. For Copernicus' sixteenth- and seventeenth-century followers,
the primary importance of the De Revolutionibus derived from its
single novel concept, the planetary earth, and from the novel astro-
nomical consequences, the new harmonies, which Copernicus had
derived from that concept. To them Copernicanism meant the three-
fold motion of the earth and, initially, that alone. The traditional con-
ceptions with which Copernicus had clothed his innovation were not
to his followers essential elements of his work, simply because, as
traditional elements, they were not Copernicus' contribution to sci-
,....
183
ence. It was not because of its traditional elements that people
quarreled about the De Revolutionibus.
That is why the De Revolutionibus could be the starting point for
a new astronomical and cosmological tradition as well as the culmina-
tion of an old one. Those whom Copernicus converted to the concept
of a moving earth began their research from the point at which
Copernicus had stopped. Their starting point was the earth's motion,
which was all they necessarily took from Copernicus, and the prob-
lems to which they devoted themselves were not the problems of the
old astronomy, which had occupied Copernicus, but the problems
of the new sun-centered astronomy, which they discovered in the
De Revolutionibus. Copernicus presented them with a set of problems
that neither he nor his predecessors had had to face. In the pursuit
of those problems the Copernican Revolution was completed, and a
new astronomical tradition, deriving from the De Revolutionibus, was
founded. Modern astronomy looks back to the De Revolutionibus as
Copernicus had looked back to Hipparchus and Ptolemy.
Major upheavals in the fundamental concepts of science occur by
degrees. The work of a single individual may play a pret\minent role
in such a conceptual revolution, but if it does, it achieves preeminence
either because, like the De Revolutionibus, it initiates revolution by
a small innovation which presents science with new problems, or
because, like Newton's Principia, it terminates revolution by inte-
grating concepts derived from many sources. The extent of the inno-
vation that any individual can produce is necessarily limited, for
each individual must employ in his research the tools that he acquires
from a traditional education, and he cannot in his own lifetime replace
them all. It seems therefore that many of the elements in the De
Revolutionibus which, in the earlier parts of this chapter, we pointed
to as incongruities are not really incongruities at all. The De Revolu-
tionibus seems incongruous only to those who expect to find the entire
Copernican Revolution in the work which gives that revolution its
name, and such an expectation derives from a misunderstanding of
the way in which new patterns of scientific thought are produced.
The limitations of the De Revolutionibus might better be regarded
as essential and typical characteristics of any revolution-making work.
Most of the apparent incongruities in the De Revolutionibus re-
flect the personality of its author, and Copernicus' pe1sonality seems
184
entirely appropriate to his seminal role in the development of astron-
omy. Copernicus was a dedicated specialist. He belonged to the
revived Hellenistic tradition of mathematical astronomy which em-
phasized the mathematical problem of the planets at the expense of
cosmology. For his Hellenistic predecessors the physical incongruity
of an epicycle had not been an important drawback of the Ptolemaic
system, and Copernicus displayed a similar indifference to cosmo-
logical detail when he failed to note the incongruities of a moving
earth in an otherwise traditional universe. For him, mathematical and
celestial detail came first; he wore blinders that kept his gaze focused
upon the mathematical harmonies of the heavens. To anyone who did
not share his specialty Copernicus' view of the universe was narrow
and his sense of values distorted.
But an excessive concern with the heavens and a distorted sense
of values may be essential characteristics of the man who inaugurated
the revolution in astronomy and cosmology. The blinders that re-
stricted Copernicus' gaze to the heavens may have been functional.
They made him so perturbed by discrepancies of a few degrees in
astronomical prediction that in an attempt to resolve them he could
embrace a cosmological heresy, the earth's motion. They gave him
an eye so absorbed with geometrical harmony that he could adhere
to his heresy for its harmony alone, even when it had failed to solve
the problem that had led him to it. And they helped him evade the
nonastronomical consequences of his innovation, consequences that
led men of less restricted vision to reject his innovation as absurd.
Above all, Copernicus' dedication to the celestial motions is re-
for the painstaking detail with which he explored the mathe-
matical consequences of the earth's motion and fitted those conse-
quences to an existing knowledge of the heavens. That detailed
technical study is Copernicus' real contribution. Both before and after
Copernicus there were cosmologists more radical than he, men who
with broad brush strokes sketched an infinite and multipopulated
universe. But none of them produced work resembling the later books
of the De Revolutionibus, and it is these books which, by showing
for the first time that the astronomer's job could be done, and done
more harmoniously, from a moving earth, provided a stable base from
which to launch a new astronomical tradition. Had Copernicus' cos-
mological First Book appeared alone, the Copernican Revolution would
and should be known by someone else's name.
6
THE ASSIMILATION Of
COPERNICAN ASTRONOMY
The Reception of Copernicus' Work
Copernicus died in 1543, the year in which the De Revolu-
tionibus was published, and tradition tells us that he received the
first printed copy of his life's work on his deathbed. The book had to
fight its battles without further help from its author. But for those
battles Copernicus had constructed an almost ideal weapon. He had
made the book unreadable to all but the erudite astronomers of his
day. Outside of the astronomical world the De Revolutionibus created
initially very little stir. By the time large-scale lay and clerical opposi-
tion developed, most of the best European astronomers, to whom
the book was directed, had found one or another of Copernicus'
mathematical techniques indispensable. It was then impossible to
suppress the work completely, particularly because it was in a printed
book and not, like Oresme's work or Buridan's, in a manuscript.
Whether intentionally or not, the final victory of the De Revolu-
tionibus was achieved by infiltration.
For two decades before the publication of his principal work
Copernicus had been widely recognized as one of Europe's leading
astronomers. Reports about his research, including his new hypothesis,
had circulated since about 1515. The publication of the De Revolu-
tionibus was eagerly awaited. When it appeared, Copernicus' con-
temporaries may have been skeptical of its main hypothesis and
disappointed in the complexity of its astronomical theory, but they
were nevertheless forced to recognize Copernicus' book as the first
European astronomical text that could rival the Almagest in depth
and completeness. Many advanced astronomical texts written during
the fifty years after Copernicus' death referred to him as a ''second
186
Ptolemy" or "the outstanding artificer of our age"; increasingly these
books borrowed data, computations, and diagrams from the De Revo-
lutionibus, at least from parts of it independent of the motion of the
earth. During the second half of the sixteenth century the book be-
came a standard reference for all those concerned with advanced
problems of astronomical research.
But the success of the De Revolutionibus does not imply the suc-
cess of its central thesis. The faith of most astronomers in the earth's
stability was at first unshaken. Authors who applauded Copernicus'
erudition, borrowed his diagrams, or quoted his determination of the
distance from the earth to the moon, usually either ignored the earth's
motion or dismissed it as absurd. Even the rare text that mentioned
Copernicus' hypothesis with respect rarely defended or used it. With
a few notable exceptions, the most favorable of the early reactions to
the Copernican innovation are typified by the remark of the English
astronomer Thomas Blundeville, who wrote: "Copernicus ... affirm-
eth that the earth turneth about and that the sun standeth still in
the midst of the heavens, by help of which false supposition he hath
made truer demonstrations of the motions and revolutions of the
celestial spheres, than ever were made before."
1
Blundeville's remark
appeared in 1594 in an elementary book on astronomy that took the
earth's stability for granted. Yet the tenor of Blundeville's rejection
must have sent his more alert and proficient readers straight to the
De Revolutionibus, a book which, in any case, no proficient astrono-
mer could ignore. From the start the De Revolutionibus was widely
read, but it was read in spite of, rather than because of, its strange
hypothesis.
Nevertheless, the book's large audience ensured it a small but in-
creasing number of readers equipped to discover Copernicus' har-
monies and willing to admit them as evidence. There were a few con-
verts, and their work helped in varied ways to spread knowledge of
Copernicus' system. The Narratio Prima or First Account by Coperni-
cus' earliest disciple, George Joachim Rheticus ( 1514-1576), remained
the best brief technical description of the new astronomical methods
for many years after its first publication in 1540. The popular ele-
mentary defense of Copernicanism published in 1576 by the English
astronomer Thomas Digges (c.1546-1595) did much to spread the
concept of the earth's motion beyond the narrow circle of astronomers.
ASSIMILATION OF COPERNICAN ASTRONOMY 187
And the teaching and research of Michael Maestlin ( 1550-1631),
professor of astronomy at the University of Tiibingen, gained a few
converts, including Kepler, for the new astronomy. Through the teach-
ing, writing, and research of men like these, Copernicanism inevitably
gained ground, though the astronomers who avowed their adherence
to the conception of a moving earth remained a small minority.
But the size of the group of avowed Copernicans is not an ade-
quate index of the success of Copernicus' innovation. Many astrono-
mers found it possible to exploit Copernicus' mathematical system
and to contribute to the success of the new astronomy while denying
or remaining silent about the motion of the earth. Hellenistic astron-
omy provided their precedent. Ptolemy himself had never pretended
that all of the circles used in the Almagest to compute planetary posi-
tion were physically real; they were useful mathematical devices and
they did not have to be any more than that. Similarly, Renaissance
astronomers were at liberty to treat the circle representing the earth's
orbit as a mathematical fiction, useful for computations alone; they
could and occasionally did compute planetary position as if the earth
moved without committing themselves to the physical reality of that
motion. Andreas Osiander, the Lutheran theologian who saw Coperni-
cus' manuscript through the press, had actually urged this alterna-
tive upon readers in an anonymous preface attached to the De Revo-
lutionibus without Copernicus' permission. The spurious preface prob-
ably did not fool many astronomers, but a number of them never-
theless took advantage of the alternative that it suggested. Using
Copernicus' mathematical system without advocating the physical
motion of the earth provided a convenient escape from the dilemma
posed by the contrasting celestial harmonies and terrestrial discord
of the De Revolutionibus. It also gradually tempered the astronomer's
initial conviction that the earth's motion was absurd.
Erasmus Reinhold ( 1511-1553) was the first astronomer to do
important service for the Copernicans without declaring himself in
favor of the earth's motion. In 1551, only eight years after the pub-
lication of the De Revolutionibus, he issued a complete new set of
astronomical tables, computed by the mathematical methods developed
by Copernicus, and these soon became indispensable to astronomers
and astrologers, whatever their beliefs about the position and motion
of the earth. Reinhold's Prutenic Tables, named for his patron, the
188
Duke of Prussia, were the first complete tables prepared in Europe
for three centuries, and the old tables, which had included some errors
from the start, were now badly out of date -the clock had run too
long. Reinhold's supremely careful work, based on somewhat more
and better data than had been available to the men who computed
the thirteenth-century tables, produced a set of tables which, for most
applications, were measurably superior to the old. They were not, of
course, completely accurate; Copernicus' mathematical system was in-
trinsically no more accurate than Ptolemy's; errors of a day in the
prediction of lunar eclipses were common, and the length of the year
determined from the Prutenic Tables was actually slightly less accu-
rate than that determined from the older tables. But most comparisons
displayed the superiority of Reinhold's work, and his tables became
increasingly an astronomical requisite. Since the tables were known
to derive from the astronomical theory of the De Revolutionibus,
Copernicus' prestige inevitably gained. Every man who used the
Prutenic Tables was at least acquiescing in an implicit Copernicanism.
During the second half of the sixteenth century astronomers could
dispense with neither the De Revolutionibus nor the tables based
upon it. Copernicus' proposal gained ground slowly but apparently
Successive generations of astronomers, decreasingly pre-
disposed by experience and training to take the earth's stability for
granted, found the new harmonies a more and more forceful argument
for its motion. Besides, by the end of the century the first converts
had begun to uncover new evidence. Therefore if the decision be-
tween the Copernican and the traditional universe had concerned
only .. astronomers, Copernicus' proposal would almost certainly have
achieved a quiet and gradual victory. But the decision was not exclu-
sively, or even primarily, a matter for astronomers, and as the debate
spread from astronomical circles it became tumultuous in the extreme.
To most of those who were not concerned with the detailed study
of celestial motions, Copernicus' innovation seemed absurd and im-
pious. Even when understood, the vaunted harmonies seemed no evi-
dence at II. The resulting clamor was widespread, vocal, and bitter.
But the clamor was slow in starting. Initially, few nonastronomers
knew of Copernicus' innovation or recognized it as more than a passing
individual aberration like many that had come and gone before. Most
of the elementary astronomy texts and manuals used during the second
half of the sixteenth century had been prepared long before Coperni-
cus' lifetime- John of Holywood's thirteenth-century primer was still
a leader in elementary training - and the new handbooks prepared
after the publication of the De Revolutionibus usually did not mention
Copernicus or dismissed his innovation in a sentence or two. The
popular cosmological books that described the universe to laymen
remained even more exclusively Aristotelian in tone and substance;
Copernicus was either unknown to their authors or, if known, he was
usually ignored. Except, perhaps, in a few centers of Protestant learn-
ing, Copernicanism does not seem to have been a cosmological issue
during the first few decades after Copernicus' death. Outside of astro-
nomical circles it seldom became a major issue until the beginning of
the seventeenth century.
There were a few sixteenth-century reactions from nonastronomers,
and they provide a foretaste of the immense debate to follow, for they
were usually unequivocally negative. Copernicus and his few followers
were ridiculed for the absurdity of their concept of a moving earth,
though without the bitterness or the elaborate dialectic which de-
veloped when it became apparent that Copernicanism was to be a
stubborn and dangerous opponent. One long cosmological poem, first
published in France in 1578 and immensely popular there and in
England during the next century and a quarter, provides the follow-
ing typical description of the Copernicans as
Those clerks who think (think how absurd a jest)
That neither heav'ns nor stars do turn at all,
Nor dance about this great round earthly ball;
But th'earth itself, this massy globe of ours,
Turns round-about once every twice-twelve hours:
And we resemble land-bred novices
New brought aboard to venture on the seas;
Who, at first launching from the shore, suppose
The ship stands still, and that the ground it goes.
So, never should an arrow, shot upright,
In the same place upon the shooter light;
But would do, rather, as, at sea, a stone
Aboard a ship upward uprightly thrown;
Which not within-board falls, but in the flood
Astern the ship, if so the wind be good.
So should the fowls that take their nimble flight
From western marches towards morning's light;
190
And bullets thundered from the cannon's throat
(Whose roaring drowns the heav'nly thunder's note)
Should seem recoil: since the quick career,
That our round earth should daily gallop here,
Must needs exceed a hundred-fold, for swift,
Birds, bullets, winds; their wings, their force, their drift.
Arm'd with these reasons, 'twere superfluous
T' assail the reasons of Copernicus;
\Vho, to save better of the stars th'appearance,
Unto the earth a three-fold motion warrants.
2
Since the author of this poetic rejection of Copernicanism was a
poet, not a scientist or philosopher, his cosmological conservatism and
his adherence to classic sources may not be surprising. Yet it was from
poets and popularizers rather than from astronomers that most people
in the sixteenth and seventeenth century, as today, learned about the
universe. Du Bartas's The Week, or the Creation of the World, from
which the preceding excerpt is taken, was a far more widely read and
influential book than the De Revolutionibus.
In any case, uncritical offhand condemnations of Copernicus and
his followers were not restricted to conservative and unoriginal popu-
larizers. Jean Bodin, famous as one of the most advanced and creative
political philosophers of the sixteenth century, discards Copernicus'
innovation in almost identical terms:
No one in his senses, or imbued with the slightest knowledge of physics,
will ever think that the earth, heavy and unwieldy from its own weight and
mass, staggers up and down around its own center and that of the sun;
for at the slightest jar of the earth, we would see cities and fortresses,
towns and. mountains thrown down. A certain courtier Aulicus, when some
astrologer in court was upholding Copernicus' idea before Duke Albert of
Prussia, turning to the servant who was pouring the Falernian, said: "Take
care that the flagon is not spilled." For if the earth were to be moved,
neither an arrow shot straight up, nor a stone dropped from the top of a
tower would fall perpendicularly, but either ahead or behind .... Lastly,
all things on finding places suitable to their natures, remain there, as
Aristotle writes. Since therefore the earth has been allotted a place fitting
its nature, it cannot be whirled around by other motion than its own.
3
In this passage Bodin looks a traditionalist, but he was not. Because
of its generally radical and atheistic tone, the book from which the
quotation is taken was in 1628 placed upon the Index of books that
Catholics are forbidden to read. Although its author was himself a
Catholic, the book remains there to this day. Bodin was quite willing
to break with tradition, but that was not enough to make a man
a Copernican. It was almost invariably also necessary to understand
astronomy and to take its problems immensely seriously. Except to
those with an astronomical bias, the earth's motion seemed very nearly
as absurd in the years after Copernicus' death as it had before.
The anti-Copernican arguments suggested by Du Bartas and
Bodin can be considerably elaborated along lines anticipated by our
discussions of the Aristotelian universe in Chapters 3 and 4. In one
or another disguise, which we need not penetrate, they appear again
and again during the first half of the seventeenth century when the
debate about the earth's motion became bitter and intense. The
earth's motion, it was said, violates the first dictate of common sense;
it conflicts with long-established laws of motion; it has been suggested
merely "to save better of the stars th'appearance," a ridiculously
minuscule incentive for revolution. These are forceful arguments, quite
sufficient to convince most people. But they are not the most forceful
weapons in the anti-Copernican battery, and they are not the ones
that generated the most heat. Those weapons were religious and, par-
ticularly, scriptural.
Citation of Scripture against Copernicus began even before the
publication of the De Revolutionibus. In one of his "Table Talks,"
held in 1539, Martin Luther is quoted as saying:
People gave ear to an upstart astrologer who strove to show that the
earth revolves, not the heavens or the firmament, the sun and the moon.
. . . This fool wishes to reverse the entire science of astronomy; but sacred
Scripture tells us [Joshua 10:13] that Joshua commanded the sun to stand
still, and not the earth.4
Luther's principal lieutenant, Melanchthon, soon joined in the in-
creasing Protestant clamor against Copernicus. Six years after Coperni-
cus' death he wrote:
The eyes are witnesses that the heavens revolve in the space of twenty-
four hours. But certain meu, either from the love of novelty, or to make a
display of ingenuity, have concluded that the earth moves; and they main-
tain that neither the eighth sphere nor the sun revolves .... Now, it is a
want of honesty and decency to assert such notions publicly, and the ex-
ample is pernicious. It is the part of a good mind to accept the truth as
revealed by God and to acquiesce in i t ~ >
192
Melanchthon then proceeded to assemble a number of anti-Coperni-
can Biblical passages, emphasizing the famous verses, Ecclesiastes
1:4-5, which state "the earth abideth forever" and that "The sun
also ariseth, and the sun goeth down, and hasteth to his place where
he arose." Finally he suggests that severe measures be taken to re-
strain the impiety of the Copernicans.
Other Protestant leaders soon joined in the rejection of Copernicus.
Calvin, in his Commentary on Genesis, cited the opening verse of
the Ninety-third Psalm- "the earth also is stablished,. that it cannot
be moved" - and he demanded, "Who will venture to place the
authority of Copernicus above that of the Holy Spirit?"
6
Increasingly,
Biblical citation became a favored source of anti-Copernican argu-
ment. By the first decades of the seventeenth century clergymen of
many persuasions were to be found searching the Bible line by line
for a new passage that would confound the adherents of the earth's
motion. With growing frequency Copernicans were labeled "infidel"
and "atheist," and when, after about 1610, the Catholic Church
officially joined the battle against Copernicanism, the charge became
formal heresy. In 1616 the De Revolutionibus and all other writings
that affirmed the earth's motion were put upon the Index. Catholics
were forbidden to teach or even to read Copernican doctrines, except
in versions emended to omit all reference to the moving earth and
central sun.
The preceding sketch displays the most popular and forceful
weapons in the arsenal arrayed against Copernicus and his followers,
but it scarcely indicates what the war was really about. Most of the
men quoted above are so ready to reject the earth's motion as absurd
or ~ conflicting with authority that they fail to show, and may not at
first have realized fully, that Copernicanism was potentially destructive
of an entire fabric of thought. Their very dogmatism disguises their
motives. But it does not eliminate them. More than a picture of the
universe and more than a few lines of Scripture were at stake. The
drama of Christian life and the morality that had been made depend-
ent upon it would not readily adapt to a universe in which the earth
was just one of a number of planets. Cosmology, morality, and theol-
ogy had long been interwoven in the traditional fabric of Christian
thought described by Dante at the beginning of the fourteenth cen-
tury. The vigor and venom displayed at the height of the Copernican
controversy, three centuries later, testifies to the strength and vitality
of the tradition.
When it was taken seriously, Copernicus' proposal raised many
gigantic problems for the believing Christian. If, for example, the earth
were merely one of six planets, how were the stories of the Fall and of
the Salvation, with their immense bearing on Christian life, to be pre-
served? If there were other bodies essentially like the earth, God's
goodness would surely necessitate that they, too, be inhabited. But
if there were men on other planets, how could they be descendants
of Adam and Eve, and how could they have inherited the original sin,
which explains man's otherwise incomprehensible travail on an earth
made for him by a good and omnipotent deity? Again, how could men
on other planets know of the Saviour who opened to them the possi-
bility of eternal life? Or, if the earth is a planet and therefore a
celestial body located away from the center of the universe, what
becomes of man's intermediate but focal position between the devils
and the angels? If the earth, as a planet, participates in the nature
of celestial bodies, it can not be a sink of iniquity from which man will
long to escape to the divine purity of the heavens. Nor can the heavens
be a suitable abode for God if they participate in the evils and im-
perfection so clearly visible on a planetary earth. Worst of all, if
the universe is infinite, as many of the later Copernicans thought,
where can God's Throne be located? In an infinite universe, how is
man to find God or God man?
These questions have answers. But the answers were not easily
achieved; they were not inconsequential; and they helped to alter the
religious experience of the common man. Copernicanism required a
transformation in man's view of his relation to God and of the bases
of his morality. Such a transformation could not be worked out over-
night, and it was scarcely even begun while the evidence for Coperni-
canism remained as indecisive as it had been in the De Revolutionibus.
Until that transformation was achieved, sensitive observers might well
find traditional values incompatible with the new cosmology, and the
frequency with which the charge of atheism was hurled at the Coperni-
cans is evidence of the threat to the established order posed to many
observers by the concept of a planetary earth.
But the charge of atheism is only indirect evidence. More forceful
testimony comes from men who felt compelled to take the Copernican
innovation seriously. As early as 1611, the English poet and divine
John Donne said to the Copernicans that "those opinions of yours may
very well be true. . . . [In any case, they are now] creeping into
every man's mind,"
7
but he could discover little except evil in the
impending transition. During the same year in which he reluctantly
conceded the probability of the earth's motion, he portrayed his dis-
comfort at the impending dissolution of traditional cosmology in The
Anatomy of the World, a poem in which "the frailty and decay of this
whole world is represented." Part of Donne's malaise derived specifi-
cally from Copernicanism:
[The] new Philosophy calls all in doubt,
The Element of fire is quite put out;
The Sun is lost, and th'earth, and no man's wit
Can well direct him where to look for it.
And freely men confess that this world's spent,
When in the Planets, and the Firmament
They seek so many new; then see that this
Is crumbled out again to his Atomies.
'Tis all in pieces, all coherence gone;
All just supply, and all Relation:
Prince, Subject, Father, Son, are things forgot,
For every man alone thinks he hath got
To be a Phoenix, and that then can be
None of that kind, of which he is, but he.
8
Fifty-six years later, when scientists, at least, had overwhelmingly
accepted the earth's motion and its status as a planet, Copernicanism
presented the same problem of Christian morality to the English poet
John Milton, though he resolved it differently. Milton, like Donne,
thought that Copernicus' innovation might very well be true. He in-
cluded in Paradise Lost a lengthy description of the two opposing
systems of the world, the Ptolemaic and the Copernican, and he re-
fused to take sides in what he described as the abstruse technical
controversy between them. But in his epic, whose object was "to
justify the ways of God to man,"
9
he was compelled to use a tradi-
tional cosmological frame. The universe of Paradise Lost is not quite
Dante's universe; Milton derives the positions of heaven and hell
from a tradition even older than Dante's. But the terrestrial stage upon
which Milton portrays man's fall is still necessarily a unique, stable,
and centrally located body, created by God for man. Though more
.....
than a century had passed since the publication of the De Revolutioni-
bus, the Christian drama and the morality that had been made de-
pendent upon it could not be adapted to a universe in which the
earth was a planet and in which new worlds could continually be
discovered "in the Planets and the Firmament."
Donne's uneasiness and Milton's cosmological choice illustrate the
extrascientific issues which, during the seventeenth century, were in-
tegral parts of the controversy over Copernicanism. These issues, even
more than its apparent absurdity or its conflict with established laws
of motion, account for the hostility that Copernicus' proposal encoun-
tered outside of scientific circles. But they may not quite account either
for the intensity of that hostility or for the willingness of both Protes-
tant and Catholic leaders to make anti-Copernicanism an official
Church doctrine which could justify the persecution of Copernicans.
It is easy to understand the existence of strong resistance to Coperni-
cus' innovation - its patent absurdity and destructiveness were not
offset by effective evidence - but it is difficult to understand the ex-
treme forms which that resistance occasionally took. Before the middle
of the sixteenth century the history of Christianity offers few prece-
dents for the rigidity with which the official leaders of major religious
groups applied the literal text of Scripture to suppress a scientific and
cosmological theory. Even during the early centuries of the Catholic
Church, when distinguished Church Fathers like Lactantius had em-
ployed the Scriptures to destroy classical cosmology, there had been
no official Catholic cosmological position to which communicants
were required to adhere.
The bitterness of official Protestant opposition is, in practice, far
easier to understand than its Catholic counterpart, because the Protes-
tants' opposition can be plausibly related to a more fundamental con-
troversy which arose in the split between the Churches. Luther and
Calvin and their followers wished to return to a pristine Christianity,
as it could be discovered in the words of Jesus and the early Fathers
of the Church. To Protestant leaders the Bible was the single funda-
mental source of Christian knowledge. They vehemently rejected the
ritual and the dialectic subtleties that successive authoritarian Church
Councils had interposed between the believer and the fountainhead
of his belief. They abhorred the elaborate metaphorical and allegori-
cal interpretation of Scripture, and their literal adherence to the Bible
in matters of cosmology had no parallel since the days of Lactantius,
Basil, and Kosmas. To them Copernicus may well have seemed a sym-
bol of all the tortuous reinterpretations which, during the later Middle
Ages, had separated Christians from the basis of their belief. There-
fore the violence of the thunder that official Protestantism directed
at Copernicus seems almost natural. Toleration of Copernicanism
would have been toleration of the very attitude toward Holy Writ
and toward knowledge in general which, according to Protestants,
had led Christianity astray.
Copernicanism was thus indirectly involved in the larger religious
battle between the Protestant and Catholic Churches, and that in-
volvement must account for some of the excessive bitterness the
Copernican controversy evoked. Protestant leaders like Luther, Calvin,
and Melanchthon led in citing Scripture against Copernicus and in
urging the repression of Copernicans. Since the Protestants never
possessed the police apparatus available to the Catholic Church, their
repressive measures were seldom so effective as those taken later
by the Catholics, and they were more readily abandoned when the
evidence for Copernicanism became overwhelming. But Protestants
nevertheless provided the first effective institutionalized opposition.
Reinhold's silence about the physical validity of the mathematical
system that he had employed in computing the Prutenic Tables is
usually interpreted as an index o{ the official opposition to Copernican-
ism at the Protestant university of Wittenberg. Osiander, who added
the spurious apologia to the beginning of the De Revolutionibus, was
also a Protestant. Rheticus, the first outspoken defender of Copernicus'
astr.onomy, was a Protestant, too, but his Narratio Prima was written
while he was away from Wittenberg and before the De Revolutioni-
bus appeared; after his return to Wittenberg he published no more
Copernican tracts.
For sixty years after Copernicus' death there was little Catholic
counterpart for the Protestant opposition to Copernicanism. Individual
Catholic clergymen expressed their incredulity or abhorrence of the
new conception of the earth, but the Church itself was silent. The
De Revolutionibus was read and at least occasionally taught at lead-
ing Catholic universities. Reinhold's Prutenic Tables, based on Co-
pernicus' mathematical system, were used in the reformation of the
calendar promulgated for the Catholic world in 1582 by Gregory
XIII. Copernicus himself had been a cleric and a reputable one, whose
judgment was widely sought on astronomical and other matters. His
book was dedicated to the Pope, and among the friends who urged
him to publish it were a Catholic bishop and a cardinal. During the
fourteenth, fifteenth, and sixteenth centuries the Church had not im-
posed cosmological conformity on its members. The De Revolutionibus
was itself a product of the latitude allowed to Churchmen in matters
of science and secular philosophy, and before the De Revolutionibus
the Church had spawned even more revolutionary cosmological con-
cepts without theological convulsions. In the fifteenth century the
eminent cardinal and papal legate Nicholas of Cusa had propounded
a radical Neoplatonic cosmology and had not even bothered about
the conflict between his views and Scripture. Though he portrayed
the earth as a moving star, like the sun and the other stars, and though
his works were widely read and had great influence, he was not con-
demned or even criticized by his Church.
Therefore, when in 1616, and more explicitly in 1633, the Church
prohibited teaching or believing that the sun was at the center of
the universe and that the earth moved around it, the Church was re-
versing a position that had been implicit in Catholic practice for cen-
turies. The reversal shocked a number of devout Catholics, because it
committed the Church to opposing a physical doctrine for which new
evidence was being discovered almost daily, and because there clearly
had been an alternative attitude open to the Church. The same de-
vices which, in the twelfth and thirteenth centuries, had permitted the
Church to embrace Ptolemy and Aristotle might, in the seventeenth
century, have been applied to Copernicus' proposal. In a limited
fashion they had already been applied. Oresme's fourteenth-century
discussion of the earth's diurnal rotation had not ignored the scriptural
evidence for the earth's immobility. He had cited two of the Biblical
passages noted above and had then replied:
To the ... argument concerning the Holy Scripture which says that
the sun revolves, etc., one would say that it is here conforming to the
manner of common human speech. just as is done in several [other] places,
e.g., where it is written that God is repentant and that he is angry and
pacified and all other things which are not just as they sound. Also appro-
priate to our question, we read that God covers the heaven with clouds:
... and yet in reality the heaven covers the clouds.lO
198
Though the reinterpretation demanded by Copernicanism would
have been more drastic and more costly, the same sort of arguments
would have sufficed. During the eighteenth and nineteenth centuries
similar arguments were employed, and even in the seventeenth cen-
tury, at the time when the official decision to prohibit Copernicanism
was being taken, a few Catholic leaders recognized that some such
far-reaching reformulation might conceivably be required. In 1615
Cardinal Bellarmine, the leader of the Church officials who one year
later condemned Copernican views, wrote to the Copernican Fos-
carini:
If there were a real proof that the sun is in the center of the universe,
that the earth is in the third heaven, and that the sun does not go round
the earth but the earth round the sun, then we should have to proceed
with great circumspection in explaining passages of Scripture which appear
to teach the contrary, and rather admit that we did not understand them
than declare an opinion to be false which is proved to be true.
11
Very probably Bellarmine's liberalism is more apparent than real.
The next sentence of his letter reads, "But as for myself, I shall not
believe that there are such proofs until they are shown to me," and
that sentence was written in full knowledge of the telescopic dis-
coveries by which Galileo had provided strong new evidence for
Copernicus' innovation. We may wonder what sort of evidence Bellar-
mine would have considered "real proof" against the literal word of
Scripture. But he was aware, at least in principle, of the possibility
of evidence that would necessitate reinterpretation. Only, by the
second decade of the seventeenth century, Catholic authorities were
giving greater weight to scriptural evidence and allowing less latitude
for speculative dissent than they had done for centuries.
Much of the increasingly fundamentalist position that underlies
the Catholic condemnation of Copernicus must, I think, be a reaction
to the pressures brought to bear upon the Church by the Protestant
revolt. Copernican doctrines were, in fact, condemned during the
Counter Reformation, just when the Church was most convulsed by
internal reforms designed to meet Protestant criticism. Anti-Coperni-
canism seems, at least in part, one of those reforms. Another cause of
the Church's increased sensitivity to Copernicanism after 1610 may
well have been a delayed awakening to the fuller theological implica-
tions of the earth's motions. In the sixteenth century those implications
had rarely been made explicit. But in 1600 they were emphasized
with a clamor heard throughout Europe by the execution of Giordano
Bruno, the philosopher and mystic, at the stake in Rome. Bruno was
not executed for Copernicanism, but for a series of theological heresies
centering in his view of the Trinity, heresies for which Catholics had
been executed before. He is not, as he has often been called, a martyr
of science. But Bruno had found Copernicus' proposal congenial to
his Neoplatonic and Democritean vision of an infinite universe con-
taining an infinity of worlds generated by a fecund deity. He had
propounded Copernicanism in England and on the Continent and
had given it a significance not to be found in the De Revolutionibus
(see Chapter 7 below). Certainly the Church feared Bruno's Coperni-
canism, and that fear may also have stimulated their reaction.
But whatever the reasons, the Church did, in 1616, make Coperni-
canism a doctrinal issue, and all the worst excesses of the battle against
the earth's motion- the condemnation of Copernican opinions, the
recantation and "imprisonment" of Galileo, and the dismissal and
banishment of prominent Catholic Copernicans - occurred in or after
that year. Once the apparatus of the Inquisition had been unleashed
upon Copernicanism it was difficult to recall. Not until 1822 did the
Church permit the printing of books that treated the earth's motion
as physically real, and by then all but the most rigidly orthodox
Protestant sects had long been persuaded. The Church's official com-
mitment to the earth's stability did irrevocable harm both to Catholic
science and, later, to Church prestige. No episode in Catholic literature
has so often or so appropriately been cited against the Church as the
pathetic recantation forced upon the aged Galileo in 1633.
Galileo's recantation marks the peak of the battle against Coperni-
canism, and, ironically, it was not delivered until a time when the
outcome of the battle could be foreseen. Before 1610, when the opposi-
tion to Copernicus' doctrine was mustering, all but the most fanatical
advocates of the earth's motion would have been forced to admit that
the evidence for Copernicanism was weak and the counterevidence
strong. Perhaps the fundamental premise of the De Revolutionibus
would have to be abandoned. But by 1633 that was not the case. Dur-
ing the first decades of the seventeenth century new and stronger
evidence was discovered, and the complexion of the battle changed.
Even before Galileo's recantation, the new evidence had transformed
the opposition to Copernicanism into a hopeless rear-guard action. The
rest of this chapter examines that new evidence drawn from the
heavens by three of Copernicus' immediate successors.
Tycho Brahe
If Copernicus was the greatest European astronomer in the
first half of the sixteenth century, Tycho Brahe ( 1546-1601) was the
preeminent astronomical authority of the second. And, judged purely
by technical proficiency, Brahe was the greater man. But comparison
is largely meaningless, because the two have different strengths and
weaknesses which would not readily have merged in a single person-
ality, and both sorts of strength were essential to the Copernican Revo-
lution. As a cosmological and astronomical theorist, Brahe displayed
a relatively traditional frame of mind. His work shows little of that
Neoplatonic concern with mathematical harmonies that had been in-
strumental in Copernicus' break with the Ptolemaic tradition and
that at the start provided the only real evidence of the earth's motion.
He propounded no enduring innovations in astronomical theory. He
was, in fact, a lifelong opponent of Copernicanism, and his immense
prestige helped to postpone the conversion of astronomers to the new
theory.
But though Brahe was no innovator of astronomical concepts, he
was responsible for immense changes in the techniques of astronomi-
cal observation and in the standards of accuracy demanded from
astronomical data. He was the greatest of all naked-eye observers.
He designed and built many new instruments, larger, stabler, and
be!ter calibrated than those in use before. With great ingenuity he
investigated and corrected many errors that developed in using these
instruments, establishing a whole series of new techniques for the col-
lection of accurate information about the position of planets and stars.
Most important of all, he began the practice of making regular obser-
vations of planets as they moved through the heavens rather than
observing them only when in some particularly favorable configura-
tion. Modern telescopic observation indicates that when Brahe took
particular care in determining the position of a fixed star his data
were consistently accurate to 1' of arc or better, a phenomenal achieve-
ment with the naked eye. His observations of planetary position s0em
normally to have been reliable to about 4' of arc, more th::m twice the
l
ASSIMILATION OF COPERNICAN ASTRONOMY
201
accuracy achieved by the best observers of antiquity. But even more
important than the accuracy of Brahe's individual observations was
the reliability and the scope of the entire body of data he collected. In
his own lifetime he and the observers he trained freed European astron-
omy from its dependence on ancient data and eliminated a whole series
of apparent astronomical problems which had derived from bad data.
His observations provided a new statement of the problem of the
planets, and that new statement was a prerequisite to the problem's
solution. No planetary theory could have reconciled the data employed
by Copernicus.
Trustworthy, extensive, and up-to-date data are Brahe's primary
contribution to the solution of the problem of the planets. But he has
another and a larger role in the Copernican Revolution as the author
of an astronomical system that rapidly replaced the Ptolemaic system
as the rallying point for those proficient astronomers who, like Brahe
himself, could not accept the earth's motion. Most of Brahe's reasons
for rejecting Copernicus' proposal are the usual ones, though he de-
veloped them in more detail than most of his contemporaries. But
Brahe gave particular emphasis to the immense waste space that the
Copernican theory opened between the sphere of Saturn and the
stars merely to account for the absence of observable parallactic mo-
tion. He himself had looked for parallax with his great new instru-
ments. Since he found none, he felt forced to reject the earth's motion.
The only alternative compatible with his observations would have
required a distance between the stellar sphere and Saturn seven
hundred times the distance between Saturn and the sun.
But Brahe was nothing if not a proficient astronomer. Though he
rejected the earth's motion, he could not ignore the mathematical
harmonies which the De Revolutionibus had introduced into astron-
omy. Those new harmonies did not convert him- they were not, for
him sufficiently stron<T evidence to counterbalance the difficulties in-
' D
herent in the earth's motion- but they must at least have increased his
discontent with the Ptolemaic system, and he rejected it, too, in favor
of a third system of his own invention. Brahe's system, the "Tychonic,"
is shown in Figure 37. Once again the earth lies stationary at the
geometric center of a stellar sphere whose daily rotation accounts
for the diurnal circles of the stars. As in the Ptolemaic system, the
sun, moon, and planets are carried westward daily with the stars by
the outer sphere, and they have additional eastward orbital motions
of their own. In the diagram these orbital motions are represented by
circles, though in the full Tychonic system minor epicycles, eccen-
trics, and equants are also required. The circles of the moon and sun
are centered on the earth; to this point the system is still Ptolemaic.
But the centers of the five remaining planetary orbits are transferred
from the center of the earth to the sun. Brahe' s system is an exten-
sion, though perhaps not a conscious one, of Heraclides' system, which
attributed sun-centered orbits to Mercury and Venus.
Figure 37. The Tychonic system. The earth is once again at the center of a
rotating stellar sphere, and the moon and sun move in their old Ptolemaic orbits.
The other planets are, however, fixed on epicycles whose common center is the sun.
The remarkable and historically significant feature of the Tychonic
system is its adequacy as a compromise solution of the problems raised
by the De Revolutionibus. Since the earth is stationary and at the
center, all the main arguments against Copernicus' proposal vanish.
Scripture, the laws of motion, and the absence of stellar parallax, all
are reconciled by Brahe's proposal, and this reconciliation is effected
without sacrificing any of Copernicus' major mathematical harmonies.
The Tychonic system is, in fact, precisely equivalent mathematically
to Copernicus' system. Distance determination, the apparent anomalies
ASSIMILATION OF COPERNICAN ASTRONOMY
203
in the behavior of the inferior planets, these and the other new har-
monies that convinced Copernicus of the earth's motion are all pre-
served.
The harmonies of the Tychonic system may be developed individ-
ually and in detail by the same techniques employed in discussing
Copernicus' system, but for present purposes the following abbreviated
demonstration of the mathematical equivalence of the Copernican and
Tychonic systems should be sufficient. Imagine the sphere of the stars
in Figure 37 immensely expanded until an observer on the moving
sun could no longer observe any stellar parallax from opposite sides
of the sun's orbit. This expansion does not affect the system's mathe-
matical account of any of the planetary motions. Now imagine that
within this expanded stellar sphere the various planets are driven
about their orbits by a clockwork mechanism like that indicated sche-
matically in Figure 38a for the earth, the sun, and Mars. In the diagram
the sun is attached to the central earth by an arm of fixed length which
it counterclockwise about the earth, and Mars is attached to
the sun by another arm of fixed length which moves it clockwise about
the moving sun. Since the lengths of both arms are fixed throughout the
motion, the clockwork mechanism will produce just the circular orbits
indicated in Figure 37.
Now 1magme that, without interfering with the gears that drive
the arms in Figure 38a, the whole mechanism is picked up and, with
the arms turning as before, put down again with the sun fixed at the
central position formerly held by the earth. This is the situation indi-
cated in Figure 38b. The arms have the same lengths as before; they
are driven at the same rates by the same mechanism; and they there-
fore retain the same relative positions at each instant of time. All of
the geometric spatial relations of the earth, sun, and Mars in the dia-
gram of Figure 38a are preserved by the arrangement of Figure 38b,
and since only the fixed point of the mechanism has been changed,
all the relative motions must be identical.
But the motions produced by the mechanism of Figure 38b are
Copernican motions. That is, the fixed arms shown in the second dia-
gram move both the earth and Mars in circular orbits about the sun
and those orbits are just the basic ones by Copernicus:
Carrying out the same argument with the hypothetical mechanism
of Figure 38 elaborated to include all the planets, demonstrates that
204
the equivalence is general. Omitting minor epicycles and eccentrics,
which have no bearing on the harmonies of Copernicus' system, the
Tychonic system is transformed to the Copernican system simply by
holding the sun fixed instead of the earth. The relative motions of the
planets are the same in both systems, and the harmonies are there-
fore preserved. Mathematically the only possible difference between
the motions in the two systems is a parallactic motion of the stars, and
that motion was eliminated at the start by expanding the stellar sphere
until parallax was imperceptible.
M
\.
{a)
s
.........
E
(b)
M
Figure 38. The geometrical equivalence of (a) the Tychonic and (b) the
Copernican systems. In (a) the sun S is carried eastward about the stationary
t!'!!rth E by the the rigid arm ES. Simultaneously, the planet Mars, M, is carried
westward about S by the steady rotation of the arm SM. Since ES rotates more
rapidly than SM, the net motion of Mars is eastward except during the brief
period when SM crosses over ES. In the second diagram (b) the same arms are
shown rotating about the fixed sun S. The relative positions of E, S, and M are the
same as those in (a), and they will stay the same while the arms in the two
diagrams rotate. Notice particularly that in (b) the angle ESM must decrease as
it does in (a) because ES rotates about the sun more rapidly than SM.
The Tychonic system has incongruities all its own: most of the
planets are badly off center; the geometric center of the universe is no
longer the center for most of the celestial motions; and it is difficult
to imagine any physical mechanism that could produce planetary
motions even approximately like Brahe's. Therefore the Tychonic sys-
'
tern did not convert those few Neoplatonic astronomers, like Kepler,
who been attracted to Copernicus' system by its great symmetry.
But It did convert most technically proficient non-Copernican astrono-
of day, because it provided an escape from a widely felt
dilemma: It retained the mathematical advantages of Copernicus'
system without the physical, cosmological, and theological drawbacks.
That is the real importance of the Tychonic system. It was an almost
perfect compromise, and in retrospect the system seems to owe its
existence to the felt need for such a compromise. The Tychonic system,
to which almost all the more erudite seventeenth-century Ptolemaic
astronomers retreated, appears to be an immediate by-product of the
De Revolutionibus.
Brahe himself would have denied this. He proclaimed that he had
taken nothing his system from Copernicus. But he can scarcely
have been consciOus of the pressures at work on him and his contem-
poraries. Certainly he knew both Ptolemaic and Copernican astronomy
before he thought of his own system, and he was clearly
aware m advance of the predicament that his own system was to re-
solve. The immediate success of the system is one index of the strength
and of the need. That two other astronomers disputed
s pnonty and claimed to have worked out similar compromise
solutiOns for themselves provides additional evidence for the role of
-?e and the resulting climate of astronomical opin-
m the of. the Tychonic system. Brahe and his system pro-
VIde the first IllustratiOn of one of the major generalizations that closed
the last chapter: the De Revolutionibus changed the state of astronomy
by posing new problems for all astronomers.
Brahe's criticisms of Copernicus and his compromise solution of
the problem of the planets show that, like most astronomers of his
day, he to break with patterns of thought about
the earths motwn. Among Copernicus successors Brahe is one of the
immense body of conservatives. But the effect of his work was not
conservative. On the contrary, both his system and his observations
forced his successors to repudiate important aspects of the Aristotelian-
Ptolemaic universe and thus drove them gradually toward the Co-
pernican camp. In the first place, Brahe' s system helped to familiarize
astronomers. with the mathematical problems of Copernican astronomy,
for geometncally the Tychonic and Copernican systems were identical.
More important, Brahe's system, abetted by his observations of comets,
to be discussed below, forced his followers to abandon the crystalline
spheres which, in the past, had carried the planets about their orbits.
In the Tychonic system, as indicated by Figure 37, the orbit of Mars
intersects the orbit of the sun. Both ~ a r s and the sun cannot, there-
fore, be embedded in spheres that carry them about, for the two
spheres would have to penetrate and move through each other at all
times. Similarly, the sun's sphere passes through the spheres of Mer-
cury and Venus. Abandoning the crystalline spheres does not make
a man a Copernican; Copernicus himself had utilized spheres to ac-
count for the planetary motions. But the spheres had, in one of a
number of modifications, been an essential ingredient of the Aristo-
telian cosmological tradition which was the principal barrier to the
success of Copernicanism. Any break with the tradition worked
for the Copernicans, and the Tychonic system, for all its traditional
elements, was an important break.
Brahe's skillful observations were even more important than his
system in leading his contemporaries toward a new cosmology. They
provided the essential basis for the work of Kepler, who converted
Copernicus' innovation into the first really adequate solution of the
problem of the planets. And even before they were used to revise
Copernicus' system, the new data collected by Brahe suggested the
necessity of another major departure from classical cosmology - they
raised questions about the immutability of the heavens. Late in 1572,
when Brahe was at the beginning of his career in astronomy, a new
celestial body appeared in the constellation Cassiopeia, directly across
the pole from the Big Dipper. \Vhen first observed it was very brilliant,
as clear as Venus at its greatest brightness; during the next eighteen
m'"onths the new occupant of the heavens grew gradually dimmer; and
finally it vanished altogether early in 1574. From the start the new
visitor drew the interest of scientists and nonscientists throughout
Europe. It could not be a comet, the only sort of celestial apparition
widely recognized by astronomers and astrologers, for it had no tail,
and it always appeared in the same position against the sphere of the
stars. Clearly it was a portent; astrological activity multiplied; and
astronomers everywhere devoted their observations and their writings
to the "new star" in the heavens.
The word "star" is the key to the astronomical and cosmological
significance of the new phenomenon. If it were a star, then the im-
mutable heavens had changed, and the basic contrast between the
superlunary region and the corruptible earth was in question. If it
were a star, the earth might more easily be conceived as a planet, for
the transitory character of terrestrial affairs would now have been
discovered in the heavens as well. Brahe and the best of his contem-
poraries did conclude that the visitor was a star. Observations like
the one illustrated in Figure 39 indicated that it could not be located
SPHERE
OF STARS
Figure 39. Diurnal parallax of a body below the stars. If S is between the
earth and the sphere of the stars, then it should appear at different positions
against the background of stars when observed by terrestrial observers at 0 and 0'.
Two observers are not required. The eastward rotation of the earth (or the
equivalent westward rotation of the observed body and the stellar sphere) carries
an observer from 0 to 0' in six hours; as a result of the rotation the body S appears
to change its position continually, returning to its starting point among the stars
after twenty-four hours. If S were as close as the moon, its apparent displacement
during six hours would be very nearly- 1 o. Bodies farther from the earth show less
displacement.
With modern instruments the technique illustrated above is useful in determin-
ing the distances to the moon and planets, but naked-eye observations are not
accurate enough for this application. The large size of the moon and its rapid
orbital motion disguise the parallactic effect. The planets are too far away.
below the sphere of the moon or even close to the sublunary region.
Probably it was among the stars, for it was observed to move with
them. Another cause for cosmological upheaval had been discovered.
The sixteenth-century discovery of the mutability of the heavens
might have been relatively ineffectual if the only evidence of super-
lunary change had been drawn from the new star, or nova, of 1572.
It was a transient phenomenon; those who chose to reject Brahe's
data could not be refuted; by the time the data were published the
star had disappeared; and some less careful observers could always
be discovered who had observed a parallax sufficient to place the nova
below the moon. But fortunately additional ancl continuing evidence
of superlunary change was provided by comets which Brahe observed
carefully in 1577, 1580, 1585, 1590, 1593, and 1596. Once again no
measurable parallax was observed, and the comets too were therefore
located beyond the moon's sphere where they moved through the
region formerly filled by the crystalline spheres.
Like the observations of the nova, Brahe's discussions of comets
failed to convince all of his contemporaries. During the first decades
of the seventeenth century Brahe was frequently attacked, occasionally
with the same bitterness displayed toward Copernicus, by those who
believed that other data proved comets and novas to be sublunary
phenomena and that the inviolability of the heavens could therefore
be preserved. But Brahe did convince a large number of astronomers
of a basic flaw in the Aristotelian world view, and, more important,
he provided a mode of argument by which skeptics could continuously
check his conclusions. Comets bright enough to be seen with the naked
eye appear every few years. After their superlunary character had
been deduced from observation and then widely debated, the evi-
dence that comets provided for the mutability of the heavens could
not indefinitely be ignored or distorted. Once again the Copernicans
were the gainers.
Somehow, in the century after Copernicus' death, all novelties of
astronomical observation and theory, whether or not provided by
Copernicans, turned themselves into evidence for the Copernican
theory. That theory, we should say, was proving its fruitfulness. But,
at least in the case of comets and novas, the proof is very strange, for
the,. observations of comets and novas have nothing whatsoever to do
with the earth's motion. They could have been made and interpreted
by a Ptolemaic astronomer just as readily as by a Copernican. They
are not, in any direct sense, by-products of the De Revolutionibus,
as the Tychonic system was.
But neither can they be quite independent of the De Revolu-
tionibus or at least of the climate of opinion within which it was
created. Comets had been seen frequently before the last decades of
the sixteenth century. New stars, though they appear less frequently
to the naked eye than comets, must also have been occasionally ac-
cessible to observers before Brahe's time; one more appeared in the
year before his death and a third in 1604. Even Brahe's fine instru-
ments were not required to discover the superlunary character of
novas and comets; a parallactic shift of 1 o could have been measured
without those instruments, and a number of Brahe's contemporaries
did independently conclude that comets were superlunary using
instruments that had been known for centuries. The Copernican
Maestlin needed only a piece of thread to decide that the nova of 1572
was beyond the moon. In short, the observations with which Brahe
and his contemporaries speeded the downfall of traditional cosmology
and the rise of Copernicanism could have been made at any time
since remote antiquity. The phenomena and the requisite instruments
had been available for two millenniums before Brahe's birth, but the
observations were not made or, if made, were not widely interpreted.
During the last half of the sixteenth century age-old phenomena
rapidly changed their meaning and significance. Those changes seem
incomprehensible without reference to the new climate of scientific
thought, one of whose first outstanding representatives is Copernicus.
As suggested at the end of the last chapter, the De Revolutionibus
marked a turning point, and there was to be no turning back.
Johannes Kepler
Brahe's work indicates that after 1543 even the opponents
of Copernicanism, at least the ablest and most honest ones, could
scarcely help promoting major reforms in astronomy and cosmology.
Whether or not they agreed with Copernicus, he had changed their
field. But the work of an anti-Copernican like Brahe does not show
the extent of those changes. A better index of the novel problems that
accrued to astronomy after Copernicus' death is provided by the re-
search of Brahe's most famous colleague, Johannes Kepler ( 1571-
1630). Kepler was a lifelong Copernican. He seems first to have been
converted to the system by Maestlin when he was a student at the
Protestant university of Tiibingen, and his faith in it never wavered
after his student days. Throughout his life he referred in the rhapsodic
tones characteristic of Renaissance Neoplatonism to the suitability
of the role that Copernicus had attributed to the sun. His first im-
portant book, the Cosmographical Mystery, published in 1596, opened
with a lengthy defense of the Copernican system, emphasizing all
those arguments from harmony that we discussed in Chapter 5 and
adding many new ones besides: Copernicus' proposal explains why
Mars's epicycle had been so much larger than Jupiter's and Jupiter's
than Saturn's; sun-centered astronomy shows why, of all the celestial
wanderers, only the sun and moon fail to retrogress; and so on and on.
Kepler's arguments are the same as Copernicus', though more numer-
ous, but Kepler, in contrast to Copernicus, develops the arguments
at length and with detailed diagrams. For the first time the full force
of the mathematical arguments for the new astronomy was demon-
strated.
But though Kepler was full of praise for the conception of a sun-
centered planetary system, he was quite critical of the particular
mathematical system that Copernicus had developed. Again and again
Kepler's writings emphasized that Copernicus had never recognized
his own riches and that after the first bold step, the transposition of
the sun and earth, he had stayed too close to Ptolemy in developing
the details of his system. Kepler was acutely and uncomfortably aware
of the incongruous archaic residues in the De Revolutionibus, and
he took it upon himself to eliminate them by exploiting fully the earth's
new status as a planet governed, like the other planets, by the sun.
Copernicus had not quite succeeded in treating the earth as just
another planet in a sun-centered system. Unlike the qualitative sketch
in the First Book of the De Revolutionibus, the mathematical account
of the planetary system developed in the later books attributed several
special functions to the earth. For example, in the Ptolemaic system
the planes of all planetary orbits had been constructed so that they
intersected at the center of the earth, and Copernicus preserved this
terrestrial function in a new form by drawing all orbital planes so
they intersected at the center of the earth's orbit. Kepler insisted
that, since the sun governed the planets and the earth had no unique
status, the planes of the orbits must intersect in the sun. By redesign-
ing the Copernican system accordingly he made the first significant
progress since Ptolemy in accounting for the north and south devia-
tions of the planets from the ecliptic. Kepler had improved Coper-
nicus' mathematical system by applying strict Copernicanism to it.
A similar insistence upon the parity of the planets enabled Kepler
to eliminate a number of pseudo problems that had distorted Coper-
nicus' work. Copernicus had, for example, believed that the eccen-
tricities of Mercury and Venus were slowly changing, and he had
added circles to his system to account for the variation. Kepler showed
that the apparent change was due only to an inconsistency in Coper-
nicus' definition of eccentricity. In the De Revolutionibus the eccen-
tricity of the earth's orbit was measured from the sun (it is the distance
SOE in Figure 34a, p. 169) while the eccentricities of all other orbits
were measured from the center of the earth's orbit (Mars's eccentric-
ity is OEOM in Figure 34b). Kepler insisted that all planetary eccen-
tricities must, in a Copernican universe, be computed in the same way
and from the sun. When the new method was incorporated in his
system, several of the apparent variations of eccentricity vanished,
and the number of circles required in computation was reduced.
Each of these examples shows Kepler striving to adapt Copernicus'
overly Ptolemaic mathematical techniques to the Copernican vision
of a sun-dominated universe, and it was by continuing this effort that
Kepler finally resolved the problem of the planets, transforming
Copernicus' cumbersome system into a supremely simple and accurate
technique for computing planetary position. His most essential dis-
coveries were made while studying the motion of Mars, a planet whose
eccentric orbit and proximity to the earth produce irregularities that
had always challenged the ingenuity of mathematical astronomers.
Ptolemy had been unable to account for its motion as satisfactorily as
for that of the other planets, and Copernicus had not improved on
Ptolemy. Brahe had attempted a new solution, undertaking a long
series of observations specially for the purpose, but surrendering the
problem as he encountered its full difficulties. Kepler, who had worked
with Brahe during the last years of Brahe's life, inherited the new
observations and, in the years after Brahe's death, took up the
himself.
It was an immense labor which occupied much of Kepler's time
for almost ten years. Two orbits had to be worked out: the orbit of
Mars itself and the orbit of the earth from which Mars is observed.
Again and again Kepler was forced to change the combination of
circles used in computing these orbits. System after system was tried
and rejected because it failed to conform to Brahe's brilliant observa-
tions. All of the intermediate solutions were better than the systems
of Ptolemy and of Copernicus; some gave errors no larger than 8' of
arc, well within the limits of ancient observation. Most of the systems
that Kepler discarded would have satisfied all earlier mathematical
astronomers. But they had lived before Brahe, whose data were ac-
212
curate to 4' of arc. To us, Kepler said, Divine goodness has given a
most diligent observer in Tycho Brahe, and it is therefore right that
we should with a grateful mind make use of this git to find the true
celestial motions.
A long series of unsuccessful trials forced Kepler to conclude that
no system based upon compounded circles would solve the problem.
Some other geometric figure must, he thought, contain the key. He
tried various sorts of ovals, but none eliminated the discrepancies
between his tentative theory and observation. Then, by chance, he
noticed that the discrepancies themselves varied in a familiar mathe-
matical fashion, and investigating this regularity he discovered that
theory and observation could be reconciled i the planets moved in
elliptical orbits with variable speeds governed by a simple law which
he also specified. These are the results that Kepler announced in On
the Motion of Mars, first published at Prague in 1609. A mathematical
technique simpler than any employed since Apollonius and Hippar-
chus yielded predictions far more accurate than any that had ever been
made before. The problem of the planets had at last been solved, and
it was solved in a Copernican universe.
The two laws that constitute Kepler's (and our) final solution of
the problem of the planets are described in detail in Figure 40. The
planets move in simple elliptical paths, and the sun occupies one of
the two foci of each elliptical orbit- that is Kepler's First Law. His
Second Law follows immediately, completing the description em-
bodied in the First - the orbital speed of each planet varies in such
a way that a line joining the planet to the sun sweeps through equal
~ s of the ellipse in equal intervals of time. When ellipses are sub-
stituted for the basic circular orbits common to Ptolemy's and Coper-
nicus' astronomy and when the law of equal areas is substituted for
the law of uniorm motion about a point at or near the center, all
need for eccentrics, epicycles, equants, and other ad hoc devices
vanishes. For the first time a single uncompounded geometric curve
and a single speed law are sufficient for predictions of planetary
position, and for the first time the predictions are as accurate as the
observations.
The Copernican astronomical system inherited by modern science
is, therefore, a joint product of Kepler and Copernicus. Kepler's system
of six ellipses made sun-centered astronomy work, displaying simul-
taneously the economy and the fruitfulness implicit in Copernicus'
innovation. We must try to discover what was required for this transi-
tion of the Copernican system to its modern, Keplerian, form. Two of
the prerequisites of Kepler's work are already apparent. He had to be
a convinced Copernican, a man who would begin his search for more
adequate orbits by treating the earth as a mere planet and who would
construct the planes of all planetary orbits through the center of the
sun. In addition, he needed Brahe's data. The data used by Copernicus
and his European predecessors were too infected with errors to be
(a) (b)
(c)
Figure 40. Kepler's first two Laws. Diagrams (a) and (b) define the ellipse,
the geometric curve in which all planets that obey Kepler's First Law must move.
In (a) the ellipse is shown as the closed curve in which a plane intersects a
circular cone. When the plane is perpendicular to the axis of the cone, the inter-
section is a circle, a special case of the ellipse. As the plane is tilted, the curve of
intersection is elongated into more typically elliptical patterns.
A more modem and somewhat more useful definition of the ellipse is given in
diagram (b). If two ends of a slack string are attached to two points Ft and F.
in a plane, and if a pencil P is inserted into the slack and then moved so that it
just keeps the string taut at all times, the point of the pencil will generate an
ellipse. Changing the length of the string or moving the foci F1 and F. together
or apart alters the shape of the ellipse in the same way as a change in the tilt of
the plane in diagram (a). Most planetary orbits are very nearly circular, and the
foci of the corresponding ellipses are therefore quite close together.
Diagram (c) illustrates Kepler's Second Law, which governs orbital speed.
The sun is at one focus of the ellipse, as required by the First Law, and its center
is joined by straight lines to a number of planetary positions P and P', arranged
so that each of the three shaded sectors SPP' has the same area. The Second Law
states that, since each of these areas is the same, the planet must move through
each of the corresponding arcs PP' in equal times. When near the sun, the planet
must move relatively quickly so that the short line SP will sweep out the same area
per unit time as is swept out by the longer line SP when the planet is moving
more slowly farther from the sun.
explained by any set of simple orbits, and even if freed from error
they would not have sufficed. Observations less precise than Brahe's
could have been explained, as Kepler himself showed, by a classical
system of compounded circles. The process by which Kepler arrived
at his famous Laws depends, however, upon more than the avail-
ability of accurate data and a prior commitment to the planetary
earth. Kepler was an ardent Neoplatonist. He believed that mathe-
matically simple laws are the basis of all natural phenomena and that
the sun is the physical cause of all celestial motions. Both his most
lasting and his most evanescent contributions to astronomy display
these two aspects of his frequently mystical Neoplatonic faith.
In a passage quoted at the end of Chapter 4 Kepler described the
sun as the body "who alone appears, by virtue of his dignity and
power, suited . . . [to move the planets in their orbits], and worthy
to become the home of God himself, not to say the first mover." This
conviction, together with certain intrinsic incongruities discussed
above, was his reason for rejecting the Tychonic system. It also played
an immensely important role in his own research, particularly in his
derivation of the Second Law upon which the First depends. In its
origin the Second Law is independent of any but the crudest sort of
observation. It arises rather from Kepler's physical intuition that the
planets are pushed around their orbits by rays of a moving force, the
anima matrix, which emanates from the sun. These rays must, Kepler
believed, be restricted to the plane of the ecliptic, in or near which
all the planets moved. Therefore the number of rays that impinged on
a planet and the corresponding force that drove the planet around the
SUI\ would decrease as the distance between the planet and the sun
increased. At twice the distance from the sun half as many rays of the
anima matrix would fall on a planet (Figure 4la), and the velocity
of the planet in its orbit would, in consequence, be half of its orbital
velocity at its original distance from the sun. A planet, P, moving about
the sun, S, on an eccentric circle (Figure 4lb) or some other closed
curve must move at a speed inversely proportional to SP. The speed
will be greatest when the planet is at the perihelion, p, closest to the
sun, and least at the aphelion, a, where the planet is farthest from the
sun. As the planet moves around the orbit, its speed will vary con-
tinually between these extremes.
Long before he began to work on elliptical orbits or stated the law
of areas in its familiar modern form, Kepler had worked out this
inverse-distance speed law to replace both the ancient law of uniform
circular motion and the Ptolemaic variant which permitted uniform
motion with respect to an equant point. This early speed law was very
much "pulled from a hat" by a strange intuition - one that was rapidly
discarded by his successors- of the forces that must govern a sun-
dominated universe. Furthermore, this early form is not quite correct.
The later law of areas, Kepler's so-called Second Law, is not quite
equivalent to the inverse-distance law, and the law of areas gives
somewhat better results. But when used to compute planetary position
the two forms of the speed law lead to almost the same predictions.
Kepler mistakenly thought the two equivalent in principle and used
them interchangeably throughout his life. For all its visionary over-
tones the early Neoplatonic speed law proved fundamental in Kepler's
most fruitful research.
p a
(a) (b)
Figure 41. Kepler's earliest speed law. Diagram (a), which shows typical
rays of the anima matrix radiating from the sun, illustrates the physical theory
from which Kepler derived the law. Diagram (b) shows how the law could be
applied to a planet moving on an eccentric circle.
Unlike his derivation of the speed law, Kepler's work on elliptical
orbits was completely dependent upon the most painstaking and
exhaustive study of the best available astronomical observations.
Trial orbit after trial orbit had to be abandoned because, after labori-
ous computation, it did not quite match Brahe's data. Kepler's scrupu-
lous attempt to fit his orbits to objective data is often cited as an early
example of the scientific method at its best. Yet even the law of
elliptical orbits, Kepler's First Law, was not derived from observation
and computation alone. Unless the planetary orbits are assumed to
be precisely reentrant (as they were after Kepler's work but not be-
fore), a speed law is required to compute orbital shape from naked-
eye data. When analyzing Brahe's observations, Kepler made constant
use of his earlier Neoplatonic guess.
The interrelation of orbit, speed law, and observation was obscured
in our earlier discussions of astronomical theory, because ancient and
medieval astronomers chose a simple speed law in advance. Before
Kepler astronomers assumed that each of the compounded circles
which moved a planet around its orbit must rotate uniformly with
respect to a point at or near its center. Without some such assump-
tion they could not have begun the elaboration of orbits to fit observa-
tions, for in the absence of a speed law the specification of an orbit
tells little or nothing about where a planet will appear among the
stars at a particular time. Neither speed law nor orbit can be inde-
pendently derived from or checked against observation. Therefore,
when Kepler rejected the ancient law of uniform motion, he had to
replace it or else abandon planetary computations entirely. In fact,
he rejected the ancient law only after (and probably because) he
had developed a law of his own- a law that his Neoplatonic intuition
told him was better suited than its ancient counterpart to govern
celestial motions in a sun-dominated universe.
Kepler's derivation of the inverse-distance law displays his belief
in mathematical harmonies as well as his faith in the causal role of
the sun. Having developed the conception of the anima matrix Kepler
insisted that it must operate in the simplest way compatible with crude
observation. He knew, for example, that planets move fastest at peri-
hdion, but he had few other data, none of them quantitative, on which
to base an inverse-distance law. But Kepler's belief in number har-
monies and the role of this belief in his work is more forcefully ex-
hibited in another one of the laws that modern astronomy inherits
from him. This is Kepler's so-called Third Law, announced during
1619 in the Harmonies of the World.
The Third Law was a new sort of astronomical law. Like their
ancient and medieval counterparts the First and Second Laws govern
only the motions of individual planets in their individual orbits. The
Third Law, in contrast, established a relation between the speeds of
planets in different orbits. It states that if 1'
1
and 1'
2
are tl:e periods
that two planets require to complete their respective orbits once, and
L
if R1 and R2 are the average distances between the corresponding
planets and the sun, then the ratio of the squares of the orbital periods
is equal to the ratio of the cubes of the average distances from the
sun, or (T1 I T2)
2
= (R1 I R2 )
3
This is a fascinating law, for it points
to a regularity never before perceived in the planetary system. But, at
least in Kepler's day, that was all it did. The Third Law did not, in
itself, change the theory of the planets, and it did not permit astrono-
mers to compute any quantities that were previously unknown. The
sizes and the periods associated with each planetary orbit were avail-
able in advance.
But though it had little immediate practical use, the Third Law is
just the sort of law that most fascinated Kepler throughout his career.
He was a mathematical Neoplatonist or Neopythagorean who be-
lieved that all of nature exemplified simple mathematical regularities
which it was the scientists' task to discover. To Kepler and others of
his turn of mind a simple mathematical regularity was itself an ex-
planation. To him the Third Law in and of itself explained why the
planetary orbits had been laid out by God in the particular way that
they had, and that sort of explanation, derived from mathematical
harmony, is what Kepler continually sought in the heavens. He pro-
pounded a number of other laws of the same kind, laws which we
have since abandoned because, though harmonious, they do not fit
observation well enough to seem significant. But Kepler was not so
selective. He thought that he had discovered and demonstrated a
large number of these mathematical regularities, and they were his
favorite astronomical laws.
In Kepler's first major work, the Cosmographical Mystery, he
argued that both the number of the planets and the size of their orbits
could be understood in terms of the relation between the planetary
spheres and the five regular or "cosmic" solids. These are the solids
shown in Figure 42a, and they have the unique characteristic that all
of the faces of each solid are identical and that only equilateral figures
are used for faces. It had been shown in antiquity that there could
be only five such solids: cube, tetrahedron, dodecahedron, icosahe-
dron, and octahedron. Kepler proclaimed that if the sphere of Saturn
were circumscribed about the cube within which Jupiter's sphere was
inscribed, and if the tetrahedron were placed just inside Jupiter's
sphere with Mars's sphere inscribed in it, and so on for the three other
solids and three other spheres, then the relative dimensions of all the
spheres would be just those that Copernicus had determined by
measurement. The construction is shown in Figure 42b. If it is to be
used, there can be only six planets, corresponding to the five regular
solids, and when it is used the permissible relative dimensions of the
(a)
(b)
Figure 42. Kepler's application of the five regular solids. Diagram (a) shows
the solids themselves. From left to right they are: cube, tetrahedron, dodecahedron,
icosahedron, and octahedron. Their order is the one that Kepler developed to ac-
count for the sizes of the planetary spheres. Diagram (b) shows the solids in this
application. Saturn's sphere is circumscribed about the cube, and Jupiter's sphere
is inscribed in it. The tetrahedron is inscribed in Jupiter's sphere, and so on.
planetary spheres are determined. That, said Kepler, is why there are
only six planets and why they are arranged as they are. God's nature
is mathematical.
Kepler's use of the regular solids was not simply a youthful ex-
travagance, or if it was, he never grew up. A modified form of the
same law appeared twenty years later in his Harmonies of the World,
the same book that propounded the Third Law. Also in that book
Kepler elaborated a new set of Neoplatonic regularities which related
the maximum and minimum orbital speeds of the planets to the con-
cordant intervals of the musical scale. Today this intense faith in
number harmonies seems strange, but that is at least partly because
today scientists are prepared to find their harmonies more abstruse.
Kepler's application of the faith in harmonies may seem naive, but
the faith itself is not essentially different from that motivating bits of
the best contemporary research. Certainly the scientific attitude
demonstrated in those of Kepler's "laws" which we have now discarded
is not distinguishable from the attitude which drove him to the three
Laws that we now retain. Both sets, the "laws" and the Laws, arise
from the same renewed faith in the existence of mathematical harmony
that had so large a role in driving Copernicus to break with the astro-
nomical tradition and in persuading him that the earth was, indeed,
in motion. But in Kepler's work, and particularly in the parts of it
that we have now discarded, the Neoplatonic drive to discover the
hidden mathematical harmonies embedded in nature by the Divine
Spirit are illustrated in a purer and more distinct form.
Galilee Galilei
Kepler solved the problem of the planets. Ultimately his ver-
sion of Copernicus' proposal would almost certainly have converted
all astronomers to Copernicanism, particularly after 1627 when Kepler
issued the Rudolphine Tables, derived from his new theory and
clearly superior to all the astronomical tables in use before. The story
of the astronomical components of the Copernican Revolution might
therefore end with the gradual acceptance of Kepler's work because
that work contains all the elements required to make the Revolution
in astronomy endure. But, in fact, the astronomical components of
the story do not end there. In 1609 the Italian scientist Galileo Galilei
( 1564---1642) viewed the heavens through a telescope for the first time,
and as a result contributed to astronomy the first qualitatively new
sort of data that it had acquired since antiquity. Galileo's telescope
changed the terms of the riddle that the heavens presented to astrono-
mers, and it made the riddle vastly easier to solve, for in Galileo's
hands the telescope disclosed countless evidences for Copernicanism.
But Galileo's new statement of the riddle was not formulated until
after the riddle had been solved by other means. If it had been an-
nounced earlier, the story of the Copernican Revolution would be
quite different. Coming when it did, Galileo's astronomical work
contributed primarily to a mopping-up operation, conducted after the
victory was clearly in sight.
In 1609 the telescope was a new instrument, though it is not clear
just how new it was. Galileo heard that some Dutch lens grinder had
combined two lenses in a way that magnified distant objects; he tried
various combinations himself and quickly produced a low-power
telescope of his own. Then he did something which, apparently, no
one had done before: he directed his glass to the heavens, and the
result was astounding. Every observation disclosed new and un-
suspected objects in the sky. Even when the telescope was directed
to familiar celestial objects, the sun, moon, and planets, remarkable
new aspects of these old friends were discovered. Galileo, who had
been a Copernican for some years before he knew of the telescope,
managed to turn each new discovery into an argument for Coperni-
canism.
The telescope's first disclosure was the new worlds in the firma-
ment about which Donne, only two years later, complained. Wherever
he turned his glass, Galileo found new stars. The population of the
most crowded constellations increased. The Milky Way, which to the
naked eye is just a pale glow in the sky (it had frequently been ex-
plained as a sublunary phenomenon, like comets, or as a reflection
of diffused light from the sun and moon) was now discovered to be
a gigantic collection of stars, too dim and too little separated to be
resglved by the naked eye. Overnight the heavens were crowded by
countless new residents. The vast expansion of the universe, perhaps
its infinitude, postulated by some of the Copernicans, seemed sud-
denly less unreasonable. Bruno's mystical vision of a universe whose
infinite extent and population proclaimed the infinite procreativeness
of the Deity was very nearly transformed into a sense datum.
Observation of the stars also resolved a more technical difficulty
that had confronted the Copernicans. Naked-eye observers had esti-
mated the angular diameter of stars and, with the aid of the accepted
figure for the distance between the earth and the sphere of the stars,
had transformed the angular diameter into an estimate of linear
dimensions. In a Ptolemaic universe these estimates had given not
r
l
unreasonable results: the stars might be as large as the sun, or there-
abouts. But, as Brahe repeatedly emphasized in his attacks upon
Copernicanism, if the Copernican universe were as large as the ab-
sence of stellar parallax demanded, then the stars must be incredibly
large. The brighter stars of the heavens must, Brahe computed, be
so large that they would more than fill the entire orbit of the earth,
and this he not unnaturally refused to believe. But when the tele-
scope was directed to the heavens, Brahe's problem turned out to be
an apparent problem only. The stars did not need to be so large as he
had estimated. Though the telescope immensely increased the number
of stars visible in the skies, it did not increase their apparent size.
Unlike the sun, moon, and planets, all of which were magnified by
Galileo's glass, the stars retained the size they had had before. It be-
came apparent that the angular diameter of stars had been immensely
overestimated by naked-eye observation, an error now explained as
a consequence of atmospheric turbulence which blurs the images of
stars and spreads them over a wider area in the eye than would be
covered by their undistorted image alone. The same effect makes the
stars seem to twinkle; it is largely suppressed by the telescope, which
gathers a larger number of rays to the eye.
The stars did not, however, provide the only, or even the best,
evidence for Copernicanism. When Galileo turned his telescope to
the moon, he found that its surface was covered by pits and craters,
valleys and mountains. Measuring the length of the shadows cast into
craters and by mountains at a time when the relative positions of the
sun, moon, and earth were known, he was able to estimate the depths
of the moon's declivities and the height of its protuberances and to
begin a three-dimensional description of the moon's topography. It
was not, Galileo decided, very different from the earth's topography.
Therefore, like the measurements of the parallax of comets, telescopic
observations of the moon raised doubts about the traditional distinc-
tion between the terrestrial and the celestial regions, and those doubts
were reinforced almost immediately by telescopic observations of the
sun. It too showed imperfections, dark spots which appeared and
disappeared on its surface. The very existence of the spots conflicted
with the perfection of the celestial region; their appearance and dis-
appearance conflicted with the immutability of the heavens; and, worst
of all, the motion of the spots across the sun's disk indicated that the
sun rotated continually on its axis and thus provided a visible paradigm
for the axial rotation of the earth.
But this was not the worst. Galileo looked at Jupiter with his
telescope and discovered four small points of light quite close to it
in the sky. Observations made on successive nights showed that they
continually rearranged their relative positions in a manner that could
most simply be explained by supposing that they revolved continu-
ally and quite rapidly about Jupiter (Figure 43). These bodies were

...

Figure 43. Three successive observations of Jupiter and its satellites separated
by intervals of several days. The constant rearrangement of the four small satellites
is most easily explained by supposing that the satellites are constantly rotating
about the larger planet.
the four principal moons of Jupiter, and their discovery had an im-
mense impact upon the seventeenth-century imagination. There were,
it appeared, new worlds "in the Planets" as well as in "the Firmament."
More important, these new worlds could not be conceived, on either
the Ptolemaic or the Copernican hypothesis, to move in roughly cir-
cular orbits about the center of the universe. Apparently they moved
around a planet, and their behavior was therefore the same as that
of the earth's moon in Copernican astronomy. The discovery of
Jupiter's moons therefore reduced the force of one more objection
to !he Copernican system. The old astronomy, as well as the new,
would have to admit the existence of satellites, governed by planets.
In addition, and perhaps most consequential of all, the observations
of Jupiter provided a visible model of the Copernican solar system
itself. Here in planetary space was a heavenly body surrounded by
its own "planets," just as the planets previously known encircled the
sun. The arguments for Copernicanism were multiplied by the tele-
scope almost as rapidly as the heavenly bodies themselves.
Many other arguments were derived from telescopic observation,
but only the observations of Venus provide sufficiently direct evidence
for Copernicus' proposal to concern us here. Copernicus himself had
noted in Chapter lO of the First Book of the De Revolutionibus that
the appearance of Venus could, if observable in detail, provide direct
information about the shape of Venus's orbit. If Venus is attached to
an epicycle moving on an earth-centered deferent, and if the center
of the epicycle is always aligned with the sun, then, as indicated by
Figure 44a, an observer on the earth should never be able to see more
than a crescent edge of the planet. But if Venus's orbit encircles the
sun as in Figure 44b, then an earthbound observer should be able to
see an almost complete cycle of phases, like the moon's; only phases
near "new" and "full" would be imperceptible, because Venus would
then be too close to the sun. Venus's phases can not be distinguished
with the naked eye, which sees the planets as mere shapeless points.
But the telescope enlarges planets sufficiently to give them shape,
(a)
(b)

(c)
Figure 44. The phases of Venus in (a) the Ptolemaic system, (b) the Coperni-
can system, and (c) as observed with a low-power telescope. In (a) an observer
on the earth should never see more than a thin crescent of the lighted face. In (b)
he should see almost the whole face of Venus illuminated just before or after Venus
crosses behind the sun. This almost circular silhouette of Venus when it first
becomes visible as an evening star is drawn from observations with a low-power
telescope on the left of diagram (c). The successive observations drawn on the
right show how Venus wanes and simultaneously increases in size as its orbital
motion brings it closer to the earth.
and, as indicated in Figure 44c, its shape provides strong evidence
that Venus moves in a sun-centered orbit.
The evidence for Copernicanism provided by Galilco's telescope
is forceful, but it is also strange. None of the observations discussed
above, except perhaps the last, provides direct evidence for the main
tenets of Copernicus' theory- the central position of the sun or the
motion of the planets about it. Either the Ptolemaic or the Tychonic
universe contains enough space for the newly discovered stars; either
can be modified to allow for imperfections in the heavens and for
satellites attached to celestial bodies; the Tychonic system, at least,
provides as good an explanation as the Copernican for the observed
phases of and distance to Venus. Therefore, the telescope did not
prove the validity of Copernicus' conceptual scheme. But it did pro-
vide an immensely effective weapon for the battle. It was not proof,
but it was propaganda.
After 1609 the main psychological force of the Ptolemaic system
was its conservatism. Those who held to it would not be forced to
learn new ways. But if the Ptolemaic system required extensive re-
visions to adjust it to the results of telescopic observation, it would
lose even its conservative appeal. It was very nearly as easy to make
the full transition to Copernicanism as to adjust to the requisite new
version of Ptolemy, and many of those who took the observations
seriously did make the full transition. These new converts may also
have been impelled by another consideration: the Copernicans, or
at least the cosmologically more radical ones, had anticipated the
sort of universe that the telescope was disclosing. They had predicted
a ~ e t a i l the phases of Venus, with precision. More important, they
had anticipated, at least vaguely, the imperfections and the vastly
increased population of the heavens. Their vision of the universe
showed marked parallels to the universe that the telescope made
manifest. There are few phrases more annoying or more effective
than "I told you so."
For the astronomically initiate the evidence of the telescope was,
perhaps, superfluous. Kepler's Laws and his Rudolphine Tables would
have been equally, though far more slowly, effective. But it is not on
the astronomically initiate that the telescope had the greatest im-
mediate impact. The first unique role of the telescope was providing
generally accessible and nonmathematical documentation for the
Copernican point of view. After 1609 men who knew only a smatter-
ing of astronomy could look through a telescope and see for them-
selves that the universe did not conform to the nai've precepts of
common sense, and during the seventeenth century they did look.
The telescope became a popular toy. Men who had never before
shown interest in astronomy or in any science bought or borrowed
the new instrument and eagerly scanned the heavens on clear nights.
The amateur observer became a well-known figure, a subject for both
emulation and parody. With him came a new literature. The begin-
nings of both popular science and science fiction are to be discovered
in the seventeenth century, and at the start the telescope and its dis-
coveries were the most prominent subjects. That is the greatest im-
portance of Galileo's astronomical work: it popularized astronomy,
and the astronomy that it popularized was Copernican.
The Decline of Ptolemaic Astronomy
Kepler's ellipses and Galileo's telescope did not immediately
crush the opposition to Copernicanism. On the contrary, as we noted
at the start of this chapter, the bitterest and most vociferous opposition
was not organized until after both Kepler and Galileo had made their
principal astronomical discoveries. Kepler's work, like Copernicus'
sixty-five years earlier, was accessible only to trained astronomers, and,
in spite of the great accuracy that Kepler was known to have achieved,
many astronomers found his noncircular orbits and his new techniques
for determining planetary velocities too strange and uncongenial for
immediate acceptance. Until after the middle of the century a number
of eminent European astronomers can be found trying to show that
Kepler's accuracy can be duplicated with mathematically less radical
systems. One tried to revert to epicycles; another consented to ellipses
but insisted that the speed of a planet was uniform with respect to the
unoccupied focus of the ellipse; still others tried orbits of another
shape. None of these attempts was successful, and as the century con-
tinued fewer and fewer of them were made. But not until the last
decades of the seventeenth century did Kepler's Laws become the
universally accepted basis for planetary computations even among the
best practicing European astronomers.
Galileo's observations met initially even greater opposition, though
from a different group. With the advent of the telescope Copernican-
ism ceased to be esoteric. It was no longer primarily the concern of
highly trained mathematical astronomers. Therefore it became more
disquieting and, to some, more dangerous. The new worlds discovered
by the telescope were a primary source of Donne's malaise. A few
years later telescopic observations provided part of the impetus neces-
sary to set in motion the ecclesiastical machinery of official Catholic
opposition to Copernicanism. After Galileo had announced his ob-
servations in 1610, Copernicanism could not be dismissed as a mere
mathematical device, useful but without physical import. Nor could
even the most optimistic still regard the concept of the earth's motion
as a temporary lunacy likely to vanish naturally if left to itself. The
telescopic discoveries therefore provided a natural and appropriate
focus for much of the continuing opposition to Copernicus' proposal.
They showed the real cosmological issues at stake more quickly and
more clearly than pages of mathematics.
The opposition took varied forms. A few of Galileo's more fanatical
opponents refused even to look through the new instrument, asserting
that if God had meant man to use such a contrivance in acquiring
knowledge, He would have endowed men with telescopic eyes. Others
looked willingly or even eagerly, acknowledged the new phenomena,
but claimed that the new objects were not in the sky at all; they were
apparitions caused by the telescope itself. Most of Galileo's opponents
behaved more rationally. Like Bellarmine, they agreed that the
phenomena were in the sky but denied that they proved Galileo's
contentions. In this, of course, they were quite right. Though the
telescope argued much, it proved nothing.
The continuing opposition to the results of telescopic observation
is of the deeper-seated and longer-lasting opposition to
Copernicanism during the seventeenth century. Both derived from
the same source, a subconscious reluctance to assent in the destruc-
tion of a cosmology that for centuries had been the basis of everyday
practical and spiritual life. The conceptual reorientation that, after
Kepler and Galileo, meant economy to scientists frequently meant a
loss of conceptual coherence to men like Donne and Milton whose
primary concerns were in other fields, and some men whose first
interests were religious, moral, or aesthetic continued to oppose
Copernicanism bitterly for a very long time. The attacks were scarcely
abated by the middle of the seventeenth century. Many important
tracts insisting on a literal interpretation of Scripture and upon the
absurdity of the earth's motion continued to appear during the first
decades of the eighteenth century. As late as 1873 the ex-president
of an American Lutheran teachers' seminary published a work con-
demning Copernicus, Newton, and a distinguished series of subsequent
astronomers for their divergence from scriptural cosmology. Even
today the newspapers occasionally report the dicta of a dotard who
insists upon the uniqueness and stability of the earth. Old conceptual
schemes never die!
But old conceptual schemes do fade away, and the gradual ex-
tinction of the concept of the earth's uniqueness and stability clearly,
if almost imperceptibly, dates from the work of Kepler and Galileo.
During the century and a half following Galileo's death in 1642, a
belief in the earth-centered universe was gradually transformed from
an essential sign of sanity to an index, first, of inflexible conservatism,
then of excessive parochialism, and finally of complete fanaticism.
By the middle of the seventeenth century it is difficult to find an im-
portant astronomer who is not Copernican; by the end of the century
it is impossible. Elementary astronomy responded more slowly, but
during the closing decades of the century Copernican, Ptolemaic,
and Tychonic astronomy were taught side by side in many prominent
Protestant universities, and during the eighteenth century lectures
on the last two systems were gradually dropped. Popular cosmology
felt the impact of Copernicanism most slowly of all; most of the
eighteenth century was required to endow the populace and its
teachers with a new common sense and to make the Copernican uni-
verse the common property of Western man. The triumph of Coperni-
canism was a gradual process, and its rate varied greatly with social
status, professional affiliation, and religious belief. But for all its
difficulties and vagaries it was an inevitable process. At least it was
as inevitable as any process known to the historian of ideas.
The Copernican universe assimilated during the century and a half
after Galileo's death was not, however, the universe of Copernicus or
even of Galileo and Kepler. Nor was its novel structure derived pre-
dominantly from astronomical evidence. Copernicus and the astrono-
mers who followed him made the first successful substantive break
with Aristotelian cosmology, and they began the construction of the
new universe. But the early <?opernicans did not fully see where their
work was leading. During the seventeenth century many other scien-
tific and cosmological currents converged to modify and complete the
cosmological framework that had directed their thought. The Co-
pernicanism that the eighteenth, nineteenth, and twentieth centuries
inherited is a Copernicanism rebuilt to suit the seventeenth-century
conception of a Newtonian world machine. That final historic integra-
tion of Copernican astronomy into the complete and coherent universe
envisaged by the seventeenth century is the subject of our final chap-
ter, though we shall treat it only with the limited detail and fore-
shortened perspective appropriate to an epilogue. In so far as the Co-
pernican Revolution was a revolution merely in astronomical thought,
its story ends here. What follows is a partial sketch of the larger revo-
lution in science and cosmology - a revolution which began with
Copernicus and through which the Copernican Revolution was at last
completed.

7
THE NEW UNIVERSE
The New Scientific Perspective
Kepler and Galileo compiled impressive evidence for the
earth's status as a moving planet. The concept of elliptical orbits and
the new data collected with telescopes were, however, only
cal evidence for the planetary earth. They did not answer the non-
astronomical evidence against it. While they remained unanswered,
each of those arguments, whether physical, or cosmological, or reli-
gious, testified to an immense disparity between the concepts of tech-
nical astronomy and those employed in other sciences and in philoso-
phy. The more difficult it became to doubt the astronomical innovation,
the more urgent was the need for adjustments in other fields of thought.
Until those adjustments were made, the Copernican Revolution was
incomplete.
Most large-scale upheavals in scientific thought produce similar
conceptual disparities. We are today, for example, in the late stages
of a scientific revolution initiated by Planck, Einstein, and Bohr. Their
new concepts and others upon which the contemporary revolution
depends show close historical parallels to Copernicus' concept of a
planetary earth. Conceptions like Bohr's atom and Einstein's finite but
unbounded space were introduced to solve pressing problems in a
single scientific specialty. Those who accepted them did so initially
because of the immense felt need in the field of their origin and in
spite of their obvious conflict with common sense, physical intuition,
and the basic concepts of other sciences. For a time they were used by
the specialist even though, within the larger climate of scientific
thought, they seemed incredible.
Cortinued use, however, makes even the strangest conception
plausible, and once plausible the new conception gains a larger scien-
tific function. It ceases, in the vocabulary of Chapter 1, to be merely
a paradoxical and ad hoc device for economically describing the
known, and becomes instead a basic tool for explaining and exploring
nature. At this stage the new conception cannot be restricted to a
single scientific specialty. Nature ought not display incompatible
properties in different fields. If the physicist's electron can leap from
path to path without crossing the intervening space, then the chemist's
electron should have the same ability, and the philosopher's concept
of matter and space demand reexamination. Every fundamental inno-
vation in a scientific specialty inevitably transforms neighbvring sci-
ences ancl, more slowly, the worlds of the philosopher and the edu-
cated layman.
Copernicus' innovation is no exception. In the early decades of the
seventeenth century it was at best an astronomical innovation. Out-
side of astronomy it raised a host of problems just as perplexing and
far more obvious than the questions of numerical detail it had re-
solved. Why do heavy bodies always fall toward the surface of a
spinning earth as the earth moves in its orbit about the sun? How far
away are the stars, and what is their role in the structure of the uni-
verse? What moves the planets, and how, in the absence of spheres,
are they kept in their orbits? Copernican astronomy destroyed tradi-
tional answers to these questions, but it supplied no substitutes. A new
physics and a new cosmology were required before astronomy could
again participate plausibly in a unified pattern of thought.
Before the end of the seventeenth century that new science and
cosmology had been created, and the men responsible were all members
of the Copernican minority. Their adherence to Copernicanism gave
a new shape and direction to much of their research. It provided a
new set of problems, one of which- what moves the earth?- has
alr;ady emerged briefly in our study of Kepler's anima matrix. In
addition, Copernicanism supplied a multiplicity of hints about the
concepts and techniques that the solution of these new problems de-
manded. By suggesting, for example, the unification of terrestrial and
celestial laws, it made the projectile a legitimate source of information
about planetary motions. Finally, Copernicanism gave a new meaning
and value to a number of cosmological doctrines which, though cur-
rent as minority views in antiquity and the Middle Ages, had previ-
ously been disregarded by most scientists. During the seventeenth
century several of these newly popular views, particularly atomism,
were a constant source of significant suggestions for science.
THE NEW UNIVERSE 231
These new problems, new techniques, and new evaluations consti-
tute the new perspective that seventeenth-century science gained from
Copernicanism. The last chapter displayed the effects of this new
perspective upon astronomy. This one will show its role in the de-
velopment of other sciences and of cosmology, for the Newtonian
universe was born in an intellectual climate that Copernicanism had
helped make fertile. But unlike Kepler's Laws, which are the astro-
nomical culmination of the Copernican Revolution, the Newtonian uni-
verse is a product of more than Copernicus' innovation. In discussing
its evolution and discovering how the concept of a planetary earth
came at last to make coherent sense we shall therefore have occa-
sionally to introduce concepts and techniques that have been neglected
to this point because they had little bearing upon the development of
astronomy or cosmology until after Copernicus' death. Our problem
now becomes larger than the Copernican Revolution proper.
Toward on Infinite Universe
The Aristotelian universe had been, in most versions, a finite
universe -matter and space ended together at the sphere of the stars -
and most early Copernicans preserved this traditional feature. In the
cosmologies of Copernicus, Kepler, and Galileo the center of the sun
coincided with the center of the finite stellar sphere; the sun simply
changed places with the earth, becoming the unique central body, the
Neoplatonic symbol of the Deity. This new two-sphere universe was
a natural revision of traditional cosmology. Since there was no concrete
counterevidence, it might well have endured until the nineteenth cen-
tury when improved telescopes showed that different stars were at
very different distances from the sun.
The role of the two-sphere construction was, however, very dif-
ferent in the Aristotelian and Copernican world views; finitude, in
particular, had essential functions in the first that were completely
absent in the second. For example, in Aristotelian science the stellar
sphere was needed to carry the stars in their diurnal circles and to
provide the push that kept both the planets and terrestrial objects
in motion. In addition, the outer sphere defined an absolute center of
space, the center toward which all heavy objects moved of their own
accord. Copernicus' universe deprived the sphere of all these functions
and of others as well. Terrestrial motion did not demand an absolute
center of space; stones fell toward the moving earth. Nor was an outer
sphere needed to produce celestial motions; whether or not located on
a sphere, the stars were at rest. Copernicans were still at liberty to
preserve the stellar sphere, but only tradition could supply a motive
for doing so. The sphere could be abandoned without disrupting either
Copernican physics or cosmology.
Copernicanism therefore allowed a new freedom to cosmological
thought, and the result was a new speculative conception of the uni-
verse that would surely have horrified both Copernicus and Kepler.
A century after Copernicus' death his two-sphere framework had been
replaced by a universe in which the stars were scattered here and there
through an infinite space. Each of these stars was a "sun," and many of
them were thought to possess their own planetary systems. By 1700
the unique earth, which Copernicus had reduced to but one of six
planets, had become little more than a speck of cosmic dust.
Though historians still know little about the way in which this
new Copernican conception was established, its origin is quite clear.
By doing away with the cosmological functions of the outer sphere,
Copernicus revitalized three earlier speculative conceptions of the
universe, conceptions associated respectively with scholasticism, Nco-
platonism, and atomism. Before the De Revolutionibus these three
cosmologies were quite different in structure and motive, and none
was relevant to celestial science. But all were transformed to scientific
cosmologies by Copernicanism, and once transformed they showed
remarkable structural similarities.
Consider first the most prevalent pre-Copernican conception of an
infinjte universe, developed by Islamic philosophers who could not
accept Aristotle's proof of the logical impossibility of a void. This uni-
verse was substantially the same as Aristotle's from the central earth
to the rotating stellar sphere, but beyond the sphere space no longer
ceased to exist with matter. Instead, the entire Aristotelian universe
was embedded as a kernel at the center of an infinite space devoid
of matter, the abode of God and his angels. Because it did not con-
strain God's power to make an infinite universe, this conception be-
came relatively popular in Europe after the thirteenth century. It was
described in several well-known elementary books current during
Copernicus' lifetime, and knowledge of the conception may have
helped him to justify the expansion of the stellar sphere that was
necessitated by the absence of observed parallax. But before Coper-
nicus, this version of an infinite universe had had little bearing upon
the practice of astronomy or any other science. So long as celestial
bodies were conceived to be in continuous motion, they could not
easily be placed in the infinite space beyond the outermost sphere.
The functions of that space were theological, not physical or astro-
nomical.
By bringing the stars to rest, however, Copernicus made it possible
to give infinite space astronomical functions, and that new freedom
was first exploited about a generation after the publication of the
De Revolutionibus. In 1576 the English Copernican, Thomas Digges,
introduced the idea of an infinite universe into an otherwise straight-
forward paraphrase of Copernicus' Book One, and the result, repro-
duced from Digges's original illustration, is shown in Figure 45. The
central core of the universe is identical with the universe of the De
Revolutionibus, but the stars have been removed from the surface of
the stationary stellar sphere and scattered outward through the in-
finite space posited by the older minority cosmological tradition.
Though few of Copernicus' immediate successors went as far as
Digges, many of them recognized that the stars no longer had to
be on a sphere and that the distances between individual stars and
the sun might vary without affecting the appearances. When Galileo's
telescope showed countless new stars where none had been seen
before, the scattering of stars through an immeasurably distant space
seemed to the less traditionally minded astronomers almost a matter
of experience.
Digges was the first to describe an infinite Copernican universe,
but he achieved infinity only by the unconscious introduction of a
paradox which, in antiquity and the Middle Ages, had provided a
principal reason for rejecting infinite space. Digges's unique central
sun is a contradiction, for it is no more "at the center" than each one
of the stars and pJanets. The center is the point that is equidistant
from all points on the periphery, and that condition is satisfied by
every point in an infinite universe or by none. This paradox had been
fully elaborated a century before Copernicus by the important Nco-
platonist, Nicholas of Cusa. He had believed that the universe was an
infinite sphere - no smaller sphere would, he said, be consistent with
God's creative omnipotence- and he had expressed the resulting para-
dox by declaring that the center of the sphere everywhere coincided
with its periphery. In his universe each body, fixed or moving, was
simultaneously at the center, on the surface, and in the interior. Be-
A perfit defcription of the Crelefiiall
tl the 1111JI J1llrint '.{tk
PJih4[.1TUIII. &t.
Figure 45. The infinite Copernican universe of Thomas Digges, reproduced
from his Perfit Description of the Caelesliall Orbcs, published in 1576. The
diagram is like all other early sketches of the Copernican universe except that the
stars are no longer restricted to the surface of the celestial sphere. No stars occur
within the sphere (if they did, there would be observable stellar parallax), but the
infinite space beyond the sphere is studded with them. Notice, however, that the
sun still retains a privileged position and that the distance between neighboring
stars is far less than that between the sun and the celestial sphere. In Digges's
universe the sun is not just another star.
cause no part of space could be distinguished from any other, the
occupants of space - the earth, planets, and stars - must all move and
must all be of the same nature.
Cusa's vision provides a second example of a cosmology that
could be transformed by the existence of Copernicanism. As Cusa
developed it, a hundred years before the publication of the De Revo-
lutionibus, the cosmology made no scientific sense at all. In the role
of cosmologist, Cusa was a mystic who joyfully rejected the appear-
ances for the sake of a transcendent apprehension of the infinite Deity
in whom all paradoxes were reconciled. Yet the Neoplatonic insistence
upon the infinite and its paradoxes was not intrinsically incompatible
with the appearances or with science. After Copernicus' death the
same insistence provided a motive and recurrent theme for the cos-
mological writings of the Italian mystic, Giordano Bruno, and in
Bruno's vision of the universe the infinite and the appearances were
reconciled by Copernicanism. Bruno's approach to cosmology was
scarcely more concerned with science or with the appearances than
was the approach of Cusa, by whom he was greatly influenced. But
whatever his motives Bruno was right. The sun need not be at the
center; in fact, no center is needed. A Copernican solar system may
be set down anywhere in an infinite universe. Providing only that the
sun is far enough from the neighboring stars to account for the ab-
sence of parallax, the appearances will be preserved.
Bruno's reconciliation of an infinite and centerless universe with
the appearances was only one part of his cosmological construction.
Beginning around 1584 he also made explicit the physical relation of
the Copernican solar system to the other celestial residents of his in-
finite space. The sun was, he thought, merely one of an infinite num-
ber of stars scattered through the infinite expanse of space; some of
the other bodies in the infinite heavens must be populated planets like
the earth. Not only the earth but the sun and the entire solar system
were transformed to insignificant specks lost in the infinitude of God's
creation; the compact and ordered cosmos of the scholastics had be-
come a vast chaos; the Copernican departure from tradition had
reached its maximum.
But though radical, this last extension of Copernicanism was
achieved almost without novelty. Two millenniums before Bruno's birth
the ancient atomists, Leucippus and Democritus, had envisaged an
236
infinite universe containing many moving earths and many suns. In
antiquity their doctrines had never rivaled Aristotle's as a basis. for
scientific thought, and their writings had disappeared almost entirely
during the Middle Ages. But the works of their Epicurus
and Lucretius, were among the principal literary recovenes made by
the Renaissance humanists. From these works, particularly from Lucre-
tius' De Rerum Natura, Bruno derived many of his most fruitful con-
ceptions. In Bruno's cosmology a third ancient
of the universe was revitalized and given new vens1m1htude by Its
affinity for Copernicanism.
.
That affinity is somewhat surprising, for both historically and logi-
cally atomism and Copernicanism seem totally different doctrines.
The ancient atomists had developed the main tenets of their cos-
mology not principally from observation but in an .attempt :o
apparent logical paradoxes. The existence and motion of bod.tes
could, they felt, be explained only if the real world conSISted of tmy
indivisible corpuscles, or atoms, swimming free in a vast empty space,
or void. The void was required to account for motion; if there were
no empty spaces, there would be no place for matter to move into.
Similarly, the indivisibility of the ultimate particles seemed .to
essential for the existence of finite bodies; if matter were mfimtely
divisible, then its ultimate parts would be mere geometric points,
occupying no space at all. From parts that, taken individually, occupy
no volume it seemed impossible to construct a finite body that would.
Zero plus zero equals zero no matter how often the addition is re-
peated. Reality must therefore, said the atomists, consist of
atoms and the void, and this premise, quite foreign to Copermcamsm,
.;as the foundation of their world view.
The premise had, however, some striking consequences which were
not so foreign. For example, the atomists' void must be infinite in
extent. It could only be bounded by matter, and the matter must,
in turn, be bounded by more void. When matter and space ceased to
go together, as they had in Aristotle's physics, there was no end .to
the process of bounding the universe. Again, there were no spec1al
positions or unique bodies in the atomists' universe. The void itself
was neutral; each position was like every other. The earth or sun
existed in one region rather than another simply because the fortui-
tous motions and collisions of the atoms happened to produce an
THE NEW UNIVERSE
237
aggregate at that position, and because once the atoms had met by
chance they became entangled and stuck. The process could equally
well have occurred somewhere else. In fact, since the universe was
infinite and contained infinitely many atoms, the process almost cer-
tainly had occurred elsewhere at some time. Many earths and suns
as well as many atoms populated the infinite void of atomistic cos-
mology. There was no possible terrestrial-celestial dichotomy. Accord-
ing to the atomists, the same sort of matter obeyed the same set of
laws everywhere in the infinite neutral void.
Since Copernicanism also destroyed the earth's uniqueness, abol-
ished the terrestrial-celestial distinction, and suggested the infinity of
the universe, the atomists' infinite void provided a natural home for
Copernicus' solar system, or rather for many solar systems. Bruno's
principal contribution was recognizing and elaborating this obscure
affinity between the ancient and modern doctrines. Once the affinity
was recognized, atomism proved the most effective and far-reaching
of the several intellectual currents which, during the seventeenth
century, transformed the finite Copernican cosmos into an infinite and
multipopulated universe. That extension of cosmological dimensions
was, however, only the first of atomism's significant roles in the con-
struction of the new universe.
The Corpuscular Universe
Early in the seventeenth century atomism experienced an
immense revival. Partly because of its significant congruence with
Copernicanism and partly because it was the only develofed cos-
mology available to replace the increasingly discredited scholastic
world view, atomism was firmly merged with Copernicanism as a
fundamental tenet of the "new philosophy" which directed the scien-
tific imagination. Donne's lament that because of the "new Philoso-
phy" the universe was "crumbled out again to his Atomies" is an
early symptom of the confluence of these previously distinct intellec-
tual currents. By 1630 most of the men who dominated research in
the physical sciences showed the effects. They believed that
the earth was a moving planet, and they attacked the problems pre-
sented by this Copernican conception with a set of "corpuscular"
premises derived from ancient atomism.
The "corpuscularism" that transformed seventeenth-century science
often violated the premises of ancient atomism, but it was atomistic
nonetheless. Some of the "new philosophers" believed that the ultimate
particles were divisible in principle, but all agreed that they were
seldom or never divided in fact. Some doubted the void, but the
aethereal fluid with which they filled all space was for most purposes
as neutral and inactive as the void. And, most important, all agreed
that the motions, interactions, and combinations of the various par-
ticles were governed by laws imposed by God upon the corpuscles at
the Creation. The discovery of these laws was, for the corpuscularian,
the first problem in the program of the new science. The second was
to apply these laws in explaining the rich flux of sensory experience.
The French philosopher, Rene Descartes (1596-1650), was the
first man systematically to apply this program to the problems of a
Copernican universe. He began by asking how a single corpuscle
would move in the void. Then he asked how this free motion would
be altered by collision with a second corpuscle. Since he believed
that all change in the corpuscular universe resulted from a succession
of free corpuscular motions punctuated by intercorpuscular collisions,
Descartes expected to deduce the entire structure of the Copernican
universe from the answers to a few questions like these. Though all of
his deductions were intuitive and though most of them were mistaken,
the cosmology that his imagination dictated to his reason proved im-
mensely plausible. Descartes's vision dominated much of science for
almost a century after its details were first published in his Principles
of Philosophy in 1644.
Descartes's answer to his first question was extremely successful.
B)'. applying contemporary. versions of the medieval impetus theory
to a corpuscle in the infinite neutral space of atomistic cosmology, he
arrived at the first clear statement of the law of inertial motion. A
corpuscle at rest in the void will, he said, remain at rest, and a cor-
puscle in motion will continue to move at the same speed in a straight
line unless it is deflected by another corpuscle. The constancy of
particle speed was a straightforward consequence of the impetus
theory (discussed in the penultimate section of Chapter 4 ), particularly
as that theory had been developed by Galileo. But the linearity of
motion was a novelty and an immensely consequential one, typifying
the fruitful suggestions which atomism offered to seventeenth-century
science. The atomists' infinite void was a space with no center and
(except in a few debased versions which were discarded early in the
century) without any intrinsic directions, like "up" and "down." In
such a space a bocly free from external influence could only stop or
move straight ahead. The self-sustaining circular motions borrowed
from the scholastic impetus theory by Copernicus, Galileo, and other
early Copernicans were impossible. After Descartes's work they played
no significant role in the construction of the Copernican universe.
In nature, Descartes recognized, all particles or aggregates of par-
ticles continually change their speed and direction. These alterations
must, said Descartes, be caused by external pushes and pulls derived
from other bodies (Figure 46). Corpuscular collisions were therefore
c
E
B
I
I
I
I

D
m A
Figure 46. The effect of an impulse upon an inertial motion. At the point A
the body m is given a sharp push toward B. If no additional pushes intervene, the
body will move straight from A toward B at a constant speed. If the body is given
a second push in the same direction when it reaches B, it will continue straight on
toward C but at an increased speed. If pushed in the opposite direction, it may
continue toward C but at reduced speed, or, if the second push is strong enough,
the body may recoil from B and move straight back t::.-ward A. Finally, if the body
when it reaches B is given a lateral push toward D, it will begin a new inertial
motion along the slant line BE. The inertial motion along BE may be thought of
as the result of two simultaneous inertial motions, one along BC, generated by the
first push at A, and the other along BD, generated by the second lateral push at B.
the second subject for investigation, and with it Descartes was less
successful. Only one of his seven laws of collision was retained by
his successors. But though Descartes's laws of collision were discarded,
his conception of the collision process was not. Corpuscularism had
again introduced a new problem, and that problem was solved within
thirty years of Descartes's death. From the solution came both the law
of conservation of momentum and, though less directly, the conceptual
240
relation between a force and the change of momentum that it pro-
duces. Both were essential steps toward Newton's universe.
In proceeding from his laws of motion and collision to the struc-
ture of a Copernican universe, Descartes introduced a concept which
since the seventeenth century has greatly obscured the corpuscular
basis of his science and cosmology. He made the universe full. But the
matter that filled Cartesian space was everywhere particulate in struc-
ture, and in determining the behavior of this particulate plenum Des-
cartes made constant imaginative use of the void. He used it first to
determine the laws of motion and collision for individual particles.
Then, to discover how these laws operated in a plenum, he seems
first to have imagined the particles swimming in a void where their
inertial motions were punctuated by collisions; after which he gradu-
ally squeezed the void out of the system, bringing the particles closer
and closer together, until finally their collisions and inertial motions
merged into a single process in the plenum. Unfortunately, in the
plenum the motions of all particles must be considered simultaneously,
an incredibly complex problem which Descartes scarcely tried to solve.
Instead he indulged in an imaginative leap from his corpuscular laws
to his final solution, pausing for none of the absolutely essential inter-
mediate steps.
To Descartes it seemed self-evident that the only enduring mo-
tions in a plenum must occur in circulatory streams. Each particle in
such a stream pushes forward its nearest neighbor until, finally, to
avoid a vacuum, the push is returned to the first particle over an
approximately circular path. Then, having filled the potential void,
t'he process starts again. Since, for Descartes, these circulatory streams
were the only possible enduring motions, he believed that, whatever
push God gave to the corpuscles at the creation, they would ultimately
circulate in a set of vortices scattered through all of space. A small set
of these vortices is reproduced in Figure 47 from an illustration for
one of Descartes's early works.
Each of Descartes's vortices was, at least potentially, a solar sys-
tem, generated and governed by the corpuscular laws of inertia and
collision. For example, corpuscular impacts just balance the centrifu-
gal tendency that inertia gives to each corpuscle in the vortex. If all
others were removed, each single particle would travel straight ahead
along a tangent to its normally circular path and thus leave the
r
Figure 47. Descartes's vortex-cosmolo d
or a Treatise on Light. The oints 5 E ~ r p r o uced from his book, The World
rapid churning motion of thep e t . t'd ' ' ani d are the centers of vortices. The
I
r s nc e centra corp 1 k h
uminous, so that they act II'ke t Th . usc es rna es t ese centers self-
.
5
ars. e vanous dott d 1 h' h
precisely circular, represent the aths of h e em: es, w Ic need not be
that compose the vortex Th pd t t e eternally rotatmg corpuscular streams
e o s around the v rt
5
swept through their orbits by the t' I H o ex center are planets,
top of the diagram is a com t v ~ r Ica ow. The body C, which crosses the
the flow is too slow' to tr 'te', passmg from vortex to vortex in a region where
ap I m a continuing ci 1 b'
0
I .
space outside the d
1
g d rcu ar or It. t 1er vortices fill the
a ram an each one I
tion of a solar system D, , represents, at east potentially, the Ioca-
m escartes s multipopulated Copernican universe.
242
vortex. It does not do so only because constant collisions with particles
outside of it in the vortex continually drive it back toward the center.
Similar impacts the stable corpuscular aggregates that form the
planets circulating in approximately circular paths about the vortex
center.
The constant rapid churning motion of each vortex center sets up
a continual vibration which is transmitted in waves from the center
through all of space. According to Descartes that vibration is light,
continually emerging from the suns or stars which are the vortex cen-
ters. Apparently an infinite multiplicity of star-centered planetary sys-
tems have been derived from corpuscular premises. And Descartes's
derivations do not stop with these celestial phenomena. For example,
he explains the motion of the moon, the tides, and projectiles by posit-
ing a set of small subsidiary vortices, one around each planet. Cor-
puscular impacts within these smaller vortices keep, the .moon in .cir-
culation and drive falling stones to earth. In Descartes s umverse weight
itself, like motion, light, and all other sensory appearances, is ulti-
mately traceable to corpuscular collisions governed by the laws of
atomic motion and interaction.
It is, today, childishly easy to discover errors and inadequacies in
Descartes's discussion of vortex cosmology and in the astronomy, optics,
chemistry, physiology, geology, and dynamics that he derived from it.
His vision was inspired, and its scope was tremendous, but the amount
of critical thinking devoted to any one of its parts was negligibly
small. His laws of corpuscular collision provide but one of countless
examples. But in the development of seventeenth-century science the
of Descartes's system were far less important than the whole.
cartes's brilliant successors, led by Christian Huyghens, found their
inspiration in his underlying conception, rather than in its detailed
development. They could and did change his laws of collision, his
description of vortices, and his laws for the propagation of light. But
they did not compromise his conception of the universe as a corpuscu-
lar machine governed by a few specified corpuscular laws. For half
a century that conception guided the search for a self-consistent Co-
pernican' universe. It seems no coincidence that this basic conception
of the structure appropriate to a Copernican universe was so largely
inspired by an ancient world view which Copernicanism itself had
helped make popular.
THE NEW UNIVERSE
243
The Mechanical Solar System
. Two quite separate historical paths lead from Copernicus'
fimte c.osmos to the Newtonian universe that gave the
Copernican Revolution its final form. One of these is the liaison
illustrated between Copernicanism and the corpuscular
phy. The IS focused series of attacks upon the single
most pressmg physical problem raised by Copernicanism: What moves
the Both paths began half a century after Copernicus' death.
Their common source is the new scientific perspective created when
Ke?ler, and others separated the truly novel from the pseudo
elements in Copernicus' work. The two paths merge again
m .Newtons terminal formulation of the structure of the Copernican
umverse, and both provided essential elements to that formulation.
But except at their initial and final termini the two paths were most
often separate, though striking parallelisms gave occasional evidence
that they were proceeding in the same direction.
The physical explanation of planetary motion was not, in the six-
teenth and seventeenth centuries, quite a new problem. Neither Aris-
totle: Ptolemy, nor the medieval astronomers had been able fully to
the cause of each minor irregularity in a planet's mo-
tion. Bu.t traditiOnal science had at least explained the average east-
ward drift of all planets around the ecliptic. Planets and the spheres in
which they were embedded were made from a perfect celestial element
whose nature expressed itself in eternal rotations about the center of
the universe.
Copernicus had tried to preserve this traditional explanation of
planetary motion. But the conception of natural celestial motions was
less suitable to a sun-centered than to an earth-centered universe, and
the incongruities of Copernicus' initial proposal did not remain hidden
for long. To explain even the planets' eastward drift, the Copernican
system demanded that each particle of earth rotate naturally about
two different centers - the fixed center of the universe and the
center of the earth. Each particle of a satellite, like the moon,
was simultaneously governed by at least three centers -the center of
the universe, the center of the governing planet, and the center of the
satellite itself. Copernicanism therefore jeopardized the plausibility
of self-sustaining circular motions by compounding them and by re-
lating them to many simultaneous fixed and moving centers. Further-
more, the multiplicity and motion of the various centers deprived Co-
pernican motions of any fixed relation to the intrinsic geometry of
space. In Aristotelian physics all natural motions had been either
toward, or away from, or about the center of the universe. That center,
though only a geometric point, could plausibly possess a special causal
role because it was unique, determined once and for all by its relation
to the boundary of space. Copernicus' proposal, on the other hand,
demanded that some natural motions be governed by moving centers,
and moving centers could not act causally by virtue of geometric posi-
tion alone.
Late in the sixteenth and early in the seventeenth century other
new astronomical doctrines converged to make the physical problem
of the planets still more acute. All the celestial spheres except the
sphere of the stars were made obsolete by the new observations of
comets and by the increasing popularity of the Tychonic system. With
the spheres went the entire physical mechanism that had previously
accounted for the planets' average circular motions. Even this dissolu-
tion of spheres did not end the influence of the classical approach.
As late as 1632 Galileo could still attempt to elaborate Copernicus'
physical doctrine, arguing in his Dialogue on the Two Principal Sys-
tems of the World that even without celestial spheres all matter would
rotate naturally, regularly, and eternally in a set of compounded cir-
cles. But the brilliance and subtlety of Galileo's dialectic- scarcely
rivaled since in a major scientific work- could not for long disguise
the fundamental implausibility of the approach. His Dialogue was
imp.prtant as a great popularization of Copernicanism, but his monu-
mental contributions to physical science are in other works. After his
death progress on the physical problem of the planets took a quite
different direction, because even before Galileo's Dialogue appeared
Kepler's researches had given the physical problems of Copernicanism
a new dimension and suggested a new set of techniques for their
solution.
By doing away with the profusion of epicycles and eccentrics, Kep-
ler made it possible, for the first time, to subject the full complexity
of the celestial appearances to physical analysis. An explanation which,
like Copernicus' or Galileo's, dealt only with the average eastward
drift of planets ceased to seem adequate even when plausible. The
r
I
geometrically simple and precise elliptical motions rather than the
average drifts now demanded explanation. But the new precision and
simplicity were achieved only at a price. In contrast to the average
circular motions of classical astronomy, elliptical motions governed
by the Second Law could not be natural motions, for they were not
symmetric with respect to any center. A planet moving uniformly on a
deferent, or even on a simple epicycle-deferent system, is in some
sense "doing the same thing" or "moving in the same way" at each
and every point in its orbit; this motion might conceivably be "natu-
ral" to it. On the other hand, the motion of a planet obeying Kepler's
Laws changes speed, direction, and curvature at each point in the
orbit. These variations seemed to demand the introduction of a force
in the heavens, a force acting continually to alter the planet's motion
at each point in its orbit. In the heavens as on the earth an asymmetric
motion was most naturally explained as the result of a continuing push
or pull.
In other words, Copernicus' innovation first destroyed the tradi-
tional explanation of planetary motion and then, as modified by Kep-
ler, suggested a radically new approach to celestial physics. That new
approach first appeared in the writings of Kepler himself during the last
decades of the sixteenth century and the first decades of the seven-
teenth. In essence it was an inversion of the technique which Coperni-
cus had already used, and which Galileo was to resurrect, in unifying
terrestrial and celestial laws. Copernicus and Galileo achieved uni-
formity by applying the traditional conception of natural circular
celestial motions to the earth. Kepler achieved the same effect more
fruitfully by applying the ancient conception of violent force-governed
terrestrial motions to the heavens. Guided by his ever-present Neo-
platonic perception of the sun, Kepler introduced forces emanating
from the sun and planets to provide a causal foundation for planetary
motion. In his writings the solar system was for the first time modeled
on a terrestrial machine. For all the crudities of its initial conception,
the future lay with Kepler's approach.
The first of Kepler's solar forces was the anima motrix, which we
examined briefly in Chapter 6. Kepler visualized it as a system of
rays projecting from the sun in the plane of the ecliptic and carried
about by the sun's continuing rotation. As the moving arms passed
a planet, they pushed against it, impelling it in a continuing circle
about the sun. To change the basic circular orbit to an ellipse, a second
force was required, one that would change the distance between sun
and planet in different portions of the orbit. That second force Kepler
identified as magnetism, whose properties had recently been thoroughly
investigated and recorded in a very influential book, On the Magnet,
published in 1600 by the English physician William Gilbert. Gilbert
had recognized that the earth itself was a huge magnet, and Kepler
extended the generalization to the other bodies of the solar system.
Not only the earth, said Kepler, but also the planets and the sun are
magnets, and the attractions and repulsions of their various poles
determine the paths in which the planets move.
Few of Kepler's successors took his physical theory, whose details
are indicated in Figure 48, as seriously as they took his mathematical
description of the planetary orbits. Some of his dynamical concepts
I
/
............ /
____ /
..
Figure 48. Kepler's mechanical solar system. The sun is shown at the center
of the diagram, and the lines radiating from it represent the anima matrix which,
in the absence of additional magnetic forces, would push the planet, P, around
the broken circle centered on the sun. The magnets which, according to Kepler,
transform this circular motion into an ellipse are represented by arrows. The sun's
south pole is buried at its center, where the pole exerts no influence. Its north
pole is distributed uniformly over the sun's surface. The earth's magnetic axis
always remains very nearly parallel to itself as the earth moves. Therefore, when-
ever the planet is to the right of a central vertical line through the diagram, its south
pole is nearer the sun than its north, and it is attracted gradually toward the sun.
During the other half of its motion, the planet is gradually repelled. At all positions
the planet's orbital speed is inversely proportional to its distance from the sun, be-
cause the anima matrix is stronger near the sun.
THE NEW UNIVERSE
247
were out of date when he wrote; the sun rotates too slowly to account
for the observed periods of the planets; when observed with magnetic
needles the direction of the earth's magnetic axis is different from that
required to account for astronomical observations. Therefore, after
Kepler's death neither the anima matrix nor the magnetic sun often
appeared in seventeenth-century scientific writings. But Kepler's con-
ception of the solar system as a self-contained and self-governing
machine recurred again and again. As Copernicanism developed dur-
ing the seventeenth century, that conception proved to have a double
importance.
In the first place, Kepler's physical system, though entirely inde-
pendent of the corpuscular philosophy, reinforced some of corpuscu-
larism's most significant conclusions. In particular it provided a second
natural route to the conception of an infinite neutral space. In Kepler's
planetary mechanism a planet's motion depended only on its relation
to another physical body, the sun. Magnetism and the anima motrix
functioned equally well whatever the sun's position. Though Kepler
himself kept the sun at the center of a finite stellar sphere, no such
center was necessary. Corpuscularism had reached similar conclusions,
but from quite different motives and by a totally different route. Appar-
ently some of Copernicanism's most striking consequences could not
be suppressed by any of the approaches that led to a coherent Co-
pernican universe.
Kepler's substitution of violent, force-determined planetary motions
for the natural, force-free, and space-determined motions of tradi-
tional celestial physics has a second major role in the development
of seventeenth-century science. Kepler's mechanical solar system is the
first in a series that culminates in the system of Newton's Principia.
Historically, the intervening developments are extremely complex.
They depend upon the tortuous evolution and laborious assimilation
of a new set of dynamical concepts and mathematical techniques,
developments that by themselves could be the subject of another book.
But conceptually the route from Kepler to Newton is relatively simple.
A few significant emendations will convert Kepler's system to one that
is qualitatively very like Newton's, and these emendations are direct
consequences of recognizing the role in celestial physics of Descartes's
conception of inertial motion. The absence of that conception is the
principal feature which distinguishes Kepler's mechanical solar system
from the similar systems designed by Newton's immediate predecessors.
Two of these later systems, designed by the Italian G. A. Borelli ( 1608-
1679) and by the Englishman Robert Hooke ( 1635--1703), will bring
us very close to the qualitative features of Newton's system.
Borelli's conception of inertial motion was far less complete than
Hooke's, and his planetary theory was therefore far closer to Kep-
ler's. Unlike Kepler, Borelli realized that no push was necessary simply
to keep the planets from coming to rest. But he retained a sort of anima
matrix to account for the variation of a planet's speed with its distance
from the sun, and on occasions he too seems to have thought of the
anima matrix as a continuing pusher. Elsewhere his break with Kep-
A
(a)
(b)
Figure 49. Borelli's planetary theory. Diagram (a) shows Borelli's model of the
planetary mechanism in which a cork C is driven counterclockwise around the
bowl AA by the rotating arm DD. When the arm revolves rapidly, the cork spirals
quickly out to the rim of the bowl under the dominant influence of its inertial
tendency to move in a straight line. If the arm moves slowly, the cork spirals
toward the center of the bowl because the slight centrifugal tendency generated
by the arm's slow revolution is more than overcome by the attraction between the
magRets mounted at C and P. At an appropriate intermediate speed the two oppos-
ing tendencies, centripetal and centrifugal, will just balance, and the cork will move
in a continuous circle, the basic Copernican orbit.
Diagram (b) illustrates Borelli's derivation of an elliptical orbit. When the
planet moves around the broken circle, the centrifugal tendency generated by the
anima matrix just balances the planet's tendency to fall into the sun, and the orbit
is therefore a circle. If the planet is now removed from the broken circle and placed
on the solid curve at position 1 outside the circle, its motion will be slower,
corresponding in diagram (a) to a slower rotation of the arm, because at a greater
distance from the sun the anima motrix is weaker. As a result the planet will begin
to spiral inward toward the sun along the solid line, crossing the broken circle at
2, and continuing to 3 where the increased strength of the anima motrix has in-
creased its speed sufficiently to overcome the inward drift. Then the planet begins
an outward spiral which carries it back to 1. Borelli claimed that tl.e resulting
orbit would be an ellipse.
THE NEW UNIVERSE
249
ler (and Aristotle) was more complete. In particular, Borelli recog-
nized (and demonstrated with an imagined model described in Figure
49) that no push like that from the anima matrix could keep a planet
moving in a closed orbit. Unless some other force were available to
pull the planets straight toward the sun, each of them would, Borelli
thought, move off along a straight line tangent to its orbit and thus
leave the solar system entirely. To maintain stable orbits Borelli there-
fore introduced a second force which constantly deflected the depart-
ing planet back toward the sun. In his model Borelli used magnets to
simulate this force; in the heavens he displayed the residual strength
of Aristotelian concepts by replacing the force with a natural tendency
of all planets to fall toward the central sun.
Borelli's conception of the solar system was elaborated in a book
published in 1666, the same year in which Robert Hooke at last
demonstrated the full parallelism between the celestial motions and
those of a terrestrial machine. Much influenced by Descartes, Hooke
began with a complete conception of inertial motion and of the iden-
tity of terrestrial and celestial laws. As a result he was able to discard
both the anima matrix and the vestiges of natural tendencies to motion.
A moving planet ought, he said, to continue its motion uniformly in a
straight line through space, because the senses reveal nothing to push
or pull it. Since its motion is not straight, but rather in a continuous
closed curve encircling the sun, the immediate evidence of the senses
must be misleading. There must be an additional attractive principle
or force operating between the sun and each planet. Such a force
would, said Hooke, continually deflect the planets' straight inertial
motions toward the sun, and that is all that their Copernican orbits
require.
Hooke's intuitive perception of a Copernican planetary motion
is indicated by Figure 50a, though in a form more explicit than any
provided by Hooke himself. The solid circle (or it might be an ellipse)
is the planet's Copernican orbit, and the planet is shown in this orbit
at P, where it is moving with a constant speed. If there were no force
between the sun, S, and the planet, then the planet should move straight
out along the broken line tangent to the orbit,- always at the same
speed. But if, when the planet is at P, it is suddenly pushed sharply
toward the sun, then (remember Figure 46) it will acquire a simul-
taneous motion toward the sun, indicated by the short broken radial
line in the diagram. The result of these two motions will be a new
inertial motion along the solid arrow in the diagram, rejoining the
actual orbit at P'. If the planet at P' were again pushed sharply toward
the sun, it would begin to move along the second solid arrow, toward
P", and the process could be continued until the planet finally returned
toP.
p"'
.. --.... ,
b ' .... / ............ ___ .,
(a) (b)
Figure 50. Hooke's planetary theory (a) and his pendulum demonstration (b).
In diagram (a) the planet is given a sudden push toward the sun, S, at each of the
points P, P', P", .... Each push changes the direction of the planet's inertial
motion, and the result of all the pushes is motion along the perimeter of a polygon.
Increasing the number of points at which the planet is pushed toward the center
increases the number of sides of the polygon. In the limit a continuous push
toward the center produces motion around a circle. This circular motion is demon-
strated physically with the apparatus sketched in diagram (b). Unless given a
lateral push the pendulum bob, b, will simply be drawn by its weight toward a
point close to the center of the dotted circle. But when the bob is given an initial
horf'zontal push in a direction perpendicular to the string, the weight of the bob
can only deflect the motion into a curve. If the push is of the right strength, the
bob will revolve in a horizontal circle or in an elongated orbit that closely ap-
proximates an ellipse.
The succession of pushes described above does not move the planet
around the smooth curve representing its orbit. Instead the planet
moves around a polygon. But the polygonal series of solid arrows
approximates the planet's orbit, and the approximation can be im-
proved indefinitely. Suppose, for example, that the magnitudes of
the impulses delivered at P, P', P'', ... are reduced so that the planet
is deflected less at each of these points and therefore rejoins its curved
orbit sooner; and suppose that the original series of impulses (now
reduced in strength) are supplemented by a new series delivered at
the points between P and P', F" and P", ... where the planet now re-
joins the curve. The resulting motion will still describe a polygon rather
than an ellipse or circle, but the polygon now approximates the circle
more closely. As the individual impulses are further decreased in
strength and increased in number, the approximation becomes still
closer. Finally, when the individual impulses become infinitely small
and infinitely numerous, the planet is deflected toward the sun at
each point in its trajectory, and if the deflecting force is always of the
proper strength, the resulting curve will be just the desired ellipse or
circle.
This was Hooke's hypothesis, and it remained quite vague. Hooke
did not know how to relate the magnitude of a force to the size of
the deflection it would produce, and he did not know how to generate
an ellipse from a continuous series of deflections. He did not and
could not show that his hypothesis would work. That job was left for
Newton. But Hooke was able to give his idea a concrete and plausible
form by setting up a model that produced motions like the planetary
motions under the influence of a single centrally directed force. In
1666 he concluded the lecture whose content we have just sketched
by showing his colleagues at the Royal Society a so-called conical
pendulum (Figure 50b), constructed by suspending a heavy bob from
a wire free to move in any direction. When the bob was pulled slightly
aside from its lowest position, the only effective force exerted on it was
a pull directed approximately toward the low point of the pendulum,
one wire length under the point of suspension. Released from a posi-
tion of rest away from this low point, the bob was simply pulled back,
and it oscillated steadily to and fro in a plane like an ordinary pen-
dulum. But when the bob, after being displaced, was given a sharp
horizontal push in a direction perpendicular to the line connecting it
with the low point, then it did not necessarily return to its lowest
point at all. Instead it could swing steadily in a horizontal plane about
the low point, executing a continuous orbit like that of a planet. When
started in the appropriate direction with the appropriate speed, the
bob revolved in a horizontal circle. With a slightly different initial
speed it moved in an elongated orbit, quite like an ellipse. The cen-
trally directed force was unable to pull the moving bob to the center.
It only deflected the motion toward the center, thus producing a
continuous curve. A single centrally directed force had produced in
the laboratory a closed orbit of the proper shape. A similar force in the
heavens should, said Hooke, have the same effect.
Hooke's model made his vaguely enunciated theory both clear and
plausible, but it has a larger importance as well. To us the model
provides a forceful illustration of the immense and fruitful change that
the physical problem of the planets had undergone through the in-
fluence first of Copernicanism and then of corpuscularism and Coperni-
canism combined. In Hooke's work, even more than in Kepler's and
Borelli's, the explanation of planetary motion has become a problem
in applied mechanics, identical in principle with the terrestrial prob-
lems of the pendulum and the projectile. Terrestrial experiments yield
direct knowledge of the heavens, and celestial observations give in-
formation immediately applicable on the earth. The breakdown of
the terrestrial-celestial dichotomy, demanded by the De Revolutionibus
and facilitated by the corpuscular philosophy, is at last complete. Crys-
tal spheres and all other special celestial devices have been banished
and replaced by a mechanism of terrestrial type, and that mechanism
has been shown to function as adequately as Aristotle's spheres.
Gravity and the Corpuscular Universe
One other pressing problem raised by Copernicus' innova-
tion played an essential role in the evolution of the new universe: Why
do heavy bodies fall to the surface of a moving earth whatever the
earth's position in space? Though philosophers have maintained that
scie!ltists should not ask questions of this sort - questions beginning
with "Why"- they were certainly asked, and asked fruitfully, during
the seventeenth century. Descartes, for example, provided one answer
to this particular question -loose bodies are driven to earth by the im-
pact of aerial corpuscles in the earth-centered vortex - and that answer
was widely believed until after Newton's death. A competing solution
had, however, been developed by earlier Copernicans: heavy bodies are
drawn to earth by an intrinsic attractive principle which acts between
all pieces of matter. Once modified to fit at least a few of corpuscu-
larism's major premises, the answer based on an intrinsic principle
of attraction triumphed over the purely corpuscular explanation of
Descartes and his followers. By the end of the century that attractive
I
THE NEW UNIVERSE
253
principle, known today as gravity, had provided the key to most mo-
tions on the earth and all those in the heavens.
Like most conceptions current in seventeenth-century science, grav-
ity has precursors reaching back to antiquity. Some of Plato's prede-
cessors, for example, had believed that similar substances must attract
or, alternatively, repel each other. But, except in the study of mag-
netism and electricity, these innate attractive and repulsive principles
had had few concrete applications until the concept of a planetary
earth called them forth. The obscure connection between these two
apparently disparate concepts, gravity and the planetary earth, appears
clearly and early in a passage we have already noted from Oresme's
commentary on Aristotle's On the Heavens (p. 114). There could,
said Oresme, be many earths in space, but in that case stones must
fall to earth because matter aggregates naturally to matter rather than
because it aggregates to the geometric center of the universe.
In the De Revolutionibus, Book One, a similar need called forth a
conception very like Oresme's. "Now it seems to me," said Copernicus,
"gravity [which here means simply weight] is but a natural inclina-
tion, bestowed on the parts of bodies by the Creator so as to combine
the parts in the form of a sphere" (see p. 153 above). Kepler, too,
elaborated the idea of an attractive principle acting between the earth
and its parts. He even suggested that the same principle might act
reciprocally between the earth and the moon. Only in considering
bodies outside of the earth-moon system did Kepler feel the need of
special celestial forces like the anima matrix. Until Descartes's cor-
puscular explanation of weight was published in 1644, most Coperni-
cans contir.ued to explain the descent of stones by some device like
Kepler's. Either there was an intrinsic attractive principle, like mag-
netism, by which the earth attracted stones and stones attracted the
earth, or else (the equivalent only for present purposes) stones pos-
sessed an intrinsic tendency to move toward the physical center of
the earth.
After the middle of the century these Copernican explanations of
the fall of a stone were rapidly applied to the new Copernican prob-
lem raised by the assimilation of the concept of inertial motion. First
Descartes and then Borelli, Hooke, Huyghens, and Newton, all recog-
nized that to pursue a closed orbit about the sun a planet must con-
tinually "fall" toward the sun, thus transforming its linear inertial
motion into a curve. As the need for an explanation of this "fall" was
recognized, each Copernican adapted some variant of his explanation
for terrestrial fall to the celestial case. Descartes's planets were pushed
toward the sun by corpuscular impact; Borelli's possessed a natural
tendency to move toward the sun; and Hooke's were drawn to the sun
by an intrinsic mutual attraction. This much we have already seen.
Hooke, however, and Newton at about the same time, made one
other immensely consequential step. Led, perhaps, by Descartes's idea
that the same mechanism governed terrestrial and celestial fall, they
suggested that the force which drew planets to the sun and the moon
to earth was the same gravitational attraction which caused the fall of
stones and apples. We shall probably never know which of the two
reached the conception first. But Hooke was at least the first to an-
nounce it publicly, and his statement of 1674 is still worth reading as a
clear description of the vision which, made quantitative and corpus-
cular by Newton, guided the scientific imagination of the eighteenth
and nineteenth centuries. Hooke wrote:
[At a future date] I shall explain a System of the World differing in many
particulars from any yet known, [and] answering in all things to the com-
mon rules of mechanical motions. This depends upon three suppositions:
first, that all celestial bodies whatsoever have an attraction or gravitating
power towards their own centers, whereby they attract not only their own
parts, and keep them from flying from them, as we may observe the earth
to do, but that they do also attract all the other celestial bodies that are
within the sphere of their activity; and consequently that not only the sun
and moon have an influence upon the body and motion of the earth, and
the earth upon them, but that Mercury, also Venus, Mars, Jupiter, and
Saturn, by their attractive powers, have a considerable influence upon its
motiOn as in the same manner the corresponding attractive power of the
earth hath a considerable influence upon every one of their motions also.
The second supposition is this: that all bodies whatsoever that are put into
a direct and simple motion, will so continue to move forward in a straight
line, till they are by some other effectual powers deflected and bent into
a motion, describing a circle, ellipse, or some other more compounded
curve line. The third supposition is: that these attractive powers are so
much the more powerful in operating, by how much the nearer the body
wrought upon is to their own centers. Now what these several degrees are
I have not yet experimentally verified; but it is a notion, which if fully
prosecuted as it ought to be will mightily assist the astronomer to reduce
all the celestial motions to a certain rule, which I doubt will never be done
true without it.l
Hooke's first two "suppositions" are fundamental premises of the
new universe. Inertia plus a single attractive force, gravity, governs
both celestial motions and the motions of terrestrial projectiles. By
implication, at least, planets and satellites are simply terrestrial pro-
jectiles, cannon balls fired with a muzzle velocity so great that they
never quite fall to the earth's surface but instead circle it continually.
Newton himself made this image explicit and familiar in his System of
the World (Figure 51). But Hooke's remarks supply something more
Figure 51. Newton's description of the cannon ball as a satellite. As the
velocity of the ball is increased the length of its trajectory also increases, so that
it travels farther around the curved surface of the earth. When the velocity is
great enough, the ball does not fall to earth at all, but instead revolves continually
in an approximately circular orbit.
than conceptual foundations. The passage just quoted also makes ex-
plicit two of the main problems that remained to be solved before the
new universe would be complete. How does the gravitational force
vary with the distance between the attracting bodies, and how can
a knowledge of this law of attraction be used to predict both terres-
trial and celestial motions?
With these problems Hooke himself could do nothing. He was not
enough of a mathematician to deduce the law of attraction from
Kepler's description of the planetary orbits; the instruments that he
carried to the top of St. Paul's cathedral and to the bottoms of mines
were far too insensitive to detect the small variation of gravity near the
earth's surface. But Hooke was not the only scientist working in the
field. Though neither he nor his other contemporaries knew it, Isaac
Newton ( 1642-1727) had already arrived independently at an impor-
tant part of Hooke's qualitative conception. Furthermore, if his own
subsequent dating of the discovery is reliable, Newton had used the
conceptions to determine Hooke's "several degrees" of gravitational
attraction eight years before the passage above was written.
When Newton turned to the problem around 1666, he succeeded in
working out mathematically the rate at which a planet must "fall"
toward the sun, or the moon toward the earth, in order to remain in a
stable circular orbit. Then, having discovered how this mathematical
rate of fall varied with the planet's speed and with the radius of its
circular orbit, Newton was able to deduce two immensely important
physical consequences. lf the speeds of the planets and their orbital
radii were related to each other by Kepler's Third Law, then the attrac-
tion that drew planets to the sun must, Newton found, decrease in-
versely as the square of the distance separating them from the sun. A
planet twice as far from the sun would require only one-fourth the
attractive force to remain in its circular orbit at its observed speed.
Newton's second discovery was equally far-reaching. The same inverse
square law that governed the attraction between sun and planets
would, he found, account quite well for the difference in the rate at
which the distant moon and a nearby stone fell to the earth. Thirteen
years later, recalled to the problem by a controversy with Hooke, he
generalized his results still further and showed that an inverse-square
law would account precisely for both the elliptical orbits specified by
Kepler's First Law and the speed variation described in the Second.
These mathematical derivations were without precedent in the
history of science. They transcend all the other achievements that stem
from the new perspective introduced by Copernicanism. Our inability
to pursue them in an elementary treatise produces the single worst
distortion in this foreshortened epilogue to the Copernican Revolution.
From Newton's inverse-square law and the mathematical techniques
that related it to motion, both the shape and the speed of celestial and
terrestrial trajectories could be computed for the first time with
immense precision. The resemblance of cannon ball, earth, moon, and
planet was now seen, not in a vision, but in numbers and measurement.
With this achievement seventeenth-century science reached its climax.
Yet strangely enough that climax was not quite the end of the Coper-
nican Revolution. Despite its scope and power, neither Newton nor
many of his contemporaries were satisfied with the concept of gravity
-
and its operation. By 1670 the corpuscular philosophy provided the
metaphysical background for almost all progressive research, and the
concept of gravity violated corpuscular premises in two significant
respects. Another half-century of research and argumentation were
required for the reconciliation. In the new universe that finally emerged
both corpuscularism and Newton's concept of gravity had been altered
once more.
Newton, whose continuing allegiance to corpuscularism is repeat-
edly attested by his letters and college notebooks, was himself intensely
aware of the metaphysical inadequacy of his working concept of
gravity. That awareness probably accounts for at least part of his
delay in announcing the results of his early work in celestial physics.
In fact, the Principia did not appear until Newton, in 1685, succeeded
in resolving one of the apparent conflicts between gravity and the
corpuscular philosophy and until he had expended much fruitless effort
in attempting to resolve the other.
The first conflict between corpuscular premises and Newton's early
theory of gravity appears in the calculation of 1666, which compared
the earth's attraction for the distant moon and a nearby stone. By
comparing the stone's rate of fall with the moon's, Newton concluded
that the earth's attraction for a unit mass outside its surface varied
inversely as the square of the distance between the exterior mass and
the earth's center. The conception was simple and in adequate agree-
ment with experiment. Furthermore it applied brilliantly to the entire
solar system. But it was not corpuscular. According to a corpuscularian
the earth's attraction for an external corpuscle can be discovered only
by adding together the attractions that each individual corpuscle within
the earth exerts upon the single exterior corpuscle (Figure 52). If
the exterior corpuscle is very distant from the earth, the addition is
easy. In that case the exterior corpuscle is approximately equidistant
from each corpuscle in the earth; each earth corpuscle, wherever
located, exerts approximately the same force on the exterior corpuscle;
and the total force must be very nearly the same as it would be if all
the earth corpuscles were moved slightly to their mean position and
concentrated together at the earth's center. Therefore, if the attraction
of individual particles is governed by the inverse-square law, the
attraction of gross bodies at large distances must be governed by the
same law.
The addition of microscopic forces is not, however, nearly so easy
when the exterior corpuscle is close to the earth's crust. In this case the
gross inverse-square law seems most unlikely to apply. When close to
the earth (Figure 52), the external corpuscle is millions of times nearer
to the nearby earth corpuscles than to those on the opposite side of
the earth. The nearby corpuscles therefore exert an immensely stronger
force than the more distant ones. Apparently they will exert almost all
the force, and the total attraction will increase extremely rapidly as
the external corpuscle draws close to the earth's surface. The distance
R
Figure 52. Making the inverse-square law corpuscular. If gravity is a cor-
puscular attraction, then the total attraction of the earth for an exterior corpuscle
must be just the sum of the attractions between each earth corpuscle and the
external corpuscle. It is by no means clear that this total attraction varies in any
simple way with distance. Newton, however, succeeded in showing that if the
attraction between individual corpuscles varied inversely as the square of the
distance between them, then the total attraction between the earth and an exterior
corpuscle would vary inversely as the square of the distance from the earth's center
to the exterior corpuscle.
to the earth's center seems almost irrelevant in calculating the total
force exerted on, say, an apple. Newton was able to show that it is not
irrelevant. In 1685 he proved that, whatever the distance to the external
corpuscle, all the earth corpuscles could be treated as though they were
located at the earth's center. That surprising discovery, which at last
roofed gravity in the individual corpuscles, was the prelude and per-
haps the prerequisite to the publication of the Principia. At last it could
be shown that both Kepler's Law and the motion of a projectile could
be explained as the result of an innate attraction between the funda-
mental corpuscles of which the world machine was constructed.
Yet even this corpuscular conception of gravity did not satisfy
Newton. Indeed, until the eighteenth century, it satisfied very few
scientists. To most seventeenth-century corpuscularians gravity, as an
innate attractive principle, seemed far too much like the Aristotelian
"tendencies to motion" which they were unanimous in rejecting. The
great virtue of Descartes's system had been its complete elimination
I
-
of all such "occult qualities." Descartes's corpuscles had been totally
neutral; weight itself had been explained as the result of impact; the
conception of a built-in attractive principle operating at a distance
therefore seemed a regression to the mystic "sympathies" and "poten-
cies'' for which medieval science had been so ridiculed. Newton himself
entirely agreed. He repeatedly attempted to discover a mechanical ex-
planation of the attraction, and though forced at last to admit his
failure, he continued to maintain that someone else would succeed, that
the cause of gravity was not "uncapable of being discovered and made
manifest."
2
Again and again he insisted that gravity was not innate in
matter. "To tell us," he proclaimed in his scientific testament at the end
of his Opticks, "that every Species of Things is endow'd with an occult
specifick Quality [like Gravity] by which it acts and produces manifest
Effects, is to tell us nothing."
3
It does not, I think, misrepresent Newton's intentions as a scientist
to maintain that he wished to write a Principles of Philosophy, like
Descartes, but that his inability to explain gravity forced him to restrict
his subject to the Mathematical Principles of Natural Philosophy.
Both the similarity and the difference of titles are significant. Newton
seems to have considered his magnum opus, the Principia, incomplete.
It contained only a mathematical description of gravity. Unlike Des-
cartes's Principles it did not even pretend to explain why the universe
runs as it does. It did not, that is, explain gravity, or so Newton thought.
But though twentieth-century science has justified Newton's misgivings
- gravity can be explained without recourse to an innate attractive
principle that acts at a distance- few of Newton's contemporaries and
successors were willing to preserve his subtle distinctions. Either they
rejected the whole idea of gravity as reversion to Aristotelianism, or
they accepted the concept and insisted that Newton had shown gravity
to be an intrinsic property of matter.
The resulting battle was anything but trivial. It was forty years
before Newtonian physics firmly supplanted Cartesian physics, even
in British universities. A few of the ablest physicists of the eighteenth
century continued to seek a mechanical-corpuscular explanation of
gravity. But no such explanation was found; meanwhile the power of
the Principia made it indispensable to scientists. Gravity was, there-
fore, gradually accepted, and despite Newton's disclaimers it became
an intrinsic property of the ultimate corpuscles of matter.
As a result the corpuscular philosophy was remade and the search
for forces began. Near the beginning of the Principia Newton had said,
I am induced by many reasons to suspect that . . . [the phenomena of
Nature] may all depend upon certain forces by which the particles of bodies,
by some causes hitherto unknown, are either mutually impelled towards
one another, and cohere in regular figures, or are repelled and recede from
one another.
4
And toward the end of his Opticks he added to a long series of
"Queries" about the results of corpuscular action:
All these things being consider'd, it seems probable to me, that God in
the Beginning form'd Matter in solid, massy, hard, impenetrable, moveable
Particles, of such Sizes and Figures, and with such other Properties, and
in such Proportion to Space, as most conduced to the End for which he
form' d them. . . . And therefore, that Nature may be lasting, the Changes
of corporeal Things are to be placed only in the various Separations and
new Associations and Motions of these permanent Particles. . . . It seems
to me farther, that these Particles have not only a Vis inertiae [inertial
Force], accompanied with such passive Laws of Motion as naturally result
from that Force, but also that they are moved by certain active Principles,
such as is that of Gravity, and that which causes [chemical] Fermentation,
and the Cohesion of Bodies.
11
These statements and others like them describe the N ewtonianism
that played so large a role in the thought of the eighteenth and nine-
teenth centuries. After Newton's death in 1727 most scientists and
educated laymen conceived the universe to be an infinite neutral
space inhabited by an infinite number of corpuscles whose motions
were governed by a few passive laws like inertia and by a few active
like gravity. From these premises Newton had deduced with
unprecedented precision most of the known phenomena of optics and
all the known phenomena of celestial and terrestrial mechanics, in-
cluding both the tides and the precession of the equinoxes. Beginning
where he had left off, his successors tried to discover the additional
force laws required to account for the remaining natural phenomena:
heat, electricity, magnetism, cohesion, and, above all, chemical combi-
nation. At last the crumbling Aristotelian universe was replaced by a
comprehensive and coherent world-view, and a new chapter in man's
developing conception of nature was begun.
l
THE NEW UNIVERSE
261
The New Fabric of Thought
The construction of Newton's corpuscular world machine
completes the conceptual revolution that Copernicus had initiated a
I century and a half earlier. Within this new universe the questions
raised by Copernicus' astronomical innovation were at last resolved,
and Copernican astronomy became for the first time physically and
cosmologically plausible. The earth's relation to the other bodies in the
universe was once again defined. Men again knew why a shot fired
into space would return to the point from which it had departed,
though they now understood that the shot must not be fired quite
vertically. Only as Copernicanism became credible, through the dis-
semination and acceptance of this new conceptual framework, did the
last significant opposition to the conception of a planetary earth disap-
pear. Newton's universe was not, however, merely a framework for
Copernicus' planetary earth. Far more important, it was a new way of
looking at nature, man, and God - a new scientific and cosmologic
perspective which, during the eighteenth and nineteenth centuries,
repeatedly enriched the sciences and reshaped both religious and politi-
cal philosophy.
The same Newtonian principles which, by providing an economical
derivation and a plausible explanation of Kepler's Laws, closed the
astronomical revolution also supplied astronomy itself with a host of
powerful new research techniques. For example, when improved quan-
titative techniques of telescopic observation showed that the planets do
not really quite obey Kepler's Laws, Newtonian physics made it pos-
sible first to explain and then to predict the planets' minor deviations
from their fundamental elliptical orbits. As Newton's derivation had
shown, Kepler's Laws should apply rigorously only if the sun exerts
the sole attractive force on each planet. But planets also attract each
other, particularly when they approach and pass, and this extra attrac-
tion draws them out of their fundamental orbits and changes their
speeds. During the eighteenth century mathematical extensions of
Newton's work enabled astronomers to predict these deviations with
great precision, and during the nineteenth century the inversion of
this predictive technique was responsible for one of astronomy's
greatest triumphs. In 1846 Leverrier in France and Adams in England
independently predicted the existence and the orbit of a previously
unsuspected planet which they believed was the cause of unexplained
irregularities in the orbit of the known planet Uranus. When the tele-
scope was turned to the heavens, the new planet, Neptune, was dis-
covered, dimly visible, within a degree of the position predicted by
Newtonian theory.
Examples of N ewtonianism' s fruitfulness in astronomy could be
multiplied almost endlessly, and astronomy was not the only science
affected. To examine but one of the possible examples from one of
many sciences, consider the impact of Newton's work upon chemical
experimentation during the eighteenth century. In spite of his explicit
intention Newton led most of his successors to believe that gravity, and
therefore weight, were intrinsic properties of matter. He thus gave
weight a new significance in science. It became, for the first time, an
unequivocal measure of quantity of matter, and as a result the balance
became a fundamental chemical instrument. The balance alone could
tell the chemist how much matter was put into a chemical reaction and
how much was evolved. Since antiquity chemists had believed that
quantity of matter was conserved during chemical reactions, but there
had been no widely accredited measure of "quantity of matter." In the
climate of Aristotelian or even Cartesian thought weight was usually
regarded, like color, texture, or hardness, as a secondary characteristic
of matter - one that might be altered by the process of chemical reac-
tion. The concept of weight as a universally accredited tool for "bal-
ancing" chemical reactions and for determining whether matter is lost
to or gained from an unsuspected source during such reaction was
therefore a partial product of Newtonianism. That new tool was one of
the aeveral important roots of the revolution in chemical thought that
centered about the Frenchman Lavoisier during the last decades of
the eighteenth century.
A whole book would be required to transform and multiply these
two isolated examples- the discovery of Neptune and the new sig-
nificance of weight - into a balanced discussion of the effects of the
new universe upon science, and the discussion would still be incom-
plete. In the conceptual fabric that grew up around the new universe
nonscientific thought was transformed as well. In the infinite and multi-
populated universe conceived by seventeenth-century scientists and
philosophers the localization of heaven in the skies and of hell beneath
the earth's crust became mere metaphors, dying echoes of a symbolism
that had once had concrete geographic significance. Simultaneously
the conception of a universe constructed of atoms which move forever
in accordance with a few God-given laws changed many men's image
of the Deity Himself. In the clockwork universe God frequently ap-
,peared to be only the clockmaker, the Being who had shaped the
atomic parts, established the laws of their motion, set them to work,
and then left them to run themselves. Deism, an elaborated version
of this view, was an important ingredient in late seventeenth- and
eighteenth-century thought. As it advanced, the belief in miracles de-
clined, for miracles were a suspension of mechanical law and a direct
intervention by God and his angels in terrestrial affairs. By the end of
the eighteenth century an increasing number of men, scientists and
nonscientists alike, saw no need to posit the existence of God.
Other reflections of the new science can be discovered in the politi-
cal philosophy of the eighteenth and nineteenth centuries. Several re-
cent writers have pointed to the significant parallels between the seven-
teenth-century conception of a mechanically functioning solar system
and the eighteenth-century conception of a smoothly running society.
The system of checks and balances incorporated in the Constitution of
the United States, for example, was intended to give the new American
society the same sort of stability in the presence of disruptive forces
that the exact compensation of inertial forces and gravitational attrac-
tion had given to the Newtonian solar system. Also, the eighteenth
century's determination to derive the characteristics of a good society
from the innate characteristics of the individual man may well have
been fostered in part by the corpuscularism of the seventeenth century.
In eighteenth- and nineteenth-century thought the individual appears
again and again as the atom from which the mechanism, society, is
fabricated. In the opening paragraphs of the Declaration of Independ-
ence, Jefferson derived the right to revolution from the God-given or
inalienable rights of the social atom, man, and his derivation seems to
parallel the one in which Newton, a century earlier, had derived the
mechanism of nature from the God-given or innate properties of the
individual physical atom.
Even these few disparate and undeveloped examples indicate that
with the creation of the Newtonian universe our story has come full
circle. What the Aristotelian universe had done for earth-centered
astronomy the Newtonian universe was to do for Copernican astron-
omy. Each was a world view that tied astronomy to other sciences and
related it to nonscientific thought as well; each was a conceptual tool,
a way of organizing knowledge, evaluating it, and gaining more; and
each dominated the science and philosophy of an age. Having
traversed this circle from world view to world view, we may at last
realize the sense in which it turns upon Copernicus' astronomical inno-
vation. The conception of a planetary earth was the first successful
break with a constitutive element of the ancient world view. Though
intended solely as an astronomical reform, it had destructive conse-
quences which could be resolved only within a new fabric of thought.
Copernicus himself did not supply that fabric; his own conception of
the universe was closer to Aristotle's than to Newton's. But the new
problems and suggestions that derived from his innovation are the
most prominent landmarks in the development of the new universe
which that innovation had itself called forth. The creation of the need
and the aid supplied in its fulfillment are the contributions to history
that constitute the Copernican Revolution.
Its historical contributions do not, however, exhaust the Revolu-
tion's significance. Because it illustrates the continuing cyclic process
by which knowledge grows, the Copernican Revolution has a larger
importance as well. The last two and one-half centuries have proved
that the conception of the universe which emerged from the Revolu-
tion was a far more powerful intellectual tool than the universe of
Aristotle and Ptolemy. The scientific cosmology evolved by seven-
teenth-century scientists and the concepts of space, force, and matter
that underlay it, accounted for both celestial and terrestrial motions
\Yith a precision undreamed of in antiquity. In addition, they guided
many novel and immensely fruitful research programs, disclosing a
host of previously unsuspected natural phenomena and revealing order
in fields of experience that had been intractable to men governed by
the ancient world view. These are permanent achievements. As long
as the continuous tradition of Western learning survives, scientists will
be able to explain the phenomena first elucidated by Newtonian con-
cepts, just as Newton was able to explain the more restricted list of
phenomena previously elucidated by Aristotle and Ptolemy. That is
how science advances: each new conceptual scheme embraces the
phenomena explained by its predecessors and adds to them.
But though the achievements of Copernicus and Newton are per-
r
THE NEW UNIVERSE
265
manent, the concepts that made those achievements possible are not.
Only the list of explicable phenomena grows; there is no similar cumu-
lative process for the explanations themselves. As science progresses,
its concepts are repeatedly destroyed and replaced, and today New-
~ o n i n concepts seem no exception. Like Aristotelianism before it,
Newtonianism at last evolved- this time within physics- problems
and research techniques which could not be reconciled with the
world view that produced them. For half a century we have been in
the midst of the resulting conceptual revolution, a revolution that is
once again changing the scientist's (though not yet the layman's) con-
ception of space, matter, force, and the structure of the universe.
Because they provide an economical summary of a vast quantity of in-
formation, Newtonian concepts are still in use. But increasingly they
are employed for their economy alone, just as the ancient two-sphere
universe is employed by the modem navigator and surveyor. They are
still a useful aid to memory, but they are ceasing to provide a trust-
worthy guide to the unknown.
Therefore, though more powerful than its predecessors, the New-
tonian universe is not proving more final. Nor is its history, considered
as one of many chapters in the development of human thought, very
different in structure from that of the earth-centered universe which
Copernicus and Newton destroyed. This book is one long chapter in a
continued and continuing story.
TECHNICAL APPENDIX
1. Correcting Solar Time
In the central chapters of this book we have assumed that if the
apparent solar day is defined as the time interval between one local noon
and the next, then the time required by the stars to complete their diurnal
circles is invariably just 4 minutes (more accurately 3 minutes 56 seconds)
shorter than this solar day. But as noted in a footnote to Chapter 1, this is
not quite the case. If the intervals between successive local noons are per-
fectly regular, then the stars must move at an irregular rate. Conversely, if
the stars complete successive diurnal circles in equal intervals of time, then
the lengths of successive solar days must vary. This fact was recognized in
antiquity, at least by the time of Ptolemy and very probably before. To
understand it let us assume, as the ancients did, that the apparent motion
of the stars is perfectly regular, so that the stars provide a fundamental time
scale. We shall then discover two distinct reasons why the intervals between
the instants when the sun achieves its maximum daily elevation must vary.
The first cause of the irregularity of apparent solar time is the variation
in the rate at which the sun seems to move through the zodiacal constella-
tions. As we discovered in Chapter 2, the sun moves more rapidly along
the ecliptic from autumnal to vernal equinox than from vernal to autumnal.
In its daily race with the stars the sun therefore seems to lose ground more
in winter than in summer, so that if time is measured by the stars,
the sun must take longer to regain maximum elevation during the winter
than it requires in summertime. It follows that the apparent solar day should
be longest in midwinter and shortest in midsummer, and this would be the
case if another cause of irregularity did not intervene.
The second source of the apparent solar day's variability is the angle at
which the ecliptic intersects the equator on the celestial sphere. To under-
stand its effect look again at Figure 13. Chapter 1, and imagine that equally
spaced lines of celestial longitude are drawn on the sphere, just as lines of
longitude are drawn on any terrestrial globe. For the sake of simplicity,
assume in addition that the sun's apparent motion along the ecliptic is per-
fectly regular and at the rate of 1 o along the great circle per day. It then
turns out that, because the ecliptic is tilted with respect to the equator, the
net eastward motion of the sun varies from one day to the next. When the
TECHNICAL APPENDIX 267
sun is at or near one of the solstices, its apparent motion with respect to
the stars is very nearly parallel to the celestial equator. In addition, it is
moving on a part of the sphere where the lines of longitude are somewhat
closer together than they are at the equator. As a result the net eastward
motion of the sun is somewhat more than 1 o of celestial longitude per day,
ald the celestial sphere must therefore turn westward through slightly
more than 361 o in order to carry the sun from maximum elevation to
maximum elevation. At the equinoxes the situation is quite different. There
the lines of celestial longitude have their maximum spacing on the sphere.
Furthermore, the sun's constant total motion is to the northeast or southeast
rather than due east, and it therefore does not move eastward as much as
1 a day. As a result the celestial sphere need not rotate through quite 361
to return the sun to maximum elevation. This effect, considered alone, makes
the apparent solar day longest at the solstices and shortest at the equinoxes.
In order to correct for these two irregularities modern civilizations have
adopted a time scale known as mean solar time, whose fundamental time
unit is the average length of the apparent solar day. On this time scale the
stars do, by definition, move perfectly regularly, completing their diurnal
circles in just 23 hours 56 minutes 4.091 seconds. But the scale that makes
the stars regular makes the sun irregular. For example, the sun's maximum
elevation rarely occurs at local noon, mean solar time. The time kept by
sundials, the only instruments that directly measure apparent solar time,
does not pass at the same rate as the time kept by our watches or announced
by time signals on the radio. During December and January, when both
the effects discussed above act to shorten the apparent solar day, the interval
between successive maximum elevations of the sun is very nearly 0.5 minute
less than the mean solar day. Furthermore, the effect of this small discrep-
ancy is cumulative - apparent time runs slower than mean time for many
days in a row - so that at one season of the year the sun reaches maximum
elevation (apparent noon) almost 20 minutes before mean solar noon. At
other seasons apparent time runs fa'Oter than mean time, and over the years
the two stay together. But they are rarely together during any one day. In
order to keep accurate time by the sun it is therefore necessary to correct
the sundial by using a table, or a diagram like the one shown in Figure 53.
IS JAN
+I
+
10 /
lSI/
..
.. +
,_
=>
z
:1!
0
IS
I
liS
FEB MAR APR
h
\
"
-
JUN JUL AUG !EPIOCTIIOI DEC
I
-
SUN SLOW
/ '\ I
r\ I I
\ 17
1\IJ
""'""
53. A graph of the equation of time, indicating the annual variation of
the difference between mean and apparent solar time.
268
The preceding discussion of time employs the apparent motion of the
stars as its standard of regularity. Clearly this choice of standard is arbitrary,
at least from a logical viewpoint. Logically we might equally well have
chosen the sun's apparent motion as our standard of regularity and shown
that on the corresponding time scale the stars move at a continually varying
rate. But the choice of the sun as a standard of regularity would be im-
mensely inconvenient to both science and civil life. The diagram of Figure
53 would have to be applied to clocks and watches rather than to sundials.
Astronomers and physicists would be forced to describe the earth's axial
rotation as occurring at a constantly varying rate. The stellar standard avoids
these awkwardnesses. It is well to all civil and most scientific
functions.
Yet it has not turned out to be quite adequate for science, at least not
for scientific theory; the time scale implicit in Newton's Laws of Motion does
not quite correspond to the stellar standard. From Newton's Laws, as they
are now understood, it is possible to show that the earth's axial rotation is
being gradually slowed by effects like tidal friction and that, as a result, the
apparent stellar motions are very gradually slowing down. Either the Laws
or the stellar standard must therefore be adjusted, and considerations of
scientific convenience suggest the search for a new standard. To date the
theoretical of the standard is without practical signifi-
cance. But It has an Immense Importance to science, and it has therefore
led to a renewed search, which continues actively today, for a
clock that will conform to the time scale of scientific theory more accurately
than the celestial machine itself.
2. Precession of the Equinoxes
A second technical sin1plification introduced in the body of this
book was the neglect of the precession of the equinoxes. This is the effect
mentioned briefly in Chapter 1, that results in a slow motion of the celestiai
pole thr?ugh the stars: If we been concerned only with naked-eye
obi_ervahons made durmg the lifetime of a single astronomer, our simplifi-
cation would have been entirely appropriate - naked-eye observations can-
_disclose its unless they are made at widely separated points
m But observatiOns made, for example, two centuries apart show that,
while the stars themselves retain constant relative positions, the celestial
pole _about they move gradually shifts its position among them at a
JUSt over 0.5 per century. Observations repeated over far longer periods
disclose the pattern of this precessional motion; as the centuries pass the
pole of the moves gradually through the stars in a circle, completing
26,00_0 years. The center of this circle is the pole of the
-:- the pomt at wh1eh a diameter perpendicular to the plane of the
mtersects the celestial_ - and the radius of the circle is just
, . the same as the angle m which the celestial equator intersects the
ecl1pt1c on the sphere of the stars (Figure 54a).
l
TECHNICAL APPENDIX
269
The precessional motion seems to have been noticed first by the Hellen-
istic astronomer Hipparchus during the second century B.C., and, though not
widely known at first, it was discussed by a number of subsequent astrono-
mers, including Ptolemy. Most of Ptolemy's Moslem successors described
form of the effect, and by adding a ninth sphere to the ancient system
they succeeded in explaining it physically. Their most popular explanation
is indicated diagramatically in Figure 54b, which shows the three outermost
spheres of the set; N and S are the north and south celestial poles, and the
exterior sphere rotates westward about them, once every 23 hours 56
minutes, as the sphere of the stars had rotated in the older system. The next
(a)
PRECESSIONAL
PATH OF
CELESTIAL POLE
STELLAR
SPHERE
P<>LE OF
ECLIPTIC ON
STELLAR
SPHERE
s
(b)
Figure 54. The precession of the equinoxes. Diagram (a) shows the circle on
the celestial sphere around which the celestial pole moves once in every 26,000
years. The center of the circle is the pole of the ecliptic, and all points on the circle
are just 23% from this center. Diagram (b) shows how the Moslems explained
precession with the aid of a ninth sphere, the outer sphere in the drawing. This
ninth sphere rotates once every 23 hours 56 minutes, as the sphere of the stars had
rotated in the older eight-sphere systems. The eighth sphere, on which the stars are
set, rotates about its own poles once in 26,000 years, thus slowly changing the
position of the celestial pole among the stars. Inside the eighth sphere is the
sphere of Saturn, which encloses the remaining planetary spheres as in the older
systems.
sphere, the middle one in the diagram, is the sphere that carries the stars,
and it is joined to the outermost sphere by an axis which passes through the
poles of the ecliptic on the sphere of the stars and through two points

from the poles on the outer sphere. This new sphere of the stars is whirled
around daily by the outermost sphere (this accounts for the diurnal stellar
circles). In addition it has a slow motion of its own, one rotation every
26,000 years, which gradually changes the relations between the individual
stars and the celestial poles. The innermost of the three spheres is Saturn's,
and it is drawn as a thick shell to allow space for the epicyclic components
of Saturn's motion. By itself this thick sphere, connected to the sphere of
the stars by an axis through the poles of the ecliptic, accounts for Saturn's
average circular motion through the stars.
In the context of ancient and medieval astronomical thought this ninth-
sphere explanation of precession seems both simple and natural. In fact, it
compares relatively well with the Copernican explanation - a gradual conical
motion of the earth's axis which, during the course of 26,000 years, is directed
successively to all the points on a circle of radius 23Ji o about the pole of the
ecliptic. Until Newton explained precession as a physical consequence of the
moon's gravitational attraction for the equatorial bulge of the earth, both
Copernican and Ptolemaic astronomers required one extra and physically
superfluous motion in order to account for it.
0
Precession has, therefore, no
logical bearing upon the transition from an earth-centered to a sun-centered
universe.
Historically, however, the problem of explaining precession had a signifi-
cant role in inaugurating the Copernican Revolution. It helped to make
Ptolemaic astronomy seem monstrous. The observational consequences of
precession are very small even when observations extend over several
centuries, and a small error in the data can therefore result in a radical
change in the description of the over-all phenomenon. Both Hipparchus and
Ptolemy described precession in a way qualitatively equivalent to the one
represented by Figure 54, but many of their contemporaries denied the
existence of the effect entirely or else described it quite differently. Particu-
larly in the Moslem world a number of divergent descriptions of precession
were prevalent. There was no agreement about its rate - in fact, many
astronomers believed that the rate varied. In addition, there was an im-
portant school which believed that even the direction of precession changed
periodically, an effect known as trepidation. Brahe's observations were re-
quired before astronomers could again recognize the true simplicity of the
phe1.1omenon. Copernicus himself did not improve the situation in the
slightest. He added extra circles to his system in order to account for the
gradual change in the precessional rate and for other nonexistent effects. But
though Copernicus did not improve the account of precession given by
ancient and medieval astronomers, he was immensely concerned to do so,
and that concern provided an important impetus to astronomical reform. In
Copernicus himself did not require an extra motion to account for precession,
because he had already introduced one in another connection. He used an annual
conical motion to keep the earth's axis parallel to itself throughout the year (Figure
3lb), and he could therefore explain precession by giving this conical motion a
period very very slightly less than a year. But Copernicus' successors, who thought
that a single orbital motion would keep the earth's axis perpetually in alignment,
did need an additional conical motion with a period of 26,000 years in order to
explain the changing position of the celestial pole.
TECHNICAL APPENDIX 271
Copernicus' day an adequate account of precession was the principal pre-
requisite for the most pressing problem of practical astronomy, the reform of
the Julian calendar.
To discover the effect of precession upon the design of calendars, return
once more to Figure 54. As the diagram shows, the position of the ecliptic
upon the sphere of the stars is fixed once and for all. But though the chang-
ing positions of the celestial poles do not affect the ecliptic, they do change
the position of the celestial equator and therefore of the equinoxes, the
points at which the ecliptic and the celestial equator intersect. During the
precessional period, 26,000 years, each of the equinoxes moves slowly and
steadily around the ecliptic at the rate of about llio per century. Therefore,
the length of time required by the sun to move once around the ecliptic
(the so-called sidereal year) is not quite the same as the length of time it
requires to move on the ecliptic from vernal equinox to vernal equinox (the
tropical year). The latter, which is more than 20 minutes shorter than the
former, is vastly more difficult to measure, because it refers the sun's motion
to an imaginary and moving point rather than to a fixed star. But the
tropical year is the year of the seasons, and it is this that must be measured
with precision before an accurate long-term calendar can be designed. Coper-
nicus' concern with the calendar therefore led him to a serious study of
precession, and thus to an intimate knowledge of that aspect of astronomy
about which Ptolemaic astronomers were in the greatest disagreement. It
is the problem of precession which underlies Copernicus' remark that "the
mathematicians . . . cannot even explain or observe the constant length of
the seasonal year" (p. 137), and it is this remark which heads his list of
motives for innovation.
3. Phases of the Moon and Eclipses
Because it is identical with the modern explanation, the ancients'
account of the cause of the moon's phases played no role in the Copernican
Revolution, and it could therefore be omitted from the earlier chapters of
this book. But the phases of the moon play a direct role in the ancient
measurements of the dimensions of the universe, and these measurements,
as we have repeatedly noted, helped make the ancient two-sphere universe
seem concrete and real to scientists and nonscientists alike. Besides, the
ancient explanation of phases, as well as the correlated explanation of
eclipses, provides an important additional illustration of the scientific ade-
quacy of the ancient world view.
The explanation with which we are concerned was well known in Greece
by the end of the fourth century B.C., though it may have originated con-
siderably earlier. With the acceptance of the two-sphere universe came the
larger and less well documented assumption that all the celestial wanderers
were spheres as well. In part this assumption derived from analogy to the
spherical shape of the earth and heavens, and in part from the conception
of the perfection of the spherical shape and its appropriateness to the perfect
(a)
V>
z
1-V>
0 wz
zw 0
-V>
c:t>
0:
1-z
...J<I: ow
a
n.w <1 ::::e>

:X:
3:
<1:
I.L:x: ...Jw
OCI ..... <t:x;
z=>
(/)
z
b
oo 41 a::I.L
_a::

=0 1-:X:
01- 0
::::e
(b)
d ) ex lanation of the moon's phases.
Figure 55. The ancient (and mo f etr;: s rhce of a sphere is illuminated by
Diagram (a) indicates that half o( b) e hu th ortion of this illuminated
h d
. t Diagram s ows e p
the rays of t e !Stan sun. f . relative positions of the sun,
bl
t estrial observer or vanous
f II hemisphere visi e to .err . .
2
is the waxing half moon; 3, u
earth, and moon. PositiOn 1 IS new moon,
moon; and 4, the waning half moon.
TECHNICAL APPENDIX
273
A westward rotation of the entire diagram, excepting the central earth,
accounts for the diurnal motion of the sun and moon, so that an observer at
a sees the sun just setting and one at b sees it rising. Only the eastward
orbital motions of the sun and moon are motions with respect to the diagram.
When the moon is at position 1 in the diagram it rises with the sun, but,
since its dark side is pointed toward the earth, it can scarcely be seen by a
terrestrial observer. This is the position of new moon. Slightly more than a
week later the moon's rapid orbital motion has carried it 90 east of the
slow-moving sun where it appears, relative to the sun, in position 2. It now
rises at noon and is near the zenith at sunset. Only half of the disk is clearly
visible from the earth, so that this is the position of first quarter. After
another week or a bit more, the moon is full and rises as the sun sets ( posi-
tion 3) . Third quarter is shown at position 4, corresponding to a moon that
rises around midnight and is near the zenith at sunrise.
The diagram used in deciphering the moon's phases can also be used in
the explanation of eclipses: as the moon moves from position 2 to position 4,
it may pass through the earth's shadow, in which case it grows dim and is
eclipsed. If the moon always appeared on the ecliptic, it would be eclipsed
each time it reached position 3, but, since it continually wanders north and
south, the full moon, earth, and sun rarely lie on a straight line. Full moon
must lie close to the ecliptic for a lunar eclipse, and this cannot happen
more than twice a year and seldom happens that often. Solar eclipses occur
whenever the moon, at position 1, casts its shadow on the earth, and this
happens relatively frequently, at least twice a year. Yet solar eclipses are
rarely seen by terrestrial observers. The moon's shadow on the earth is
extremely small, and an observer must be in the shadow to see the eclipse.
Besides, the moon rarely blocks off more than a small fraction of the sun's
disk. Therefore, an observer at any one location can seldom see even a partial
eclipse of the sun and may never see a total one. For him it will be a rare,
striking, and so.metimes terrifying phenomenon.
4. Ancient Measurements of the Universe
One of the most interesting technical applications of ancient
astronomy was its use in the determination of cosmological distances and
sizes which could not be measured directly, that is, by ordinary measuring
sticks. These distance measurements illustrate the world view's fruitfulness
with greater immediacy than most of its other applications, because the
mathematical operations upon which they depend lose all physical signifi-
cance unless certain essential elements in the conceptual scheme are true.
For example, whether the earth is a flat disk or a sphere, the stars do appear
to move in diurnal circles, and techniques that describe this apparent motion
are therefore useful whatever their conceptual basis. But only if the earth
is really a sphere can it be said to have a circumference that can be deter-
mined from the observations of the skies discussed below.
The first reference to measurements of the earth's circumference appears
in Aristotle's writings, so that such measurements were probably made
by the middle of the fourth century B.C. But we know only the results of
these earliest measurements, not the method employed; the first measure-
ment of which we have a relatively complete, though second-hand, account
is the one made by Eratosthenes, the librarian of the great manuscript col-
lection at Alexandria, during the third century B.C. Eratosthenes measured
the angle a (Figure 56) between the rays of the noon sun and a vertical
Figure 56. Eratosthenes' measurement of the earth's circumference. If S
is due south of A on the earth's surface, then the distance AS must be the same
fraction of the earth's circumference as the angle a is of 360.
gnomon located at Alexandria, A, on a day when the noon sun was directly
overhead at Syene, S, a second Egyptian city located 5000 stades due south
of Alexandria. This angle he found to be 1/50 of a full circle (or 7 1/5).
Since all the rays striking the surface of the earth from the very distant sun
may be considered parallel, the angle a, which is the sun's distance from
the zenith at Alexandria, is equal to the angle AOS subtended by S and A
at the center of the earth, 0. Furthermore, since this angle is just 1/50 of a
circle, the distance from Alexandria to Syene must be 1/50 of the circum-
ference of the earth, and the total circumference must be 50 X (distance
from Alexandria to Syene) = 50 X 5000 = 250,000 stades. Most modem
students believe that Eratosthenes' figure is approximately 5 percent lower
than the result given by modem measurement ( 24,000 miles), but unfor-
tunately it is impossible to be sure. The length of the unit "stade" used by
Eratosthenes is unknown, and the known location of Alexandria and Syene
cannot be used to define the unit, because both the "5000" and the "1/50"
used in the computation above have clearly been "rounded off" to make the
report easier to read.
A second group of measurements was made during the second century
B.c. by Aristarchus of Samos, now more famous for his anticipation of the
Copernican system. He estimated the distance to and the sizes of the sun
and moon in terms of the angle MES subtended by the centers of the sun
l
TECHNICAL APPENDIX
275
moon at the earth when the moon is exactly half full (Figure 57) .
Smce the moon can be half full only if the angle EMS subtended by the
earth and the sun at the moon is exactly a right angle, the size of MES
must completely determine the shape of the right triangle whose vertices
are the moon, the earth, and the sun. Aristarchus' measurement gave
MES = 87, which corresponded to a triangle in which ES:EM::l9:1. Ac-
MOON

Figure 57. Aristarchus' measurement of the relative distances from the earth
to the moon and the sun. When the moon is exactly half full the angle EMS
be just go. Therefore a measurement of the angle MES wlll determine the
ratio of EM to ES, that is, the ratio of the moon's distance from the earth to that
of the
cordingly, he reported that the sun was 19 times as far from the earth as the
moon that, the moon and the sun subtend the same angle at the
earth ( F1gure 58), It was also 19 times as large.
. Modern measurements, made by quite different techniques and with the
a1d of telescopes, indicate that Aristarchus' ratio was too small by a factor of
m_ore than twenty; the ratio ES:EM is very nearly 400:1, not 19:1. This
discrepancy arises from the measurement of the angle MES which should be
51', tha_n 87. In practice that measurement is very difficult, par-
ticularly With the mstruments known to have been available to Aristarchus.
centers of the sun and moon are very hard to determine: in addi-
tion It IS difficult to be sure when the moon is just half full. Given these
to have chosen the smallest angle compatible
h1s observations, presumably in order to keep the resulting
ratio credible. Similar considerations must have motivated his successors for
' ,
Figure 58. The large but distant sun and the smaller but nearby moon subtend
the same angle at the earth's surface.
though appreciably improved, estimates of the relative distances to the sun
and the moon remained too small throughout antiquity and the Middle Ages.
The preceding measurements yield only the ratios of astronomical dis-
tances, but by an immensely ingenious argument Aristarchus was able to
convert them to absolute magnitudes, that is, he was able to determine the
diameter of and distances to the sun and the moon in stades. His results
were derived from observations of a lunar eclipse of maximum duration, an
eclipse during which the moon lies squarely on the ecliptic and therefore
passes through the very center of the earth's shadow. First he measured the
time that elapsed between the instant when the edge of the moon first
entered the shadow and the instant when the moon was totally obscured
for the first time. This figure he compared with the length of time during
which the moon was totally obscured, and he thus discovered that the
period of total obscurity was approximately the same length as the period
required for the moon to enter into the earth's shadow. He concluded that the
breadth of the earth's shadow in the region where it is crossed by the moon
is very nearly twice the diameter of the moon itself.
SUN
OON
-====-----M P.u:F:.d
I
X
I R
I
19R
I
Figure 59. Aristarchus' construction for computing the absolute distances to
the moon and sun in terms of the observations made during a lunar eclipse.
Figure 59 shows the astronomical configuration which Aristarchus
analyzed. In the diagram the moon is shown immediately after it has fully
entered the shadow of the earth. The diameter of the moon is d (an un-
known) and the diameter of the earth's shadow at the moon is therefore 2d;
the diameter of the earth is D (known in stades from Eratosthenes'
measurement of the circumference of the earth) ; and the distance from the
moon to the earth is R (again an unknown to be determined). Finally, since
the sun's diameter and distance from the earth are just 19 times the moon's,
the diameter of the sun's disk is just 19d and its distance from the earth is
just 19R. So Aristarchus' problem, and ours, is to determine d and R, the
unknown distances, in terms of the earth's diameter, D, a quantity whose
value in stades has already been determined.
The diagram shows immediately three similar triangles whose bases are
of length 2d, D, and 19d, and whose altitudes are respectively x (an un-
known), x + R, and x + 20R. (Actually the bases of the three triangles are
very slightly shorter than the diameters with which they have been
equated above, but if the triangles are extremely acute, as they are, this
r
TECHNICAL APPENDIX
277
discrepancy is too small to affect the result.) The ratio of the altitude to the
base of the smallest triangle must be the same as that of the largest, or,
X X+ 20R
2d = 19d
Multiplying both sides of the equation by 38d yields a new equation:
19x = 2x + 40R, so that x = 40R/17. In other words, the earth's shadow
extends beyond the moon for a distance about times the distance from
the earth to the moon.
. Co.mparing the smalle.st of the triangles with the triangle of intermediate
Size giVes another equatiOn, from which d may be determined. The first
comparison gives:
x x+R
2d =v-
Substituting 40Rj17 for x and multiplying both sides by 17 /R yields:
20 40 + 17
-y= D
From the last equation, d = 20D/57 = 0.35D. That is, the diameter of the
moon is just greater than one-third the diameter of the earth, and since the
sun's diameter is just 19 times the moon's, the sun must have just over 6%
times the diameter of the earth.
Since D, the diameter of the earth, is known, the actual sizes of the sun
and moon are given by the computation above. Their distances can be de-
termined by a small additional computation. Because both the sun and the
moon an angle of 0.5 at the earth, each could be placed 720 times
on the Circumference of a full ( 360) circle with its center at the earth
The distance of the moon from the earth must therefore be the radius of
a circle whose circumference is 720 times the moon's diameter now known
and the sun's distance must be just 19 times as great. the
ference of a circle is 211" times its radius, the moon's distance from the earth
must be just over 40 diameters of the earth and the distance to the sun
should be approximately 764 earth diameters.
The methods employed in computations are brilliant, typifying
the very best of Greek scientists, but the numerical results, particu-
larly those concemmg the sun, are uniformly inaccurate because of the initial
error in the determination of the angular separation of the sun and the
half moon. measurements give the moon's diameter as just over
the diameter of the earth and its distance as approximately 30
diameters, neither of which is far from the values computed by
Anstarchus. But the sun's diameter is now thought to be almost 110 times
that of the earth and the distance to the sun is about 12,000 earth diam-
eters, both very much larger than Aristarchus supposed. Though various
corrections to .me.asurements were made during antiquity and
though the possibihty of significant error in the measured distance to the
sun was often recognized, all ancient and medieval estimates of this cos-
mological dimension remained vastly too small.
Because it depends only upon the relative positions of earth, moon,
and sun, Aristarchus' techniques for determining size and distance can be
applied with equal accuracy or inaccuracy in the Ptolemaic and Copernican
universes. The ancient determinations of astronomical dimensions could,
therefore, have no direct role in the Copernican Revolution. But they did
have several indirect ones, all of which helped to strengthen the Ptolemaic
system. The possibility of making astronomical measurements illustrated the
great fruitfulness of the Aristotelian-Ptolemaic universe. In addition, the
results of the measurements helped to make the ancient cosmology seem
real by increasing the concreteness with which its structure was specified.
Finally, and most important, the measurement of the distance to the moon
provided an astronomical yardstick which, during the Middle Ages, was
used to provide an indirect measure of the size of the entire universe.
As indicated early in Chapter 3, medieval cosmologists often supposed
that each crystalline shell was just thick enough to contain the epicycle of
its planet and that the shells as a group nested so that they filled all of
space. Using these hypotheses mathematical astronomers were able to de-
termine the relative sizes and thicknesses of all the shells. These relative
dimensions were then converted to earth diameters, stades, or miles, by
using Aristarchus' method of determining the distance to the moon's sphere.
A typical set of the cosmological dimensions that resulted was included in
the original discussion. It indicates the detail with which the universe was
investigated and understood by pre-Copernican scientists.
REFERENCES
Chapter 1. The Ancient Two-Sphere Universe
I. Sir Thomas L. Heath, Greek Astronomy, Library of Greek Thought (Lon-
don: Dent, 1932), pp. 5-7.
2. Benjamin Jowett, The Dialogues of Plato, 3rd ed. (London: Oxford Uni-
versity Press, 1892), III, 452--453.
Chapter 2. The Problem of the Planets
I. Vitruvius, The Ten Books on Architecture, trans. M. H. Morgan (Cam-
bridge: Harvard University Press, 1926), pp. 261-262.
2. Sir Thomas L. Heath, Aristarchus of Samos (Oxford: Clarendon Press,
1913)' p. 140.
Chapter 3. The Two-Sphere Universe in Aristotelian Thought
I. Aristotle, On the Heavens, trans. W. K. C. Guthrie, The Loeb Classical
Library (Cambridge: Harvard University Press, 1939), p. 91 (279a6-17).
2. Ibid., pp. 243-253 ( 296h8-298a 13).
3. Sir Thomas L. Heath, Greek Astronomy, Library of Greek Thought (Lon-
don: Dent, 1932), pp. 147-148.
4. Aristotle, On the Heavens, p. 345 (310b2-5).
5. Aristotle, Physics, trans. P. H. Wickstead and F. M. Comford, The Loeb
Classical Library (Cambridge: Harvard University Press, 1929), I, 331 (213a31-
34).
6. Aristotle, On the Heavens, pp. 23-25 (270b1-24).
7. Jean Piaget, The Child's Conception of Physical Causality, trans. Marjorie
Gabain (London: Kegan Paul, Trench, Trubner, 1930), pp. 110-111.
8. Heinz Werner, Comparative Psychology of Mental Development, rev. ed.
(Chicago: Follett, 1948), pp. 171-172.
9. Aristotle, Physica, trans. R. P. Hardie and R. K. Gaye, in The Works of
Aristotle, II (Oxford: Clarendon Press, 1930), 208h8-22.
Chapter 4. Recasting the Tradition: Aristotle to the Copernicans
I. St. Augustine, Works, ed. Marcus Dods (Edinburgh: Clark, 1871-17), IX,
180-181.
2. St. Thomas Aquinas, Commentaria in libros Aristotelis De caelo et mundo,
in Sancti Thomoe Aquinatis ... Opera Omnia, III (Rome: S. C. de Propaganda
Fide, 1886), p. 24. My translation.
3. St. Thomas Aquinas, The "Summa Theologica," Part I, Questions L-LXXIV,
trans. Fathers of the English Dominican Province, 2nd ed. (London: Burns Oates
& Washbourne, 1922), p. 225 (Q. 68, Art. 3).
4. Aquinas, Summa Theologica, Part Ill, Questions XXVII-LIX (London:
Washbourne, 1914), pp. 425, 433 (Q. 57, Arts. 1, 4). Quoted by pern1ission of the
publishers and of Benziger Brothers, Inc., New York.
5. Charles H. Grandgent, Discourses on Dante (Cambridge: Harvard Uni-
versity Press, 1924), p. 93.
6. Dante, The Banquet, trans. Katharine Hillard (London: Routledge and
Kegan Paul, 1889), pp. 65-66.
7. Ibid., pp. 69, 79-80.
8. Nicole Oresme, Le livre du ciel et du monde, ed. A. D. Menut and A. J.
Denomy, in Mediaeval Studies, 111-V (Toronto: Pontifical Institute of Mediaeval
Studies, 1941-43), IV, 243.
9. Ibid., IV, 272. In translating this and the following passages from Oresme's
commentary I have often been guided by the longer English selection in the
mimeographed pamphlet, "Selections in Medieval Mechanics," prepared by Marshall
Clagett of the University of Wisconsin and kindly made available to me by its
author. Professor Clagett's translations from Oresme and many other medieval
scientific writers will soon appear under the title Mechanics in the Middle Ages.
10. Ibid., p. 273.
11. Condensed with permission from Marshall Clagett's "Selections in
Medieval Mechanics" (see n. 9), pp. 35-39. The original is Jean Buridan,
Quaestiones super octo libros physicorum (Paris, 1509), Book VIII, Question 12.
I have introduced a few purely stylistic changes and dropped one set of italics.
12. Clagett, "Selections," p. 40, from Buridan, Quaestiones.
13. Mediaeval Studies, IV, 171.
14. Alfred North Whitehead, Science and the Modern World (New York:
Macmillan, 1925 ), p. 19.
15. Quoted by John Herman Randall, Jr., The Making of the Modern Mind,
2nd ed. (Boston: Houghton Miffiin, 1940), p. 213.
16. Sir Thomas L. Heath, A History of Greek Mathematics (Oxford: Claren-
don Press, 1921), I, 284.
17. Quoted by Edward W. Strong, Procedures and Metaphysics (Berkeley:
University of California Press, 1936), p. 43, from Thomas Taylor, The Philosophical
and Mathematical Commentaries of Proclus on the First Book of Euclid's Elements
(London, I [1788] and II [1789] ).
18,; Marsilio Ficino, Liber de Sole, in Marsilii Ficini Florentini, ... Opera
(Basel: Henrie Petrina, [1576]), I, 966. My translation.
19. Quoted and translated by Edwin A. Burtt, The Metaphysical Foundations
of Modern Physical Science, 2nd ed. (New York: Harcourt, Brace, 1932), p. 48,
from a fragment of one of Kepler's early disputations.
Chapter 5. Copernicus' Innovation
I. All the quotations in Chapter 5 are from the Preface and Book I of Coperni-
cus' De Revolutionibus Orbium Caelestium ( 1543). The translation was pre-
pared by John F. Dobson and Selig Brodetsky and published as Occasional Notes
of the Royal Astronomical Society, vol. 2, no. 10 (London: Royal Astronomical
Society, 1947). In reprinting the translation I have consistently replaced the word
"orbit" by either "sphere" or "circle" (see Edward Rosen, Three Coprrnican
Treatises [New York: Columbia University Press, 1939], pp. 13-16, for the diffi.
culties inherent in Copernicus' use of the Latin orbis). In perhaps a dozen other
i
REFERENCES
281
4Maces I have suppressed other examples of modem terminology or have made
similar minor changes for the sake of increased clarity. In making such changes
I have rPpeatedly been guided by the very useful Latin-French edition of Co-
pernicus' Book I prepared from the authoritative Thorn edition ( 1873) by Alex-
andre Koyre (Paris: Felix Alcan, 1934). I am indebted to the Royal Astronomical
Society for permission to reprint so large a portion of their translation.
Chapter 6. The Assimilation of Copernican Astronomy
I. Quoted by Francis R. Johnson, Astronomical Thought in Renaissance Eng-
land (Baltimore: Johns Hopkins Press, 1937), p. 207. I have modernized spell-
ing and punctuation in this and many of the other quotations in this chapter.
2. Ibid., pp. 188-189, from the translation (1605) by Joshua Sylvester.
3. Translated and quoted by Dorothy Stimson, The Gradual Acceptance of
the Copernican Theory of the Universe (New York, 1917), pp. 46-47, from
Bodin's Untversae Naturae Theatrum (Frankfort, 1597).
4. Translated and quoted by Andrew D. White, A History of the Warfare of
Science with Theology in Christendom (New York: Appleton, 1896), I, 126.
5. Ibid., pp. 126-127, from Melanchthon's Initia Doctrtnae Physicae.
6. Ibid., p. 127.
7. John Donne, "Ignatius, his Conclave," in Complete Poetry and Selected
Prose of John Donne, ed. John Hayward (Bloomsbury: Nonesuch Press, 1929),
p. 365.
8. Ibid., p. 202.
9. John Milton, Paradise Lost, Book I, line 26.
10. Nicole Oresme, Le livre du ciel et du monde, in Mediaeval Studies, IV,
276.
11. Translated and quoted by James Brodrick, The Life and Work of Blessed
Robert Francis Cardinal Bellarmine, S.]. (London: Bums Oates and Washbourne,
1928), II, 359.
Chapter 7. The New Universe
I. Robert Hooke, An Attempt to Prove the Motion of the Earth from Observa-
tions (London: John Martyn, 1674), reproduced in R. T. Gunther, Early Science
in Oxford, VIII (Oxford: privately printed, 1931), pp. 27-28.
2. Newton, Opticks, 4th (1730) ed. (New York: Dover, 1952), p. 401.
3. Ibid.
4. Newton; Mathematical Principles of Natural Philosophy, ed. Florian Cajori
(Berkeley: University of California Press, 1946), p. xviii.
5. Newton, Opticks, pp. 400-401.
BIBLIOGRAPHICAL NOTES
Introductory
The following notes provide both an indication of my most essen-
tial debts to previous studies and a convenient entrance to the vast labyrinth
of scholarly literature dealing with the history of astronomy and related
fields before 1700. Where possible I have restricted the discussion to books
readily available in English. Articles, monographs, and studies in foreign
languages are, with few exceptions, cited only if they have contributed sub-
stantially to my own account of the Copernican Revolution or if they are
omitted (as many recent studies are) from the principal bibliographical
sources below. A few minor sources, included in the list of references, are
omitted below.
Fuller bibliographies for several portions of the field will be found in
M. R. Cohen and I. E. Drabkin, A Source Book of Greek Science (New
York, 1948) ; E. J. Dijksterhuis, De M echanisering van het W ereldbeeld
(Amsterdam, 1950); F. Russo, Histoire des sciences et des techniques: bibli-
ographie (Paris, 1954); and George Sarton, A Guide to the History of Sci-
ence (Waltham, Mass., 1952) . Exhaustive bibliographies for several rele-
vant topics will be found in George Sarton, Introduction to the History of
Science, 3 vols. in 5 (Baltimore, 1927-1947) and in the annual bibliogra-
phies in the journal Isis. Many of the books cited in other connections below
also contain useful bibliographic information. The more recent works, par-
ticularly A. C. Crombie, Augustine to Galileo (Cambridge, Mass., 1952),
and A. R. Hall, The Scientific Revolution, 1500-1800 (London, 1954), will
prove especially useful.
All the general histories of science discuss the period and many of the
problems covered by this book, but only Herbert Butterfield, The Origins
of Modern Science, 1300-1800 (London, 1949), has had particular influ-
ence on the structure of this book. Marshall Clagett, Greek Science in An-
tiquity (New York, 1955) and Hall, Scientific Revolution (above), give
extremely useful background for their respective periods, though neither
was available to me until my manuscript was in substantially final form.
Dijksterhuis, M echanisering (above), is also an essential source for those
who can read Dutch.
Bertrand Russell, A History of Western Philosophy (New York, 1945),
and W. Windelband, A History of Philosophy, trans. J. H. Tufts (New
York, 1901), have provided useful background on the development of
philosophy. J. L. E. Dreyer, A History of Astroncrmy from Thales to Kepler,
2nd ed. (New York, 1953); Lynn Thorndike, A History of Magic and Ex-
perimental Science, 6 vols. (New York, 1923-1941); and Sarton, Intro-
duction (above), have been consulted so often in the composition of this
book that I cite them, below, only for those parts of it that follow their
treatments closely. Pierre Duhem, Le systeme du monde, 6 vols. (Paris,
1913-1954), which could be used in the same way, I have consulted for
special topics.
Chapters I and 2
R. H. Baker, Astronomy, 5th ed. (New York, 1950) is an ex-
cellent source for the requisite information about technical astronomy.
George Sarton, A History of Science: Ancient Science through the Golden
Age of Greece (Cambridge, Mass., 1952), surveys Egyptian, Mesopotamian,
and Hellenic astronomy within the context of ancient science and culture.
Neugebauer, The Exact Sciences in Antiquity (Princeton, 1952), pro-
vides a more detailed introduction to Egyptian and Babylonian astronomy
from their beginning through the Hellenistic period, though its selectivity
may mislead some readers about the important role played by the Hellenic
astronomical tradition. Neugebauer's second edition (Providence, R.I., 1957)
includes in an appendix a very useful description of the astronomical devices
elaborated in Ptolemy's Almagest. Sir Thomas L. Heath, Aristarchus of Samos
(Oxford, 1913) , is the standard source for Greek astronomy to the third cen-
tury B.c. Chapters 7-9 of Dreyer, History (above), discuss Greek astronomy
from Apollonius through Ptolemy.
Many selections from ancient astronomical writings can be found in Sir
Thomas L. Heath, Greek Astronomy (London, 1932), and in Cohen and
Drabkin, Source Book (above). Other relevant passages are Plato, Timaeus,
in Tile Dialogues of Plato, ed. Benjamin Jowett, 3rd ed. (London, 1892),
III, and Vitruvius, The Ten Books on Architecture, trans. M. H. Morgan
(Cambridge, Mass., 1926), though the translation of the latter is occasionally
handicapped by ignorance of astronomical fact and theory. Ptolemy, The
Almagest, has recently been translated by R. Catesby Taliaferro in Great
Books of the Western World, vol. XVI (Chicago, 1952). Detailed studies
will, however, continue to depend upon the standard edition, Syntaxis
mathematica, ed. J. L. Heiberg, 2 vols. (Leipzig, 1898-1903).
Much information about ancient calendars is contained in many of the
secondary sources above. More detailed studies are F. H. Colson, The Week
(Cambridge, Mass., 1926), and R. A. Parker, The Calendars of Ancient
Egypt (Chicago, 1950). Stonehenge as a primitive observatory is discussed
in Sir Norman Lockyer, Stonehenge and Other British Stone Monuments
Astronomically Considered, 2nd ed. (London, 1909), but see also Jacquetta
Hawkes, "Stonehenge," Scientific American CLXXXVIII (June 1953), 25-
BIBLIOGRAPHICAL NOTES 285
31. For the role of the heavens in primitive cosmological thought see Henri
Frankfort et al., The Intellectual Adventure of Ancient Man (Chicago,
1946), and Heinz Werner, The Comparative Psychology of Mental De-
velopment, rev. ed. (Chicago, 1948).
Chapter 3
The principal sources for this chapter are Aristotle's writings on
the physical sciences, particularly his Physics, Metaphysics, On the Heavens,
Meteorology, and On Generation and Corruption. All are translated in The
Works of Aristotle Translated into English, ed. Sir William David Ross,
12 vols. (Oxford, 1928-1952) and all except the last in the editions of
The Loeb Classical Library. The annotations and text in Sir W. D. Ross,
Aristotle's Physics (Oxford, 1936), make it a particularly useful edition.
John Burnet, Early Greek Philosophy, 3rd ed. (London, 1920); Theodor
Gomperz, Greek Thinkers, trans. Laurie Magnus (I) and G. A. Berry (II-
IV) (New York, 1901-1912); and Kathleen Freeman, The Pre-Socratic
Philosophers (Oxford, 1946) make it possible to set Aristotle's thought in
the tradition established by his predecessors. Sir W. D. Ross, Aristotle, 3rd
ed. (London, 1937) and Werner Jaeger, Aristotle: Fundamentals of the
History of His Development, trans. Richard Robinson (Oxford, 1934) are
important biographical studies of the works. F. M. Cornford, The Laws of
Motion in Ancient Thought (Cambridge, Eng., 1931), deals incisively with
a number of special problems treated in this chapter.
The post-Ptolemaic computations of cosmological dimensions from the
principle of plenitude are discussed by Edward Rosen, "A Full Uni-
verse," Scientific Monthly LXIII ( 1946), 213-217, and in Chapters VIII
and XI of Dreyer, History (above, Introductory section). The evidence for
the experiment at Pisa is analyzed by Lane Cooper, Aristotle, Galileo, and
the Leaning Tower of Pisa (Ithaca, 1935), but Cooper's work should be
supplemented by the discussions of the development of Galileo's laws cited
below for Chapters 4 and 7. Primitive conceptions of space and motion are
discussed by Werner (above, Chapter 1) and in the numerous works of
Jean Piaget, particularly The Child's Conception of the World, trans. Joan
and Andrew Tomlinson (London, 1929); The Child's Conception of Physical
Causality, trans. Marjorie Gabain (London, 1930); and Les notions de
mouvement et de vitesse chez renfant (Paris, 1946).
Chapter 4
The most salient aspects of the transition from Hellenic to Hel-
lenistic science are sketched in George Sarton, Ancient Science and Modem
Civilization (Lincoln, Neb., 1954). Far more detail is provided in the same
author's Introduction (above, Introductory section).
Henry Osborn Taylor, The Mediaeval Mind, 4th ed., 2 vols. (Cam-
bridge, Mass., 1925) discusses the early Christian depreciation of pagan
science, and Dreyer, History (above, Introductory), provides much rele-
286
vant illustration for astronomy. Important primary sources are Augustine,
Confessions and Enchiriliion, in Works, ed. Marcus Dods (Edinburgh,
1871-1877).
My account of the reconciliation of Aristotelian cosmology with Biblical
history derives from St. Thomas Aquinas, The "Summa Theologica," trans.
Brothers of the English Dominican Province, 22 vols. (London, 1913-1925)
and from the Commentaria on Aristotle's physical treatises in vols. II and
III of St. Thomas Aquinas, Opera Omnia, 12 vols. (Rome, 1882-1906). The
result of the new integration is illustrated in Dante, The Banquet, trans.
Katharine Hillard (London, 1889), and The Divine Comedy (many English
editions). The effects of the cosmological metaphor on medieval and Renais-
sance thought are sketched in Charles H. Grandgent, Discourses on Dante
(Cambridge, Mass., 1924), and inS. L. Bethell, The Cultural Revolution of
the Seventeenth Century (London, 1951).
Arabic and medieval European astronomy are treated by Dreyer, His-
tory; by Duhem, Le systeme; by Sarton, Introduction (all above, Intro-
ductory); and by Lynn Thorndike, Science and Thought in the Fifteenth
Century (New York, 1929). Thorndike feels that previous scholars have
dated the emergence of an erudite European astronomical tradition too
late, but, at least with respect to the problem of the planets, I find his evi-
dence unconvincing. Additional valuable information is contained in Derek
J. Price, ed., Equatorie of the Planetis (Cambridge, England, 1955).
A. C. Crombie, Augustine to Galileo (above, Introductory), gives the
best substantive and bibliographic survey of medieval science. My own dis-
cussion is also indebted to numerous special studies, particularly Carl Boyer,
The Concepts of the Calculus, 2nd ed. (Wakefield, Mass., 1949); Marshall
Clagett, Giovanni Marliani and Late Medieval Physics (New York, 1941),
and "Some General Aspects of Physics in the Middle Ages," Isis XXXIX
( 1948), 29-44; Alexandre Koyre, 1:tudes Galileennes, 3 vols. (Paris, 1939);
Annaliese Maier, Studien zur Naturphilosophie der Spiitscholastik, 4 vols.
(Rome, 1951-1955); and John Herman Randall, Jr., "The Development of
Scie!l'tific Method in the School of Padua," Journal of the Histo,;y of Ideas
I ( 1940), 177-206. Koyre and Randall give particularly useful illustrations
of the transmission of scholastic ideas to the early founders of modern
science. Primary sources for the study of scholastic theories of motion in-
clude Thomas Bradwardine, Tractatus de Proportionibus, ed. and trans.
H. Lamar Crosby, Jr. (Madison, Wis., 1955); Marshall Clagett, "Selections
in Medieval Mechanics" (Madison, Wis., mimeographed, no date); Jean
Buridan, Quaestiones super lihris quattuor de caelo et mundo, ed. Ernest
A. Moody (Cambridge, Mass.: Mediaeval Academy of America, 1942);
and Nicole Oresme, Le livre du ciel et du monde, ed. A. D. Menut and A. J.
Denomy, Mediaeval Studies III-V (Toronto, 1941-1943).
The relation between science and a variety of social, economic, and in-
tellectual currents of the Renaissance is discussed by John Herman Ran-
dall, Jr., The Making of the Modern Mind, rev. ed. (Boston, 1940), and
287
by Myron P. Gilmore, The World of Humanism 1453-1517 (N y k
195<?) A . . , ew or ,
. ncient and Renaissanoe.Neoplatonism are discussed in Lynn Tho _
dike, and Experimental Science (above, Introductory), and
0. LoveJoy, The Great Chain of Being (Cambridge, Mass., 1948). Henry
Osborn Taylor, and Expression in the Sixteenth Century, 2 vols.
(New, 1920) mcludes a description of Renaissance Neoplatonism.
Plato. s attitude toward mathematics is treated by Sir Thomas L. Heath,
A HU:tory _of Greek Mathematics, 2 vols. (Oxford, 1921), and the effects
of attitude, in its Neoplatonic form, on science are discussed from a
v.ariety of points of view in Edwin Arthur Burtt, The Metaphysical Founda-
of Modern Science (New York, 1932); Alexandre Koyre,
Gahleo and Plato, Journal of the History of Ideas IV ( 1943), 400-428;
and Edward W: Strong, Procedures and Metaphysics (Berkeley, Calif., 1936).
The Is the only one that gives adequate emphasis to the mystical
and .ascienhfic tenor of Neoplatonic thought, but it may go too far in con-
that no point of view so fundamentally irrational could have had a
fruitful effe.ct on practice of science. For N eoplatonism see also the
works relatmg to Nicholas of Cusa and Giordano Bruno cited below for
Chapter 6.
Chapter 5
Copernicus' life and work are well described in Angus Armitage
Copernicus, The Founder of Modern Astronomy (London, 1938), but thi;
should supplemented by the far fuller one in Ludwig Prowe,
Coppermcus, 2 vols. (Berlin, 1883-1884). Copernicus' minor astro-
and Rheticus' Narratio prima are translated with an excel-
lent mtroductwn and notes in Edward Rosen, Three Copernican Treatises
(N.ew. York, .. The only complete English translation of Copernicus'
pnncipal work lS Nicolaus Copernicus, On the Revolutions of the Heaven[
Spheres, trans. Charles Glenn Wallis, in Great Books of the Western W orJ
vol. XVI (Chicago, 1952). Anyone using this edition should first
the highly crit.ical review by 0. Neugebauer in Isis XLVI (1955), 69-71.
Enghsh translation of the Preface and Book I of the De Revolu-
by John F. Dobson and Selig Brodetsky has been published as
Occaswnal Notes of the Royal Astronomical Society, vol. II, no. 10 (Lon-
don, .1947). Alexandre Koyre, Copernic: Des Revolutions des Orbes Celestes
( 1934), provides a convenient French and Latin edition of Book 1
;-vith all the materials as well as a penetrating and provocative
discussiOn. The standard edition of the complete text is Maxi-
milian Curtze, Nicolai Copernici Thorunensis: De revolutionibus orbium
caelestium libri ':'I (Torun, 1873). Important special aspects of Copernicus'
are discussed by Dreyer, History (above, Introductory), and of
his physics and cosmology by Edgar Zilsel, "Copernicus and Mechanics"
Journal of the History of Ideas I (1940), 113-118. '
288
Chapter 6
Much useful material on the sixteenth- and seventeenth-century
reactions to Copernican astronomy is contained in Francis Johnson, Astro-
nomical Thought in Renaissance England (Baltimore, 1937); Grant McCol-
ley, "An early friend of the Copernican theory: Gemma Frisius," Isis XXVI
(1937), 322-325; Dorothy Stimson, The Gradual Acceptance of the Co-
pernican Theory of the Universe (New York, 1917); Lynn Thorndike,
Magic and Experimental Science (above, Introductory), particularly vol.
V, chap. 18, and vol. VI, chaps. 31 and 32; and Andrew D. White, A His-
tory of the Warfare of Science with Theology in Christendom, 2 vols.
(New York, 1896). Thorndike's material is the richest and most balanced,
but it must be used cautiously because it is occasionally predicated on im-
portant elementary errors about the technical relations between Copernican
and Ptolemaic astronomy (see, for example, the sentence connecting pp.
424 and 425 in vol. V) .
The fullest and most recent account of Galileo's conflict with the Church
is Giorgio de Santillana, The Crime of Galileo (Chicago, 1955). Some of
the older discussions are, however, still useful, particularly Karl von Gebler,
Galileo Galilei and the Roman Curia, trans. Mrs. George Sturge (London,
1879), and James Brodrick, S. J ., The Life and Work of Blessed Robert
Francis Cardinal Bellarmine, 2 vols. (London, 1928).
On Tycho Brahe see J. L. E. Dreyer, Tycho Brahe (Edinburgh, 1890),
and also his Opera Omnia, ed. J. L. E. Dreyer, 15 vols. (Hauniae, 1913-
1929). The often underestimated popularity of the Tychonic system has
been effectively documented by Grant McColley in "Nicolas Reyrners and
the Fourth System of the World," Popular Astronomy XLVI (1938), 25-
31, and "The Astronomy of Paradise Lost," Studies in Philology XXXIV
(1937), 209-247.
There is no adequate English study of Kepler's life or work, but Carola
Baumgardt, Johannes Kepler: Life and Letters (New York, 1951), includes
s ~ useful quotations from sources. The standard study, Max Caspar,
Johannes Kepler (Stuttgart, 1948), should soon be available in translation.
The relevant works must still be read in the fine Gesammelte werke, ed.
Max Caspar, 12 vols. (Munich, 1938--1955). R. H. Baker, Astronomy
(above, Chapter 1), gives a technical account of Kepler's Laws from a
modern viewpoint. Much information about their technical development is
included in Dreyer, History (above, Introductory), and in A. Wolf, A His-
tory of Science, Technology, and Philosophy in the XVI and XVII Centuries,
rev. ed. prepared by Douglas McKie (London, 1950). Other important stud-
ies of Kepler are cited for Chapter 7, below.
Galileo's telescopic observations are discussed in many of the preceding
studies. They are, however, best approached directly in Galileo Galilei,
The Sidereal Messenger, trans. Edward Stafford Carlos (London, 1880),
and Dialogue on the Great World Systems, ed. Giorgio de Santillana (Chi-
cago, 1953). Some indication of the telescope's immense impact on the sci-
L
289
entific and popular imagination is provided in Marjorie Hope Nicolson, "A
World in the Moon," Smith College Studies in Modern Languages XVII,
no. 2 (Northampton, Mass., 1936); in Martha Ornstein, The Role of Scien-
tific Societies in the Seventeenth Century (Chicago, 1938) ; in some selec-
tions from The Portable Elizabethan Reader, ed. Hiram Haydn (New York,
1946); and in Edward Rosen, The Naming of the Telescope (New York,
1947). Most of Galileo's work is beyond the scope of this book; some other
important studies are, however, cited for Chapters 4 and 7.
Chapter 7
Pre-Copernican and post-Copernican ideas about the infinity of
the universe are discussed in Francis R. Johnson and Sanford V. Larkey,
"Thomas Digges, the Copernican System, and the Idea of the Infinity of
the Universe," Huntington Library Bulletin V (April 1934), 69-117; Alex-
andre Koyre, "Le vide et l'espace infini au XIV siecle," Archives d'histoire
doctrinale et litteraire du moyen age XXIV (1949), 45-91; Lovejoy, Great
Chain (above, Chapter 4); Grant McColley, "Nicolas Copernicus and an
Infinite Universe," Popular Astronomy XLIV (1936), 525-533, and "The
Seventeenth-Century Doctrine of a Plurality of Worlds," Annals of Science
I (1936), 385-430. McColley's articles are particularly informative, though
he seriously overstates the case for Copernicus' belief in an infinite universe.
Johnson's article reprints the relevant portions of Digges's Perfit Description.
Other useful primary sources are Nicholas of Cusa, Of Learned Ignorance,
trans. Germain Heron (London, 1950), and excerpts from De ludo globi
chosen and translated by Maurice de Gandillac in Oeuvres choisies de Nico-
las de Cues (Paris, 1942). See also the annotated translation of Bruno's On
the Infinite Universe and Worlds in Dorothea Waley Singer, Giordano Bruno,
His Life and Thought (New York, 1950). For some time to come the standard
source on this problem will surely be Alexandre Koyre, From the Closed
Worlrl to the Infinite Universe (Baltimore, 1957).
Despite the wealth and range of this literature, there still seems to be
one important gap in our knowledge of the evolution of the concept of an
infinite Copernican universe. From Bruno's death in 1600 to the publication
of Descartes's Principles of Philosophy in 1644, no Copernican of any promi-
nence appears to have espoused the infinite universe, at least in public. After
Descartes, however, no Copernican seems to have opposed the conception.
This silence during the first half of the seventeenth century is understandable,
but it leaves a puzzle about the development and propagation of the belief
in a physically infinite universe.
Frederick A. Lange, The History of Materialism, trans. E. C. Thomas,
3rd ed. (New York, 1950), and Kurd Lasswitz, Geschichte der Atomistik
2nd ed., 2 vols. (Hamburg, 1926), include much essential information abou;
the development of atomism since antiquity. Seventeenth-century atomism
is thoroughly surveyed in Marie Boas, "The Establishment of the Mechanical
Philosophy," Osiris X (1952), 412-541, a monograph which also provides a
fine bibliography on this topic. Significant special studies of atomism's role
290
in the development of modern science include H. The
Philosophy of Francis Bacon (Chicago, 1948); Mane Boas, Boyle as a
Theoretical Scientist," Isis XLI ( 1950), 261-268; Thomas S. "Robert
Boyle and Structural Chemistry in the Seventeenth Century, XLIII
(1952), 12-36; and Paul Mouy, Le developpement de la physzque Car-
tesienne (Paris, 1934) . Important and representative primary sources for
the central tenets of this seventeenth-century tradition are Rene Descartes,
Les principes de la philosophie and Le monde ou le traite de la lumiere in
vols. IX and XI of the Oeuvres de Descartes, ed. Charles Adam and
Tannery (Paris, 1904 and 1909), and Robert Boyle, Origin of Qual1t1es
and Forms, in vol. II of The Works, ed. A. Millar (London, 1744). .
For the problems presented to terrestrial physicists by the
theory see Alexandre Koyre, 1:tudes Galileennes, 3 vols. (Pans,,
"Galileo and the Scientific Revolution of the Seventeenth Century, Philo-
sophical Review LII ( 1943), 333-348, and especially "A Documentary His-
tory of the Problem of Fall from Kepler to Newton," Transactions of the
American Philosophical Society ( n.s.) XXXXV, no. ( 1955), 329-395.
Kepler's celestial mechanics is discussed in D\eyer, History (above: Intro-
ductory); in Gerald Holton, "Johannes Kepler s Universe: Its Physics and
Metaphysics," American Journal of Physics, XXIV ( 1956), 34?-351;
in Alexandre Koyre, "La gravitation universelle, de Kepler a Newton,
Archives internationales d'histoire des sciences XXX (1951), 638-653.
Borelli's system is described in Angus Armitage, "'Borelli's Hypothesis' and
the Rise of Celestial Mechanics," Annals of Science VI (1950), 268-282,
and Alexandre Kovre "La mechanique celeste de J. A. Borelli," Revue
d'histoire des scier:ces' V ( 1952), 101-138. The work of Robert Hooke is
treated in its relation to Newton by Louise D. Patterson, "Hooke's Gravita-
tion Theory and its Influence on Newton," Isis XL (1949), 327-341, and
XLI ( 1950), 32-45, and more incisively with the aid of a new document by
Alexandre Koyre, "An Unpublished Letter of Robert Hooke to Isaac New-
ton," Isis XLIII ( 1952), 312-337. Many documents illustrating Hooke's
wol"k will be found in R. T. Gunther, Early Science in Oxford, 14 vols.
(Oxford, 19-20-1945), particularly vols. VI and VIII. .
Guidance to the vast literature on Newton can be found m almost all
the bibliographical sources listed in the Introductory, above. What-
ever is novel ih my own approach concerns Newtons the re-
lated metaphysical substructure of the Principia. This analysis Is at least
partially derived from and supported by Florian Cajori, "Ce que Newton
doit a Descartes," L'enseignement mathematique XXV (1926),
"Newton's Twenty Years' Delay in Announcing the Law GravitatiOn,
in Sir Isaac Newton, ed. History of Science Society (Baltimore, 1928);
A. R. Hall, "Sir Isaac Newton's Note-Book, 1661-65," Cambridge Historical
Journal IX ( 1948), 239-250; Alexandre Koyre, "T?e. Si?nificance the
Newtonian Synthesis," Archives internationales d htstotre des sczences
XXIX (1950), 291-311; Thomas S. Kuhn, "Newton's '31st Query' and
291
the Degradation of Gold," Isis XLII ( 1951), 296-298, and "Preface to
Newton's Optical Papers," in I. B. Cohen, ed., Isaac Newton's Letters and
Papers on Natural Philosophy (Cambridge, Mass., in press), and S. I. Vavi-
lov, "Newton and the Atomic Theory," in The Royal Society Newton Ter-
centenary Celebrations (Cambridge, Eng., 1947). Convenient primary
sources are Sir Isaac Newton, Mathematical Principles of Natural Philoso-
phy, ed. Florian Cajori (Berkeley, Calif., 1946), and Opticks, reprint ed.
(New York, 1952).
Technical Appendix
R. H. Baker, Astronomy (above, Chapter 1), discusses the equa-
tion of time, precession of the equinoxes, eclipses, and the moon's phases
from a modern viewpoint. Heath, Aristarchus (above, Chapter 1), and
Dreyer, History (above, Introductory) have much historical information
on all these subjects except the first, on which see A. Rome, "Le probleme
de !'equation du temps chez Ptolemee," Annales de la societe scientifique de
Bruxelles (Series 1) LIX (1939), 211-224. Heath and Dreyer also treat
the ancient determinations of astronomical dimensions, on which see also
Aubrey Diller, "The Ancient Measurements of the Earth," Isis XL (1949),
6-12. Additional details about the Moslem complications of the treatments
of precession will be found in Francis J. Carmody, Al-Bitru;i: de motibus
celorum (Berkeley, Calif., 1952), and "Notes on the Astronomical Works of
Thabit b. Qurra," Isis XLVI (1955), 235-242.
INDEX
The more important page references are in italic type
Absolute space. See Aristotelian space
Adams, J. C., 261
Aether, 79-82, 91-92
AI Fargani, 81-82, 160. See also Dis-
tance, astronomical
Alpetragius, 177
Amici, C. B., 138
Anaximander, 26-27
Anima matrix, 214-216, 230, 245-249
Aphelion, 214-215
Apogee, 147
Apollonius, 59, 72-74
Aquinas, Thomas, 109-111
Aristarchus, 42, 144, 160, 274-278
Aristotelian cosmology, 59, 79-84, 91-
92, 109, 112. See also Cosmology,
Christian
Aristotelian space, 79, 87-88, 97-98, 231
Aristotelian theory of motion, 82, 84-87,
95-98, 118-119, 151, 244
Aristotle, 78, 105, 111-112; criticism,
8.3, 106, 109, 115-123, 151-153,
206, 208; authority, 78, 83-84, 95-
99, 103, 111, 115, 117, 127
Astrology, 25, 49-50, 92-94, 98, 107,
206; relation to astronomy, 94
Astronomical tables, 103, 105, 187-188,
196,219
Astronomy, prehistoric, 9-11, 13-14,47,
49-50; Hellenic, 5, 26-27, 50, 52,
104-105; Hellenistic, 66, 72-73,
104-105; Moslem, 72, 74, 102, 112,
269-270; medieval, 74, 112-113,
123-124; Renaissance, 67, 125-126.
See also Cosmology
Atomism, 41, 199, 2.'30, 2.32, 235-237.
See also Corpuscular philosophy
Augustine, 107, 128
Bellarmine, Cardinal, 198, 226
Blundeville, Thomas, 186
Bodin, Jean, 190
Bonamico, F., 120
Borelli, G. A., 248-249,
Brahe, fycho, 93, 211, 221; role in
Copernican Revolution, 201, 204-
206; observations, 160, 200-201,
212-214. See also Tychonic system
Bruno, Giordano, 199, 220, 235-237,243
Buridan, Jean, 119-122, 124
Calendars, 8-12, 46-47, 125-126, 196,
271
Callippus, 58, 80
Calvin, 125, 192, 195-196
Capella, Martianus, 178
Catholic Church, attitude toward astron-
omy, 93, 106-112, 197; and Coper-
nicanism, 106, 192, 195-199, 226.
See also Copernican cosmology
Celestial mechanisms, 27, 52, 79-80, 89,
105, 114, 121-122, 214-216, 240--
242, 245-252, 254-260
Celestial sphere. See Sphere of stars
Celestial-terrestrial dichotomy, 43, 91-
92, 94, 121-122, 153, 207-208,
221-222, 237, 252. See also Sublu-
nary region; Superlunary region
Collisions, atomic, 238-240, 242
Columbus, Christopher. See Voyages
Comets, 45, 93, 208, 244
Compass points, 9
Conceptual schemes, commitment to, 39,
75-77, 83; economy, 36-37, 54, 74,
77, 90, 169, 212-213, 226, 265;
fruitfulness, 39-41, 54, 64-66, 77,
208-209, 212-213, 261-262, 264,
273, 278; distinguished from ob-
servations, 25-26, 35; psychological
functions, 38-39, 105, 113. See also
Scientific revolutions
Conjunction, 49, 178
Constellations, 13-16; Big Dipper, 13-
16, 20--21, 206; zodiacal, 25, 46, 93,
266
Copernican astronomy, 135, 143-144,
149, 155-172, 179-180, 210-214,
270n; construed as computing de-
vice, 187-188, 196, 226; harmonies,
142, 172-181, 184; compared with
Ptolemaic, 75, 162-163, 165-166,
169-175, 178-179, 181-182, 223;
reception, 131, 172, 182-183, 185-
188, 201, 209-210, 224, 227-228;
compared with Tychonic, 202-205.
See also Copernican cosmology
Copernican cosmology, 7, 40-41, 94,
145-155, 220, 224, 231-232; antic-
ipations, 41-42, 115-116, 142, 144,
149, 233-237; and 86-87,
115-123, 150-155, 214-215, 230-
231, 238, 243-245, 252-254, 261-
294
262; reception, 182-183, 186, 189-
201, 209, 224-228, 237-238; and
religion, 92, 130--131, 191-199,
262-263. See also Copernican as-
tronomy; Size of universe
Copernican Revolution, 1-4, 41, 172,
181-184, 228, 261-265; causes, 8,
11, 45, 50, 55, 74-76, 83, 100, 105,
124-133, 140-141, 270; plurality,
1-3, 77, 84, 87, 94, 132-133, 141,
195,227-231,261-264
Copernicus, Nicolaus, 42, 72-73, 129;
and the calendar, 11, 125-126, 142,
271; Commentariolus, 138, 172;
conservatism, 30, 59, 84, 90, 135,
145, 148, 155, 164-165, 182,
243; De Revolutionibus, 59, 134-
139, 141-143, 145-155, 172, 177-
180, 182-184, 185-187. 192, 196;
on equants, 69-70, 72, 139, 148;
motives, 126, 138-139, 142, 144-
145, 183, 271; and Neoplatonism,
129,131,141-142,179-181,200
Corpuscular philosophy, 237-242, 247,
252, 257-260. See also Atomism
Cosmology, functions, 6-7, 38-39; rela-
tion to astronomy, 5-8, 25-27, 104-
105, 114; primitive, 4-8, 27; Greek,
26-29, 41-43; medieval, 38, 42, 54;
Christian, 92, 108-110, 112-114,
194-195. See also Aristotelian cos-
mology; Copernican cosmology
Counter Reformation, 198-199
Crystalline spheres, 79-82, 89, 206, 278.
See also Aether; Celestial-terrestrial
dichotomy
Cusa, Nicholas of, 144, 197, 233-235
Dante, 112-114, 192, 194
Data. See Conceptual schemes
Day, apparent solar, 9, 266-267; mean
solar, 267
Deferent, 59, 61. See also Epicycle-def-
erent system; Ptolemaic system
Deism, 263
Democritus. See Atomism
Descartes, Rene, 238-242, 247, 249,
252-254, 258-259
Thomas, 186, 233-234
Distance, angular, 15n, 19, 23, 25, 220-
221; astronomical, 54, 81-82, 160,
201, 271, 273-278. Sec also Size of
universe
Diurnal circle, 16-17, 28, 31, 34, 36, 269
INDEX
Diurnal motion. See Sun; Planets;
Sphere of stars
Donne, John, 194-195, 226, 237
Du Bartas, G. de S., 189-190
Earth, size and shape, 29, 54, 82, 85-86,
146, 273-274. See also Copernican
astronomy; Two-sphere universe
Eccentric, 70--72. See also Ptolemaic sys-
tem
Eclipses,26-29,50,54,93,271-273,276
Ecliptic, 21-25, 34-37, 46-48, 56-58,
157, 266; plane, 51, 53, 60--61, 65,
161-162; pole, 268-270; obliquity,
37,266,268
Einstein, Albert, 4, 99, 229
El1ipse, 212-213, 225, 246, 248, 250--
251
Elongation, 48, 65, 171, 173, 177
Empyrean sphere, 112-113
Epicurus. See Atomism
Epicycle, 59, 61; major, 67-69, 74;
minor, 67
Epicycle-deferent system, 59-64, 67, 73-
7 4, 77, 80. See also Ptolemaic sys-
tem
Equant, 70--71, 72
Equant point, 71
Equation of time, 267
Equator, celestial, 31, 33, 35, 37, 157,
266-267; terrestrial, 32-33, 36
Equinoxes, 9-10, 12, 24-25, 35-36, 66-
71, 157, 159, 162, 266-267; preces-
sion of, 16n, 112, 114, 260, 268-271
Eratosthenes, 274, 276
Eudoxus. See Homocentric spheres
Evening star, 49, 65, 223
Explanation, 38-39. See also Conceptual
schemes
Ficino, Marsilio, 130
Finite universe. See Size of universe
Foscarini, P. A., 198
Fracastoro, G., 138
Full universe. See Plenum
Galileo, 127, 173, 199, 231; and Coper-
nican Revolution, 224-226; theory
of motion, 95, 117-118, 122, 238-
239, 244; observations,
198, 219-226, 229
Gilbert, William, 246
INDEX
Gnomon, 8-9, 12, 274
Gravity, 1.54, 252-260, 262
Great circle, 33, 266
Hellenic civilization, 5, 104
Hellenistic civilization, 104-105
Henry the Navigator. See Voyages
Heraclides, 42, 75, 86, 142, 149, 157,
202
Hicetas, 142, 148
Hipparchus, 59, 72-74, ll8, 269-270
Homocentric spheres, 55-59, 79-80,
138
Hooke, Robert, 248-256
Horizon, 9-10, 12, 15-23, 25, 157
Horizon plane, 30-32, 35-36, 156-159
Horizon-window, 20--22, 24
Humanism, 127-131
Huyghens, Christian, 242, 253
Impetus theory, 118-122, 238-239
Inertial motion, 238-240, 247-250, 253-
255
Infinite universe. See Size of universe
John of Holywood, 124, 189
Jupiter, satellites of, 222. See also
Planets
Kepler, 93, 127, 129, 187, 205-206,
209-211, 224-227, 229, 231, 243-
247, 253; Laws, 212-217, 225,
245-246, 256, 258, 261; Neopla-
tonism, 131, 209, 214-219, 245
Kosmas, 108, 196
Lactantius, 107-108, 143, 195-196
Latitude, 17
Leap year, ll-12
Leucippus. See Atomism
Leverrier, U. J. J., 261
Longitude, celestial, 266
Lucretius. See Atomism
Lunar. See Moon; Calendars
Luther, Martin, 125, 191, 195-196
Maestlin, Michael, 187, 209
Magellan. See Voyages
Magnetism, 246-247
295
Mars, 49, 170-171, 211-212. See also
Planets
Melanchthon, Philipp, 191-192, 196
Milton, John, 7, 194-195
Momentum, 122, 239
Moon, 45-46, 54, 221; phases of, 271-
273. See also Calendars; Distance,
astronomical; Eclipses; Planets
Morning star, 49, 65
Miiller, Johannes, 124, 126-127
Navigation, 37, 125
Neoplatonism, 128-132, 197, 199, 231,
233--235
I\ewton, Isaac, 183, 228, 247, 251, 253-
260, 262, 270
Newtonian dynamics, 122-123, 256,
268
Newtonian space, 97, 235-237, 239, 260
Newtonian universe, 1, 74, 99, 231, 243,
260-261, 263-265
Ninth sphere, 1lln, 269
Normal motion, 47, 167. See also Planets
North Star, 13, 15-16
Novara, D. M. de, 129
Novas, 206-209
Observations in anti(Juity, 8, 13-14, 26.
See also Conceptual schemes
Omens. See Astrology
Opposition, planetary, 49, 178, 180
Oresme, Nicole, 115-117, 120-121, 124,
154, 156-157, 197, 25.'3
Osiander, Andreas, 187, 196
Parallax, 157-160, 163-164, 180, 201-
204, 207-208
Paul III, Pope, 137, 143
Perigee, 147
Perihelion, 214-216
Petrarch, 127
Peuerbach, Georg, 124, 126
Philoponus, John, 119
Planets, 45; apparent motion, 45-50;
diurnal motion, 45-46, 53-64;
early theories, 7, 26-27, 41-42; in-
ferior, 42, 48-49, .53-54, 65-66,
166--167, 171, 173, 17.5-176, 178,
202-203; order, 52-54, 173-175,
177-180; periods, 47, 168, 174;
problem, 8, 50-.51, 55, 66, 72-74,
94, 136-137, 172, 184, 201, 2ll-
296
213; superior, 49, 166--167, 174,
176, 178; variation of brightness,
49, 59, 62, 73, 167. See also Co-
pernican astronomy; Distance, as-
tronomical; Epicycle-deferent sys-
tem; Homocentric spheres; Kep-
ler's Laws; Two-sphere universe;
Tychonic system
Plato, 28-30, 50, 55, 79, 97, 128, 178
Plenum, 79-80, 87-90, 240. See also
Size of universe; Vacuum
Plurality of worlds, 41, 79, 89-90, 144,
199,222,235-237,240-242
Pole, celestial, 15-18, 20-21, 30-33, 36,
55-56, 268-270; terrestrial, 32-33,
36
Proclus, 129-130
Projectiles, 119, 230, 242, 252, 255, 258
Protestantism. See Copernician cosmol-
ogy; Reformation
Pru.tenic Tables. See Astronomical
tables
Ptolemaic system, 7, 66--73, 75, 83, 139-
140, 210, 278; decline, 224-228.
See also Copernican astronomy
Ptolemy, 50, 54, 71-72, 80, 93, 100, 104,
125, 178, 269-270; and Aristote-
lian science, 86; criticism, 69-70,
151-152; on crystalline spheres,
80, 105; Almagest, 71-72, 86, 93,
102, 123; Hypotheses, 105; Tetra-
biblos, 93. See also Copernican
astronomy
Pythagoreans, 41-42, 128, 141
Reformation, 124-125, 195-196
.,Regular solids, 217-218
Reinhold, Erasmus, 187, 196
Renaissance, 124-128; Twelfth Century,
102-103
Retrograde motion, 47-49; in homocen-
tric spheres, 56-59; in epicycle-
deferent system, 60-64; in Coper-
nican system, 150, 165-167. See
also Planets, problem of
Rheticus, G. J., 138, 186, 196
Royal Society, 251
Rudolphine Tables. See Astronomical
tables
Scholasticism, 102-106, 123, 232
Scientific Revolution, 2, 227-228, 230-
231,260-261
Scientific revolutions, 3-4, 94, 134-135,
INDEX
183-184, 229-230, 263-265; struc-
ture of, 75-76
Seasons, 10-11, 25, 162-163, 271. See
also Equinoxes; Solstices
Size of universe, 81-82, 90, 132, 160,
180, 201; finite, 79, 89-90, 155,
231; infinite, 41, 89-90, 144, 157,
193, 199, 220, 232-237, 247, 260,
285. See also Distance, astronom-
ical; Parallax; Plenum; Vacuum
Solstices, 9-10, 12, 24-25, 35-36, 67-72,
157, 158, 162, 267
Sphere of stars, 27, 31-37, 51-56, 60,
231-234, 267; diurnal motion, 28,
33-34. See also Distance, astronom-
ical; Equinoxes, precession of
Star maps, 20-25
Stars, apparent motion, 13-21, 25, 32,
266, 268; circumpolar, 6, 17, 19-
20, 31-32. See also Sphere of stars
Star trails, 18-19, 156
Stonehenge, 10-11
Sublunary region, 82-84. See also Aris-
totelian cosmology
Sun, 23, 45; annual and diurnal motion,
21-25, 34-35; apparent motion,
8-13, 266-268; irregular motion,
66--67, 169, 266--267; in two-sphere
universe, 33-36. See also Calendars;
Distance, astronomical; Eclipses;
Ecliptic; Planets
Sun spots, 221
Sun worship, 10-11, 129-130. See also
Astrology
Superlunary region, 80, 89, 91-92. See
also Aristotelian cosmology
Symmetry, 28-30, 51, 70, 91, 129, 180-
181, 205. See also Two-sphere uni-
verse
Telescope, 46, 198, 219-226, 231, 233,
261-262
Theory. See Conceptual schemes
Time measurement, 9, 266--268. See
also Calendars
Trepidation, 270
Two-sphere universe, 25-37, 44, 54, 74-
77, 271; arguments for, 27-33, 43,
84-87, 90, 94, 109, 193-194; plan-
ets and stars in, 30-33, 50-53. See
also Aristotelian cosmology
Tychonic system, 74, 201-206, 214, 224,
227, 244
Universities, 102-1();3, 127
INDEX
Vacuum, 41, 79, 88-90, 110, 119, 236-
238, 240. See also Plenum
Venus, 6!5; ~ h s e s of, 177, 222-224;
prehistonc observations, 49. See
also Planets
Vitruvius, 52-53
Vortex cosmology, 50, 240-
242
Voyages, 40, 125, 146--147
Weight, 242, 253 259 262
Whitehead, A. N.', 123'
297
Year, 11, 47, 160; sidereal "71
ical, 271 ' - ' trop-
Zodiac. See Constellations

Kuhn, Thomas - Copernican Revolution, The (Harvard, 1985) PDF

Uploaded by

Copyright:

Available Formats

Kuhn, Thomas - Copernican Revolution, The (Harvard, 1985) PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Kuhn, Thomas - Copernican Revolution, The (Harvard, 1985) PDF

Uploaded by

Copyright:

Available Formats

What is the book about?

What aspects of the Copernican Revolution does the book aim to examine?

THE COPERNICAN REVOLUTION

Planetary Astronomy in the Development of Western Thought

it contains all the points at which the center

You might also like