Des Chene, Dennis - Physiologia - Natural Philosophy in Late Aristotelian and Cartesian Thought-Cornell University Press (2000)

PHYSIOLOGIA
Unauthenticated
Download Date | 3/18/19 2:07 AM
Unauthenticated
PHYSIOLOGIA
Natural Philosophy in
Late Aristotelian and
Cartesian Thought
'i
Dennis Des Chene
Cornell University Press
Ithaca and London
Unauthenticated
PUBLICATION OF THIS BOOK WAS ASSISTED BY A GRANT
FROM THE PUBLICATIONS PROGRAM OF THE NATIONAL ENDOWMENT
FOR THE HUMANITIES, AN INDEPENDENT FEDERAL AGENCY.
Copyright© I996 by Cornell University
All rights reserved. Except for brief quotations in a review, this book, or parts thereof, must not
be reproduced in any form without permission in writing from the publisher. For information,
address Cornell University Press, Sage House, 5I2 East State Street, Ithaca, New York I485o.
First published I996 by Cornell University Press
First printing, Cornell Paperbacks, 2000
Library of Congress Cataloging-in-Publication Data
Des Chene, Dennis.
Physiologia : natural philosophy in late Aristotelian and
Cartesian thought I Dennis Des Chene.
p. em.
Includes bibliographical references and index.
ISBN o-8oi4-3072-o (cloth: alk. paper)
ISBN o-8oi4-8687-4 (pbk. : alk. paper)
1. Physics-Philosophy. 2. Aristotle-Influence. 3· Descartes,
Rene, I596-I65o-Influence. 4· Philosophy, Medieval. 5· Philosophy
of nature. I. Title.
QC6.D43 I995
II3--<iC20
Printed in the United States of America
Cornell University Press strives to use environmentally responsible suppliers and materials to
the fullest extent possible in the publishing of its books. Such materials include vegetable
based, low-VOC inks and acid-free papers that are recycled, totally chlorine-free, or partly com
posed of nonwood fibers. Books that bear the logo of the FSC (Forest Stewardship Council)
use paper taken from forests that have been inspected and certified as meeting the
highest standards for environmental and social responsibility. For further informa
tion, visit our website at www.cornellpress.cornell.edu.
I 3 5 7 9 Cloth printing IO 8 6 4 2
I 3 5 7 9 Paperback printing 10 8 6 4 2
Unauthenticated
Contents
Preface vii
Abbreviations and Orthographical Conventions xi
Introduction
1. Interpretations of Nature 1
2. Aristotelianism 7
3· Leitfaden 1 1
PART 1 Vicaria Dei

1. Natural Change
2. Motus, Potentia, Actus 21
2 .1. Potentia and Actus 24
2.2. Independent Existence of Motus 34
2.3. Action and Passion 40
2.4. Active and Passive Potentim 46
3· Form, Privation, and Substance 53
3.1. Principles of Change 55
3.2. Substantial Form and Prime Matter 64
3·3· Form as Substance 76
4· Matter, Quantity, and Figure

4.1. The Essence of Matter 83
4.2. Quantity and Prime Matter 97
4·3· Figure and Other Qualities 109
[v]
Unauthenticated
[vi] Contents
5· The Structure of Physical Substance 122

5.1. Matter and Form Distinguished 124
5.2. Substantial Union 134
5·3· Conditions for the Reception of Form: Dispositions 138
5-4- Substantial Form and Active Powers 157
6. Finality and Final Causes 168

6.1. Varieties of End 171
6.2. Existence of Ends 177
6.3. Character of the Final Cause 186
6.4. Teleological Reasoning 200
7· Nature and Counternature 212

7.1. The Uses of Nature 213
7.2. Individual Natures 227
7·3· Artifacts, Human and Divine 239
PART n: Bodies in Motion

8. Motion and Its Causes 2 55
8.1. The Definition and Mode of Existence of Motion 257
8.2. Persistence, Conatus, and Quantity of Motion 272
8.3. Natural and Divine Agency: The Problem of Force 312
g. Parts of Matter 342

9.1. Extensive Quantity and the Nature of Matter 345
9.2. Substance and Space 354
9·3· Physical Questions: The Sufficiency of Extension 377
10. World without Ends 39 1
Bibliography 399
Primary Sources 399
Secondary Sources 406
Index
Unauthenticated
Preface
I t sometimes happens that the book one sets out to write is not the
book one eventually writes. So it was with this book: what began as an
introductory pair of chapters in a book on Aristotelian and Cartesian
psychology augmented itself into a book devoted to Aristotelian and
Cartesian philosophy of nature. There was, it turned out, no compendious
study of the sixteenth- and early seventeenth-century Aristotelian philoso
phers who, at least until the time of Locke, dominated the teaching of the
universities of Europe. Despite the efforts of Gilson and some of his suc
cessors, their commentaries and cursus, immense and forbidding, have re
mained largely unknown territory. Especially among analytic philosophers,
the Latin world of the early modern period has until recently been all but
forgotten. This work is an initial effort to make that world more familiar.
The disadvantages of ignorance are several. It is impossible to discern
what is new and what is not in Descartes's work without a thorough knowl
edge of those earlier works from which he first learned his metaphysics and
natural philosophy. Or rather, since a distinction of new and old is too
crude, it is impossible to assess his use of the resources available to him; nor
can one securely divine from the Cartesian corpus alone the significance to
be attached to terms and propositions whose intended audience often con
sisted of Aristotelians. It is, moreover, on any but the most naive view of the
timelessness of philosophical questions and solutions, fruitless to evaluate
his arguments without such knowledge. I would add, finally, that the Aristo
telians, contrary to the caricatures often painted of them by their oppo
nents, were careful and serious thinkers; some, like Suarez, are worthy of
study independent of their importance to later philosophers.
One aim of this work is to make ignorance of Aristotelianism disreputa
ble. There are indeed obstacles in the way of anyone now who undertakes to
[vii]
Unauthenticated
[viii] Preface
study the Aristotelians: the language, the disputational format of their writ
ings, the dependence, explicit or not, of later texts upon earlier texts, and
the consequent requirement that one should have also studied the domi
nant Medieval figures. But those obstacles are effectively no more difficult
to overcome than those that present themselves, say, to the student of Kant
or Hegel, and overcoming them opens the way to five centuries of rich and
profound philosophical argument.
A work whose gestation is lengthy acquires many debts, personal and intel
lectual. One of the pleasures of finishing is to acknowledge those debts.
Every student of Renaissance philosophy will, first of all, build on the work
of his or her predecessors. Historians like Etienne Gilson, Anneliese Maier,
Charles Schmitt, Tullio Gregory, and Edward Grant have brought to light
and interpreted for modern readers the enormous and complex world of
Medieval and Renaissance philosophy. Closer to home, Gilson, Dan Garber,
Roger Ariew, Alan Gabbey,John Schuster, Genevieve Rodis-Lewis,Jean-Luc
Marion, Pierre Costabel, Jean-Louis Armogathe, and many others have en
larged the horizons of the study of Cartesian philosophy of nature. This
work-and for that matter the idea of this work-would have been impossi
ble without theirs.
Among those who have read drafts of the manuscript, or who have pro
vided comments and supports along the way, I thank first my lectrix prima
Mary Des Chene, whose sharp eye no infelicitous phrase evades. She now
has more Aristotelian philosophy than any anthropologist needs. I have
benefited from comments from and discussions with Mark Rigstad, Thanos
Raftopoulos, Mauro Dorato, and Natalie Brender. Natalie has become an
able answerer of obscure queries while assisting me in my research. Ed
Minar's conversation and questions helped me as I was thinking out the
larger project of which this book is the first segment. Ira Singer read and
commented on Part I. Peter Achinstein and Rob Rynasiewicz will, I hope,
see echoes of bygone lunch conversations in Part II. Jerry Schneewind's
belief in the value of a history of philosophy not subservient to present-day
conceptions has reinforced my own. The late David Sachs presented to all
who knew him a day-to-day exemplum of philosophical curiosity and rigor; I
regret that he was not able to see his encouragements come to fruition.
Intermittent conversations with Steve Menn have proved fruitful, as has the
reading of some of his work in manuscript. Roger Ariew and Emily Grosholz
offered knowledgeable advice and criticism. Roger Haydon, the editor for
Cornell University Press, has supported the work since its earliest stages:
sine qua non.
The project from which this book has grown began with work at Stanford
University under Arnold Davidson, Peter Galison, and Stuart Hampshire.
Unauthenticated
Preface [ix]
During that time I received support from the philosophy department; a

· fellowship from the Mrs. Giles S. Whiting Foundation in 1986-1987 funded
my fifth year of graduate study. In 1984-1986 a graduate fellowship at the
Stanford Humanities Center yielded a fine office and the intellectual stim
ulation of meeting scholars from diverse disciplines. An Eli S. Lilly Founda
tion teaching fellowship funded one semester's leave-replacement in the
spring of 1993, a crucial time in the writing of this book. Jacqueline Mitchell
coordinated the program atJohns Hopkins University; she was, moreover, a
source of encouragement through the year of the fellowship. Research
materials without which this book would have been impossible were ac
quired for me through the alchemical skills ofAlan Braddock and others at
the interlibrary loan office at the Milton S. Eisenhower Library. I thank them
for their efficiency and patience.
Earlier versions of material from §6 were given as talks at the University of
British Columbia and at Yale University. I thank Paul Russell and Paolo
Mancosu for their invitations and hospitality; I have since benefited also
from reading Paolo's work on Descartes's geometry. A version of §8.1 was
presented, at the invitation of Sharon Kingsland, at the History and Philoso
phy of Science Colloquium at Hopkins in the fall of 1993.
DENNIS DEs CHENE
Baltimore, Maryland
Unauthenticated
Unauthenticated
Abbreviations and
Orthographical Conventions
R
Citations
eferences to primary sources are by author, short title, and editor
or edition (where necessary). Secondary sources are referred to
by author and year. I have tried where possible to note original
dates of publication or composition. Texts from primary sources
are cited first by the smallest numbered or named subdivision, and then by
volume (if there is more than one) and page or column number. The
edition will be indicated only if I have referred to more than one. In page
and column references, the following symbols are used:
p page or folium
r, v recto, verso
a, b first, second column
A,B,C ... top-to-bottom divisions of the page, when marked
his Coimbra In Phys.
f ff as usual, but superscripted to avoid confusion
'
I have converted Roman numerals, when they indicate the primary pagina
tion ofa text (as in Buridan's works), into Arabic numerals. Misprinted page
numbers, chapter numbers, and so forth are silently corrected. Line num
bers are separated from page numbers by a period.
Medieval and Renaissance texts, by reason of their manuscript origins
and their mode of presentation, are often articulated into a hierarchy four
or five layers deep. Commentaries are divided by book, sometimes by chap
ter, and then into lectiones or textus, which are short passages to which a
portion of commentary is devoted. Commentaries on the Sentences of Peter
Lombard, a mainstay in the curriculum of the thirteenth and fourteenth
[xi]
Unauthenticated
[xii] Abbreviations and Orthographical Conventions
centuries, are divided, like the Sentences themselves, into books, distinctiones,
qumstiones, and articula, with further subdivisions as required. Many com
mentaries also include qumstiones on standard topics associated with some
part of the work commented on. References to them are by book, chapter,
and question. Such references, I should note, are largely independent of
edition; readers should be able to locate them in whatever text is available.
Names for the parts of texts are abbreviated as follows:
a article, articulum
A,B,C .. . subsection of question (in Ockham)
adl, ad2 .. . subsection of article
c chapter, caput
com commentary, commentarium
disp disputatio
d, dist distinction, distinctio (in Sentences commentaries)
ex exercitatio
lect lectio
lib book, liber
pr proposition
q question, qutestio
resp responsum
text textus
tr tractatus
Thus the passage in Fonseca In meta. that begins at 1:710C of the 1615
edition will be referred to as '4c2q 1§ 1'.
Since the titles actually given to commentaries on Aristotle (and, in the
case of Calcidius, on Plato) vary from edition to edition (and from manu
script to manuscript) and are typically bestowed by their editors, not by their
authors, such works are cited under uniform titles of the form 'In X, where
X is the abbreviated title of the work commented on: e.g., In meta. for a
Metaphysics commentary, In de An. for a De anima commentary. Commen
taries on the Sentences of Peter Lombard are cited under the uniform title In
Sent.
Orthography and Translations

In reproducing printed Latin texts, I have observed the following
conventions:
1. The letters 'u', 'v', 'i', T have been changed to accord with modern
practice. The first letters of sentences and proper names have been
capitalized; other capitalization (e.g., of words like 'Philosophy') has been
Unauthenticated
Abbreviations Orthographical Conventions [xiii]
preseiVed. I have also kept the accentuation occasionally used to distin

guish, e.g., cum meaning 'with' from rum meaning 'when'.
2. Manuscript abbreviations, which are often carried over into printed
editions, have been expanded. Other abbreviations-for example, those of
the names or authors or works-have been left as printed, or expanded with
brackets, as 'Alb[ertus]. Mag[nus].' Philosophus, as a proper name, always
denotes Aristotle; Commentator always denotes Averroes.
3· Obvious misprints have been silently corrected. Where a variant might
be significant, it has been noted.
4· When modern editions are quoted-the Adam-Tannery Descartes, for
example-the orthography and accidentals of the edition are preseiVed.
All translations are mine unless otherwise indicated. The desiderata have
been to preseiVe sentence structure wherever possible and to be consistent
in terminology. Certain words, however, notably ratio and ordo in certain
constructions, require fairly wide departures if the result is to make sense in
English. Brackets are used to indicate insertions. All emphases are in the
original. Since sentences, marked with a period, can extend for haifa page
or more, I have sometimes silently broken them up, replacing colons or
semicolons with periods.
The generic pronoun in the sixteenth and seventeenth centuries was
'he'. To use 'she' in paraphrase or exegesis of authors from that period
would be anachronistic. So too the use of 'she' to denote God. I have used
'she' as a generic pronoun only when writing in propria persona.
Unauthenticated
Unauthenticated
PHYSIOLOGIA
Unauthenticated
Unauthenticated
Introduction
Si les phenomenes ne sont pas enchaines les uns aux autres, il n'y a point de
philosophie.
-Diderot, De l'int1!1j»itation de la nature, 158
T
1. Interpretations of Nature
he physicist who in an idle moment opens Aristotle's Physics is
likely to be disconcerted by the content of that work. Only in the
seventh book will she feel entirely at home, when Aristotle at last
gets down to the business of stating quantitative rules for the
comparison of motions. But the Physics deals with such things only glan
cingly; its primary and essential subject is what Aristotle calls the "principles"
of nature: matter and form, the four causes, natural change, and nature
itself. To those principles the Aristotelian commentaries and cursus I study
in the first part of this work devoted much of their effort. For them physica
covered a little of what we call physics and quite a bit of what we call
metaphysics. The Aristotelians had their own principled distinction be
tween the two: metaphysics was the study of being qua being, physics the
study of corporeal or natural being. But in practice the topics just referred
to were treated by works in both disciplines. The rather fluid boundary
region between them, which is populated not by particular facts or general
izations but by the vocabulary and schemes in which those are cast, is what I
call the philosophy of nature.
It is a somewhat neglected region in the history of science and philoso
phy, especially in the period I examine. 1 Historians have tended to magnifY
the importance of the parts of Aristotelian natural philosophy which corre
spond to the concerns of modern physical science. Large efforts have been
devoted to the medieval theory of impetus and the quantitative rules of the
1. A brief list of general works on Aristotelian natural philosophy in the Middle Ages and
Renaissance is given in §1 below. For the relation of Descartes to the Aristotelians of the
Schools, see especially Gilson 1913, Gilson 1984, and Ariew 1992.
[1]
Brought to you by | University of Warwick
Authenticated
[2] Physiologia
Oxford Calculators, and to the cosmologies of Kepler and Copernicus. The

consequence, perhaps unintended, has been to displace the center of grav
ity of Aristotelianism. Although questions that exhibit some continuity with
what are now regarded as genuinely physical questions were certainly not
unimportant to them, the Aristotelians spent as much, and sometimes
more, labor and ink on the definition of natural change, the intension and
remission of qualities, the existence of substantial form.2 Aristotelian phi
losophy of nature, moreover, contains the principles common to all natural
philosophy, and not just to the part that became our physics. Though the
Aristotelians believed, as many philosophers do now, that physics provides
the foundation of all the sciences, other branches, including the study of
living things, were as well developed. In disputed questions about the phi
losophy of nature, the generation and corruption of living things and mix
tures, and the peculiarities of the human soul, figure as prominently as do
properly physical instances.
An interest in quickening rather than in moribund science may well be
excused. But it does not merely put Aristotelianism in false perspective. It
also inclines the historian to partiality in delineating the task that Descartes
and his peers confronted. One aim of this work is to urge that the profoun
dest and historically most effective part of Descartes's project has to do
neither with method (whose relation to Descartes's practice is at times
tenuous, and which was in any case not the most significant part of his
legacy), nor with the geometrization of nature (a means, not an end), nor
yet with experiment (which Descartes did not make central to his strategies
of persuasion, as Boyle and the Royal Society did later), but with construct
ing, from prime matter upward and from God downward, a functional
equivalent to the Aristotelian philosophy of nature.
To understand how the project was conceived and executed, one must
reinsert it into the· dense and turbulent context of the early seventeenth
century. For Descartes the textbooks and commentaries of Aristotelianism,
his chief resource and competitor, provide much of that context. The com
prehension of his vocabulary will be helped by reference to them. That is
important enough: even where Descartes innovates he starts from the tan
gled, dispute-ridden network inhabited by those terms, and is obliged, if he
would make himself understood, to take it into account. But beyond that, a
second level of significance becomes accessible through the study of the
2. The Coimbrans' commentary on the Physics devotes its entire first volume, some 400
pages, to the first three books (out of eight). In their second volume, almost a third of the 374
pages is devoted to questions about the eternity of the world, its creation, and the existence ofa
creator. Toletus is more evenhanded; nevertheless Physics 1-3 receives 106 out of 245 folia,
with book 8 (the questions about eternity, and so forth) taking up 34 more. In Arriaga's widely
circulated Cursusof 1632, material coming from Physics 1-3 occupies 172 of 250 pages in the
Disputationes physicce.

Authenticated
Einführung
textbooks: the marking of territory in the intellectual landscape accom

plished by asserting a systematic array of positions. Not all of those positions
need be explicitly stated. To affirm that matter is res extensa rules out, for
example, the Thomist thesis that matter is pura potentia. It also leads to grave
discord with the Catholic dogma of the Eucharist. Not for nothing did the
Theological Faculty at Louvain censure in 1662 the assertion that in bodies
there is only "motion, rest, place, figure, and size," which "appears to sub
vert the Holy Sacrament of the Altar. "3 The sense of the intellectual position
a philosopher chose to take up or created, like the sense of social position
conveyed by one's choice of dress, cannot be grasped except by understand
ing its implications, explicit or implicit, in relation to those of other posi
tions seriously tenable at the time.
The philosophy of nature, in fact, was a kind of clearinghouse in which
physics, metaphysics, and theology could meet and negotiate their claims,
much less needed now that those disciplines have gone their own ways.
Historians of ethics have described the project of moral philosophy in the
seventeenth and eighteenth centuries as one of devising a secular ethics.
The civic ideal of tolerance promoted endeavors to devise a theory of moral
obligation requiring few presuppositions about God and his works. But even
though philosophers eventually succeeded in divorcing ethics from theol
ogy, the forms ethics took in the interim bear the impress of the particular
religions and religious disputes they were struggling to accommodate them
selves to. Similarly the project in natural philosophy begun by Descartes and
others eventually resulted in a secular and unmetaphysical physics. Al
though the result now bears few signs of its gestation, its earlier stages can be
understood only in relation to the particular religious context that sur
rounded them. The content of specific generalizations, like the law of fall
ing bodies or the finite velocity oflight, may itself not require that context to
be understood; but the implications such generalizations were thought to
have for the philosophy of nature do.
In the two parts of this work I will trace a number of themes. The first and
most basic is that of natural change and agency. Aristotelianism understood
natural change as the expression of potentialities. That expression could be
frustrated, naturally in some instances, only by miracle in others. But it is
always directed, always from one state to another. It is also always directed in
the sense of proceeding from an agent to a patient. Sheer becoming has no
place in Aristotelian physics; to change is to be changed. The history of an
individual, then, is ordered by the relation of its potentim to their actualiza
tions; among individuals there is a second order determined by relations of
3· DuPlessis d'Argentre Collectio 3pt2:303b. The third condemned thesis reads: "Extensio
Corporis est attributum ejus naturam essentiamque constituens," with a reference to PP 1§53
(ib. 304a). The Rector of Paris issued a similar condemnation in 1691 (gpu:149b).

Authenticated
Physiologia
agency and patiency. Those relations were definitive of natural kinds. Heat,
for example, is the most active of elemental qualities, dryness the most
passive. More generally, substantial forms are classified first according to
what they do, and only secondarily according to what can be done to them.
The second theme is that of the structure of material substance. Matter and
form, the two principles ofsubstance, were ramified by the Aristotelians into
a rather complicated layering of prime matter, quantity and dispositions,
substantial form, and active powers. A host of questions in the commen
taries lay out the definitions of and distinctions among these components.
In those questions it is clear that the positions taken had not only to save the
phenomena but to achieve a balance between what one might call the
Manichaean and Platonist extremes. The material world had to be, on the
one hand, not so divorced from God that it could not share in his goodness
or submit to his rule; it had, on the other hand, to be not so dependent as to
be deprived of genuine agency altogether. Questions of structure lead
sooner or later to questions about power.
The third theme is that of finality. Aristotle, notoriously, included ends
among the four causes laid out in the second book of the Physics. By the end
of the sixteenth century, however, Aristotelians had come to limit the scope
of the final cause in a number of ways; in particular, it can act on inanimate
things only insofar as they are instruments of rational agents. But they do
not, and could not, abandon finality altogether. It is embedded in the very
notion of natural change; it is essential to distinguishing per se causes, which
alone admit of scientific understanding, from per accidens causes, or what
Aristotle calls 'chance'; it is the ordering principle not only of the histories
of individuals but of nature as a whole.
Descartes and the other proponents of the new science regarded the
Aristotelian entities, especially prime matter and substantial form, as not
only superfluous but incoherent. Yet the textbooks show that, however gra
tuitous such entities may seem from an empirical standpoint, they could be
defended on a priori grounds. Even the egregious doctrine of real accidents
founders not in contradiction but in obscurity. But even if a priori argu
ments are set aside, matter and form were applicable to a vast range of
phenomena, celestial, meteorological, biological, and psychological. If, as
some philosophers of science have argued, some explanation proceeds by
way of unification, then Aristotelianism, by its standards at least, had ex
plained a great deal.
Descartes could have ignored the Aristotelian definition of change, its
notions of matter and form, its appeal to ends, had he been content to
remain within what was called mixed or middle mathematics, under which
his early work in music and optics would have been subsumed. But he

Authenticated
Introduction [5]
wanted to refashion not just those subordinate disciplines, but the master
discipline itself. Le Monde, his first major attempt at an interpretation of
nature, and above all the Principles were intended to supplant Aristotelian
physics. He had, therefore, explicitly to confront the themes I have just laid
out. To define matter as res extensa had, no doubt, the epistemological
advantage of permitting the admirable methods of geometry to be applied
to natural things. But because it forced, in Descartes's view, all other acci
dents of matter to be mere modes of extension, it also had the effect of
radically simplifYing the structure of material substance. There had already
been a tendency, visible in Suarez, to ascribe to matter alone such accidents
as quantity and figure, along with the qualities that "disposed" or "propor
tioned" matter to receive substantial form. Active powers, on the other
hand, remained with form-at least those that could not be explicated in
terms of dispositive qualities. Descartes goes much further: powers are re
duced to dispositions, and dispositions, which in Aristotelianism are con
stituted by but not reducible to primary qualities, are reduced to "configura
tions" of quantity. The avenue to that reduction is simulation: if a
configuration answers to the phenomena, if the world of Le Monde appears
to its inhabitants as this world appears to us, then a physics of res extensa and
its modes has been proven sufficient.
A further consequence of the reduction is to eliminate, at least prima
facie, agency from material nature. In Aristotelian physics, figure and quan
tity are prototypically passive. For that reason the nominalists, though they
held that matter and quantity are not distinct, did not propose also that
qualities be reduced to figure. That would be, as the Aristotelians reiterated
in their criticisms ofatomism, to deny agency in nature altogether. In Carte
sianism, the identification of agents and patients becomes a matter of con
venience, because there are no active powers and because motion is relative.
But paring away active powers, however attractive an economy it may have
seemed, had its price. The trouble brought by simplifYing a traditionally
asymmetric notion of causal efficacy, and evidenced in Leibniz and Mal
ebranche, came partly because Cartesian physics yielded only symmetrical
interactions. Although Newtonianism did reintroduce attractive and repul
sive forces, by then the philosophical damage, to which Hume's doubts
testify, was well advanced.
The directedness of natural change, and with it the contrast of potential
and actual, are likewise banished in the Cartesian restriction of natural
properties to figure, size, and motion. Even in Aristotelianism, figure is an
anomaly among qualities, in part because, unlike hot and cold, it provides
no ground for attributing tendencies. Figure is changed only as a side
effect-per accidens, as the Aristotelians would put it-consequent upon

Authenticated
[6] Physiologia
other, directed changes. Cartesian matter, moreover, is, from an Aristotelian

standpoint, at every instant entirely actual. Hence even at the most basic
level, that of particular changes to individuals, the Aristotelian concept of
natural change fails to apply.
Directedness also fails to apply, at the highest level, to nature as a whole.
The Aristotelian maxim "Nature does nothing in vain," because it attributes
to nature an intrinsic economy, has no place in Cartesian physics. The ap
pearance of economy must be an accidental consequence of the operation
of bodies according to laws, or else referred not to nature but to God. The
absence of directedness might even go so far as to rule out providence.
Leibniz was quick to criticize the suggestion in the Principles that the uni
verse might proceed through all possible states: it would ruin, he said, any
notion of progress. Indeed the relation of God to the world in Cartesianism
is curiously static, limited to the creation of matter endowed with a fixed
quantity of motion and thereafter to conservation, entirely uniform, of its
existence and concurrence in its operations.
Descartes does, of course, appeal to notions of perfection. Perfection
figures largely in the causal proofs of the existence of God in the Discourse
and the Principles. The perfection of nature as a whole serves in the fourth
Meditation as a tentative excuse for God's having given him an imperfect
soul. But in such uses the notion is divorced from becoming: substances, for
example, are more perfect than attributes, but substances are not attributes
made perfect. Hence even when Descartes uses a notion that had been
imbued with finality, it seems purged of its associations, leaving only a fixed
hierarchy of modes of existence.
Desmond Clarke begins his work on Descartes's scientific method by
proposing to treat Descartes as a "practising scientist who, somewhat unfor
tunately, wrote a few short and relatively unimportant philosophical essays"
(Clarke 1982:2). That is, of course, an exaggeration. It is closer to the truth
to treat Descartes as what his contemporaries would have called a physiolo
gist: in our dialect, a philosopher of nature. His metaphysical forays, even
his reluctant theological interventions, are all in keeping with the scope of
physics as one finds it in early seventeenth-century textbooks. Far from
being unfortunate or inappropriate to a "practising scientist," they are en
tirely in keeping with the profession of a natural philosopher. More than
that, the epistemological concerns- that philosophers now often take to be
central in Descartes's work are, I think, subordinate to his philosophy of
nature. If a revolution in method had been his sole ambition, we would have
had a second Ramus, not a Descartes. Ramus hoped to supplant the Aristo
telian dialectic and rhetoric with his own. Descartes not only devised a new
method, but a new interpretation of nature.

Authenticated
Introduction [7]
2. Aristotelianism
In keeping with a number of recent works on Descartes, then, I emphasize
the specifics of the Aristotelian context in which he was educated and
against which he defined himself. 4 Renaissance Aristotelianism is a complex
phenomenon. A history even of interpretations of the Physics or De anima
would require a volume to itself. To keep this book within reasonable
bounds I have restricted my attention to a small group of central texts. Writ
ten by university professors for the required series of lectures on the Aristo
telian corpus, these works include running commentaries on Aristotle's
text, or qucestiones on more or less standard topics suggested by it, or both. 5
The qucestio, a form inherited from the Scholastics, was originally the
record of a disputatio or formal debate on a given topic. 6 It began with a
question, usually in yes-or-no form, like 'Does prime matter exist?'. In the
most traditional version, found in Thomas's Summa, for example, what
followed would be a statement of the wrong answer, marked by phrases like
'videtur quod' ('it seems that'), with one or more arguments for it. Next
would come a brief statement of the right answer, marked by phrases like
'sed contra' ('but against this') or 'respondeo' ('I answer'), accompanied
traditionally by an authority or two to back it up. The heart of the question
was next: a laying out of distinctions required for understanding the discus
sion, and a series of conclusiones, with arguments, which together comprise
the true view. Sometimes-in the Scholastics almost always-the question
would conclude with refutations of the arguments for the wrong view.
The form could be elaborated in various ways, but the basic framework
persists. It has the advantage of clarity. Arguments pro and con, conclusions,
objections are often numbered; refutations follow the order of the argu
ments they refute. One could no more mistake one's place in the discussion
than mistake the nave for the apse in a Gothic church. The disadvantage of
the form for the modern reader is that in large doses it is tiring to read, and
4· Throughout this work the term 'Aristotelian' will denote the Western Christian philo
sophical tradition that was predominant in the Schools and elsewhere from the thirteenth
century until the mid-seventeenth century. More specifically, it will denote the later part of that
tradition, from around 1550 until its dissolution. Phrases like 'of Aristotle' or 'in Aristotle' will
denote Aristotle's own texts. I should note once and for all that the Aristotle here presented
makes no pretension to be the real Aristotle. The views attributed to him are, except where
otherwise indicated, those he was believed to hold, and not necessarily those he actually held.
On the use of the term 'Aristotelian', see Grant 1987. I am inclined to agree that 'Aristotelian'
denotes a papulation in something like the sense that word has in evolutionary biology, rather
than a species having a fixed essence. But it is not clear what will seiVe as the analogue to
reproductive isolation.
5· On textbooks in Renaissance Aristotelianism, see Brockliss "Aristotle," 1981a, Brockliss
"Philosophical teaching," 1981b, Grafton 1981, Brockliss 1987, Schmitt 1988.
6. On the earlier history of the qut11stio and the disputatio, see Marenbon 1987.

Authenticated
[8] Physiologia
an author's systematic views will be parceled out among responses to dozens

or hundreds of qua!stiones. In commentaries, moreover, qua!stioneswere scat
tered at more or less conventional sites along the way. In Physics commen
taries, for example, a standard question on natural limits to quantity was
typically put in the first book, long before the machinery needed to answer
it has been encountered in the Physics itself.
Here the systematic rearrangement of qua!stiones according to topic in
Suarez, Eustachius, and John of St. Thomas shows its advantage. Such
works, often called cursus, came in the seventeenth century increasingly to
take the place of commentaries like those of the Coimbrans and Toletus. 7
The new arrangement made it possible for an author, if he so chose, to
incorporate extended discussions of the new science. In the central texts,
however, there are limited signs even of sixteenth-century developments.s
The jesuit teachers, following the declarations of the Council ofTrent, were
preoccupied with the production of texts that would counter the teachings
of schismatics and that would provide a firm philosophical foundation for
Catholic theology. Later on, works like Eustachius's Summa, because they
presented an accessible Aristotelianism, gave to the opponents of the
Schools the advantage of a definite target.
The central texts were written to serve the teaching of Aristotle in the
universities. The curriculum of the first three years of study consisted in a
systematic exposition of the Logic, the Physics, De generation et corruptione, De
anima, the Metaphysics, the Nicomachean Ethics, and selected other works in
natural philosophy, including De ca!lo and the biological texts. 9 We may
lament the rigidity of the curriculum, but we can only envy its coherence. A
Physics course could presuppose knowledge of the Logic, a De anima course
that of the Logic and the Physics, and so forth. Toletus, for example, can
presuppose in his Physics commentary the terminology that would have been
laid out in a Logic course. In reading a commentary, one must keep in mind,
then, that terms not explained on the spot are likely to have been explained
7. Such cursus continued to be produced for use in Catholic seminaries into the twentieth
century, receiving new impetus from the encyclical JEtemi Patris promulgated by Leo XIII in
1879 (see Denzinger 1976, *3135-3140; Hickey 1919, 1:vi; Steenberghen 1974).
8. There is, for example, an exceptional reference to Vesalius in a question on the eye in
Toletus's De anima commentary; Pomponazzi, who argued in the early 1500s that reason alone
could not prove the immortality of the soul, is mentioned by Toletus in a question on that
subject. The Coimbrans occasionally refer to such near-contemporaries as Scaliger and Fra
castoro. One recent development that is reflected in the central texts is the recovery and
translation into Latin of the ancient Greek commentators, notably Johannes Philoponus and
Alexander of Aphrodisias.
9· On the curriculum of the colteges de plein exercice in France, of which Descartes's alma
mater, LaFleche, was one, during the seventeenth century, see Brockliss 1981 b. On LaFleche
in particular, Rochemonteix 1899 is still the standard reference. The philosophy curriculum is
outlined at 4:21 rr.

Authenticated
Introduction [g]
elsewhere and that often a phrase like per se or formaliter was shorthand for
distinctions or doctrines expounded at length elsewhere.
The coher~nce of the curriculum reflects that of the material. The Aris
totle of the Schools was the product of several centuries of refinement,
whose guiding principle was that all the genuine works of Aristotle were
effectively one, as if written in a day. One finds no trace in Aristotelian
interpretation of any notion of development, no Higher Criticism. The
Categories, now thought to be an early work, are treated on the same footing
as the later Metaphysics or the still later biological works. If two passages seem
to be in disagreement, that cannot be put down to a change of mind.
Instead they must be reconciled. Determined exegesis had long noted and
rendered harmless the "contradictions" sedulously cataloged by Gassendi in
his Exercitationes.l 0 It is indicative of the amnesia in which Aristotelianism
was already beginning to be shrouded that such criticisms were taken se
riously and repeated ad nauseam by the lesser lights of the new science.
The rationale of each work, and thus of each branch of natural philoso
phy, was given by its place in Aristotle's system. Though there were con
troversies about the division and order of the sciences, it was agreed that
each had its allotted place. The structure of natural philosophy, moreover,
reflected that of nature it-self. Toletus's description ofits division is at once a
classification of the sciences and of their objects:
What is contained in natural Philosophy is either about the principles or

about the things which are composed out of them. The Physics is about the
principles of all natural things and their common properties; the rest of
the works are about what is composed out of them. Now what is composed
is either a simple body, which is not constituted from others, or composite
and mixed. If it is simple, it is either incorruptible, like the heavens, which
are treated in the first two books of De cado, or corruptible, like the ele
ments, which are the concern of the last two books. [ ...] As for com
posites, because generation and corruption, and not only they, but also
the simple elements themselves, are common to all, De generatione first
discusses the one and then the others. Of composites, some are inani
mate, some animate. Inanimate composites are treated first, and then
animate. Among inanimate things some are sublime, and are called "me
teors," and occur above us, like winds, rain, rainbow, haloes, and the like.
The books of the Meteors are about them. Some are beneath us in intrinsic
parts of the earth, like metals, stones, which are treated in the Mineralia.
As for animate things, because the soul is ·common to them, it is treated
first of all in the three books of De anima, and then certain things that
proceed from the soul, namely, sleep, waking, youth, age, life, death, and
10. See, for example, the Solutiones contradictionum of Marcantonio Zimara, included in
Averroes In Phys., opera 4:464vff, and In Meta., opera 8:401rr.

Authenticated
[10] Physiologia
the like are treated in the Paroa naturalia. Mter those subjects, animate
things themselves: of which some are animals, some plants. Animals and
their kinds are extensively discussed in the Historia animalium, and in De
partibus animalium. Finally there is De plantis. (Toletus In Phys. "Pro
legomenon," q2; opera 4:6rb) 11
I quote at length in part to show how comprehensive was the philosophy

whose foundations Descartes hoped to overturn. The passage hints also at
the degree to which, from the eleventh century forward, the Aristotelian
sciences had been systematically integrated so as to form a nearly seamless
whole. The resulting natural philosophy had no gaps; it omitted nothing of
importance. To a degree that the Vienna Circle could only dream of, Aristo
telianism already had been a unified science, comprehending everything in
this world and out of it.
Such was the imposing monument that virtually all seventeenth-century
philosophers confronted-perhaps not throughout their careers, but cer
tainly so long as they remained in school. It was more than comprehensive.
Through centuries, it had, with the occasional prodding of anathemas,
been reshaped until its parts gave at least the appearance of decent confor
mity to one another, to theology, to common experience and ancient au
thorities. That much we can say in general about the character that Aristo
telianism would have presented to the young Descartes. Specifics are,
because of his near-obsession with denying any provenance to his thought
but that which traced its principles to God and its phenomena to experi
ence, somewhat difficult to come by. What we know is that at LaFleche the
texts ofToletus anc~ Suarez were taught, that he very likely read Suarez in the
late 162os, that he certainly did read Eustachius, and that he had looked at
Abra de Raconis. The Coimbrans, along with Toletus, he mentions among
the texts he recalls from his school days. Arriaga and Fonseca each are
mentioned once in his correspondence.I2
Though all of my central texts have Aristotle at their core, they are not, as
one might think from reading Descartes, an undifferentiated mass. Of these
texts, Suarez, Fonseca, and Arriaga have the greatest philosophical interest.
The Coimbra commentaries vary in quality, the Physics commentary being a
more polished work than the Logic. The Physics exhibits, by contrast with
Toletus, greater concern with theological questions and more use of patris
tic and Neoplatonist sources, including pseudo-Dionysius and Hermes Tri
1 1. The Mineralia and the De plantis are pseudo-Aristotelian. Other works attributed to
Aristotle in the Middle Ages, like the De causis and the Problemata mechanica, are omitted. The De
causis, an Arabic compilation from Proclus's Theology, received very few commentaries in me
sixteenth century; the Problemata several dozen, mostly by secular autllors. On pseudo-Aristotle
in the Renaissance, see Schmitt 1g86.
12. In addition to the references cited in n.1, see SiiVen 1928, c. I.

Authenticated
Introduction [11]
smegistus. Toletus's Physics commentary is a solid piece of work that exhibits

originality on certain questions. Abra de Raconis occasionally diverges from
most of the other authors: on the relation of action to motus, for example,
he favors a more Scotist view than the jesuits. Eustachius's Summa quadripar
tita, which Descartes called "the best book ever written on this matter," is, to
put it bluntly, not. It is a kind of Cliff's Notes condensation, mainly of the
Coimbrans, from whom Eustachius sometimes takes whole sentences ver
batim. It is extremely sparing in its citation of authorities (though that was
becoming standard in cursus), it often gives no arguments for its conclu
sions, and it rarely considers alternatives or objections. I am inclined to
think that Descartes, who had no patience for details, little regard for au
thority, and an aversion to dialectic, liked it because it was unequivocal,
comprehensive, and short (he regrets that the Coimbra commentaries are
too long for him to use in his project-never completed, perhaps never
started-of a blow-by-blow Auseinandersetzung with Aristotelianism).
Though the Summa is a good listing of many of the statements Aristotelians
thought were true, it is rather less useful for understanding why they
thought those statements were true, or what the stakes were in affirming or
denying them. For that one must go elsewhere.
The exposition ofAristotelian natural philosophy that occupies the lion's
share of this work is synoptic and, despite its length, strictly limited in scope.
It does not purport to be more than an exposition in depth of what the
various figures listed above thought about the themes I am interested in. In
particular I do not attempt to find a line of development over the half
century or so between Toletus and Arriaga, nor do I trace filiations among
these thinkers. The exposition covers, moreover, with such detours as are
needed to make the arguments comprehensible, only material correspond
ing to the first three books ofAristotle's Physics: matter and form, substance,
nature, and motus.
That, as the reader will see, is more than enough. Arguments on motus,
matter and form, finality, and nature were complex and the positions even
tually taken often quite intricate. But it is there that one must locate the
issues upon which Descartes believed, correctly I think, that he differed
from the Aristotelians most radically.
3· Leitfaden
This work has two parts. The first, on Aristotelian natural philosophy, is
structured around the three themes I have mentioned: agency, the structure of
material substance, and finality. I examine the notions of potentia, actus, and
motus and the contrast of active and passive in §2, completing that discus

Authenticated
[ 12] Physiologia
sion in the introduction to matter and form in §3.1. The rest of §3 is

devoted to substantial form, one of the two "incomplete" substances that
make up individual material substances. In §4 I study the other incomplete
substance, matter, and its relations-evidently of great importance to
Descartes-with quantity; §4. 3 is devoted to the role offigure in Aristotelian
physical explanation. The exposition of the structure of material substance
is completed in §5, which issues in a blueprint, so to speak, of ens naturale,
the object of physics. I also look at what was to become a key notion in
Descartes's physics, dispositio. In §6 I take up the appeal to ends and to final
causes in Aristotelian natural philosophy, examining the arguments for
their existence and their mode of causality, and then considering some
instances of their use. The last section, §7, brings together all three themes
in a study of the Aristotelian notions of nature-Nature great, or the
cosmos, and nature small, the essence or defining character of individual
substances. That section concludes with a treatment of the analogy between
artifacts and natural things which Descartes would so heavily promote, an
analogy not without peril for traditional views about the causal efficacy of
natural agents.
The second part takes up the same three themes in Descartes's natural
philosophy. As in Part I, I again start with change. In §8 I examine
Descartes's definition of motion, the laws of motion, and the ontology of
force. The structure of material substance is taken up in §g, where I study
the relations between matter, quantity, and space. In § 10 I conclude with a
brief look at Descartes's arguments against finality and the final cause.
The Descartes that emerges from these investigations is certainly not the
character portrayed at the beginning of the Meditations, who divests himself
/
of his rotten old beliefs, to begin afresh with the I think. But I doubt that
anyone thought he was. The Aristotelians who corresponded with him after
the publication of the Discours do not exhibit the shock of the new (Garber
1988). They disagree with him, certainly, and occasionally there are misun
derstandings, but, as Part I will make clear, in one way or another, they had a
place ready for him in their spectrum of possible philosophical positions.
Later, after the publication of the Meditations, there were harsher confronta
tions, with Bourdin and Voetius. Descartes's works were eventually placed
on the Index of prohibited books, with the notation donee corrigantur-until
corrected. The radical divergence that Descartes thought to obtain between
his thought and that of his predecessors only gradually gained recognition
and force, though already in 1640 his sometime disciple Regius was bran
dishing Cartesian ideas against the professors of Utrecht. 13
Historians have come to realize that fractures in intellectual history are
13. On condemnations ofCartesianism, see Ariew 1994; on Regius, see Verbeek 1992, c.2.

Authenticated
Einführung
seldom clean, seldom uniform. Entities like Aristotelianism are rarely so

homogeneous as their opponents make them out to be. Descartes did be
lieve that the philosophers of the Schools all shared certain foundations-a
metaphor more suited to his philosophy than theirs-and he saw clearly
enough that he disagreed with those foundations. But the "Aristotelianism"
he argued against was, like most such isms, a polemical construction. Even if
he had managed not to agree with them on any matter of importance, still
his views would have been defined partly by their opposition to a specific set
of doctrines.
The attempt to place Descartes, or rather to re-place him, in his intellec
tual setting, has been in progress for at least a century. Already in 1897
Georg Hertling was writing on "Descartes's relations to Scholasticism";
Etienne Gilson's first ground-breaking work and the indispensable Index
cartesio-scolasticus are now eighty years old. And for that matter Leibniz had
already insinuated that Descartes had not left his predecessors so far behind
as he had claimed. What the present work contributes to that project is the
specifics of those relations in one relatively limited, though fundamental,
area of inquiry: not Descartes and Scholasticism-another invention-, or
Descartes and Thomas Aquinas, or Descartes and Aristotle, but Descartes
and the texts through which he and his audience came to know Aquinas
and Aristotle. The difference, as we will see, matters.

Authenticated
Authenticated
[1]
Natural Change
Al istotelianism never doubted that there was a natural order. The

phenomena it studies are in large part the regularities ofeveryday
ife: fire heats, water left alone cools, animals reproduce accord
ng to kind, excess is painful to the senses. There are phenomena
that lie outside that sphere-miracles, monsters, the fortuitous. But the
scientific study of change relies in the first instance on their regular effects.
We know nature first of all not when it is frustrated, but when it succeeds.
The task of natural philosophy is to understand that order, which Aristo
telianism founded on the natural necessity of the progression from what is
potentially so to what is actually so. That necessity was conceived not as the
instantiation of universal law but as that of tendencies to ends. Natural
things are divided into kinds according to those tendencies, or, to use an
Aristotelian term, their active potenti<e. Behind the set of potentice belonging
to each individual lies a substantial form, that at once individuates it, unites
its powers, and determines its destiny. That vinculum of identity, efficacy, and
ends, which Descartes did his best to unravel, together with the preponder
ance of powers over laws, is what sets Aristotelian natural philosophy apart
from modern science. 1
I start in §2 with natural change. Aristotelianism distinguishes intransitive
change from transitive change. Intransitive change, or becoming, was fa
mously, if opaquely, defined in the third book of the Physics in terms of actus
and potentia. Around that definition there arose a controversy whose pri
1. SUiveys of Medieval natural philosophy can be found in Lindberg 1978, Wallace 1981,
Kretzmann eta!. 1982, Schmitt and Skinner 1988, and Copenhaver and Schmitt 1992. Mercer
1993 is a useful survey of key figures, both in the Schools and out. The five volumes of
Anneliese Maier's Studien are indispensable; abridged English translations of a few of the
Studien are in Maier 1982.
[ 17]
Authenticated
Vicaria Dei
mary impetus was an attempt to find a place for becoming in a system of

substances and accidents that had no obvious place for it. The central texts
take note of a reductive view according to which change is nothing other
than the successive states of the thing that changes, and a realist response
that insists on a real distinction between change and the successive states.
Rejecting both, they argue for a compromise: motus is not a distinct thing,
but neither is the difference between the motus and the successive states
solely a matter of conception. The two are "formally" or "modally" distinct:
the mobile can exist without moving, but not the moving without the mobile.
Transitive change introduces a second kind of directedness. Just as the
states of one thing are bound together by the directedness of becoming, so
too are distinct things bound together by the relation of agent to patient.
That relation is not a matter of how we look at things: the Aristotelian thesis
holds that change is in the patient alone. Active powers and passive powers
are thus distinct as well, a distinction essential to defining natural kinds. In
Aristotelian physics active powers characterize substantial forms, and thus
natural kinds; in Cartesian physics there are neither active nor passive
powers, and the classification of substances must therefore take another
path. I will examine the arguments by which the Aristotelians distinguish
active and passive powers, and briefly consider the puzzling "power of re
sistance," in which we find an Aristotelian counterpart to inertia.
In §3 and §4 I turn to what Aristotle calls the principles of natural change:
matter, form, and privation. Matter is what persists, form what changes.
Privation, the third principle, is the state of the thing at the beginning of
change. It is not the mere absence of the form eventually attained at the end
of change, but the contrary of that form. The notion of contrary, I will argue,
presupposes that properties fall into natural groups, or genera, so that
change-with the significant exception of the production of new
substances-is between properties in the same group. At the end of §3.1 I
combine the principles of change with the earlier notions of actus and
potentia to produce what I call the Aristotelian scheme of natural change. A
scheme, I should note, is not an explanatory theory in the current sense. It is
a template for generating suitably described explananda.
The remainder of §3 and the first two parts of §4 examine the two "in
complete" substances that together constitute a material thing: substantial
form and prime matter. In §3.2 I look at arguments for substantial form. In
physics it sexved primarily to provide the unifYing ground for the nonar
bitrarily conjoined powers of natural substances. We will see later that the
phenomena adduced in favor of substantial form were not always handled
more satisfactorily in the new science. In §3.3 I show how the Aristotelians
could speak of form as substance and suggest one reason for the in
comprehensibility of the phrase 'substantial form' to later philosophers.

Authenticated
Natural Change [lg]
The next section takes up similar problems about prime matter. In §4.1 I
will prepare the way for a new interpretation of the Cartesian thesis that the
essence of corporeal substance is extension. That thesis, I will later argue,
though it is often thought to rest mainly on methodological or epistemolog
ical grounds, also accomplishes two tasks in the philosophy of nature. It lays
to rest the Manichaean heresy of a matter capable of acting independently
of, and perhaps in opposition to, its creator; it precludes, at least until
Spinoza, the later heresy of a matter identical to God. However antiquated
such worries may now seem, it is clear from the central texts that a physics
that did not allay them, even if only by implication, would not have been
acceptable to the Church.
In §4.2 I examine the controversy over the nominalist thesis that prime
matter and quantity are not really distinct. Though the central texts deny
the thesis, some of them-notably Suarez-allot to quantity a privileged
place among physiCal properties: it is the "first accident" of matter, prior
even to substantial form. Much of the argument turns on the interpretation
of the Eucharist, an issue that, because it divided Catholics and Protestants,
was the object of particular scrutiny by the Church. Here we have explicit
evidence that Descartes not only knew the arguments in some detail, but
that he was eager to avoid even the appearance of heresy.
I conclude §4 with a survey of Aristotelian doctrine on figure. For the
Aristotelians, figure was perhaps the least important of physical qualities.
Though earlier philosophers like Ockham had insinuated an identity be
tween form and figure, the central texts all regard it as ineligible to serve as a
physical principle. Indeed figure is not only entirely lacking in active power,
it is not even the effect per se of any natural change, but always a by-product
of change whose end is something else. There are, however, in Nicole
Oresme's geometrical representations of intensive quantity, and in the so
called middle sciences of optics and astronomy, hints of a meatier role.
Descartes, as will be seen in Part II, promoted the middle sciences and their
geometrized ontology from their subordinate place among the sciences of
nature.
In §5 I study the causes and effects ofform, and in particular the notion of
disposition. The term dispositio, or 'disposition', has a special role in
Descartes's reworking of Aristotelian physics. It is what I call a transfer term,
used in ways that enable Descartes to speak with the vulgar, but ultimately
divorced from what in the Aristotelians' use had been essential to it.
Dispositio indeed signifies for them what Descartes means by it-the arrange
ment of parts. But it signifies in addition the arrangement of parts to some
end, and in particular to the reception of a form. Of Descartes's use the most
that can be said is that dispositio, like 'machine', denotes an arrangement of
parts that will, given the laws of nature and a certain action upon it, yield

Authenticated
[20] Vicaria Dei
known effects. Its finality, merely apparent, lies not in the thing itself but in
the intention of the philosopher who, constructing it hypothetically, wishes
to explain those effects, or in the intentions of the divine artificer who made
it.
The last three sections take up the notion of end. The Aristotelians be
lieved that both individual things and nature as a whole act toward ends,
and they defended, though with serious restrictions, Aristotle's inclusion of
ends among the four causes (§6). The restrictions amount to the require
ment that every attribution ofan end must eventually lead back to a rational
agent that recognizes the end as good. Inanimate things have ends by virtue
of being divine instruments. In contrast with Descartes, the Aristotelians did
not take the analogy between art and nature to its limit. Inanimate things,
though they are instruments, are not thereby reduced to the condition of
artifacts, which the Aristotelians, like Descartes, held to be devoid of activity
or stolidm, as the Coimbrans say. Instrumentality is not-so Suarez argues at
some length-inconsistent with genuine efficacy. It is therefore within
God's power to endow his instruments with active powers; we, on the other
hand, being limited to moving bits of matter here and there, cannot.
What is perhaps most striking, once one understands the Aristotelians'
arguments, is that they would agree that the consequences Descartes draws
out of his principles, and in particular from the definition of matter as res
extensa, are validly inferred from them. If, for example, the only accidents of
material substance were modes of extension, then material "forms," being
nothing more than figures and magnitudes, would not be natures as the
Aristotelians understood that term. If substances without rational souls liter
ally were artifacts, instead ofjust being similar to them in figure, then they
would indeed lack active powers. Most of the views Descartes adheres to,
and many of his arguments against substantial form, real accidents, and the
other entities he rejected, are to be found in the Aristotelians themselves,
but with the crucial difference that an Aristotelian will take the conclusions
of such arguments to be reductiones ad absurdum of the very same premises
that Descartes would have us believe are certain.

Authenticated
[2]
Motus, Potentia, Actus
F or the Aristotelian, common sense, authority, and reason con

curred in affirming two compendious maxims about the world
around us. According to the first, there are not continuous grada
tions among things, but distinct kinds, marked off from one an
other by differences in structure and patterns of behavior. "What could be
more known to us," writes Suarez, "than that the sun shines, fire heats, and
water cools?" (Suarez Disp. t8§t~6, opera 25:594). Such truths are so obvi
ous that merely to point out that a doctrine would have the effect of denying
them was sufficient to prove its absurdity.
The other maxim concerns the changes that help us to discern natural
kinds. The Aristotelians, seconded again by common sense and authority,
distinguish within the goings-on around us a class of specifically natural
changes. Natural change has two notable features: once begun, it proceeds
spontaneously; an~, at a certain stage, it ceases spontaneously. This is most
evident in the actions of animals. But even among inanimate things there
are many kinds of events that seem to have an inherent terminus and that can
sometimes be said to be interrupted by other events, and so to be in
complete. A stone, if nothing hinders it, keeps falling until it reaches the
ground and then ceases to move. Water cools off to a point and then no
more. It is an indication of how thoroughly the concept of nature has been
transformed since 16oo that we are not struck by the ubiquity of the pattern:
things have powers that, when circumstances are right, are triggered, oper
ate, and cease. To that pattern Aristotelianism was exquisitely attuned.
The regularities we find in nature govern natural change so understood.
Such regularities will admit of exceptions. Stones naturally fall to the
ground, but sometimes an instance of falling will be interrupted by another
natural change. Humans give birth to humans, but once in a while the
[21]
Authenticated
[22] Vicaria Dei
formation of a human may be frustrated through the action of celestial

agents. Still it remains true that the complete or perfected action of human
seed issues in a human being and that the existence of monstrous births is
derivative upon that of natural births. The same applies generally: the rare,
the prodigious, the deviant presuppose an order that is not only regular but
normative, the standard against which they are measured and found
wanting.
The two maxims support each other. 1 The changes characteristic of natu
ral kinds are just the natural changes they undergo. More precisely: if we
take the explanatory principles of particular occurrences to be the natural
kinds whose existence is affirmed in everyday experience (and confirmed by
more careful study), then the occurrences that are, to use an Aristotelian
expression, first of all and per se explicable by reference to them are just
those changes that exemplify the order also affirmed by experience.
Distinguish, first of all, between the inception of a change and the change
itself. The spontaneity of natural change is a spontaneity of the change itself
and of its cessation. The Aristotelian does not suppose that the inceptions of
a thing's changes can be referred to that thing's nature alone. Self
movement, or what Averroes called the vis initiativa, is in fact the delimiting
feature of animate things. But however a natural change begins, it seems to
proceed and cease in a regular fashion without any further intervention
from outside. The entire change, once begun, can be explained in terms of
the natural kind to which the subject of that change belongs.
Consider, on the other hand, a change which, though resembling a natu
ral change up to some point, is interrupted, or which misses the mark, as
Aristotle says of monsters (Phys. 199b4), or which has no regular point of
cessation. The entire change, if it can be explained at all, must be explained
in terms of the natural kinds both of its subject and of whatever interferes
with it. But in general the class of things that can interfere with a certain sort
of change seems quite arbitrary. The causes of monstrous births include too
much or too little seed, too much heat or too little space in the womb, vivid
imaginings by the mother, and unusual configurations of celestial powers
(§6.4). Those things would seem to have nothing in common, except of
course that they can all interfere with human generation. Although there is
some hope for a theory of the normal process of birth, there is no hope for a
theory of monsters as such.
Taking directed change to be the primary explanandum in natural philoso
phy thus brings together the two pieces of commonsense wisdom. What the
philosopher seeks, in attempting to understand the goings-on of this world,
is to redescribe them so that their ends are apparent or, failing that, to
analyze them as conjunctures of goings-on that can be redescribed. The
1. In the following discussion, I have drawn on Waterlow 1982 and Burnyeat 1981.

Authenticated
argument just made suggests a scheme2 for that redescription: in any natu
ral change suitably described, there will be a subject, the mobile, a point of
inception, the terminus a quo, and a point of (natural) cessation, the terminus
ad quem. This section, together with §3.1, has as one of its aims the develop
ment of this scheme. In §2.1 I begin with Aristotle's definition of motus. In
that definition the directedness of natural change is not so much proved as
presupposed. 3 Change starts from potentia, which can be glossed roughly as
capacity or power, and ends in actus, the manifestation or exercise of poten
tia. It is to those key terms that I will devote most of §2 .1.
But the start and end ofchange are clearly not all there is to change. What
of the change itself? It is, it would seem, neither the terminus a quo nor the
terminus ad quem, but something distinct from both: neither Paris nor Rome,
but the way between. The central texts summarize, in questions of some
complexity, the traditional standoff between nominalist and realist posi
tions, and offer a kind of compromise between them, which I will develop in
§2.2.
The definition of motus is silent about the inception of motus. It concerns
the change itself only. Yet Aristotle, unlike some of his successors, believed
that every natural change must be at once an intransitive becoming and a
transitive changing and being changed: no inception without an inceptor.
The scheme outlined above must be elaborated: "in Physical change all
these are found: an agent, a patient [...] , and furthermore an acquired
form, and a way or medium by which it is acquired (Toletus In Phys. 3C4q 1;
opera 4:84ra). The "way" is the motus; the "acquired form," as we will see
when the basic scheme is completed in §3. 1, is the terminus ad quem and actus
of the change. What is new is the distinction between agent and patient.
Added with them are two other notions: action and passion. To them §2.3 is
devoted.
In an Aristotelian setting, the addition of new terms inevitably raises the
question of whether the entities denoted by those terms are distinct from
entities already acknowledged, and if so how. The central texts, with which
Descartes for once entirely agrees, hold that motus, action, and passion are
not "really" distinct, but only distinct a ratione, one and the same entity
conceived under various relations.
Since, in the more elaborate scheme, motus is referred to two things,
2. A precise sense will be given to this term in §3.1. For now it suffices to obseiVe that a
scheme is not an explanation or an explanation-type; it is instead a form that through specifica
tion yields appropriate descriptions of putative explananda.
3· Waterlow, concluding that "Aristotle's preoccupation with nature and purpose appears
even in his choice of definiendum," refers the choice to a "conceptual bias" (Waterlow 1982:95).
I'm not sure that 'bias' is the right word. As against the world ofEmpedocles or the world of
Plato, in Aristotle's world change is, first of all, explicable and, second, explicable in relation
not to something outside the world but to natures within it. That, I take it, is not a bias-if by
that one means unreasoned opinion-but a choice founded on the premise that a science of
nature qua nature is possible.

Authenticated
Vicaria Dei
rather than one, one may sensibly ask to which it belongs. Aristotle's answer
is that motus inheres in the patient alone. There is a genuine asymmetry
between the agent and the patient in any natural change. That asymmetry,
which in Cartesian physics finds no basis, is essential to Aristotelian physics,
above all because natural kinds are distinguished primarily by their active
powers (§7.3).
2 .1. Potentia and Actus

Philosophers have recently come to treat notions like power and capacity
more hospitably than did their immediate predecessors in the heyday of
ontological economy. 4 But there remains a residue of distaste potent
enough to lead one commentator still to speak disparagingly of "dormitive
virtue" when discussing the use of potentia in De generatione et corruptione. 5 The
notion of potentia has no doubt been sullied at times by its use in "pseudo
explanations and bad theories" (Hussey 1983:xiii). So too, by the end of the
seventeenth century, was the notion of mechanism, and by the end of the
eighteenth that of attractive or repulsiveforce. Such has been the fate of many
successful notions, especially when their scientific virtues are amplified by
the indulgence of authority.
One reason later philosophers rejected Aristotelian notions is because
those notions were, or were thought to be, not just empirically unfounded
or useless, but unintelligible. In later sections there will be several instances
of such claims. Aristotelian definitions were thought to lead from the ob
scure to the more obscure. The authors of the Port-Royal Logique, arguing
that a definition must seiVe to give us "a clearer and more distinct idea of the
thing it defines," add that clarity and distinctness is "lacking in a great many
of Aristotle's definitions" (Arnauld Logique 166). That such definitions, far
from laying controversy to rest, only abetted it, reinforced the criticism. 6
4· See Cartwright 1989, Salmon 1984. One litmus test for "Aristotelicity" is that a philoso
pher should take the laws of nature to depend on the causal properties-capacities, processes,
"influences"-of individuals or types rather than the other way around. Descartes comes out as
spectacularly non-Aristotelian concerning the physical world, decidedly Aristotelian concern
ing the spiritual.
5· See C.J. F. Williams, in Aristotle Degen. (Williams 1982), 138: "As an explanation of how a
thing becomes r; the statement that previously it was potentially F is useless: it is the old tale
about 'dormitive virtue'." I would not argue, of course, that Aristotle is right; but only thatjust
so stories about "passages" (Empedocles) or atoms (Democritus) could not have been obvi
ously better-except, of course, if the only proper answer to the how question is a mechanism.
See §5.3 and §5.4 below.
6. On the role of definitions in argument, see Toletus In Log., Introductio 1c5, opera 1:8. If
potentia turns out to be no clearer and no better known to us than motus, Aristotle's definition
will have sinned against the first law of good definition (see Aristotle Organon, Topics 6,
141a26).

Authenticated
The definition of motus is a case in point. Arnauld and Nicole' single it out,
asking ''Who has better understood the nature of movement by this defini
tion?" (166). Descartes chides the Philosophers for having explained motus
in terms so obscure that he is "compelled to leave them in their own lan
guage" because he "cannot interpret them."7 Before that, Gassendi had
defied anyone to make intelligible the last phrase quatenus in potentia of the
celebrated definition (Exercitationes 1ex6a5, opera 3:134b). Ramus, for his
part, singled out the neologism entelechia. 8 Behind those jibes there was, no
doubt, severe disagreement about the purpose of definitions. But method is
not my concern here. My interest is in understanding the variety of items
comprised under the headings motus, potentia, actus and the view of exis
tence and change that underlies them, and in an eventual contrast of that
view with Descartes's.
The extension of Aristotelian motus is suggested in its remarks preceding
the definition in the Physics. Unlike its cognates in English, motus (which
translates Aristotle's KL VTJUL s-) denotes not just motion in space but several
other kinds of change as well. In the Categories Aristotle sets forth a classifica
tion of the ways in which "being is said. "9 Change, properly speaking, can
occur, for reasons I will discuss later, in only four of the ten members of that
classification. Change of substance is corruption or generation, change of quan
tity is augmentation or diminution, change of quality is alteration, and change of
place is local motion. For the first two, because there is a unique dimension, so
to speak, along which change can occur, there are two terms, one positive,
one negative. Change of substance, I should note, was for a number of
reasons set apart and designated by the term mutatio.IO We thus have the
classification in Figure 1.
Aristotle's definition seems to apply to all such changes, whether they are
natural in the sense described earlier or not. Part of the difficulty in under
standing the definition lies in knowing how broadly the defined notion is to
7· Le monde 7, CE (AT) 11 :39; cf. Regul(J! 12 (AT) 10:426. Fonseca, at least, had anticipated
such objections, as will be seen shortly.
8. "And, by God, you may swear by all the maggots and bookworms of Apellicon, but you
will never come up with a pearl so unfit [aAoyos-] as entelechia to show tlle true genus of physical
motion" (Ramus Schol. phys. 3c2, p81). Apellicon was a philosopher and bibliophile whose
collection of Aristotle manuscripts was removed from Athens to Rome by Sulla in 83 B.C. On
the novelty of entelechia, see Coimbra In Phys. 3c2q1a2, 1:334*.
g. The Categories distinguish substances from accidents, whose distinguishing character is
that they are said to be "in" or to "inhere" in something else. Accidents are in tum divided into
quantity, quality, relation, place, time, position, possession, action, and passion (Aristotle Cat.
4, 1b25ff). On substance, see §3.3; on inherence, §5.1.
10. See Aristotle Phys. 5c1, 225a. The heart of the argument is that "absolute" generation
(generation of substance) is from nonbeing to being: but "non-being cannot be moved, and if
so, generation cannot be movement" (225a26). Fonseca examines at some length the reasons
why generation and corruption are not motus, except improperly with respect to prime
matter-"improperly" because prime matter, being pure potentia is not, properly speaking, tlle
privation of any form in particular (Fonseca In meta. 5C13qg§1, 3:712).

Authenticated
Vicaria Dei
Motus
Category
Positive Negative
Substance Generation Corruption Mutatio
Quantity Augmentation Diminution
Motus
Quality Alteration
proper
Place Local motion, latio
Fig. 1. Classification of motus
be taken. There turns out in fact to be a range of notions of change which

· can be understood in terms of the definition, based on a range of interpreta
tions of potentia, from logical potentia, which amounts to nothing more than
logical possibility, to natural potentia, the primary sense in physics.
Aristotle introduces the definition of motus by asserting three "divisions,"
as the commentators put it. The first and most pertinent is that "there is
something that is in actu alone, and something that is in actu and in potentia."
That division will be the main topic in what follows. The second division is
that of the categories. Since motus does not exist outside of things (2oob32:
motus absque rebus non est), it must, presumably, belong to one of the catego
ries. Which category, however, was far from obvious; I will return to that
question in §2.2. The third and least important division is of things that are
relative: some are relative by virtue of being greater or less in quantity, some
by virtue of being agents and patients. Aristotle here hints at an argument
made later against those who thought that motus was inequality (see 3c2,
20lbt8ff).
The stage is thus set for the celebrated definition: motus is the actus of
what is in potentia,ll insofar as it is in potentia. Interpreters customarily
11. See Thomas In Phys. 3lect2 (Physics 201agff); Maier 1958:3, 32. The translatio nava of
William of Moerbeke, used by Thomas and many after him, reads: "Potentia existentis en
telechia secundum quod huiusmodi est, motus est." An Arabo-Latin version, probably due to
Michael Scotus, and included with Latin translations of Averroes, reads: "Motus erit perfectio
eius, quod est in potentia, secundum quod est tale" (Averroes In Phys. 3c2; 87vb; I should note
that the modern division of the books of the Physics into chapters differs slightly from that of
the Latin translations). The translation included with the Coimbrans' commentary reads: "Eius
quod potestate est, quatenus tale est, actus motus est" (Coimbra In Phys. 3c2, 1:350).
The Aristotelians would paraphrase the definition in several ways: "Motus est actus entis in
potentia, ut in potentia est" (Coimbra InPhys. 3c2q1a1, 1:332*) and "Motus est actus entis in
potentia, prout in potentia" (John of St. Thomas Nat. phil. tlib3CI summa, Cursus 2:288;
Suarez Disp. 49§2'12, Opera 26:go1; Eustachius Phys. 1tr3d4q1, Summa 2:161, with quatenusfor
prout) were the two most common. Descartes's citations are of the second form, but since it was
common property, and might have suggested itself to memory even if he had been taught a
different one, his wording cannot be traced with certainty to any source in particular.

Authenticated
added f:\vo other passages. One, from the same chapter, reads: "Motus is the
aaus of a mobile, insofar as it is mobile"; the other, from the next chapter,
reads: "Motus is the actus of that which can act and be acted on, insofar as it is
such" (Phys. 3c2, 201b5; 3c3, 202b26). 12 The three formulations do not
differ much; Aristotle puts them all forward without quite settling on any
one of them.
The obscurity of the definitions, and the failure of the surrounding text
to elucidate the key term potentia, made them an obvious subject for dispute.
But the definition itself, as was often the case in ostensibly exegetical
debates, was not the real issue. In her study of Medieval explications of the
definition, Anneliese Maier concludes that although philosophers from
Averroes to Buridan all agreed that the definition was "good," their con
sensus was illusory. They all find a way to achieve a nominal concord with
Aristotle. But in fact each of them "begins the explication of the Aristotelian
definition with the analysis of ens in potentia, and each tacitly brings to that
concept his own interpretation of motion" (Maier 1958:s6-57). The same
can be said of the central texts. In questions on the definition, the stakes lie
not in agreeing or disagreeing with Aristotle, who on this as on most ques
tions is right. They lie in understanding potentia, actus, and the mysterious
compound actus entis in potentia.
Start with potentia, which is far more problematic than actus. 13 If potentia
were simply logical possibility, and the difference between potentia and actus
merely the difference between what is possibly so and what is actually so,
then motus would be the becoming-so of that which was not so, but whose
being so implies no contradiction. The Aristotelians indeed recognized a
kind of potentia, which they called potentia logica and is just logical possibil
ity.14 But potentia logica is clearly not the notion appealed to in the defini
tion. Potentia logica does not admit of degrees, and yet Aristotle speaks of
motus as an "imperfect" actus (2o1b32). Nor is there any halfway house
between being possibly but not actually so and being actually so.
12. Coimbra In Phys. 3c2q1a1, 332*.

13. For a thorough discussion ofvarious kinds of potenticeand their corresponding actus, see
Fonseca In meta. gCiq2, 3:51411". One recent discussion of the issue is Freeland 1986. She
concludes that OUVUIJ.L s in the physical sense of'capacity' should be distinguished from ouvaiJ.LS
or TO 8uvaT6v (the possible) in some larger sense that would encompass outcomes of violent or
accidentally conjoined causes, and perhaps even all that is logically possible (85-86).
14. Suarez Disp. 42§3tg; 26:613; 43 ante §1, Opera 26:633, Toletus Opera 2:268. "In which
is called 'possible', two things seem to be included: one is negative, as it were-namely, non
repugnance [i.e., non-contradictoriness] of being, and this is usually called 'the logically
possible', and to it corresponds potentia logica, which is so-called because it does not consist in
any simple and real faculty, but in the mere non-repugnance of the extremes, and thus is
understood in relation to the composition and division by the mind that logic is concerned
with" (Suarez 26:613).

Authenticated
Vicaria Dei
A second kind of potentia recognized by some Aristotelians, especially the

followers of Scotus, and denied by others, was potentia neutra.15 The marble
out of which statues are made can be chiseled indifferently into many
different shapes. Similarly any natural body in its natural place is indifferent
to staying at rest or moving in a circle whose center is the center of the
world. In both cases it is moving neither upward nor downward. Some
Aristotelians, including Suarez, would say that the marble and the body
were in potentia neutra to those states, any of which could be called the
corresponding actus.16
Potentia neutra was used with respect only to natural causes of change. A
block of marble is in potentia neutra to having the shape of Michelangelo's
David provided that some natural, rather than supernatural, cause could
give it that shape. The potentia so defined is not potentia logica. It is, moreover,
plausible to speak of being shaped as a change, and even a natural change,
since it is brought about by natural causes. Aristotle's stock example of the
statue made of bronze and the Aristotelians' of water being heated are both
of this sort. One could, then, take the actus of potentia neutra to be one kind
of motus.
The corresponding notion with respect to divine action is potentia obedien
tialis.17 Some authors use the term to denote the readiness of any created
substance to receive changes that are not naturally possible, as when the
host is converted into the Body of Christ. In that use potentia obedientialis is, as
Fonseca argues, nothing other than potentia logica restricted to supernatural
actus (In meta. goq4§8, 3:537bB). But Suarez, following Thomas, uses the
term to denote the potentia that "created things have toward receiving
certain actus of divine and supernatural power, which by natural causes they
cannot receive. "18 To those who object that if the essence of the soul were
the subject in which grace was received, then all souls-not just human
could receive grace, Thomas answers that the human soul differs in species
from others; being able to receive grace "agrees with the essence of the soul
insofar as it is of such a species." The human soul, but not those inferior to
15. See Fonseca In meta. 9C1q4§1, 3:523; Suarez Disp. 43§d;10rr, opera 26:647rr, citing
Scotus In sent. 2d2q6. Suarez defines potentia neutra thus: "When, therefore, a passive potentia is
not only innate and connatural with its subject, but also has an inclination to the actus which it
can receive by natural causes, then it is most properly and in every respect a natural potentia,
and thus matter is a natural potentia to form, vision to [visible] species or to the actus of seeing,
etc. But sometimes there is a potentia [i.e., potentia neutral which is indeed innate and intrinsic
to its subject, but whose actus is not such as to aim at the perfection of that potentia, or which lies
outside its inclination, even if the actus can be induced by created causes within the order of
nature."
16. See, for example, Disp. 43§4, 14, 17, opera 26:649, 650.
17. Fonseca (In meta. 9Ciq4, 3:521rr) devotes a long question to various aspects of potentia
obedientialis, some of which I will return to in §7·3·
18. Suarez Disp. 43§4116, opera 26:649; cf. Thomas ST2pt1quo~, opera (Parma) 2:439.

Authenticated
Motus, Potentia, Actus [2g]
it, is in that sense in potentia toward grace. Since no natural cause can bring
about grace, the soul is not in potentia neutra toward grace; but neither is it
merely in potentia logica toward grace. 19
In both the heating of water and the reception of grace by the soul
something obviously changes. Yet in neither is there an inclination toward
the final state. The form of water, as we will see, opposes being heated; the
soul untouched by revelation or the light of God knows nothing of grace.
When water spontaneously cools, on the other hand, not only does the
potentia to become cool belong to the water by its nature, but also the actus is
one to which the water, even in the absence of any external agent, is in
clined. We thus arrive at potentia in the most properly physical sense: potentia
naturalis, the potentia correlative with natural change, and which is alone
acknowledged to be a genuine quality in the Categories. 20
Actus, which is sometimes translated as 'actuality', looks far less mysterious
than potentia. There is a metaphysical sense of the word according to which
actus is existence, whether in relation to a potentia or not. The other sense,
which is the only one relevant to the definition of motus, is that in which
actus is existence with respect to some potentia. That, I take it, is what Suarez
had in mind when he wrote that physical actus is so-called "because it actual
izes something, as form is the actus of matter" (Suarez Disp. 13§5 t 8, opera
25:416). The basic notion is not actus but actus of
The definition says of motus not that it is the actus ofa potentia but the actus
of something that "is in potentia." To be an actus is, in the physical sense, to
be "an accidental or substantial form, in succession or enduringly," while to
be in actu is to "participate in a form," or to have that form. To be a potentia is
to be a "principle of acting, or undergoing something," while to be in
potentia is to have such a principle (Toletus In Phys. 3c 1text3, opera 4:78va).
The central texts emphasize that the phrases' ens in actu' and' ens in potentia'
do not differ "as thing from thing, but as the same thing from itself accord
ing to modes of being" (Fonseca In meta. goq1§2, 3:513; Toletus 4:78vb).
19. Suarez notes that the potentia of natural things to be made in to artifacts is sometimes
also called potentia obedientialis. The reason is presumably thatjust as certain effects are beyond
the power of nature as a whole to produce, so some effects-the making of a book, say-are
beyond the power of nonhuman nature to produce (1)7, 26:650).
20. Potentia in the strict sense of the Categories is the "proximate principle ofan operation, to
which it is by its nature instituted and ordered" (Suarez Disp. 43§3'12; 26:644). The operation
in question can be either an action or a passion. The word 'proximate' contrasts potentil1!with
substantial forms, whose activity is always mediated by accidental forms (see §5.4), and with
prime matter, which is pure potentia but only in a broader sense. The phrase 'instituted by its
nature' selVes to distinguish potentil1! from other species of quality, which though they are
principles of operation are not defined primarily in terms of their operation. Such are the
sensible qualities, color, light, and so forth; and the elemental qualities, hot, cold, wet, dry.

Authenticated
Vicaria Dei
The water that is in potentia hot and the water that is in actu hot are the same
water, whether hot or having only the power to become so.
Motus is the actus of a being in potentia. So far so good. But then Aristotle
adds the phrase Gassendi didn't like: insofar as it is in potentia. The
Coimbrans, citing Averroes, note that two entities seem to be touched on by
the definition: "the heat, which is acquired in actu, and the acquisition or
flow [fluxes] of the heat. "2 1 The obvious question is: which is the motus? In
terms used by the Aristotelians themselves, is motus the forma Jluens, the form
itself acquired in passing, or the fluxus formr:l!, the "way" or "tending" of that
form toward another? Each view had weighty authorities to cite on its behalf.
But the central texts tend to favor the second. That question tended to be
taken up in close connection with a second question. Suppose that motus is
indeed the Jluxus and not the form. In an Aristotelian context one still must
decide whether, and how, the jluxus and the form differ. They could be two
things (res), in which case they would be said to be "really" distinct. Or they
could be one thing in every sense, but conceived in different ways. Or,
finally, they could be what was called "formally" or "modally" distinct, in the
way that the spherical shape of the earth differs from the earth itself not
merely as one conception of a thing from another, but as a modification
differs from the thing modified. That question is the topic of the next
subsection. Here I examine the question of whether motus denotes a Jluxus
rather than a form. 22
The common opinion, as I said, is that the entity defined in the definition
of motus is the jluxus, not the form. 23 The Coimbrans, rejecting Averroes'
21. Coimbra In Phys. 3c2q1a1, 1:332 (part of which is included in Gilson 1912:187-188);
Averroes In Phys. 5c3, com9, opera 4:215B. Averroes frames the question as one about the
category to which motus should be assigned (Maier 1958:21ff). Maier concludes that in the
understanding ofAvicenna and Averroes, "the ens in potentia of the original formulation [of the
definition] is no longer an existens in potentia but an exiens [going out, issuing] de potentia in
actum." The shift prepared the way toward interpreting motus as jluxus rather than form. Some
recent commentators argue that such a shift makes the definition circular (Kosman 1969:41;
Maier 1958:23). But it rests, as the Coimbrans' argument shows, on the reasonable assumption
that no form, considered in itself, changes or is a change; in the formulations favored by the
central texts,fluxus is not a process. See Toletus's remark, cited below, distinguishing the "flow"
of oil from the jluxus formO!.
22. On the controversy over forma fluens and fluxus jormfl!, see Maier 1958, c.2; Wallace
1972; Murdoch and Sylla 1978:213-222; Wallace 1981, c.4; Samowsky 1989, §4.2.
23. See Toletus InPhys. 3qq1, opera4:84ra; Eustachius Phys. 1tr3d4q1, Summa 2:161. Abra
de Raconis, citing Suarez, identifies motus as a "successive passion," rather than a jluxus. But
Suarez's position is not quite that motus is a passion (i.e., belongs to the category of passion)
rather than being ajluxus. It is that he cannot see any real difference between them: "illud [i.e.,
the claim that motusdoes not belong to the category of passion] semper mihi visum est difficile,
quia non video in quo constituatur differentia inter passionem et motum, si cum proportione
et secundum completas rationes suas sumantur" (Suarez Disp. 49§2'l[4, opera 26:901). Later
Suarez writes that if one considers motus/passio in abstraction from subject and source, that can
be called a "fluxus," but it is in no way distinct from the passio ("[ 14, 904).

Authenticated
view that motus is the form (which is also supported by Albert and by at least
one passage in Thomas), argue:
Nevertheless the contrary opinion is truer and more common [...] ,

namely that motusis the acquisition itself, or the flux, ofform. [...] Motus
according to its own account [ratio] is not a form per se; nor is it the form
and the way or acquisition together, but the acquisition or tendency to
form itself. The proof of this is that form per se is not change [ mutatio], and
so it is not motus. Again, form and flux together [...] is something com
posed per accidens out of flux and form. Such composites are not the object
of proper philosophical definitions [ ...] Since nothing else is left over, in
which the ratio of motus would consist, it therefore consists in a flux or
tendency (In Phys. 3c2qta1, 1:332)
Toletus likewise concludes that the motus can only be the via ad formam (In
Phys. 3C4q 1, opera 4:84ra,vb). I should note that jluxus does not mean 'flow'
in the sense that oil is said to flow. Forms don't flow in that sense. The fluxus
is, as the next subsection will show, rather a mode of existence that a form
can have, as warm-on-the-way-to-hot is the quality warm existing, so to speak,
toward hot, which is different from being warm-on-the-way-to-cold.
Toletus, more thorough than the Coimbrans, relies on a certain kind of
divisibility peculiar to motus to make his case. There are, he says, four ways in
which motus can be called "divisible" (84rb-va):
(i) by virtue of the extension of the mobile;

(ii) by virtue of proceeding through spatial parts or degrees of intensity;
(iii) by virtue of being the successive acquisition of places, or of qualita
tive or quantitative forms;
(iv) by virtue of occurring in successive instants of time.
Of these four, the third belongs to motus and to motus alone. The first is
simply borrowed from the mobile; the second, because it applies to quantities
or qualities independent of change, also applies to motus only at second
hand; the last is distinct because the successive acquisition of the same
quantity (in growth) or quality (in alteration) can occur over different
lengths of time. The successive acquisition offorms, in other words, cannot
be identified with the succession of instants through which a motus occurs.
Now to finish with the definition. It must exclude what is in potentia X but
not now becoming X, as well as what was in potentia X but is now wholly X in
actu.2 4 The qualifying clause 'insofar as it is in potentia' is supposed to do
24. The phrase entis in potentia (being in potentia) excludes not only all instantaneous
changes, like generation, corruption, and creation, but also "all perfected actus, like maximum
coldness, maximum heat, which leave no power [potestatem] in the subject to acquire any
further part [of the perfected accidental form]" (Coimbra In Phys. 3c2q1a1, 1:333*). Such
actus can only be the termini of change.

Authenticated
Vicaria Dei
that, as Aristotle's subsequent discussion shows (2ota2g-2o1 b5). The actus

in the definition is the "imperfect" actus of hotness in a being that was hot
entirely in potentia and yet is still not yet hot entirely in actu. It is not the
perfect actus of heat in water that has turned to steam, or the actus of the
water qua water before it began to be heated. In the former, nothing is in
potentia hot any longer; in the latter, nothing is in actu hot yet. Water is
indeed the actus of a potentia, as we will see, namely the potentia of matter to
form. But as such it is, like the heat of the steam, entirely in actu, and not in
potentia any longer. There seems, in other words, to be something that is the
actus of something in potentia, so described, when and only when there is
heating, and not before or after.25
The "imperfect" heat in water that is being heated is, I should note,
imperfect or "inchoate," as Fonseca puts it, not merely by virtue of being an
intermediate degree of heat: "For if the heat which is introduced into a piece
of wood is considered solely with respect to its definition, from which fol
lows a certain perfection in this instant, it will clearly not satisfY the defini
tion of motus or of the acquisition of form [...] But since it is taken [in the
definition of motus] in such a way that not only is it viewed with respect to
being introduced but also with respect to being about to be introduced, it
will now satisfY in every way the true and genuine definition of motus. "26 If,
for example, there is an interval of rest in the middle of the heating, only at
the moment of its introduction will the intermediate degree of heat reached
in that interval fully satisfY the definition of motus, since only then will it have
been '1ust introduced."
One can well understand why later philosophers thought that something
that should be obvious was being shrouded in mystery. The discussion has
an air of the via negativa about it. Aristotle at one point remarks that it is
"indeed difficult to grasp the nature [of motus], for it is necessary that one
should put it either with privation, or with potentia, or with what is in every
way and plainly actus, but it seems that motus can be none of these" (Phys.
3C2text15, Coimbra In Phys. 3c2 explanatio, 1:352). Fonseca forestalls that
objection: "First of all [one might object to] the obvious extreme obscurity
of the definition, which is so great, that while a definition should be a thing
notjust equally known but more known than the thing defined, this [defini
25. 'The only time at which the buildability of the heap is essential to what is happening is
when it is getting built into a house [ ... ] So, ifthe definition makes any sense at all, it gives the
right result in the case of building and by analogy in all other cases" (Hussey 1983:62, ad
2o1bsl
26. "Nam, si calor, qui introducitur in lignum, ea tantum ratione consideretur, qua certam
perfectionem in hoc instanti consecutus est, non habebit sane rationem motus, seu acquisi
tionis form.e, sed form.e cert.e perfectionis iam acquisit.e. At, cum ita accipitur, ut non tantum
spectetur, qua ex parte introductus est, sed etiam qua est immediate introducendus, iam
omnino habebit veram & germanam motus rationem" (Fonseca In meta. 503qg§2, 2:719).

Authenticated
tion] brings so much darkness that after it is proposed and we want to

explain what it is, right away we seem to be ignorant of what we knew well
enough before" (Fonseca In meta. 503q9§3, 3:719). His answer is (i) that
motus is well enough known in its species (e.g. running, walking, etc.), but
not in general; (ii) that because it is a sort of mixture of the actual and the
potential (Phys. 201 b33) but with more of the potential, since its terminus, so
long as the motus exists, is in potentia only, and since we know the actus of a
thing before and better than we know its potentia, motus is particularly hard
to know; (iii) it is, moreover, known only by analogy (see Aristotle's remarks
on matter at Phys. 1c7, tgta8ff) and is "variable and inconstant" in nature
(§4, 3:72l).
Grasping what cannot be fixed is no doubt a perennial problem, but for
the Aristotelians it took an especially acute form. The categories of being
established at the very beginning of Aristotle's logic have no place for
change, but only for agents, subjects of change, and the results of change.
The difficulty only grew worse when, in the aftermath of nominalism, the
furniture of the universe was increasingly limited to "individual substances,
qualities, and quantities. "27 In that setting, one could, it would seem, do no
more than to say what change is of, and not what change is. The tortuous
exegetical maneuvers of the Aristotelians, the aporetic close to Aristotle's
own discussion in Physics 3c2, are witnesses to the severity of the problem.
One further point can, however, be gleaned from the Aristotelians'
discussions. Actus, as I said, is existence, but existence viewed in relation to a
potentia from which the actus is thought to issue. It is thus also the perfection
or completion of what had been imperfect or incomplete. The coincidence
of existence and perfection, which is perhaps the most fundamental expres
sion of the teleological orientation of Aristotelian natural philosophy, must
confront the fact that some changes, or what seem to be changes, lead to
imperfection or even, as in corruption, to nonexistence.
Fonseca and the Coimbrans both take note of this, but Fonseca sees the
problem more clearly:
Diminution, or decrease of quantity, and remission of quality are not

inchoate and imperfect actus tending to greater perfection. On the con
trary [they are] perfect actus progressing toward less perfection. The same
can be said of corruption, which does not progress from the imperfect to
the perfect, but from the perfect to the imperfect and in all ways toward
nonbeing. (Fonseca In meta. 5C13q9§3, 3:719)
27. Murdoch and Sylla 1978:215. To their list, however, one must add intrinsic place, or
Ubi. Suarez, for one, argues that Ubi is a genuinely intrinsic property (Disp. 51§1, opera
26:g72ff).

Authenticated
Vicaria Dei
Of the several not entirely satisfactory responses Fonseca offers to the objec
tion, the one that interests me is this:
[To the objection] one may say that it is quite probable that Aristotle did
not want to include in the definition motus that are, so to speak,
deleterious [o/1ectivos], namely decrease, remission, and corruption: [ ...]
because they are not per se intended by Nature, which imitates the condi
tion of its Author (He creates, says Sapientia c. I, so that all things should be,
and not so that they should give up their existence, unless per accidens and
by a certain consequence [i.e., of other changes that are intended by
Nature]). (Fonseca In meta. sc13q9§4, 3:721aC)
The Coimbrans likewise suggest that perhaps such changes "are not true
and positive actus, but merely corruptive," and are therefore not comprised
in the definition (Coimbra In Phys. 3c2q2a2, 1:334*).
One response, then, to the observation that there are changes that do not
tend to perfection is just to deny that they are, properly speaking, motus.
Though all changes, natural or not, are imperfect actus, natural changes
or the actus of natural potentia-are, so to speak, the most perfect among
changes; unlike violent changes, for example, they can sustain themselves
without the constant influx of an external efficient cause. The terminus of
natural change is, by comparison with that of violent change, good not only
with respect to Nature as a whole, but also with respect to the thing that
changes. It is not surprising that Christian philosophy, resonant to a theol
ogy of providence and salvation, and anxious to "overcome," as Blumenberg
puts it, the threat of gnosticism (Blumenberg 1g88, pt.2 c. I), should have
found the agathism of Aristotle, intensified into optimism, to its liking.
2.2. Independent Existence of Motus
The definiendum in Aristotle's definition, then, is the jluxus of the form. But
Jluxus need not signify anything other than the form itself. That question
remains to be settled. It had, as Aristotelian questions often do, a lengthy
genealogy, rather more complicated than usual. A quick sketch of it will be
necessary to understand the terms in which the Aristotelians laid out the
alternatives and the arguments brought to bear on them.2 8
28. What follows is a history of the question as it could have been gleaned from the central
texts. A full history of the question from Avicenna to Blasius of Parma, which quotes many of
the relevant passages, is found in Maier 1958, c.2. Brief expositions are found in Wallace 1981,
c.4 (through Soto), Adams 1987, c.tg (a concise and clear account, relying on Maier, of the
question up through Ockham and Buridan), Sarnowsky tg8g, §4.2 (Albertus Magnus, Bur
idan, and Albert of Saxony).

Authenticated
That there was such a thing as change, and even natural change as Aris
totle understood it, no one doubted. 2 9 Aristotle himself dismisses from the
domain of natural philosophy the arguments of Parmenides and Melissus
(Phys. 1c2, 185a13ff, 2c1, 193a1ff). 'Is there motus?' isnotagenuinephysi
cal question. The genuine question is 'How is there motus?'
The underlying issue is certainly not peculiar to Aristotelianism.
Descartes scholars have argued over a quite similar issue in Cartesian
physics. 3D Though it is often put in terms of the relativity of motion, and
though even in Scholastic discussions one can find philosophers who notice
the perceptual relativity of motion, 31 in the questions I am about to discuss,
a less proleptic way of putting the issue is this: when a thing changes, it has
one state at the beginning, another at the end. 3 2 If the change is uninter
rupted, the same holds of each segment into which the change can be
divided. Hence it is reasonable, if change is indefinitely divisible, to suppose
that at each moment the thing has a state distinct from those it had or will
have. Change therefore entails a succession of states. But is it that succes
sion? And is the change at each momentjust the state at that moment? If so,
then once we are given things, their states, and the order of those states, we
have all we need to describe the world and what happens in it. Change
certainly occurs: in that sense it is real enough, but it is derivative upon
other things.
Such would be the view, I imagine, of most philosophers now. 33 It was
essentially Ockham's, and perhaps Descartes's. But some philosophers have
held that change is something in addition to things, their states, and the
succession of those states. 3 4 The questions I am about to examine concern
the truth of that claim, both secundum Aristotelem and secundum veritatem.
29. Abrade Raconis includes a question on "whether there is in fact motus," in which he
refutes the "sophisms" of Zeno (Abrade Raconis Phys. 240). That seems to have been unusual.
30. For a thorough airing of the issues, see Garber 1988, c.6; I return to the question in
Chapter 8, below.
31. Buridan, afterhe has demonstrated that local motus is distinct from the thing that moves
and the places it moves through, argues that local motus is not called 'local' because locus is
necessary to it but because "it cannot be perceived unless there appears to be change of place
or orientation [... ] just as those who are on ships travelling simultaneously and at the same
velocity do not perceive that they are moving" (Buridan In Phys. 3q7, P5lra). Clearly Buridan
does not believe that the perceptual relativity of motus argues against its being distinct from the
mobile and its termini.
32. By 'state' I mean the set of properties possessed by a thing at a given time. This is not
intended to be a rigorous definition. If so-called Cambridge changes (the change that occurs
to the North Pole when I move ten feet south) are to be ruled out, then some relational
properties will have to be excluded. Adjustments would also have to be made to accommodate
periodic changes. But these would not affect the argument.
33· SoW. H. Newton-Smith writes that "a change is [...] constituted by an alteration in the
properties of a persisting object or by an alteration of the properties of a region of space"
(1g8o:14ff). I take it that most philosophers would now, quibbles aside, find that
unexceptionable.
34· One notable instance is Bergson: "One would have to abstract completely [from the
"cinematographic" view of change as a series of states] in order to dissipate at one blow the
theoretical absurdities raised by the question of movement [...] There is more in a transition

Authenticated
Vicaria Dei
The Physics would have change occur in three categories-quantity, qual

ity, and place-or four if substantial change is included. Yet the Categories
put motus in the category of passio: "Acting and undergoing also admit
contrariety [the implied comparison is with qualities]. Heating and cooling,
being heated and being cooled, and being affected with pleasure or pain
are contraries: and for that reason they admit contrariety. They also admit of
degrees, since they are said to be more or less" (Cat. g, 11 b1 ff; cf. Coimbra In
Log. 1:514 and Toletus In Log., Opera 2: 166a for Latin versions). 35 Worse yet,
the Metaphysics seem to say that it is a continuous quantity (Meta. 5Cl3,
1020a27ff). That passage at least can be dismissed, since it says that motus is
quantity only by way of occurring in space and time, not in itself (see
Fonseca In meta. 5C13q8, 2:7o5ff). But the other two are not so easily
reconciled.
Averroes acknowledges both possibilities. The Categories view, that motus is
a passio, he calls the "more famous." Nevertheless, it is false. The true view is
that of the Physics, according to which motus is the "flowing form itself with
respect to its being acquired part by part [forma ipsaJluente secundum acquisi
tionem partis post partem]. "36
Returning to the question in Physics 5, Averroes introduces a rather un
happy distinction between the "matter" and the "form" of motus, by which
he means the entity that is the motus, and its definition. He restates the
earlier position thus:
It seemed that this being [i.e., motus], which is between actus and potentia
of that genus [i.e., the category of the terminus], should belong to that
genus, since what is intermediate between a potentia and actus which be
long to that genus, is necessarily of the genus of the actus which is the
completion [of the intermediate], differing however according to more
or less. [...] And it is true to say, according to this way of speaking, that
motus is in the genus of that toward which it is motus, according to its
matter. But accor~ingly as motus is a form in that matter [i.e., the so-called
matter of the motus itself], one mustjudge that motus is a category by itself
[i.e., passion]. (Averroes InPhys. 5c1com9, Dpera4:214L-215B; cf. Maier
1958:65 and Adams 1987:803)
than a series of states-that is, of possible cuts-, more in movement than the series of
positions-that is, of possible stopping-points" (L'euolution criatrice, c.4, Bergson 1959:760).
35· One chapter of the so-called Postpr(l!dicamenta in the Categories is also devoted to motus.
The chapter shows that (i) the contrary of motusgenerally is quiesor rest; (ii) alteration, though
it cannot occur except if local motion or augmentation also occurs, is nevertheless distinct
from them (Coimbra In Log. 1:563~"). It figures in the questions I am discussing only byway of
providing one argument to show that motus does not belong to the category of passio (Suarez
Disp. 49§2,2, opera 26:901).
36. Toletus In Phys. 3c3q 1, opera 4:82vb; cf. Averroes In Phys. 3com4, opera 4:87D, which is
quoted in Maier 1958:63.

Authenticated
Motus according to its "matter" belongs to the category of its terminus, al
though according to its "form" it is a passion. Certain later authors took up
that distinction to propose a reconciliation of the two views, 37 but most
philosophers were not inclined to let the matter rest there. By the end of the
thirteenth century two well-defined positions had been marked out: the
reductive view of Ockham and the realist view argued for by, among others,
Buridan, the Thomist Capreolus, and Paul of Venice.
Ockham's position was that motus is in no way distinct from the terminus.
In his words, "Motus is not something else in addition to enduring things
[rebus permanentibus]" (Ockham Summula 3c5 and 6, opera philos.
6:262,265). In order for a thing to be said to move, "it suffices that the mobile
should continually and without interruption or rest, acquire successively
and part by part one thing after another or lose something." Since the
arguments he gives are to some extent taken over by the Aristotelians on
behalf of their view, I omit them here.
The realist view is that "motus is distinguished in re from the terminus
acquired by it" (Coimbra In Phys. 3c2q3a1, 1:340). The strong flavor of
that claim can be tasted in the following passage from Buridan. 38 Buridan
is answering the objection that if motus-in this context, local motion
were really distinct from the thing moved, then "God could, having an
nihilated them, separate and separately conserve the motus without the
mobili and the terminal place, which seems implausible since then there
would be motus and nothing moving" (Buridan In Phys. 3q7, psora). His
reply:
I say that it is no more a problem that there should be motus and nothing
moved or changed than that there should be whiteness and nothing being
white. Neither is naturally possible, both are supernaturally possible. But
of [the objection] that for there to be local motus without place is a
contradiction I say that the motus of the outermost sphere or of a ship in a
river is called 'local' not because it is necessary that according to [that
motus] they should be changed in place, but because according to the
common course of nature all that is moved by this motus in fact varies in
local or postural habitation [i.e., in place or in orientation]. Yet every
37· Maier 1958:84f (on John ofJandun, who agrees however that the first view is truer), 95
(on William of Alnwick); cf. Toletus 83rb, which mentions the view without ascribing it to
anyone. The formal/material distinction is sometimes used in a logical sense to distinguish a
definiendum from the subject in which it inheres or which it informs; what is 'formal' in that
sense need not be a form in the physical or metaphysical sense. Cf. Chauvin Lexicon s.v.
'Formale'.
38. On Buridan's view, see Maier Igs8:I17-133; a brief summary of this is found in Adams
1987:824ff; Sarnowsky Ig8g:137rr. On all motus except local motus, Buridan agrees with
Ockham.

Authenticated
Vicaria Dei
motuswe call 'local' could fail to be local insofar as no place or orientation

would be changed to some other. (51ra) 39
Although Ockham argued that ifmotuswere really distinct from the terminus
there could be heating without the form of heat, and from that concludes
that motus is not really distinct, Buridan argues that since everyone agrees
that there are real accidents, which can exist apart from their subjects
(§4-1), there is nothing untoward in supposing a motuswithouta mobile. Nor
does all local motus require a change of place. The outermost sphere, for
example, which contains every corporeal thing and every place, cannot
change its place relative to anything else, since everything else moves with it.
Yet it can and does move: it rotates with unimaginable speed around the
earth's axis, and it could move in a straight line.40
The central texts, then, arrive at the end of a rich tradition. The conclu
sion typically reached in them is a compromise between the realism of
Buridan and the reduction practiced by Ockham. I should note that in
these conclusions the word terminus may denote any intermediate form or
place during a change. 41
Motus is not really, but only formally distinct from the terminus it aims at.
(Coimbra In Phys. 3c2q3a2, 1:341)
Therefore let this be the first conclusion. That way [i.e., toward form] and
motus, with respect to three of its species, namely alteration, augmenta
tion, and generation, along with their opposites, is not really distinct from
the terminus or the form [acquired in passing]. [...]
Second conclusion. Local motus is not really distinct from the mobile which
moves. [ ...]
39· The following question (3q8) argues a similar point at greater length: it is not necessary
that a local motus should have, in addition to its fluxus, a terminus d quo and a terminus ad quem.
Every rectilinear movement, however, and every movement caused by heaviness or lightness
does have extrinsic termini (52ra).
40. See In Phys. 3q7, conclusioprima. This was Buridan's most telling example, since among
the propositions condemned by Etienne Tempier, the Bishop of Paris, in 1277 was the proposi
tion "that God could not move the heavens in a straight line. And the reason is because this
would leave a vacuum" (see Hissette 1977=118ff, no. 66; Denifle and Chatelain 1889, no. 49;
Buridan refers to the proposition at sova). No movement of the heavens or outermost sphere
could consist in the successive acquisition of new places, because the heavens contain all the
places there are. This is true whether 'place' is defined in terms of circumambient bodies or
intrinsic Ubi. The heavens might seem to be a unique instance: but for any body God could
annihilate all bodies except that body, and thereby make that body as if it were the outermost
sphere.
41. "If by the total [movement of] heating eight degrees of heat are acquired, the terminus
of the entire heating will be those same eight degrees; while to the single parts of the heating
there correspond the single parts of the heat, namely those which by the single parts of the
motus were acquired" (Coimbra In Phys. 3c2q3a2, 1:341).

Authenticated
Third conclusion. Nevertheless motus and its terminus, or local motus and
the mobile are distinct in reason and in definition. (Toletus In Phys. 3c3q3,
opera 4:87ra-b)
I. Conclusion: Motus does not really differ from the terminus toward which
it tends. [. . .]
II. Conclusion: Motus is formally distinct from its terminus. (Eustachius
Phys. 1tr3d4q2, Summa 2:163)
The arguments against a real distinction are, by and large, those of Ockham
and other nominalists; the arguments in favor of a formal distinction are
suitably modified versions of the realists' arguments. 4 2
Now for the arguments. There are two cases: the distinction between the
motus and its (intermediate) termini; the distinction between the motus and
the mobile. Against a real distinction between motus and terminus, the texts
offer essentially two arguments. The Coimbrans argue, following Ockham,
that "if motus differed in re from the terminus, it could be divinely conseiVed
without it." But motus is just the acquisition of the terminus, and it would be
contradictory to suppose that it could exist without the terminus. The second
argument is a regress: if motus existed separately from the terminus, it would
itself be caused by some action. That action would amount to a second motus
and require yet another action, and so on, ad infinitum. Every motus would
comprise an actual infinity of motus, which is impossible. 43
So much for the real distinction. As for the formal distinction, many things
can be asserted of the motus that cannot be asserted of its termini: that it is fast
or slow, successive, has parts (Coimbra a1, 1:340). The motus, moreover,
exists before its termini do, since it is the way toward them. The Coimbrans, I
should note, do not hold that a difference in properties entails that two
things should be entirely distinct. They may be discernible even if one is
merely a mode of the other, as acquisition or tendentiais ofform (a2, 342). 44
42. Suarez is exceptional in holding that motus and passio are identical except in reason
(only Avicenna had taken that position before him; Abrade Raconis is the only seventeenth
century text that agrees with him). But except for that, he too agrees that motus is not really but
only formally distinct from the terminus.
43· Coimbra In Phys. 3c2q3a2, 1:341 *; condensed by Eustachius Phys. 1tr3d4q2, Summa
2:163; cf. also Suarez Disp. 49§2'115-6, opera 26:go2. For the regress argument, see Toletus In
Phys. 3c3q3, opera 4:87rb.
44· "Respondemus in eorum sententia, qui motum nihil aliud esse, quam ipsam formam
fluentem opinantur, non distingui motum a termino, nisi penes diversum modum eiusdem
essentia::; est enim motus eademmet forma, qua: acquiritur, eademque essentia, nihilo a se ipsa
plus differens, quam quod aliter ac aliter se habeat [cf. Buridan In Phys. 3q7, sova], prout ab
esse imperfecto ad perfectum progreditur [...] At iuxta aliorum opinionem, quam statuimus,
arbitrantium motum proprie, ac formaliter esse ipsum fluxum, seu progressionem ad formam,
dicendum erit motum distingui essentia atermino. Enim vero motus secundum suam naturam
est quidpiam successivum [...]; forma autem, qua: acquiritur [...] non ita se habet, etiam
interim dum acquisitionem subit; quia si tunc secundum suam essentiam foret successivus,

Authenticated
Vicaria Dei
It follows readily from the second conclusion that motus is really distinct
from the mobile, contrary to what Ockham had argued. As we will see, the
Aristotelians held that accidents are really distinct from the substances they
inhere in (§5.1). Since motus is a mode of the terminus, and the terminus is
(except in generation, which is excluded from these arguments) an acci
dent of the mobile, it is distinct from the mobile to whatever degree the
terminus is distinct.
One case, however, requires separate treatment: change of place, or latio.
Latio is indeed really distinct from the mobile, for the reasons I have just
mentioned. But because the place of a thing, which is the outermost con
taining surface of a body, is only formally distinct from quantity, latio too is
only formally distinct from quantity. 45 If, as Ockham famously argued, quan
tity does not differ from the thing that has quantity, then latio turns out to be
only formally distinct from the mobile.46
The authors of the central texts, though they were not nominalists, were
certainly cognizant of the nominalist view. They agreed with Ockham that
motus is not something in addition to res permanentes, if by that is meant
something really distinct from them. But they were also well aware of the
difficulties, some of which I have mentioned, that had been urged against
Ockham by Buridan and others. In the formal distinction, which they inher
ited from Scotus, they found the middle way they were looking for. Motus is
not a res, but a mode of a res, which consists not just in being one stage in a
succession, but in being a stage that is about to become something else. Motus,
like any mode, depends on something else for its existence; nevertheless it
exists in things, independent of the ways in which we conceive them.
2.3. Action and Passion

The definition of motus does not appear to presuppose that for every motus
there must be not only a subject in which that motus inheres, but an agent or
utique etiam post acquisitionem, finito motu eius natura in successione consisteret, quod Ionge
aberrat averitate": ''We answer the view of those who hold that motusis nothing other than the
forma fluens itself by saying that motus is not distinct from the terminus except insofar as it is a
different mode of the same essence. Motusisjust the same as the form which is being acquired
and the same essence, in no way different except that it is different at different times-namely,
it proceeds from imperfect to perfect existence [...] Concerning the opinion of the others,
already mentioned, according to which motus is properly and formally the fluxus itself, or the
progression to the form, it should be stated that motus is essentially distinct from the terminus.
Motusaccording to its nature is something successive; but the form which is acquired is not such,
even while it is being acquired, since if it were successive according to its essence, it would also
be successive after it was acquired, and its nature would consist in succession after the motuswas
over, which is very far from the truth" (Coimbra In Phys. 3c2q3a2, 1:342).
45· Aristotle Phys. 4C4· 212a5f. Cf. Ockham Q in phys. 72, opera philos. 6:5g8.
46. Ockham Q. in phys. 14, opera philos. 6:430. On quantity and matter, see Summula 1c13,
opera philos. 6:191 n· and §4.2 below.

Authenticated
an inceptor of that motus. Yet Aristotle and the Aristotelians do not doubt
that there must be. 4 7 Toletus, echoing Aristotle's final words on motus in
Physics 3, writes that "motus is twofold: it is a certain actus, but from one it
proceeds, as from its principle: in another it is received, as in a subject. "48 It
is, or entails, both action and passion.
The distinction between acting and being acted on is already presup
posed in the definition of 'nature', and a distinction between agents and
patients in the explication ofthe four causes in Physics 2. An efficient cause
is "that from which comes the first beginning of change or rest, [...] and in
general the agent is the cause of what is done, the changer of what is
changed" (194b29-32). Out of that, perhaps, one can extract a reason, if
not an argument. If there were no agents, there would be no "first begin
nings" of change, and no changes. An Aristotelian, then, could no more
doubt that there are agents and actions than that there is change in the
world, and of that there can be no doubt: it is "so evident from the perpetual
vicissitudes of things [...] that to prove it by argument would be
superfluous. "49
Among Aristotelians, that is perhaps all that needed to be said. They seem
not to have worried much about the relation of the intransitive motus
defined in Physics 3c2 and the transitive motus presupposed throughout.
Common experience showed, in any case, that in most if not all changes
there is in fact an agent and a patient. We thus arrive at a scheme for the
description of change more elaborate than that suggested in the definition
of motus. 'These five are to be considered," Toletus writes, "in every action:
the agent, the patient, the form which is brought about, the Jluxus of the
form, and the various respects or relations consequent upon them" (In Phys.
3c3q2, opera 4:86rb).
One indication of the degree to which the Aristotelians found the scheme
of transitive action essential is the effort some of them make to apply it even
in unpromising cases. Suarez, for example, accepts the standard distinction
between immanent and transeunt action. A transeuntaction is one "that has
47· Waterlow notes that "Aristotle cannot or will not recognize such a thing as intransitive
change that is neither an acting upon nor a being acted upon" (1982:170). According to Emile
Brehier, the Stoics did. He writes that "when knife cuts flesh, the one body produces upon the
second not a new property but a new attribute, that of being cut. [... ] This manner of being
exists in a way at the limit, the surface of being, whose nature it cannot change: it is truly
neither active nor passive, since passivity would suppose a corporeal nature that undelWent
some action" (Brehier 1928:11-12). This notion, new to philosophy, Brehier identifies with
our notion of 'event' or 'fact'.
48. Toletus In Phys. 3c3texu8, Opera 4:81 va; cf. Phys. 3c3, 202 b26, paraphrased thus by the
Coimbrans: "motus est actus eius, quod agere, & pati potest, quatenus tale est." This was
quoted earlier as one of the "definitions" of motus. Physics 3c3 does not argue the equivalence
of this and the earlier definitions; it presupposes from the start that in every motus there is not
only a mobile but a motor:
49· Disp. 18§101[5, 25:681ff; 12§3'l[2, 25:388.

Authenticated
Vicaria Dei
an effect outside the agent itself," while an immanent action is one that "has
no effect outside the agent. "50 The question he sets himself is this. In a
transeunt action-the sun heating a rock, say-the agent brings about a
motus in something else; that motus has a terminus that is, since it is outside
the agent, distinct from any property of the agent. The warmth of the rock is
evidently not also in the sun. The terminus is said to be the terminus of the
action as well as of the motus. In immanent action, on the other hand, since
there is no effect outside the agent, it is not obvious that there need be a
terminus, or that if there is, it must be distinct from the action. Suarez
therefore asks whether every action requires a terminus.
There was, to judge by his account, some agreement that immanent
actions are qualities of the potentice in which they inhere: desire, for exam
ple, is a quality of the will; cognition a quality of the intellect. The question
was whether immanent actions just are such qualities, and so not really
actions, or whether desire, say, is the terminus of an interior motus of the
will.51 To make a long-though by Smirezian standards rather brief-story
short, the answer is yes. After arguing that God could produce the quality of
loving without the aid of the will, and that the quality could be produced in
several ways, each ofwhich would constitute a different action, he concludes
that desire is (also) an action, "because it is a tendency toward a quality, and
by its nature distinct from it" (Disp. 48§2'l[ 13, opera 26:877).
Contrary to what many Thomists thought, and what some commentators
now believe, Suarez holds that in what Aristotle calls EVEPYELa (Meta. gc6,
1048b23ff), as well as in what he calls KLVT]OLS', there is a terminus that must
be attained if the change is to be called complete ('l[2o, 87g).52 The
difference between desiring a cake and baking it is not that desiring is not
an action or without a terminus, but that the terminus-the quality of soul
which consists in that desire-lasts only so long as the action does ('l[ 21,
879). The scheme, but for that one anomaly, suits immanent actions as
nicely as it suits transeunt actions.
50. Suarez Disp. 48§2, 1, opera 26:874; cf. Toletus In Phys. 3qq2, opera 4:8svb. Toletus,
however, defines an immanent action as one that does not exist outside its "effective principle,"
which is to say the potentia of the agent. The standard examples of immanent action are acts of
will or acts of the intellect, like contemplation.
51. Suarez's representative for the view he opposes is the Thomist Soncinas (Q, meta. 9q21,
P254blf).
52. 'EvepyELa is usually interchangeable with EvTD.EXELa. The Metaphysics passage is the
only place where Aristotle makes explicit the distinction between EVEpyELa and KtVT]ULS', al
though there are other texts that seem to call for it. One modem interpreter who emphasizes
the passage is Waterlow 1982:184. Fonseca notes, and a look at the commentaries of Thomas
and Averroes confirms, that this passage is missing in the earlier Latin translations (Fonseca In
meta. 9c6 explanatio, 3:647; see also the translator's note in Averroes In Meta. 9 ante com 11,
opera 8:235H). It had no role, therefore, in Medieval philosophers' interpretations.

Authenticated
Suarez's was far from the only view. 5 3 But his opponents, those who held
that immanent actions are not actions at all but instead qualities, typically
did so on the grounds that immanent actions have no termini. They did not,
therefore, depart from the scheme. They only disagreed about its
applicability.
From the scheme of transitive action the central texts drew three conclu
sions. Together those conclusions show that agency and efficient causality
differ significantly under the Aristotelian conception from their counter
parts in Cartesian or classical physics. The first is that motus is in the patient
alone. The second, which follows in short order from the first, is that the
agent is impassive in action except per accidens. (That consequence is
consistent with the claim in De generatione that every agent is acted upon in
return [1c7, 324b]. In reaction, the agent is not moved qua agent, but qua
patient of the coincident re-action of its patient.) The third consequence is
that motus, action, and passion all inhere in the patient and are at most
formally distinct from one another. 54
1. Motus is in the patient. Since motus is a mode of its terminus, and so must
inhere in a subject, and since in the scheme of transitive action two things
take part, it makes sense to ask to which of them the motus belongs. Aristotle
answers unequivocally: motus inheres in the patient (Phys. 3c3, 202a13f).
But since motus is the actus of a potentia, the implication is that the actus of
the agent's potentia inheres in the patient.
Naively one might wonder whether it isn't absurd to say that the actus of
one thing is in another (202b6). The quick answer is that it all depends: an
actus is "of' whatever its potentia inheres in, "not because it always inheres in
it, but because it either inheres in it or is from it" (Toletus In Phys. 3c3q2,
4:86vb).
But the objection can be pushed further. There do seem to be actions,
like walking for one's health, that perfect the agent. More generally, in De
ca!lo Aristotle writes that each thing is "on account of' its operations
(286a8). Thomas therefore calls operation the "ultimate perfection" of
things (In de Ca!lo 2lec4; opera [Parma] 1g:87). But a thing must change if
it is to be perfected, and so some actions must change their agents.
Among self-perfecting actions, some are immanent. For them agent and
patient coincide, or are parts of the same whole (as when one chooses to
53· Fonseca agrees with Scotus, Thomas, and various Thomists that intellection, which is an
immanent action of the rational soul, has no terminus and is therefore a quality, not an action
(Fonseca In meta. 7c8q3§3, 3:3ooaD).
54· Every relation between two things must rest on nonrelational accidents of those things,
which are called the fundamenta of the relation. My being taller than Tom Thumb rests on my
being such and such a height and Tom's being such and such a height. See, for example,
Coimbra In Log. 1:465.

Authenticated
Vicaria Dei
walk for one's health), or are at most formally distinct. Such actions present
no problem. But Aristotle's claim is general: even transeunt actions in some
way perfect their agents. Yet that cannot follow from the nature of action
itself, or else God, some of whose actions are transeunt, would be perfected,
which is impossible.
2. The agent is impassive. Toletus agrees that "in general the actus of natu
ral agents are also perfections [of those agents] , or some sort of way toward
their perfection" (4:86vb). The only exceptions he makes are for agents
already perfect, like the "perfect craftsman" or God. Yet he has earlier
concluded that transeunt action "is in the patient per se, but in the agent per
accidens"-either because it is from the agent, or because the patient reacts.
So it seems that without some further explanation, the most one could say is
that the agent may be perfected by its actions, not that it will be.
Suarez addresses that point. Transeunt actions of created agents bring
only a "certain extrinsic perfection," either because the external things they
perfect in turn act on them, or because they preserve themselves in acting
(which covers the case of nourishment). The perfection of the agent in
acting, then, is properly that of something else, and only denominatively
that of the agent (Disp. 48§4~ 18; on 'denominative', see below) .55 It under
goes no change in acting transeuntly.
The impassivity of Aristotelian agents reinforces the original asymmetry
in transitive action: motus is in the patient alone, the agent moves but is not
itself moved. Here we see a distinctly unmodern element in the Aristotelian
theory of efficient causation. In Cartesianism, and in classical physics gener
ally, there is no agent that is not acted upon in acting. 56 Or rather: because
the ground of that distinction-the fact that the motus inheres in the pa
tient alone-is effectively denied, the very ideas of acting and being acting
upon lose their grip on the world.
3· Action, passion, and motus are not really distinct. The third conclusion
drawn from the scheme of transitive action is that action, passion, and motus
not only inhere in the patient alone but are at most formally distinct from
each other. Opinions differ about the precise kinds of distinction to be
55· The Coimbrans rescue Thomas's claim by shifting the sense of 'perfection': "That
transeunt action also is a perfection of created agents is clear from the fact that, because the
greatest embellishment of creatures is to imitate the First Cause, they excel when in acting they
communicate something ofthemselves to others; as Dionysius [...] says, the most divine thing
of all is to become a co-operant with God [omnium divinissimum esse Dei cooperatorem fieri]"
(Coimbra In Phys. 2c7q q, 1:281; cf. Dionysius Coel. hier. 3§2, Opera 156: "divinius est omnium,
ipsius etiam Dei [...] cooperator existat, divinamque in semetipso demonstret operationem,
quoad potest, elucentem": ''The most divine thing of all is that it should exist as a co-operant
with God himself [...] , and in its very self show forth the shining divine operation").
56. For a survey of action and reaction which includes some Renaissance Aristotelian
authors, see Russell 1976.

Authenticated
drawn in the three cases. Suarez, for example, holds, unlike most other
authors, that motus and passion are only distinct in reason, or even just in
name (Suarez Disp. 49§2!4, Opera 26:904). But since a detailed census of
views would not, I think, yield further insight into the Aristotelian concep
tion of agency, I will not undertake one.
It suffices to note that the central texts (with the exception of Abra de
Raconis) all reject the Scotist contention that "transient action, regarded
formally, is the emanation of the form from the agent," and is therefore
"received not in the patient, but in the agent." The Scotists' argument is that
action differs from passion just by virtue of being a relation of the agent to
the patient. But that relation is in the patient, since its fundamentum is the
potentia of the agent (Coimbra In Phys. 3c3q1a1, 1:347). In plainer terms:
action-many philosophers now would agree-is a relation; that relation
holds just because the agent's potentia is manifested in the motus of the
patient. So the relation, even though it is between the agent and the patient,
resides in the agent.57
To that the Coimbrans reply that to distinguish action from passion it is
enough to say that the motus, with respect to its source, is action; and with
respect to its subject, is passion. For the rest, Toletus's quick way suffices: the
ground of the relation of agent to patient is just that the actus of the patient
is from the agent, not in it (1:349; cf. Suarez Disp. 48§2120, Opera 26:873).
The arguments are abstruse, as often happens when Scotus is in the
wings, but they bring to light two points. The first is that action and passion
are not relations in the modern sense. A logician now might represent
'Socrates teaches Plato' by the formula 'Tsp'. Asked what the "action" of
Socrates' teaching Plato is, she is likely to reply that it consisted in Socrates'
occupying the first slot in the (asymmetric) two-place relation T, or, more
formally, in his satisfying the open formula 'Txp', where xis a variable-name.
The Aristotelian tends to think in reverse fashion. There is a motus, a teach
ing, construed nonrelationally; Socrates' action consists in its being with
57· A relation, contrary to what anyone trained in logic now would think, inheres, like any
other accident, in one subject: "A relation is a being [ens], whose whole being [esse] is with
respect to another [ aliud respicere]." The "peculiar and proper nature of relation, which
distinguishes it from other [categories], is constituted by its tending to another as to a termi
nus," while the other categories have their essence in themselves (as substance does) or in a
subject (as quality, for example, does). My resemblance to a dog is an accident of me, whose
fundamentum is, let us say, warm-bloodedness, and the ratioofthat fundamentum is the identity of
warm-bloodedness in me and in the dog. Scotus's claim, then, is that what we call 'action' is an
accident of the agent whose fundamentum is the active potentia whose actus is the motus, and
whose ratio is the identity of that actus with the actus of the passive potentia in the patient, which
is also the motus. No more than anyone else in his period would Scotus have dreamt of calling
action a relation in the sense in which the word is now used.

Authenticated
Vicaria Dei
respect to him-the subject in whom the motus does not inhere-that the
motus comes to exist.
The second point is that motus is the actus of the patient quite differently
than it is the actus of the agent. It can be, and in natural change is, not only
an accident inhering in the patient but a perfection of the patient. However
obscure Aristotle's definition may be, it is reasonably plain what that defini
tion is about, so long as we consider motus with respect to the patient. It is
about processes like growth, nourishment, the descent of falling bodies
changes in their respective subjects. But it is not at all clear how a process
going on in something else can be thought to be the actus of a potentia in the
agent (or of the agent existing in potentia). Suarez at one point says that in
acting the agent "expresses or reveals [declarat] its perfection" (Disp.
48§4'118, opera 26:893). Though he goes on to say that it does so best in
immanent actions or in converting something to its use, the root notion is
that of a manifestation ofpower. It is true, we may suppose, that in doing so an
agent assimilates itself to God, and thus to the ultimate good. But if that can
be sensibly thought of as a kind of becoming, of being-perfected, at all, it is,
one would imagine, only tenuously related to the genuine becoming that
occurs in the patient.
What holds the two manners of becoming-perfect together, I think, is
some notion of unfolding, of coming fully into existence. In the patient that
unfolding is instigated, and governed by the ends, of another. Patiency is
therefore, however much it may be an expression of some part of one's
being, a kind ofsubordination. In the agent, on the other hand, that unfold
ing is both spontaneous and autonomous. In successful transeunt action, it
amounts to mastery over the patient.
2-4- Active and Passive PotentitE

Every change, being transitive, is the actus of two potentia, one in the agent,
one in the patient. Except in self-action, the potentite belong to distinct
things, and are therefore numerically distinct, just as your soul and mine,
though one in form, are distinct. But are active and passive distinct classes of
potentite? Or is every active potentia also passive, and conversely? Since the
distinction of matter and form, and with it the classification of natural
substances, rests on the distinction between active and passive potentice, the
question is far from idle (see §3.2 and §7.1 below).
It was important also to know whether active and passive included be
tween them all potentite. From the point of view of the new science of
Descartes, the most significant case was potentia resistendi, sometimes called

Authenticated
potentia resistiva. Some Scholastics had argued, for example, that change
takes time because the passive potentia of the patient to receive an action
must overcome a contrary potentia to resist it. 5 B The central texts were
inclined to show that potentia resistendi was either passive potentia under
another name or else not really a potentia at all. The basic pair suffice: the
distinction of active and passive is, if not disjoint, at least exhaustive.
In general the difference between the two classes was uncontroversial.
The Metaphysics simply defines active potentia as the "principle of change in
another, as other," passive potentia as the "principle of being changed by
another, as other" (gc2,1046aiiff). The only difficulty is the phrase 'as
other'. Aristotle added it, according to Thomas and many others, to cover
the case of self-action, in which the agent and patient, though numerically
one, are, under the designations 'agent' and 'patient', as if distinct. 5 9 It
seems to have been obvious that the classes so defined are distinct. We have
seen some examples from the Physics. In De generatione et corruptione the
elemental qualities hot and cold are active, while dry and wet are passive
(2c2, 329b24). De anima argues, among other instances, that sensation is
passive, while the faculty of movement is active.
Suarez, nevertheless, considers the question. Since action and passion are
not really distinct, and since agent and patient sometimes coincide, one
might well wonder if active and passive potentia are really distinct classes.
Suarez holds that they are. The argument has two parts. In one Suarez
considers whether the classes are coextensive; in the other whether each
potentia taken by itself is both active and passive. I will detail only the first.
There are, he argues, purely active, but no purely passive, potentice. The
class of active powers properly includes the class of passive potentice. Experi
ence shows that some potenticealways act transeuntly and do not intrinsically
require the action of anything else in order to act (Disp. 43§21J[ 1). Among
them are the attractive power of magnets and the explosive power of hot air.
Since such potentice do not undergo their own action, they need not be
supposed passive for that reason. Nor does their action presuppose that
they themselves are acted on by something else. Skills too are purely active,
since once learned, they need not be acted upon in order to act. There is no
58. Maier cites a formulation of the principle in William of Auvergne: "[ ...] every opera
tion of this sort [i.e., that takes time] occurs through conflict and a victory of the agent over the
patient and of the efficient power [virlus] over its contrary [...]" (Maier 1955:227, n.1).
59· See Thomas In Meta. glecu, §1776-1777· Averroes reads the passage differently: Aris
totle adds 'as other' to the definition of 'active potentia' because "it is manifest that nothing acts
on itself' (In Meta. gc2, p227ra[B]). Although it would be out of place to argue the point,
Thomas's reading is preferable, since Aristotle explicitly says of animate things that they move
themselves (e.g., at Phys. Bq, 255a6).

Authenticated
Vicaria Dei
reason to suppose that such a potentia is other than active, since it is changed
neither by itself nor by another.
So there are purely active potentice. What of passive potentice? Suarez agrees
that if potentia is taken in the broad sense (Disp. 43§3'12), there are purely
passive potentice. The potentia of matter to receive form is passive. But among
potentia in the strict sense, every passive potentia is also active. Suarez's strat
egy, in brief, is to insist on the definition of potentia. 60 A passive potentia is not
merely any accident that enables a thing to receive a form. It must be
"ordered" or "instituted" to that form, and Suarez interprets that strictly.
Wetness and dryness are said to be passive (De gen. et carr: 2c2, 329b24-25).
But they are passive qualities, not passive potentice: "first because [Aristotle]
holds them to be [also] active qualities, although not as efficacious as cold
and hot, in comparison with which they are said to be "passive."[...] Finally
because dryness and wetness are not the sort of qualities that in themselves
receive other qualities or forms by which they are actualized-which is the
character [ratio] proper to passive potentice-but instead dispose matter for
the reception of substantial form, or also other accidents, and according to
that reason [sub ea ratione] are called not passive potentice but passive
qualities, belonging to the third species" (Disp. 43§2'15, 26:639). 61 What
matters here is that although any accident by which a thing is enabled to
undergo the action of another is, in the broad sense, a passive potentia, the
proper sense is much more restrictive. The only unequivocal instances ad
mitted by Suarez are psychological: sensation and the passive intellect. But
even these are not purely passive, since every "vital" power includes some
degree of activity. All putative instances are passive, because they are not
ordered by their nature to the reception of a particular form, referred to
other species of quality, or-in the case of quantity-to other categories.
In the Metaphysics and (for some interpreters) the Physics, a simple picture
is suggested: for each active potentia there is a corresponding passive poten
tia. 62 Rene Le Bossu, the would-be conciliator of Aristotle and Descartes,
6o. "A potentia is the proximate principle of some operation, to which by its nature it is
instituted and ordered" (DMp. 43§3,2; 26:644). Cf. n.2o.
61. According to Suarez, passive qualities, the third species of quality in the Categories, are
called "passive" not because they are principles by which a thing undergoes the action of
another, or receives its form, but primarily because they are forms of a sort which are received
through passions and because they make a thing receptive to various actions (Disp. 42§4, 14).
Toletus's account is somewhat different: qualities of the third species are called "passive" either
because they cause passions (and are therefore active powers) or because they are the result of
passions (In Log., Prcedicamenta 8, opera 2:156b-157a). What matters here is that if they are
qualities of the third species, then their passivity is incidental to their nature. The four species
of quality are discussed below (§4.3).
62. Thomas, referring to Meta. gc 1 (lect2), writes that "an active potentia would be in vain if
no passive potentia corresponded to it" (De potentia Dei 1art2, ps). Since no natural active
potentia could be in vain, each must have a corresponding passive potentia.

Authenticated
Matus, Potentia, Actus
writes for example that "if a cannonball has the Active Power to overturn a
wall, the wall has likewise the Passive Power to be overturned [... ] So that
Movement is equally the Act of a Being that has an Active Power, and of a
being that has a Passive Power" (Paraltele o6, p214). Late Aristotelianism
does not in fact have so simple a view, and indeed would deny that being
able to be overturned by violent motion is a passive power. Suarez, arguing
that not every active power has a corresponding passive power, concludes
that "it is not necessary that, [even] if active potentia were a quality instituted
per se to act, passive potentia would also be a quality instituted per se and first
of all to receive [form]. Heat, as an instrumental virtue of fire, acts imme
diately in the substantial potentia of prime matter [i.e., its potentia in the
broad sense to receive form], drawing out a substantial form from its poten
tia; but as the principal virtue of heating, it acts on the body insofar as the
body is affected with quantity [i.e., it acts on the body just because the body
has length, width, and breadth], and no other peculiar passive potentia
corresponds to it" (Disp. 43§21 12, 26:641). Quantity, which is not ordered
toward any particular quality and which is always present in corporeal sub
stance, suffices for a thing to be in potentia hot. More generally, passive
powers, ordered by their nature to the reception of a form, are supplanted
by equivocal dispositiones, which, though "proportioned" to the forms whose
reception they enable, are not defined solely in relation to them. To that
degree, matter-by which I mean all that is needed for the reception of
form-becomes plastic, defined less in terms of ends, more in terms of
"material" qualities lacking teleological implications.
'Active' and 'passive', then, are not contrary. Far from it: when potentia is
taken narrowly, 'active' includes 'passive'. But do the two exhaust the field?
Experience shows that bodies do not always change readily. Every medium
resists more or less the motion of things through it (Phys. 4c8, 215a25ff).
Materials variously resist change of shape (De gen. et corr. 100, 328a36-b5;
Zabarella De rebus nat. 436D). Air resists the action of fire (Soncinas Q. meta.
gq6, p234b). Some philosophers inferred that there is, in addition to the
potentia passiva or the dispositio which assists and receives the action of an
agent, a potentia resistendi opposing it. 63
One argument for distinguishing potentia resistendi from active and passive
potentia was that resistance to motus was neither action nor passion. 54 It is
63. Neither Soncinas nor Suarez mentions any names. Zabarella cites Pomponazzi (De
reactione 5, in De rebus nat. P436D). The most important case was the resistance of a medium to
local motion, since it figured in Aristotle's mles for the comparison of motions (a general
discussion is found in Maier 1955, c.4; for the views of Albert of Saxony and others in the mid
fourteenth-century Paris School, see also Sarnowsky 1g8g:211 ff; for those ofMutius Vitelleschi,
ajesuit professor at the Collegia Romano in the 1590s, see Wallace tg8t:Ji6-t20).
64. Soncinas Q. meta. gq6, secunda ratio; p234.

Authenticated
[so] Vicaria Dei
contrary to both. 65 Soncinas's answer is to argue that "being active and

being resistive are relative, since they are said in relation to something,
[and] nothing is active or resistive in itself'. So coldness in a patient, insofar
as it lessens the heat of something acting on it, is active; but insofar as it
retards the action of the heat, is resistive. Zabarella rejects that answer. His
proposal is that each form not only acts on its contrary (as heat on cold, and
cold on heat) but strives to conserve itself. It impedes the action ofwhatever
would change it, where 'impede' is understood simply to denote a diminu
tion of that action rather than a contrary action. Diminution is not a positive
thing; it is a "privation," a nonbeing. Privations do not, in general, require
distinct causes, and the self-conserving nisus of forms is thus not a potentia
(Zabarella De rebus nat. p44oE).
Suarez likewise rejects Soncinas's answer. With Soncinas he notes that one
thing may react against the action of another, thereby diminishing the
active power of the other. Resistance, properly speaking, is not, however,
reaction. As Zabarella says, it is simply a diminution or retardation of the
agent's action, and thus mere privation. The reason for it is that any form
refuses the forms and actions contrary to itself (formr£ et actioni sibi contrarir£
Jormaliter repugnat; Disp. 43§2110, 26:636). Hence even relatively passive
qualities like dryness resist being turned into their contraries.
Soncinas makes potentia resistendi a respect under which active power can
be considered. This is of course one common way to trim unwanted entities.
We have seen it at work already in the treatment of action and passion. But it
fails, as Suarez quickly notes, to account for the generality of the phenome
non. Whiteness is never active (no color is); yet whiteness too (so Suarez
believes) resists the introduction of its contrary. Suarez's own proposal,
oddly enough, makes of what would seem the merely logical feature of
repugnantia (contrariness or contradictoriness) something that, even if only
negatively, functions to alter the course of natural change. He himself rec
ognizes that resistance so conceived is not easily seen to admit of degrees.
Yet experience confirms that it does. Though he has answers to that objec
tion, a modern reader will undoubtedly feel more comfortable with
Zabarella's proposal. That proposal, with due allowance for a new notion of
form, could easily be interpreted as Descartes's first law of motion.
Zabarella, however, speaks of the form in the thing as "striving" to conserve
65. One might wonder why potentia resistendi is not just an active potentia in the patient
reacting against the agent. The refutation of that view cited by Soncinas rests on the single
example of dryness. Air resists the action of fire by virtue of dryness (its only other elemental
quality is heat, which clearly will not resist the further heat of the fire). If dryness were active, it
would drive out wetness, being contrary to it. Yet a dry piece of iron, if cold, will not make a wet
thing dry. Only if heated will it do so. Heat is, therefore, the active power that drives out
wetness, and dryness is not active ( Q meta. p234). Soncinas's response, considered in the text
above, is to make 'active' a relative rather than an absolute feature of qualities.

Authenticated
itself.66 In Descartes's world there is no such striving. The tendance amouvoir

is, as we will see, defined so as to avoid the appearance of finality.67
In the basic Aristotelian scheme, every change comprises an agent and a
patient, with the patient receiving, upon the completion of the change, a
form from the agent. That picture remains unchallenged. But the corre
sponding distinction of active and passive potentia becomes more problem
atic. It is too simple to suppose that for every active potentia there is a
corresponding passive potentia. If by 'passive potentia' one means trivially that
the patient is in some way fit to be acted on, then indeed the correspon
dence holds. But that would be to overlook the difference Suarez insists on:
a passive potentia is a quality "instituted by its nature" to receive a particular
kind ofaction. That notion is not trivial. We have seen that passive potentia so
defined tends to be displaced by qualities whose definition does not refer
specifically to the actions they make a thing fit to receive; and by the striving
after self-conservation, which, though neither potentia nor quality, neverthe
less has consequences.
The Aristotelian theory of motus includes not one but two kinds of directed
ness. The first points from potentia to actus. In intransitive change, the
actualization of a potentia tends toward completion. Even if frustrated, it
points, like an undelivered postcard, to a destination, to its end. This kind of
directedness is of course well known. The second points from agent to
patient. Aristotelianism, like common sense, sees the arrow hit the target
and not a collision of bodies that are equal partners to the event. The
agent's action, identified with the motus in the patient, ceases with the
completion of the form that the agent donates to the patient. Active powers,
just because in them that form, belonging to another, is implicated, presup
pose the other; and since (see §5.3) the patient must be suitably disposed to
receive the form, active powers presuppose such patients as well.68 The
66. In De Anima 2q, Aristotle argues that every animate form, at least, participates in
"divine and immortal being" so far as it can. Hence "all desire it, and on account of it all things
do whatever they do, according to nature" (415b1f). Thomas explains that just as each thing
[...] when it is in potentia, is ordered to the actus, and naturally desires it, and when it is in a less
perfect actus desires a more perfect: so each thing that belongs to an inferior rank [gradus]
desires to be assimilated to the superior ranks, so far as it can be" (In de An. 2lect7; Pirotta
1315, Leonina 45/ 1:g7). Such explanations make it clear that in general the self-conservation
of forms cannot be regarded as an anticipation of inertia.
67. In an early manuscript of Galileo, resistance is defined simply as "permanence in a
proper state against a contrary action"; he adds that "I do not differentiate resistance from the
thing's very existence whereby it endures." On the other hand, he argues that resistance always
has a cause, and that just as animals guard themselves "through appropriate action," so too
"even qualities and similar things have a certain natul'<\1 property by which they more or less
resist a contrary" (Galileo Early notebooks 244; opere 1:171; the translation is Wallace's).
68. Mark Smith writes concerning vision: ''The object itself cannot possibly express its
visible nature without the physical intervention of a transparent medium, which, lacking

Authenticated
Vicaria Dei
directedness of transitive change, or better, of active powers, is less often

recognized than the directedness of intransitive change. Yet it is one tie by
which different natural kinds are bound together-luminous bodies and
the medium that receives their light, animals and the stuffs that nourish
them, humanity and nature.
In Cartesian physics, both connections are refused. The only glue that
could hold together successive states of a particle is the persistence implicit
in the first law of motion. But the ground of that persistence, which is the
fixity of the divine will, is extrinsic to Nature, and thus to the successive
states. The possibility of distinguishing agents from patients likewise falls
away, and with it the bond between them. With a change of reference frame,
the arrow becomes stationary and the target a projectile. Who hits whom is
thus a matter of convenience. The distinction between agent and patient,
and of active from passive powers, was, as I will argue shortly, the basis in
Aristotelianism for the classification of natural kinds. Without that distinc
tion, it would seem, the very idea of natural kind is in jeopardy. What
replaced active power in classification was the notion of structure.
transparency, could not take on the color that causes visibility" (1981:576). If one believes, as
the Aristotelians did, that no active power exists in vain-which is to say without acting
(n.62) -then the colored thing's having the power it has entails the existence of the means by
which to exercise those powers.

Authenticated
Form, Privation, and Substance
F ew notions met with such uniform rejection among major

seventeenth-century philosophers as that of substantial form.
Descartes was among the more categorical. In Le Monde he writes
that "all the forms of inanimate bodies" can be explained "without
the need of supposing any other thing in their matter than movement,
magnitude, figure, and the arrangement of their parts. " 1 Merely by suppos
ing, he wrote some years later, that the insensible particles of which all
bodies are composed have the very same properties we perceive in larger
bodies, one can do without "prime matter, substantial forms, and all the
great baggage of qualities that some philosophers have been accustomed to
supposing. "2 Descartes was not the first to declare that forms were super
fluous. Others had already issued similar challenges to Aristotelianism.
Most of the seventeenth-century philosophers who cast their lot with the
new science concurred in banishing substantial forms from natural philoso
phy.3 Their efforts met, as everyone knows, with great success. The notion of
a mechanism now seems obvious and indispensable; that of substantial form
obscure and gratuitous.
Two things stand out when the charge and the facts of the case are
compared. Those who objected to substantial form seem frequently to have
misunderstood it. It was often held that the very idea of substantial form was
incoherent. No form could be a substance, because substances are self
subsistent and form is not. Yet the Aristotelians were careful to explicate
'substance' as applied to form in such a way as to avoid that inconsistency.
The misreading was abetted by Aristotelian affirmations about the human
1. AT 11:26; Alq.1:338.
2. PP 4'l12o1; AT g/2:319-320 (added in the French edition).
3· Leibniz is the great exception (Discours §10).

Authenticated
VZcaria Dei
soul, which was indeed a self-subsistent material form. There were, more
over, among writers outside Aristotelianism, uses of the term 'form' that
aligned it with spirit, making it entirely distinct from matter (see §3.2). The
temptation, in the later seventeenth century, to conceive the relation of
substantial form to matter on the analogy of the relation of Cartesian soul to
Cartesian body was strong. Strong but misguided. So little were substantial
forms in nature capable of separate existence that delicate arguments had
to be brought forward to demonstrate that the human soul was, alone
among them, an exception.
The second noteworthy fact about the charge is that certain of the prob
lems that substantial form was intended to deal with were handled no more
successfully-and sometimes with less success-than they had been by the
Aristotelians. Chief among them were those of the unity of individuals,
especially of animals, and of defining natural kinds. 4 Showing what makes
an organism one was, as recent studies have shown, a primary motivation for
Aristotle's theory ofform. 5 Descartes saw that the unity ofthe human body,
at least, needed an account. But it cannot be said that his view is satisfactory.
As for the unity of animals and plants, he seems hardly to have noticed the
problem, as Leibniz soon pointed out. Nor did he have anything like a
taxonomy of plant and animal kinds.
A related question was that of the unity of active powers, or-more
neutrally-of essential properties, in a natural kind. Boyle, more per
spicuously than Descartes, recognized the difficulty. His solution was to
reject the commonsense view of kinds. 6 He was led thereby to reject yet
another distinction that substantial form had been invoked to explain.
Some change, as will be explained shortly, was held to be substantial, and its
result a substitution of one kind for another. Change not so radical was
called accidental. Boyle, and Descartes at least by implication, denied the
distinction. With it went one of the more commonly adduced arguments
against corpuscular and atomist physics.
One last question was that of preferred states. Water is by nature cold, the
Aristotelians believed, and thus will return to its natural state when a source
4· Emerton notes, citing Boyle, that some mechanists appealed to a "reinterpreted" notion
of form to supply a "cause of the specificity of bodies." Boyle, who was not alone in this,
admitted an internal "plastic" principle "implanted by the most wise Creator in certain parcels
of matter, that does produce in such concretions as well the hard consistence as the determi
nate figure" (Emerton 1984:72, 73; cf. Boyle Works 1:275-276). 'Hard consistence', or impen
etrability, was soon recognized as not comprehended under the Cartesian conception of
matter; 'determinate figure' refers in particular to the regular shapes of crystals. Gassendi had
earlier concluded that "gems seem in a special way to come from a seed" (Emerton 1984:43
44).
5· See the studies collected in Gotthelf 1985 and Gotthelf & Lennox 1987, as well as Furth
1988, esp. §§u, 15.
6. See Boyle Papers 61 rr.

Authenticated
of heat is withdrawn from it. Similar arguments could be made concerning

the natural places of the elements. Again the short way was to deny the
phenomena, as Boyle did. But the question was much deeper than they, or
indeed the Aristotelians, recognized. The cooling of hot water to room
temperature cannot be wholly explained without the second law of ther
modynamics. In many such instances the tools of seventeenth-century
physics fell short. Nevertheless substantial form, at least in name, was not
kept to explain them.
I will first consider the general question of matter and form, and then
substantial form in particular. In Physics I, Aristotle argues that the princi
ples of change are form, matter, and privation. The argument, a familiar
one, I will call the "substrate argument." What distinguishes Aristotle's ver
sion is the notion of privation, and two fundamental themes: that the muta
ble properties of things fall naturally into genera whose species are contrary
to one another, and thus fit to change into one another; that privation of
form can be distinguished from its mere negation precisely as a condition
tending to it. It is thus that the description of the termini of change joins with
the definition of motus already examined.
The introduction of substantial form and prime matter into natural phi
losophy was not, whatever later philosophers cared to say, gratuitous. I will
examine a number of empirical arguments on its behalf. They center on the
conditions under which change is reversible or not, and more fundamen
tally on the common fate of certain accidental properties conjoined in individ
ual substances. Supposing, for the moment, that in material things substan
tial form corresponds to essence or nature, the arguments indicate that to
argue for essences it is not enough to argue for essential properties. Only if
essential properties have a common ground or ratio in a single form do they
comprise an essence.
I then turn to the logical objection, according to which 'substantial form',
denoting an item at once subsistent in itself and dependent on another, is
an incoherent notion. The incomprehension implied in that criticism
stems, I suggest, from the suppression of the 'is' of specification or of
informatio in favor of the 'is' of identity and the 'is' of inherence, a suppres
sion linked in Descartes's thought with the rejection of substantial forms.
3.1. Principles of Change

The task of describing change, as I said, can be divided into two parts. One
part concerns change in itself. It issues in the definition of motus and the
explication of related notions. The other concerns the difference between
what was before and what now is. Wherever we speak of change

Authenticated
[s6J Vicaria Dei
understood intransitively-there is something that remains the same and

can be reidentified after the change, and something that differs. The ques
tion, then, is how to conceive that identity and difference. Aristotle's well~
known answer was to regard the thing that changes as a composite of matter
and form. These, together with privation, are the principles of change.
The argument of Physics I begins, like many ofAristotle's arguments, with
a sutvey of the opinions of his predecessors. For all of them, Aristotle con
cludes, change is between contraries. The atomists have their solid and
empty, their angular and nonangular, Empedocles his love and strife, and so
for the rest ( 1cs, 188agff). The conclusion can be argued empirically as
well. Not just anything acts on, or is acted on by, or comes from anything.
The white comes from the nonwhite, not the cultured ( musicus), save by
accident; it comes, moreover, notjust from any nonwhite, but from black or
some other color.
One step in the argument desetves notice. Few would deny that when a
thing becomes white, what had not been white is now white. Now a cultured
thing may be either white or not. If it is not, then-supposing it to be
colored at all-it must have been nonwhite, which is to say, a color which is
not white and which cannot coexist with it. To be cultured is not a way of
being colored, even if every cultured thing is colored, and even if being
cultured is, like being white, a quality. The argument presupposes that
accidents can be arranged into natural groups, or genera, such that within
each group the members, or species, are contrary to one another. Color is
one such group, temperature another.?
It is not easy, however, to see how Aristotle can make his point without
begging the question. One might suppose that it would be sufficient to have
noticed that black and white never occur together in the same individual.
But of course there are many collocations of properties we may never come
across. Some simply are rare. Others, more significantly, are ruled out be
cause of regularities holding among properties. No ruby is green; yet 'ruby'
and 'green' are not contraries.
The treatment of contraries in the Metaphysics does not resolve the
7· Since Aristotle speaks of white as coming from "black or from intermediate [colors],"
'contrary' is here used in the broad sense in which any two colors may be said to be contrary. In
a stricter sense, only those species in the same genus that are "maximally distant" are contraries
(Meta. wq, 1055a5f). The terms genus and species, more commonly used ofsubstantial than of
accidental forms, are here used of qualities. Species, I should note, are produced out ofgenera
by differentia. Furth argues that "the original, 'logical' concept of differentia is basically that of
a 'difference from'," and thus that of a "contrariety or opposition" (Furth Ig88:Ioi, citing
Meta. 4c2, 1004a2I, which in Moerbeke's translation reads "Differentia namque qmedam
contrarietas est, et differentia diversitas"-cf. Thomas In Meta. ad Joe.). The differentia by which
species are distinguished within a genus are contraries, as for example 'two-footed' and 'four
footed' (on differentia in biology, see Pellegrin 1987:320-322).

Authenticated
difficulty (see 10C4, wssa7-g). Suarez, referring to this passage, concludes

that "between contraries there is a certain order per seordered [quidam ordo
per se in ordine] toward transmutation; from the nature of the thing there
occurs a transition from one to the other and conversely" (Disp. 45§2!5,
26:742). Contraries are, in other words, defined in terms of change, which
must be understood to be nonarbitrary in just the way Aristotle is arguing for
in the Physics passage cited above.
Rather than fault Aristotle for circularity, it would be better to conclude
that the notion of 'contrary' and that of natural change are nourished from
the same source. That source is the common experience of particular kinds
of alteration, whether from hot to cold, from wet to dry, or-less commonly,
though the example is often used-from black to white. 8 There is no expla
nation for the fact that properties can be sifted out into genera the species
of which are transmutable into one another. 9 There is only the testimony of
cases, and abstraction from them to the notions of genus, species, and
contrary.Just as in the definition of motus Aristotle seeks not an explanation
but a perspicuous scheme under which to bring together acknowledged
phenomena, here too the claim that change is between contraries provides
not an explanation or an analysis but a scheme.
By 'scheme' I mean a general form into which the phenomena must be
cast so as to be suitable for explanation. The definition can be interpreted as
containing schematic letters, to be filled for each genus: X-ing is the actus of
the X-able qua X-able. The result is a perspicuous definition of X-ing. A
remark by Aristotle not long after the definition of motus exemplifies what I
have in mind: "What motus is has been told in general and in detail: for it is
clear how each of its kinds is to be defined; alteration is the actus of the
alterable, insofar as it is alterable" (Phys. 202b24). Toletus adds, "and aug
mentation is the actus of the augmentable, as such: and so for the rest" (In
Phys. 3c3text23, opera 4:82va). So Aristotle defines light as the actus of the
transparent qua transparent (De anima 2c7, 418bg). 10 Since 'light' here
8. For the moment I am leaving generation and corruption aside, since they present
special problems.
g. The case of figure, which seems to fit the criteria I have listed, and yet is said not to
undergo alteration, will be taken up in §4·3· The general account given here undergoes certain
restrictions in Physics 7, where Aristotle argues that only the so-called passive qualities of the
third species (sensible and elemental qualities) are alterable per se. Others are either inalter
able (virtue and vice), or else alterable only as a consequence of alteration in passive qualities
(245b3-g, 246b2-20, etc.). In the latter case, it seems to me, the scheme still applies. Aristotle
is not denying that there are motus of color or illumination; he is, rather, attempting to show
that such motus have other, more fundamental, motus as their principle.
10. The word here translated 'actus' is EbvoipyELa, not EbVTEAEXELa, but the distinction
made elsewhere between these terms is not crucial to my point. Hicks translates it as 'actuality'
(De anima, Hicks 1907:79); Moerbeke's Latin translation has 'actus' (see Thomas In de An.
2lecu4, Leonina 45/1:123).

Authenticated
[s8J Vicaria Dei
means 'illumination', and 'transparent' means 'illuminable', the definition

is simply a version of the scheme presented in the Physics.! I
Something similar holds for the proposition that change is between con
traries. Setting aside certain complications that I will come to later, the
scheme suggested is this: in any natural change, at least, when a subject S,
having been A, becomes B, the forms A and B must be so described as to be
species of the same genus. Only then will the change be amenable to expla
nation. A perspicuous description of the transmutation of earth into fire, as
we learn in De generatione et corruptione, is that it is the change of the cold and
dry into the hot and dry. 'Earth' and 'fire' themselves are in that respect
opaque, one might say nonnatural, descriptions of their referents.
So in change one contrary "expels" another. Yet something remains: al
though "the not-cultured or the uncultured do not subsist as simple or as
united to their subject," "the man subsists when he becomes cultured and
he is still a man" (Phys. 1goauf). There are two cases to consider, corre
sponding to substance and accident:
According to the diverse things that come to be, different subjects must be
set up: For in accidental [change] we suppose a first substance, because
that is the thing in which all accidents inhere, and which itself inheres in
nothing. 12
But in substantial [change], in which substances themselves change

[...] some other subject must be established that persists in the change,
which cannot be unless we say the substances themselves are composites
of matter and form, and the matter is the subject which according to the
various forms of substance is changed. (In Phys. 1qtext62, 4:31va)
The argument continues with an induction: the generation of plants and

animals is from seed, and in general substances come about by transfigura
tion, by accretion, by subtraction, by composition, and finally by alteration
"according to matter" (Phys. 1gobg). In each ofthese processes some mate
rial is presupposed, and persists through the change. Supposing that the list
is complete, it follows that all change presupposes matter. 13
The correlative to matter, form, is introduced by way of the notion of
definition. The subject of change is "not distinct numerically from itself, but
11. Hussey notes that some ancient commentators treated the definition of motus as a
"definition-schema, to be filled out differently for each type of change" (Hussey 1983:60).
12. This is the definition of 'first substance' in the Categories (c5, 2a11lf and 3a7lf). A first
substance is an individual concrete thing, a 'this something' (hoc aliquid, To& n); a second
substance is a genus or species (2a14lf).
13. Clearly "altered according to matter" (secundum materite, KaTa uA.rJV) is a wild card,
especially since Aristotle argues in De generatione et corruptione that the elements themselves can
be transformed into each other. Matter, as the subject of substantial change, was supposed to
remain fixed during change. See §4.1 below.

Authenticated
distinct according to its ratio and definition." When the uncultured be
comes cultured, "'the uncultured' denotes the subject together with a form,
or a privation, and the ratio of one [i.e., the subject] remains, while the ratio
of the other [i.e. the unculturedness of the subject] goes away" (Toletus In
Phys. tqtext6o; cf. Phys. tgoaq-16). The ratio of the subject is the human,
that of the form or privation is the uncultured. Those rationes are evidently
different (Phys. 1goa16), even though their subject is one. Hence matter
and form are distinct, at least in reason.
The argument-call it the substrate argument-is familiar, its conclusion
almost anodyne. Few natural philosophers would have disagreed with it.
The alternative, Suarez argues, is to suppose that in each change there is an
annihilation and a creatio ex nihilo, which is not only extravagant but also
leaves no reason to explain what is produced: "Here also [in generation] it is
necessary that a common subject persist, since otherwise the entire preced
ing alteration, or the heating of the oakum [stupa], would be irrelevant to
the procreation of the fire, because it would in no way contribute to the fire,
if it and its whole subject perished" (Disp. 13§1,6, 25:396). Nor would there
be any reason why the destruction of the fuel should be necessary to the
creation of the fire, and vice versa (ib. tg; 397). 14 At most one could
suppose that the fuel was destroyed to make room for the fire; but why it
should perish instead of moving out of the way, a much less drastic response,
would still be a mystery. To conceive of change as annihilation and creation,
while it yields no inconsistency, is in effect to make physics impossible, or
relegate it to a study of phenomena.
In the sixteenth and seventeenth centuries, at least, the substrate argu
ment was unchallenged. Some sort of contrast between form and matter was
standard. 15 Descartes uses the term to denote the figures of his three ele
ments, figures that are contrasted with the underlying extended substance
of which they are modes. For Boyle, the 'form' of a material particle is "an
essential modification and, as it were, the stamp of its matter," specified as
size, shape, motion, and "contexture" (Papers, "Origin of forms and
qualities," 6g). Yet despite the coincidence of terms, Descartes and Boyle
14. It should be noted that these are physical arguments. God could, according to his
absolute power, destroy and recreate the world at each moment, while making it appear as if
the regularities of this world still held. It is, in other words, not logically impossible that change
could consist in annihilation and creation; but to suppose so is not only extravagant but
renders change inexplicable, or at least requires us always to limit our assertions to the
phenomena.
15. See Emerton 1984, c.2. 'Form' was used to denote both "the cause of the specificity of
bodies," or structure, and "the physical mode in which order is established in matter," or
formative cause (72). Though both uses are found among Aristotelians, the first is primary.
Form as formative cause, semen, or vital spirit, owes more to the various forms of Platonism (sg
6s, citing Daniel Sennert and Sebastian Basso). But even ifwe exclude the latter, there are still
a great many non-Aristotelian versions of form.

Authenticated
[6o] Vicaria Dei
thought they were contradicting the Schoolmen. What, then, is the

difference?
It is with the third of Aristotle's three principles-privation-that the
Aristotelian twist on the substrate argument appears. The unculturedness of
what becomes cultured, the blackness of what becomes white, are not the
mere negation of being cultured or white, but their contraries.l6 Un
culturedness in a person who is fit to become cultured is indeed the absence
of culture. But trees and marmosets lack culture too. As Toletus puts it,
"privation means not only the negation of a form but [its negation] in a
certain subject [... ] Privation supposes not only a subject, but an aptitude
in the subject to the form, ofwhich it is the privation, so that the negation of
the form in a subject apt for it is privation. (In Phys. 1q 2 1, 4:42va). Suarez
explains that privation, unlike form and matter, is a principle "not by a
positive influx [of its being] but only according to a necessary habitus per se
toward the other" (Disp. 12§ 1, 6, 25:374). It is in one respect more general
than contrariety. The stuff of which a human is made does not have a form
contrary to the human form. Though of course many things are nonhuman,
the human form, like all substantial forms, has no contrary. But it is sup
posed to be the sort of stuff that is fit to become part of a human. In that
rather weak sense, it has the "privation" of humanness. Privation thus covers
not only contraries, which figure in alteration and other motus, but in a
certain way the matter-form distinction itself, which figures only in mutatio
or generation.
Like potentia it includes an essential though unspecific reference to some
thing other than itself-its completion or end. In the previous section the
result of change was said to be the actus of a potentia. That potentia we may
now identify with the privation of the form received in change, and the actus
with the form itself. In alteration, augmentation, and local motion the
potentia is a foil? contrary to the actus. In generation it is a disposition or
habitus of the matter whose actus is a new substantial form. As the slogan has
it, form is the actus of matter.l7
Those identifications, however, signal certain complexities in the notion
of directedness that were left unexplored in §2.1. Though each potentia
must somehow be defined in terms of the corresponding actus, that actus
16. "Since an object's lack can be rectified by means of a change, the privation is not just
any state other than the goal but one on a path leading to it" (Gil11991:247). The notion of a
'path' leading to the terminus must be understood quite broadly. It is rather the agent that
determines the path than the patient, as will be seen shortly.
17. The slogan applies, of course, only to forms that can exist only as part of a composite
substance. To distinguish them from the forms of angels and God, such forms were sometimes
called "material forms." Perez-Ramos calls the notion "lexically self-contradictory" ( 1988:74).
It would be self-contradictory if it meant 'form that is also matter'. But it does not. It means
'form of the sort that must be joined with matter to exist', or 'form that is an actus of matter'. In
that sense matter and form are not contrary but complementary.

Authenticated
Form, Privation, and Substance [61]
cannot simply be called its end, nor can the relation between them be
described simply as a tendency of the potentia to a particular actus.
1. In alteration the hot comes from its contrary, the cold. Yet it cannot be
that the cold tends to the hot, since then the cold elements earth and
water would tend to become the hot elements fire and air. Hence to
describe the cold as the hot in potentia means simply that the cold and
the hot, being extremes of the same genus, are suited thereby to be
come one another. That is a much weaker directedness than one sees
in the paradigmatic instance of acorn and oak. It is, all the same, not
quite trivial, since it presupposes, as we have seen, that qualities fall
naturally into genera.
2. In substantial change, the matter that is about to receive a new form
does not lack a form. Matter cannot naturally exist without a substantial
form. Hence the same matter must be capable of receiving more than
one form, since it loses one and takes on another. Prime matter, the
substrate of substantial form, must be capable of receiving all forms. Its
potentia, then, far from being directed to any one form, is directed
indifferently to all.
3· Finally, in certain instances of accidental change the potentia of the
patient exhibits a similar indifference. The same youth is in potentia
skilled at grammar and at soldiering. Each skill or habitus is an actus of
the rational soul, which is indifferently in potentia to them all. There
may be differences in aptitude among pupils, but qua humans they are
all equally fit to learn.
These remarks all come from the side of the patient. Passive potentice vary
enormously in the specificity oftheir corresponding actus. IS Active potentice
in general do not. The actus of the teacher who exercises her power to teach
grammar is precisely the grammatical skill that her pupil acquires. The
active power of the semen has precisely the end of introducing the form of
the father into the matter provided by the mother. The activity of heat is just
to produce more heat. Unlike passive potentice, active potentice are, with one
important qualification, unambiguously specified in terms of their actus.
The qualification concerns so-called rational powers, which are precisely
those that can act in more than one way. The vis motiva of animals can bring
about local motions of all sorts in their parts; the will can produce volitions
to any action. Since rational powers, as their name implies, are manifested
only in animate things, the study of them belongs to psychology. In physics,
active powers are univocal. It is their job, so to speak, to disambiguate
privations.
18. 'Passive potentia' is here used in the broad sense of 'any principle by which a thing is
changed', not the narrow sense in which it is a species of quality (§2.3).

Authenticated
Vicaria Dei
A finished intransitive change, then, can be described either as the com

plete actus of a potentia or as the fulfillment of a privation by a suitable form.
Transitive change is a joint actus of agent and patient (§2.2). Although the
privation belongs to the patient alone, the form, since it is the actus of both
agent and patient, must pertain to both. The Aristotelians say that the form
is received in the patient from the agent. Yet it cannot always be straightfor
wardly identified with any prior form in the agent. When water is heated by
fire, the form of heat-or better a form, more or less intense-is received by
the water from the fire. The elemental heat of the fire is the active power by
which it acts. That same active power, when brought into contact with wood,
not only heats the wood but consumes it, making more fire, which in turn
has the same power to produce more. But water, however much heat it
receives, never acquires the power to consume wood so long as it remains
water. Hence the heat received by the water cannot be quite the same as the
heat in the fire. How, then, does the form received by a patient differ from
the form identified as that by which the agent gives or imposes that form on
the patient?
The answer seems to vary with the agent and the form. In the examplejust
offered, the answer given by Zabarella is in effect that the elemental heat of
the fire and the heat of the water differ only in intensity.I9 But in the
transmission of colors, the "intentional species" received from colored
things by the senses are different in kind from the colors in the things
themselves, since they lack even the feeble activity of colors. It is in general
not true that the form received from the agent is the same as that by which it
acts; nor is it true without qualification that the two are similar.
What can definitely be affirmed is that, despite the language of giving and
receiving, nothing is literally transmitted from agent to patient. The form
that is said to be "received" is actually "educed" by the agent from the
potentia or the privation of that form in the patient (see §5.4). 'Educe'
means nothing more than that the agent initiates and sustains the motus
that, as we have seen, inheres in the patient alone. The formal cause of
change is always intrinsic to the patient; the role of the efficient cause is not
to impose a form on the patient from outside (as Aristotle's own occasional
analogy ofwax and stamp, or statue and mold, might lead one to think), but
to determine just how a certain mode of being2° in the patient, as yet
19. "The natures and forces [vires] proper to these qualities [sc., the four elemental
qualities] are the same in all things; for every [instance of] heat warms, and disperses hetero
geneous things, and gathers together homogeneous things, and melts and thins congealed
things, without any difference except that which is according to the more and the less" ("De
qualitatibus elementaribus" 2c6, De relnts nat. 531C). Toletus takes the same position, his
grounds being simplicity and likeness Of effects (In de Gen. 2q7, Dpera 5:31grb).
20. 'Mode of being' is designedly vague: I use it as a placeholder for any of the categories in
which change can occur. In substantial change the mode of being in potentia some-substance

Authenticated
potential and indeterminate, will become actual. Much better than the
analogy of wax and stamp is that of the teacher who cultivates virtue in her
student. Virtue is conceived not as itself an active power but as a persistent
modulation, so to speak, of the activity of the will. The teacher transmits
nothing, but by word and example elicits from the student habits ofjudging
wisely and acting well that were always potentially there.
We have now, finally, the entire scheme of natural change. It will be useful
to summarize it in outline:
1. Intransitive change.
1.1. Every natural change has a subject in which it occurs, a terminus a quo from
which it begins, and a terminus ad quem (i) which it will attain if not prevented
by an external agent, and (ii) at which it will cease.
1 .2. The terminus ad quem is a form X*, the terminus a quo a privation among whose
fulfillments is the terminus ad quem. The subject, relative to the form and its
privation, is matter.
1.3. The process of X-ing is the imperfect actus of X in an X.able thing.
1 + To be X.able is to have a privation of, or a potentia to, the form X* elicited in
X-ing. The perfected actus of that potentia is X*, while an imperfect actus is a
form contrary to X* (and thus of the same genus) regarded as tending to X*.
1.5. To have a privation of X* is either (i) to have a form contrary to X* or (ii) to
lack X* and yet be suitably disposed to receive it.
2. Transitive change.
2.1. Every natural change is the joint actus of two things: (i) a patient in which it
occurs as an intransitive change, and in which it is the actus of a passive
potentia; (ii) an agent, of which it is the actus of an active potentia.
2.2. The (perfected) actus of both potentia is the terminus ad quem X*.
2 ·3· It is an actus of the passive potentia by virtue of being a specified form from
among those that the passive potentia is contrary to or suited to receive.
2-4. It is an actus of the active potentia by virtue of being specified by the agent
insofar as the agent has that potentia.21
2.5. Since the actus inheres in the patient alone, the agent is never changed, and
thus never perfected, by its action, except accidentally (e.g., by reaction).
2.6. Action is nothing other than the change itself, considered as the actusofthe
agent; and similarly for passion.
The summary captures what is common to the conclusions about natural

change affirmed in the central texts. They record, of course, a wide variety
of dissenting views. But I think it reasonable to suppose that Descartes, when
or-other (but as yet none in particular) is that of prime matter; in other categories, it is a
contrary or a disposition in an existing substance.
21. The qualification is to forestall examples such as that of a medicine that cures by virtue
of one of its active ingredients and not by virtue of another. The actus is that of the first
ingredient alone.

Authenticated
Vicaria Dei
he contemplated what the Philosophers thought about change, would have

attributed to them the propositions just listed. His own views, as we will see,
if seen through the lens of the central texts, would have registered as tend
ing to the nominalist side. There is, however, a significant difference be
tween Cartesian physics and Cartesian metaphysics (including the study of
the soul apart from the body). In the physics virtually all of this machinery is
dismantled, or retained in name only. To answer a question raised earlier:
the difference between the "form" recognized by Boyle and Descartes and
Aristotelian form is that by 'form' they mean 'figure' or 'configuration'. No
figure has any intrinsic tendency to become any other figure. No figure is, in
other words, the privation of another.
In the metaphysics, on the other hand, a great deal of the machinery is
retained. It may have been inevitable that where the two met-at what we
now call the "mind-body problem"-clarity would give way to obscurity, and
certainty would founder in doubt. But even within the physics, as we will see
in Part II, there are tensions between the ideal of a science purged of activity
and finality, and the science actually practiced, in which the repressed ele
ments continually threatened to return.
3.2. Substantial Form and Prime Matter
It is one thing to admit a distinction between matter and form, quite an
other to admit substantial forms and prime matter. Since substantial form,
with certain qualifications to be added below (§5.3), precedes all others in
the constitution of material substance, the corresponding matter was a
matter bereft of all forms. We have seen that form is the actus of matter, and
matter form in potentia. The matter that receives substantial form has, there
fore, no other actus. It is, as the slogan has it, pure potentia (but see §4.1).
To argue for substantial form was to argue against a range of alternatives
specific to the period. Chief among these were the mixture theory of Em
pedocles, the atomism traditionally associated with Democritus, the Scotist
hypothesis of a forma corporeitatis common to all material substances and
preceding specific forms, and finally various Neoplatonist theories ascribing
the individuation of material substances to a "seminal spirit" or "formative
soul."22 From the standpoint of Aristotelian qucestiones on substantial form
and prime matter, the alternatives can be divided into (i) those that denied
the distinction between substantial and accidental form and (ii) those that
22. 'Seminal spirit' is found in Etienne de Claves's Paradoxes (1635), where he is recount
ing the views of earlier philosophers, including, for example, the chemist joseph Duchesne
(Quercetanus). 'Formative soul' (animaformatrix) and 'formative faculty' occur in Kepler's
work (De Nive sex., Werke 4:275r, 278). See Emerton 1984:41, 61, 16g.

Authenticated
Form, Privation, and Substance [6sJ
while maintaining the distinction, denied the uniqueness of substantial form

(and thus its role in specification), or denied that matter and form were
genuinely one (thus making each a complete substance). 23 Under (i) fall
atomism and mixture theories, under (ii) the Scotist and Neoplatonist
theories.
The position argued for, then, was that there is a distinction between
substantial and accidental form, and that substantial form is unique in each
individual, specifies that individual, and must be united to a matter without
which it cannot subsist. That characterization, it seems to me, suffices to
pick out the Aristotelian notion of form in my central texts from the be
wildering multitude of sixteenth-century options. That is the notion against
which Descartes directs by far the greater part of his polemical force.
Toletus defines substantial form as "a formal simple actus, forming with
matter [a thing that is] one per se, the principle of the proper operations of
the thing" (see Fig. 2) .24 Form is the actus of matter, matter the potentia of
form, because form perfects and determines matter.2 5 It is called 'simple' to
distinguish it from the whole composite substance, which is in a sense also
the actus of the matter. The term 'formal' designates the manner in which it
perfects the matter, which is to say, by informing (or determining) the
matter and uniting with it. That serves to distinguish the form from an
efficient cause, which does not unite itself with the matter. 26 The phrase
'one per se' is the key to distinguishing substantial from accidental form:
accidental forms are all formal simple actus of matter, but substantial form
23. Thus Gilson's verdict (1984:163) that the "monstrous" notion of substantial form,
which Descartes could conceive only as a "substance immaterielle [... ] qui s'ajoute a une
substance corporelle [... ] pour composer avec elle une substance purement corporelle," is a
monster of his own making, must be softened. Such monsters abounded, if not in textbook
Aristotelianism, at least in precincts not far from it. The French alchemist Quercetanus writes
that the "three formal! beginnings" (the Paracelsian Mercury, Sulphur, and Salt) "are more
spirituall than corporall, yet[ ... ] they make a materiall body" (Emerton 1984:182, quoting
the 1605 English translation of Quercetanus's major work)-the very contradiction that
Descartes saw in the notion of substantial form. Descartes did not, I should add, confuse the
Philosophers of the schools with the likes ofQuercetanus. But there are, even in them, traces of
the "spiritualization" of form, especially in the theories of light and intentional species.
24. "Forma est actus simplex formalis, unum per se cum materia componens, principium
propriarum operationum rei. H<ec conclusio est velut definitio explicans ipsius form<e natu
ram [... ]"(In Phys. 1c9q19, Opera 4041ra). The context makes it clear that the form being
defined is substantial, not accidental, form. Almost identical definitions are found in Eu
stachius (Physica r§2q5, Summa 3: 123), the Coimbrans (In Phys. 1c9q9), and Suarez (Disp.
1 s§sl.
25. "Forma dicitur actus, quia dat esse materi<e perfectum, & determinatum in specie certa:
Nam actus materi<e est imperfectus, & communis, ac potentialis, forma determinat & perficit
ad certam speciem" (Opera 4:41ra).
26. "Dicitur formalis, ut intelligas, quomodo forma det esse materi<e: nempe si ipsa infor
mando, & uniendo sibi materiam [...] & hoc ad discrimen caus<e efficientis, qu<e dat esse,
puta, ignis facit lignum ignem, sed non dando seipsum, sed aliquid aliud extra se producendo"
(Opera 4:41rb).

Authenticated
-----
[66] Vicaria Dei
Actus
Complex ~Simple
complete substance
~
Separate from Unified with
affected substance affected substance
efficient cause
/ ~
Per accidens Perse
accidental form substantial form
Fig 2. Definition of substantial form
alone enables matter to subsist. 27 It is, finally, the principle (by implication
unique) of all the active powers ofa thing, and thus of its "operations." That,
as we will see, is the focus of the empirical arguments: substantial form is, in
a manner yet to be delineated, the cause or origin of all other forms, and
uniquely so.
There are, in keeping with that complexity of the concept, a number of
standard qumstiones devoted to the defense of substantial form. Those that
concern the Scotist view I will examine in the next section. Here I take up
only those that concern the distinction between substantial and accidental
form. Such arguments are directed in part against atomism, in part against
skepticism about substantial form within Aristotelianism itself.
The arguments for distinguishing substantial from accidental form fall
roughly under four headings: logical differences between substance and
accident; the empirical distinction between generation (or corruption) and
alteration; the unity of active powers in a natural kind; the existence of
preferred states. A detailed examination will yield several fruits. It will show,
first, that in the hands of a determined advocate like Suarez, many experi
menta can be brought forward on behalf of substantial form.28 Second, it will
make somewhat clearer the role of substantial form in natural philosophy,
and its relation to more recent notions of essence and natural kind. It will,
finally, begin to show the complexity of the Aristotelian theory of material
substance. That complexity is sometimes glossed over in brief expositions.
27. "Dicitur, unum cum materia per se faciens, ad excludendas formas accidentales, ha:
enim actus sunt simplices & formales, sed cum eo, cui us actus sunt, non faciunt unum per se,
sed per accidens: non enim complent accidentia subiectum, at forma complet materiam: est
enim materia dimidium ens, ut diximus, ac ob id dicitur facere unum per se cum materia"
(opera 4:41rb). On the phrase unum per se, see §5.2.
28. Suarez calls them indicia and signa. "Accidents are called signa when they are better
known to us than their causes" (Zabarella, De methodis 12, opera logica 2028; see also 754E,
where the dictum is ascribed to Averroes).

Authenticated
Yet it is entirely pertinent, I believe, to the responses of the earlier oppo

nents of Aristotelianism, who were still in a position to recognize it.
1. Logical differences. I have touched on such differences already. Toletus
argues:
There is the greatest difference [ maxime discrimen] between accidental

and substantial forms. Accidental forms have [with each other] not only
repugnance [i.e., incompossibility], but a determinate repugnance, as
white with black or an intermediate [color], heat with cold, and so forth.
This determinate repugnance consists in the fact that for an appropriate
subject, either one or the other is found, nor can one leave unless another
arrives. Between substantial forms there is a certain repugnance, but it is
not determinate, since every substantial form is repugnant to every other
equally: nor are they consequent upon a particular matter [nee circa mate
riam consequenter se habet], since matter can exist without the two or three
[mutually repugnant] forms. (In Phys. 5c1, opera 4:151Vb-152ra)
The argument is that the logical space, as it were, of accidents is structured

in a way that the logical space of substance-kinds is not. Accidents, as we
have seen, are supposed to fall under genera whose character is spelled out
reasonably clearly by Toletus. First, what one might call the "generic"
accident- 'being colored'-is properly predicable only of certain kinds of
subject; and a specific accident-'being red' -falling under it is properly
predicable of a subject only if the generic accident is. Angels are not col
ored, hence neither red or blue nor whatever. Second, every thing of which
the generic accident is properly predicated has at least one specific accident
from within the genus, and no more than one such accident. The apple of
Eden, being colorable, was of one and only one shade.
Now even though substance-kinds can be arranged into genera, and then
via their differentia into species, those genera do not share the characterjust
described. The notion of "proper subject" does not apply, because 'human',
though a species of animal, is not predicated of the animal as an accident.
More significantly, any matter could, in principle, exist without any of the
forms that fall under the genus 'animal': it could be a rock or a flame
instead. The only list of substance-kinds of which one may say that a subject
must be at least one such would be the list of all substance-kinds. An accident
is repugnant especially to other accidents within its genus, and that special
repugnance is just what is called 'contrariety'. A substance-kind, on the
other hand, is indiscriminately repugnant to all other substance-kinds. To
put the point linguistically: if I call a thing 'not white', the implication
normally is that it has some color, only not white; but if I say a thing is not an
apple, there need not be a genus such that my statement implies that it is
one of those, only not an apple.

Authenticated
[68] Vicaria Dei
On the face of it, the first part of the argument begs the question. Within
the genus 'animal', the differentia 'rational', predicated of a generic "animal
stuff," yields the species 'human'. Differentia are contraries (see n.7). We
therefore have a structure not unlike that of colors: every animal is either
rational or irrational, and only animals are rational or irrational. Similarly,
'animate' could be regarded as a differentia within the genus 'complex mate
rial substance'. Why then is the phrase 'rational animal' not on a par with
'white table'? Why aren't there accidental forms all the way down?
Suarez records a similar point among the arguments against substantial
form: "fire, for example, is sufficiently understood in its constituted essence
if we conceive a substance having perfect maximal hotness and dryness
together, even if the substance which is the subject of those accidents is
simple: and this also is enough [to understand] all the actions of fire we
experience, and the distinction between fire and water, and the transmuta
tion of each into the other, which seems to consist in the fact that a sub
stance changes from maximal cold to maximal heat, and vice versa. This is
therefore sufficient for the constitution, distinction, and action of the ele
ments" (Disp. 15§ 1t 1, opera 25:498). The differentia 'hot' and 'dry' suffice to
explain what fire is, how it differs from the other elements, and how it acts.
Why not, then, say that the substance fire just is those differentia, conjoined
in a material substrate?
Now it could be that whichever substance you consider, and whichever
change you consider, there is a kernel that always and everywhere persists,
and that suffices to individuate that substance from others. The kernel, by
hypothesis common to all substances, could be called indifferently matter
or form: matter insofar as it is the substrate of change, form insofar as it
makes the substance one thing. There would be a "substantial form," but it
would have nothing to do with the kind of thing it was. That would be
determined entirely by accidental forms.
Or it could be that for no substance is there such a kernel, much less for
all. Aristotle's argument for distinguishing form from matter shows only
that in each change there is a matter and a form. Nothing licenses us so far
to speak of the matter or the form of a substance. The example of fire is
telling here. Fire is maximal heat combined with maximal dryness. If the
heat changes into its contrary cold, one has earth. If the dryness changes
into its contrary, wetness, one has air. In each case there is a "matter,"
namely, that one of the two accidents is not changed. But there is no
"matter" that is the common persistent in every change.29
29. See Furth 1g88:2 21fT for an analysis of elemental mutation that follows out this obseiVa
tion to argue that prime matter is not needed in Aristotle's physics. Zabarella already obseiVes
that even in elemental mutation, one of the two elemental qualities (or symboltF) is always
preseiVed (De rebus nat. 207F).

Authenticated
Form, Privation, and Substance [6g]
I will examine prime matter in §4.1. Here I want to obseiVe that the
Aristotelians, and perhaps even Aristotle himself, wanted to preseiVe two
fundamentals. The first is that even the elements can be generated and
corrupted. One consequence of that is to preclude any common kernel-at
least the Aristotelians thought so. The second fundamental is that for each
thing there is a kernel that determines its kind or essence and whose
destruction is the destruction of the thing. That kernel is the substantial
form. Some changes are indeed annihilations and creations, but of the
thing according to its kind-secundum quid-and not of its being simpliciter.
Aristotelianism thus has to avoid two extremes. It cannot accept a princi
ple of individuation and unification which does not also tally with natural
kinds. Nor can it give up the view that there are destructions that are not
annihilations, generations that are not creations ex nihilo. So the Coimbrans
argue that "if form were an accident, there would be no substantial genera
tion if indeed matter cannot come into existence" (InPhys. 1cgqga2). Since
matter indeed cannot come into existence (except at the Creation) there
would be no substantial generation, which is contrary to experience. But if
some destructions are not annihilations, and some changes are not destruc
tions, there must be something that determines both what kind of thing an
individual is and when it starts and ceases to exist as that individual. Aristo
telianism would like to reseiVe creation for God and restrict natural agents
to generation. But it accepts the doctrine of the transmutation of the ele
ments, which seems to preclude any preseiVation of structure beyond mere
existence. Generation coincides with alteration. Hence the continual
puzzles about generation. Hence also the temptation to rase the whole
immense structure, and in particular to banish substantial form. Rather
than provide an apt descriptive scheme for natural change, it seemed to
block the way to a true physics.
Descartes certainly thought so. In the next section we will see that his
physics embraces, in a certain way, both the hypothesis of a single substantial
form and that of the reduction of generation to alteration, whose implica
tion is that there are only accidental forms. Extended substance indeed has
an irremovable form-its quantity, which the Aristotelians numbered
among the accidental forms. But even though Descartes, with misgivings,
treats extension as an attribute of substance rather than as substance itself,
extension, because it is, so to speak, the unique substantial form in nature,
admits of neither generation nor corruption. To generate it is to create it, to
corrupt it is to annihilate it. To hold that there is one substantial form is
tantamount to holding that there are none. What the Aristotelians called
"generation" is after all reduced to alteration, and all forms, save the one,
are accidental.
But cleansing the stables had its costs. Whatever the difficulties of the

Authenticated
Vicaria Dei
doctrine, it saved a great many phenomena. Chief among them, as I have

said, were the difference between generation and alteration, the unity of
powers, and the existence of preferred states. 30 The empirical arguments, I
think, must in retrospect look to yield a stronger case for substantial forms
than the logical. The phenomena they adduce cannot be, and in the full
ness of time, were not neglected in natural philosophy; and the revisions of
mechanism necessary to manage them, although they may not have ushered
in a new Aristotelianism, did temper the radical erasures effected by
Descartes.
2. Generation and alteration. Generation is ordinarily defined as change of
substance, alteration as change of quality. To argue that the existence of the
two kinds of change so defined implies the existence of substantial form
would be question-begging. What one must do is first to argue from experi
ence that there are two kinds of change, show that one kind is best
described as change of substance, the other as change of quality, and then
make the inference to substantial form.
Suarez, thorough as always, does not overlook the argument. One "index"
of "substantial composition" is that generation differs from alteration, as
"evident experience" shows. The argument deseiVes to be stated in full:
We experience alteration, as, for example, the heating of water or iron, to

be sometimes so vehement that the most intense heat is felt in them, and
yet if the action of the contrary agent [i.e., an agent impressing upon
water or iron a quality contrary to their natural coldness] these things
remain whole or almost whole in their substance, and easily revert to their
[prior] accidental state; but sometimes alteration proceeds so far that an
all-encompassing transmutation of the thing occurs, so that although the
agent is removed, the patient can never go back to its pristine state, or
recover its prior actions or accidents similar [to those it had]; sometimes
also it is changed into viler sensible substances, like ashes, scoria, etc.; and
occasionally it is entirely and insensibly consumed, because it is trans
formed into another more subtle and insensible body; it is therefore an
evident sign that alteration is sometimes pure, and remains within the
bounds [latitudo] of accidental change, but sometimes has conjoined with
it a greater change. This cannot be otherwise than because the substantial
composite, once the form has withdrawn, itself is destroyed [ dissolvitur];
therefore substantial forms exist [dantur ergo substantiates formce]. (Suarez
Disp. 15§1112, opera 25:501)
The obseiVation that things naturally cold will revert to that quality after
being heated will be the basis for another argument. Here the emphasis is
30. See Boyle Papers 59IT and Chauvin s.v. "Forma" for brief, hostile discussions of the
arguments.

Authenticated
on changes that preclude reversion: water changed to steam, iron vapor

ized, bone reduced to ash. Consider the reasons for such changes. The heat
itself cannot be the reason. Heat per se does not tend to its contrary: no
form tends to its own destruction. So in the original substance something
else must have opposed it, and if it is now not opposed, that something must
have been removed. That something cannot be any accident of the water.
Of its accidents only coldness is contrary to the impressed heat, and by
hypothesis that coldness no longer suffices, if indeed it is present at all.
What opposed the heat in the original substance must, therefore, have
been its matter or a form distinct from ordinary accidental forms. Matter,
however, is indifferent to the accidents that inhere in it. It is, after all, that
which "remains under every transmutation." It doesn't play favorites. The
thing whose loss results in irreversible change must therefore be a form. But
we have just seen that no accidental form will do, or at least no ordinary
accidental form. This form has the peculiar feature of opposing heat with
out being contrary to it. Now it may be that some accidents are inseparable
from one another, as rarity from heat, or whiteness from a certain "tempera
ment of primary qualities." One might suppose then that the loss of density,
which is inseparable from cold, is the reason for the loss of cold. But that
only defers the question, which could just as well be asked about density:
why is it irreversibly changed? Suarez concludes that the reason must lie in a
form prior to accidental forms, "a substantial, and not an accidental, form,
since it constitutes the proper essence [of the thing], in which accidental
properties connaturally and inseparably inhere" (Disp. 15§ 1t 13, opera
25:502). That form must, in other words, be the unique determining actus
of matter defined earlier (see n.24 above).
3· Unity ofpowers. One can already see in Suarez's argument that essence,
if identified with substantial form, is not a mere list of properties the loss of
any one of which must result in the destruction of the individual. Many
discussions of essentialism now, especially those that treat the issue in terms
of modalities, forego essences in favor of essential properties. The argu
ments for substantial form which I consider here require, on the contrary,
that essences be distinguished from essential properties.
That is, in fact, the heart of the problem. It is not enough to exhibit
essential properties. There could be a world in which concrete physical
individuals had essential properties but no substantial forms. In such a
world it might be true, for example, that a piece of iron cannot persist if its
coldness gives way to extreme heat; and that it cannot persist if its dryness
gives way to extreme wetness (i.e., if it rusts).31 What would not be true is
31. Iron is of course not an element but a mixture. The explanation of its changes might
not be as simple as I am making it here.

Authenticated
Vicaria Dei
that those two had any particular relation with one another. In effect each
individual would consist in a bundle of essential properties, and natural
kinds would be lists of compossible properties (hot and dry, for example,
but not hot and cold).
Such a world is not ours. Two kinds of argument can be made to show that
it is not, one concerning the possible causes of accidental forms, the other
concerning their nonarbitrary collocation in a single subject.
Suppose, first of all, that in natural individuals certain "proper and pecu
liar functions" are found, "as in man thought, in horses whinnying, in fire
heating" and so forth.32 Where do they come from? Not from matter, since
matter has no efficacy. It will not suffice to suppose that they arise out of
whatever accidents matter happens to be endowed with, "either because
[matter], on account of its innate inertia and the sterility of bare potentia
[insitam inertiam nud(l!que potenti(l! sterilitatem] brings forth no accidents of
itself, or because in sublunary things there is one and the same [matter] and
so the same accidents would promiscuously inhere in all" (Coimbra, In Phys.
1cgqg, 1:180). The very fact that there are proper functions or active
powers requires explanation. Certain capacities are associated with individ
uals of each natural kind (it is supposed to be manifest that there are kinds).
The underlying matter (which is clearly intended to be prime matter) is
common to all. So even if there were accidents that inhered in matter as
such-even if matter were not pura potentia-they too would be common to
all. But of course different kinds of thing have different proper functions.
Hence something other than matter must be supposed as a reason for their
being proper.
The Coimbrans conclude that "the origin of such accidents must be
referred to a substantial form, as their source." Why a substantial, and not an
accidental, form? We have seen that one part of the definition of substantial
form is that it be the cause or origin of all other forms. That is so here,
pending a proof of uniqueness. One could, I suppose, argue for uniqueness
on grounds of simplicity. But a more forceful argument will be established if
we consider that these forms are united in the individuals to which they are
proper. The ground of unity, after all, could hardly be other than unique.
Suarez notes, in making the argument, that some accidents may indeed
be subordinate to others in the same individual, and originate from them.
Colors arise, for instance, from certain combinations of primary qualities.
But sometimes there is no subordination among them, as with heat and

humidity in air, whiteness and sweetness in milk, or the several senses in
animals; this multiplicity and variety of properties [ ...] requires one
32. "Matus intrinsecus est via et fluxus, via autem intrinsece respicit terminum" (Disp.
49§2'l[4, 26:go1).

Authenticated
form, in which all are united; otherwise they would be merely accidentally
united in the same subject, and if one were removed, the others would not
thereby also take their leave. But experience testifies that the opposite
holds; which is therefore a sign that such accidents, required in such a
subject and being in such number, weight, and measure, do not gain their
connection only in respect of the first subject or prime matter, but in
respect of some composite, by reason of which a certain order is required
among the forms of those accidents. (Suarez Disp. 15§ 1'l[q, opera
25:502)
We have, then, properties only accidentally related in themselves that nev

ertheless share a common fate in certain individuals. Chalk and chocolate
suffice to show that whiteness and sweetness are not inseparable. Yet in milk
(fresh milk, naturally) they stand or fall together. What is the reason for
that? Matter is indifferent. Even if these two were subordinated to a third
form, the same question could be raised in relation to that form and yet
others inhering in the same individual. Whatever hierarchy there is among
accidental forms, the search for a reason for their collocation will wind up
insisting that one form stand at the top, to account for the whole. Such a
form, as I have already argued, will satisfy the definition of 'substantial
form'. It will, I should add, not be the universal kernel briefly envisaged
earlier, since it is the ground and origin ofa particular collection of accidents
belonging to individuals of a certain kind.
4· Preferred states. In water, let us suppose, the independent properties of
coldness and wetness are conjoined. Yet some water is warm, even hot,
though perhaps never as hot as air (remember that 'water' means liquid
water, not steam). On what basis, then, does the Aristotelian call it cold? The
answer to that question provides one last empirical argument for substantial
form.
Some alterations, as we have seen, are irreversible. Others are not only
reversible, but tend naturally to be reversed. Hot water "intrinsically reverts
to its pristine coldness, as experience attests" (Suarez Disp. 15§ 118, opera
25:5oof). "Some intimate principle" must restore its coldness. No extrinsic
principle will do: some are not always present, and of those that are, the air
is naturally hot, or if cold will itself endeavor to restore its original heat
rather than cooling the water, while the various celestial and universal
causes do not have such effects.
The example was in fact quite controversial. Some thought that the cold
ness remained somehow in some part of the water even as it was heated.
Suarez calls that view "frivolous and contrary to experience," but goes on in
his usual way to offer reasons against it. Others held that coldness remains in
some degree throughout, and reasserts itself once the external cause of heat

Authenticated
Vicaria Dei
is removed. Suarez argues that a less intense degree of cold has no innate
tendency to become more intense, that the greater degree of heat, say, in
water near boiling ought to conquer any remanent coldness, and that in any
case there will not be any coldness, since it will have been conquered by the
heat.
We now believe that the water, so long as it is warmer than the surround
ing air, communicates heat to the air by imparting kinetic energy to the air
molecules. 33 Water has no "natural" temperature; its cooling comes about
by an extrinsic principle. So the example fails. But the principle it illustrates
has many other instances. Many kinds of thing do seem to have preferred
states. Excited electrons in an atom "spontaneously" emit photons and
return to a "ground" state. Proteins have preferred ways offolding up. Many
animals exhibit a tendency within certain limits to regulate their internal
states around a certain norm. The distinction between disease and health is
as between a "normal" and a "pathological" state. Cure consists partly in
removing the agent of disease, partly in the spontaneous restoration of the
body to its healthy state. There is, then, a wide variety of natural phenomena
in which a thing is taken to return spontaneously from a disturbed or
unnatural state to a ground or natural state.
Both the reversibility of some changes and the irreversibility of others are
"signs" or "indices" of substantial form. In reversible change a natural prop
erty is restored in the continuing presence of others; in irreversible change
many or all of them are altered at once and no longer spontaneously revert.
The argument, then, does not rest so much on essential properties them
selves as on the common fate of the qualities proper to a natural kind. For
the Aristotelian, their common fate requires a reason, and that reason is the
substantial form.
It may seem uninformative. What more has really been said than that
what is united is so for some reason, to which the obscure label 'substantial
form' is now attached? One answer is that it is informative not so much
through what it affirms as through what it denies. What is denied is that the
kinds we encounter in nature are merely collocations of accidents; 34 and
more specifically that reversion to pristine states can be explained entirely
by reference to other accidents. Some accounts are ruled out, and that is
33· Or rather we believe that in the long run the water and air together, considered as a
closed system, will approach a condition of maximum entropy, which effectively means that
inequalities in temperature will vanish. Toletus, who ascribes the account of cooling argued for
by Suarez to Avicenna, agrees instead with Walter Burley that the water is cooled by the
surrounding air (Toletus In Phys. 2c1q2, Opera 4:48). He considers what ought to be a crucial
instance: "What if the water were placed in a vacuum, where there was no air or anything
surrounding it?" The answer is that the heat would be corrupted by the wetness of the water,
but no new cold would be generated, for lack of a generator.
34· Some ancient "bundle" theories ofform are discussed in Sorabji 1988, c.4. I do not find
that the Aristotelians mention any of them.

Authenticated
progress of a sort. Another answer comes from the theory of generation

(§5.3): something other than accidental form is needed to explain the
manner in which preexisting stuff is "disposed" so as to become a new kind
of thing. Here again it is not single changes but the coordination of many
that seems to call for an independent principle.35
Even so, such arguments will not likely satisfy philosophers now. One
wants to know what the substantial form is. That question will not find a
straightfmward answer. First, because the expected kind of answer will not
have been sought by the Aristotelians. If one wants an analysis of substantial
form into simpler components-what philosophers now call "micro
structure"-one wants what Aristotelianism did not in general aim to pro
vide. Even the theory of secondary qualities does not proceed thus. Second,
because the only "analysis" Aristotelianism was willing to provide was to
describe the active powers associated with a form and the dispositions re
quired for its reception. It is in that way that the soul is studied. Finally,
because substantial form is that in which accidental forms inhere. The
human soul is that which "lives, feels, and thinks": if one asks what it is apart
from what it does, there is no answer. "Almost never can we explicate the
essences of things, as they are in the thing [prout in re sunt], but only through
their being ordered to some property" (Suarez Disp. 40§4'll16, opera
26:547).
We find here a familiar paradox. If one asks that an ultimate subject of
predication, a hoc aliquid or 'this something' be described apart from that
which is truly predicated of it, there is, of course, no answer. Substantial
form is, as we have seen, not quite a hoc aliquid-only the composite is-but
it is that in which all accidental forms inhere. Whatever is truly said of it, save
that it is existent and one, is said of an accidental form. Hence the question
'What is it?' cannot be answered; to ask it indicates a misunderstanding
of its role. 36
35· Mary Louise Gill argues that "active forms," like the soul, "co-ordinate distinct, and
often successive, motions toward a goal whose realization depends upon that co-ordination, as
in an organism's nutrition and growth" (1991:251). "Passive forms,"which would include not
only passive potenti~ as the Aristotelians understood them, but also certain active potentiO! like
heat and gravitas, cannot do that. Though Gill's active forms are not necessarily substantial
forms in the Aristotelians' sense, the line ofargument is similar. What one needs to connect the
two are: an argument to the effect that in each thing, or at least each sufficiently complex
thing, there is a unique active form that coordinates its activities, and which is "first" in the
Aristotelians' sense; and an argument showing that the form so defined can be identified with
the thing's essence or nature, or, in other words, that what coordinates the activities of a thing
is also that which defines it.
36. "Substance-in-general is, as far as we can know, a mysterious indeterminate stuff which is
logically indescribable except that we can say that it supports qualities. However, it was postu
lated just because it was thought that qualities needed support, so the hypothesis of substance
is a paradigm ad hoc hypothesis" (Alexander 1985:213, describing the "usual view").

Authenticated
Vicaria Dei
3·3· Form as Substance

Earlier I mentioned that some philosophers held that the notion of substan
tial form was incoherent. The objection, I said, rested on a misunderstand
ing. I will now examine briefly the response to it. The response turns on two
further notions, that of an incomplete substance, and the distinction between
inherence and formation. The fate of those two notions, and of the allied
distinction between first and second substances, hang together. I will sug
gest that the misunderstanding arose in part because the 'is' of specifica
tion, essential to defining second substances, and to making sense of in
complete substances, is elided in favor of the 'is' of accidental predication.
For the Aristotelians, interestingly enough, calling matter a substance was
more problematic than calling form a substance. Questions concerning the
substantiality of prime matter are rather more common than questions on
form. But the primary difficulty in each case was the same. A substance is
self-subsistent, requiring nothing else in order to exist. Yet matter and
form-at least corporeal form-each require the other to exist.37
Aristotle uses 'substance' in two not obviously compatible ways. In the
Categories, substance is defined as that which is "neither in a thing, nor of a
thing." 'Being in' was standardly taken to denote inherence, the relation
between a subject and its properties. In other texts, notably the Metaphysics,
Aristotle argues that form is substance, and indeed more so than matter
(Meta. 7c3, 102ga, De an. 2c1, 412a6ff).3s But if the relation of form to
matter, or to the composite, is inherence, then form cannot be a substance
according to the Categories.
To resolve the inconsistency, one could deny that substance is univocal, or
that its relation to matter or to the composite is one of inherence. The
strategy of Eustachius is effectively to deny that 'substance' has one sense.
Substantial form is "a certain substantial actus, but incomplete or (so to
speak) semisubstantia:I, and which joined with matter constitutes a single
37· Suarez reports the objection thus: "A contradiction [repugnantia] seems to be involved
when form is said to be both formative and substantial: for either it is a subsistent thing, and
requires no sustaining subject, or else it requires one: if the first holds, it cannot be a formative
form, because it is contradictory that what is subsistent should be received in another. If the
second holds, it is an inherent form, and therefore accidental. Hence there is no substantial
form" (Disp. 15§112, Opera 25:498). The term 'formative' (infurmans) distinguishes material
forms from those of angels and God, whose mode of existence does not involve their being the
form of something else. To such forms the objection does not apply. Infurmans is also used in
contrast with inherens to differentiate the way in which substantial form exists in the composite
with the way that accidents exist in substantial form or in matter.
38. "He [Aristotle] says first that the subject, which is the first particular substance [i.e., a
concrete individual, a "this"] is divided into three: matter, form, and the composite [... ] The
composite as well as the matter and form is called a particular substance" (Thomas In Meta.
7lect2, Cathala ,1276; cf. 1278).

Authenticated
whole substance [ unam integram substantiam constituit]" (Physica 1§2q5,

Summa 3:123-124). Hence when Aristotle argues that form (or matter) is
substance in the Metaphysics, he means incomplete substance or "semisubs
tance." Only the composite is a substance in the sense of the Categories.
But one could object that the very idea of "incomplete substance" is
contradictory. Suarez insists that it is not: "We can suppose [...] that such a
genus of being or incomplete substance is given in the nature of things:
what contradiction in this could anyone concoct or dream up?" It is not
inconsistent, he argues, to suppose that one and the same thing is both the
actus of something else, as form is of matter, and substantial; indeed the
more severe difficulty is to show how matter, which is not even an actus,
could be substantial. Hence whoever agrees that matter is substance must
also agree that form is substance. 39
The argument, however, seems curiously beside the point. The objection
cannot be turned aside merely by arguing ad hominem against those who
hold that matter is substance. 40 What we need is a coherent definition of
'incomplete substance' or a reasoned distinction between "informing" and
"inhering."
Recall that the key phrase in the standard definition was that in a com
plete substance substantial form and prime matter are unum per se (see
n.27). Accidental forms and matter are merely unum per accidens. The sense
of that contrast can be grasped well enough for now if one considers that
substantial form is the origin of accidental forms and that accidental forms
are therefore joined with matter only through substantial form.
The composite of substantial form and matter, then, is prior to any com
bination of matter and accidental form (but see §5.3). Matter, on the other
hand, cannot exist without substantial form (nor, of course, can material
form exist without matter). The union of matter with substantial form is
necessary for there to be an existing thing at all. Union with accidental
form, on the other hand, presupposes that there is an existing composite of
matter and form. The relation of matter and substantial form, informing, is
39· "But that is clear, whether because nothing can be assigned to either that is repugnant
with the other; or because the ratio of 'actus' of itself denotes perfection: and if that can
without contradiction be conjoined with being accidental, why should it be contradictory to
couple it with being substantial? Or finally because substantial being [entitas], since it concerns
perfection simpliciter, seems rather to contradict the ratio of 'potentiality' than that of 'actu
ality': but it is not contradictory with the first, since in prime matter we find [both substance
and potentia], and so neither is it with the second" (Suarez Disp. 15§ 1~[16, Opera 25:503-504).
40. The Coimbrans make a similar argument: "[prime] matter is bare potentia[ . ..] , and yet
it is, according to our opponent, a substance; hence form is all the more justly a substance" (In
Phys. 1c9q9, 1:179). The argument has force only if their earlier argument that matter is bare
potentia is accepted. Atomists, who hold both that matter is actual and that all forms are
accidental, will not be persuaded.

Authenticated
Vicaria Dei
quite different from the relation of matter of accidental form, inherence.

To speak of prime matter and substantial form as "incomplete substances,"
then, would be to mark their special bond with each other.
It may seem at times as if the disagreement of later philosophers is not so
much with the notion as with the name. But the disfavor that substantial
form fell into, and the misunderstandings it engendered, signal differences
between the Aristotelians and their successors that are not merely verbal.
Briefly put, those differences were a shift in emphasis in the definition of
substance, and a correlative restriction on the 'is' of predication.
Descartes, advising his protege Regius, writes that by 'substantial form' he
understands "a certain substance adjoined to matter, and with it composing
a merely corporeal whole, and which is not less, and even more, a true
substance, or a thing subsisting by itself [res per se subsistens], than matter is,
because it is called Actus, the other only Potentia" (To Regius]an. 1642; AT
3:502). There is nothing here that would surprise an Aristotelian, with one
exception. Form is indeed substance, and joined to matter to form a cor
poreal thing, and it is more ofa substance than matter, precisely because it is
actus, the other merely potentia.
What would make the Aristotelian pause is the gloss of 'substance' in this
context as 'thing subsisting by itself'. It is indeed not at all hard to find
substance so defined. Toletus writes that singular substance is called 'sub
stance' "because all accidents inhere in it first, and it itself is what subsists,"
and later that it is "a nature subsisting by itself [natura per se subsistens], and
in no way in another" (In Log., Prcedicamentum 5 [q3], opera 2: 10ob, 104b).
Subsistence per se is the defining feature of individual complete substances
like Sappho. But the Aristotelians, building on a distinction made in the
Categories between "first" and "second" substances, do not make it the defin
ing feature of substance tout court.41
First substances are singulars, like 'this human' and Sappho. Second
substances are genera or species, like 'human' and 'animal'. The distinction
between second substances and accidents parallels that between substantial
and accidental forms. 'Human' is truly said of Sappho, if the substantial
form of humanness informs the individual referred to by that name; 'white'
is truly said of Sappho, if the accidental form of white inheres in her. Second
substances and substantial forms, though they do not subsist per se, share
with first substances the feature of not inhering in anything.42
We have, then, two defining features of substance: subsistence per se, and
the weaker condition of noninherence. Descartes, by making subsistence
per se the sole defining feature of substance, effectively denies that second
41. See, e.g., Arriaga Cursus 855·

42. Toletus opera 4:78va,b.

Authenticated
substances are in any way substances. That denial is of a piece with his
rejection of substantial forms in physics. Substantial forms, like second sub
stances, satisfY only the weaker condition; they do not subsist per se. The
proscription has the further consequence of restricting the 'is' of predica
tion to 'being in'. 'Being of', as the relation that holds between individuals
and their generic or specific forms, can be done without. Logic is left with
the distinction, regarded as exhaustive, between the 'is' of identity and the
'is' of predication.43
It would take me too far afield to follow out the suggestion that the denial
of substantial forms, or second substances, and the suppression of the 'is' of
specification should be taken together as aspects of a single transformation.
But I will make one observation. Earlier I noted that one alternative to the
Aristotelian theory of substantial form would be a physics in which there was
just one kind of form, by which substances were individuated but not spec
ified. In Cartesian physics that form would be extension, which in the Princi
ples Descartes calls the "principal attribute" of body. All other physical prop
erties are, or are reducible to, "modes" of that attribute-or, to revert to
Aristotelian terms, accidental forms inhering in the one substantial form.
The Categories distinction between first and second substances, and with it
the 'is' of specification, though they might have a place in metaphysics,
where there is at least the other attribute of thought, have no work to do in
physics. In natural philosophy the 'is' of identity and the 'is' of predication
suffice.
The intent of this section has been to expose a body of doctrine much of
which is terra incognita even to historians of modern philosophy, and all the
more so to philosophers generally. It is a defense, not so much of substantial
form itself, as of the inescapability of the functions it performed in natural
43· Evidence for such a development is found in the Logic of Arnauld and Nicole, first
published in 1662. Substance is defined as "that which one conceives as subsisting by itself, and
as the subject of all that one conceives in it [ ce que l'on conroit comme subsistant par soi-meme, &
commelesujet de tout cequel'on y con{oit]" (Logique 1c2, P47).Amode, on the other hand, is "that
which, being conceived in the thing, and as not being able to subsist without it, determines it to
be in a certain manner [ ce qui etant confU dans Ia chose, & comme ne pouvant subsister sans elle, Ia
determine a etre d'une certainefafon]." A "modified thing" is a substance considered as "deter
mined by a certain manner or mode," as when we speak of a round body. Aristotle's 'being of'
has dropped out. Everything is either a concrete individual or a mode.
The significant passage comes a bit later: "Our mind, being accustomed to knowing most
things as modified [...] often divides substance even in its essence into two ideas, of which it
regards one as subject, the other as mode." So "one often considers man as the subject of
humanity habens humanitatem, and consequently as a modified thing." Every manner of
being, in other words, if it is not considered to be the substance itself (as God's infiniteness is
God himself) must be conceived as a mode-as "being in" the substance, not "of' it. Thus the
copula signifies either identity or inherence, but not specification.

Authenticated
[8o] Vicaria Dei
philosophy. 44 Those functions, which include the explanation of preferred

states and the common fate of conjoined qualities, were taken up into the
new science. We will see how in Cartesian physics the notion later christened
'configuration' by Malebranche took over the last of those functions. Sub
stantial form, moreover, was the source ofactive powers (see §5.4 for further
elaborations on that theme), and thus the ground of natural spontaneity.45
Descartes refused all spontaneity to nature, preferring to bestow movement
on his world with what Pascal called a tweak of God's finger (Pensees
77/1 oo 1). But it was not, in fact, so easily disposed of, especially when, in
the eighteenth century, the divine ghost was banished from the world
machine. The resurgence, in the writings of La Mettrie and Diderot, of a
vitalism entirely absent from Cartesian physiology is one testimony to its
stubbornness.
44· I should emphasize 'in natural philosophy'. Somewhat artificially, I have neglected
arguments for substantial form which appeal to various characteristics of the human soul, and
in particular its separability from the body.
45· For a modem Thomist defense of active powers, see Weisheipl 1985:g-to.

Authenticated
Matter, Quantity, and Figure
T he substrate argument yields a straightforward distinction be

tween form and matter. With that distinction many philosophers,
Aristotelian or not, would have agreed. But the Aristotelians, as
we have seen, divided form itself into substantial and accidental.
Corresponding to that distinction is a distinction between prime matter and
matter broadly understood. Prime matter is the stuff, whatever it may be,
which, when joined with substantial form, yields an individual substance.
Matter broadly understood is either the composite substance, regarded as
the subject of accidental form, or else the already formed and disposed
matter necessary to the more perfect among substantial forms: the human
soul cannot naturally be joined with anything but a human body.
Matter broadly understood was unproblematic. Prime matter, on the
other hand, met with opprobrium even among philosophers friendly to
form and matter. Prime matter, they said, is incoherent, incognizable, super
fluous. Walter Charleton, an English follower of Gassendi, describes it as
"rather Potential, than Actual, and absolutely devoid of all Quantity; then
which we know no more open and inexcusable a Contradiction" (Physiologia
88). Corpuscularians like Descartes and atomists like Gassendi found the
hypothesis of an entirely formless stuff underlying all corporeal substances
to be useless, or worse than useless. Far better to suppose, as Descartes did,
that matter has the attribute of extension, or, as the atomists did, that the
permanent substrate underlying natural change consists of immutable indi
vidual substances, each fully specified with respect to a list of fundamental
physical properties.
The objections had long been noted within Aristotelianism itself. Stan
dard quCEstiones are devoted to the existence of prime matter, to its essence,
and to our knowledge of it. The complaint about superfluity, though not
[81]
Authenticated
Vicaria Dei
explicitly addressed, is implicitly answered in refutations of atomism. The

cavils of seventeenth-century philosophers were anticipated by Avicenna
and Averroes five hundred years earlier, and before that by the Greeks.
Aristotelianism had long since accommodated itself to them. It is, therefore,
all the more puzzling that, in the first decades of the seventeenth century,
they became decisive reasons for throwing off the whole weight of tradition.
I consider first questions about the essence of prime matter. The starting
point for all discussions is the definition at the end of Physics I· "I call
'matter' the first subject for each thing, from which each thing comes
immanently and not accidentally" (Phys. 1cg, 192a31 ff). Aristotle's version
of the substrate argument (§3.1) makes it clear that by 'first subject' he
means that which persists through substantial change, the subject of genera
tion and corruption and not merely of alteration. Thomas argued that
prime matter must be pure potentia, not only because it is indifferently
receptive to every substantial form, but because it lacks any actuality of its
own. All actuality comes from form; and since existence entails actuality,
what has no form has no proper existence.
Against that both Scotus and Ockham had argued that matter, as part of
complete substance, and as the subject of form, must have an actuality and
existence of its own. Scotus (see §3.2) held that all material things share a
forma corporeitatis, that belongs to matter qua matter and thus actualizes it
apart from substantial form. Ockham held not only that matter is an actual
entity, but that it is "of itself quantified and endowed with dimensions. "1
With the exception of the Coimbrans, whose position I find confused, the
central texts all reject the Thomist view. Broadly put, their position is that
actuality is equivocal. One sort of actus consists in existence, the other in
determination. Matter is pure potentia only in the sense that it is in potentia to
all forms (whether indifferently so was a further question). It is not pure
potentia in the sense of lacking existence independent of form. In effect
Smirez and the others reject the unrestricted claim that all existence comes
from form.
I then turn to quantity. Aristotelian physics is often said to be qualitative,
and its successor quantitative. But a better distinction, as Annaliese Maier
has argued, is not between a physics of quantity and one of quality, but
between a physics of intensive and one of extensive quantity. Intensive quan
tity is exemplified by degrees of heat, but the notion was also applied to
virtue. It is, unlike Cartesian extension, a mode of quality first of all. From
the fourteenth century onward, intensive quantity becomes increasingly the
focus of inquiry in Aristotelian natural philosophy. The importance of ques
1. Ockham Summula 1c1o, 13; the quotation is from p192. On Scotus's and Ockham's
treatment of prime matter, see Adams 1987:639-647. Ockham attributes the view to Averroes,
as does Zabarella, who endorses it.

Authenticated
tions concerning intensive quantity is apparent, for example, in the

fourteenth-century Paris philosopher Nicole Oresme's "geometric treat
ment" of alteration.
Nevertheless, quantity plays a subordinate role. The Calculatores at Oxford
and their successors at Paris excelled in thought experiments. Evidence of
their having actually measured anything is hard to come by. Even Oresme
made no attempt to measure intensities. The importance of quantity lay not
in its being measured, or treated mathematically, but in the fact that quan
tity stands to matter as quality to form. Each was the primary "assistant"
seiVing to fulfill the ends of the incomplete substance it inhered in. Because
matter cannot naturally subsist without quantity, some philosophers came to
believe that quantity is essential to matter, or at least that quantity, unlike
other accidental forms, precedes substantial form-is presupposed by it
rather than the other way around. I will consider in some detail Suarez's
rather nuanced view. The argument that quantity is not essential to matter
turns, interestingly enough, on the traditional explanation of transubstan
tiation, one of the few theological controversies in which Descartes permit
ted himself to intexvene.
Quantity could be indeterminate or determinate. Determinate con
tinuous quantity must have what we would now call boundaries. Figure is a
particular way or mode of determination. It is a real physical quality, but it is,
as a mere mode, among the most dependent of qualities. In particular
figure is passive in extremis (see§4.3). Yet since it is found in all matter, and of
infinite variety, figure might sexve-so the ancient atomists had already
surmised-as a succedaneum for form. Descartes often emphasized the
clarity of our ideas of figure and the certainty with which in geometry we
reason about it. But I think it equally important that figure, as a physical
property, is ubiquitous, infinitely variable, passive, and only modally distinct
from extension. To substitute figure for form, as Descartes did, was effec
tively to eliminate activity from nature and substantiality from form.
4.1. The Essence of Matter

Prime matter is "very like darkness, which we perceive when we see nothing,
while when we see we do not know it. "2 No doubt part of the difficulty lies in
2. Summa pt3, Physica 1tndisp2§1q1, 3:119. Allusions to Augustine were commonplace in
questions about the cognizability of matter (Suarez Disp. 13§614, Opera 25:421; Coimbra In
Phys. 1cgqn2, 1:154). In the Confessions Augustine writes that we "know it in not knowing it or
are ignorant of it in knowing it [nosse ignorando velignorare noscendo]" (Conf. 12c5, p218). The
association with darkness is brought out by A':gidius Romanus, who writes that "matter has
something in it of darkness, since in it all the remarkable splendor of form is lacking" (Hexam.
5c3, quoted in Coimbra ib.). Fonseca says that prime matter is called Tenebra since, being

Authenticated
Vicaria Dei
attempting to know that from which every mark by which we know individ
ual substances is absent. But the difficulty lies also in the paradoxes that
seem to bedevil the concept. Matter was introduced in the first place as the
substrate persisting through substantial change. It therefore exists. What
ever exists is actual (since what is in potentia only does not exist as such), and
whatever is actual is actual by virtue of having form. But prime matter,
because it is the correlate of substantial form, can have no substantial form
itself. Nor, if all accidental forms presuppose an actual composite of sub
stantial form and matter, can it have any accidental form. Prime matter is
inactual, it does not exist. 3 But it was supposed in the first place to be the
persisting substrate of change, since otherwise generation and creation
would coincide. So it would seem that prime matter is both actual and not,
both existing and not, both something and not something.
Nevertheless, the existence of prime matter was never seriously in doubt,
no more than that of substantial form. Questions therefore concentrated on
its essence. Here there were two avenues of escape from the paradoxes. One
was to grant form to matter. From that standpoint, atomism amounts to
supposing that prime matter consists of a plurality of complete substances.
Even if, as the Aristotelians believed, atomism is false, one could still hold
that there is a generic form common to all material substances. Such a form,
often called the forma corporeitatis, was proposed by Avicenna and, in a more
restricted way, by Scotus and Henry of Ghent. Matter endowed with it would
be an actual existing thing, and the paradoxes would be avoided.
Or one could deny that all accidental forms presuppose substantial form,
and that actual existence requires a substantial form. Prime matter could
then exist and be the material cause of substance by virtue of an essential
accidental form. The most obvious candidate was quantity, since no material
substance can exist naturally without it. Prime matter, the permanent sub
strate of natural change, would be a quasi substance, quantified matter. That
was, effectively, ~he way of Descartes.
without form, it cannot be known except by analogy (In meta. tqq3§8, 1:36gE). Seventeenth
century critics of the notion were pleased to repeat these commonplaces.
3· Dupleix states the dilemma succinctly: "Si Ia matiere premiere est quelque chose elle est
substance ou accident. Or elle n'est ny substance ny accident: substance parce qu'il n'y a point
de substance (pour le moins materielle et corporelle) sans forme: accident, d'autant qu'estant
accident elle ne pourroit pas estre principe ni partie des substances: car Ia substance est le
subject et le fondement des accidens [...] Partant il n'y a point de matiere premiere en
aucune sorte": "If prime matter is something it is either a substance or an accident. But it is
neither a substance nor an accident: [it is not a] substance because there is no substance (or at
least no material and corporeal substance) without form; [it is not an] accident, since if it were
an accident it could not be either a principle or a part of substances: substance is the subject
and ground of accidents [...] Hence there is no prime matter of any sort" (Dupleix, Physique
2c5a2, p13o). For a recent presentation of a similar argument, see Graham 1987.

Authenticated
Matter, Quantity, and Figure [SsJ
The view predominant in the central texts, however, is that prime matter
has neither substantial nor accidental form. Its essence consists in its being
in potentia to all forms. The paradoxes were avoided not by qualifYing the
claim that matter is potentia, but by showing that it could nevertheless enjoy
a sort of actuality. In what follows I will trace the way to the characteristic
Thomist thesis that matter is pure potentia, and the reasons why the majority
of the central texts depart from it. The question, as we will see, concerns not
merely the definition of matter, but the genesis of material individuals.
The position of the central texts can be summed up thus. Prime matter is
(i) Unique: there is but one substrate of natural change. The atomists
and various other ancient philosophers were mistaken in supposing
several or even infinitely many kinds of ultimate matter.
(ii) Simple: prime matter is not a composite of substance and form. The
Scotists were mistaken in attributing to all material substances a
generic "form of corporeality."
(iii) Actual: it is not incoherent to hold both that prime matter is pure
potentia and that it exists. But it does not exist, as the Scotists again
believed, by virtue of a special kind of actuality distinct from that
conferred by form. It exists just by virtue of being a component, with
substantial form, in complete substances. (Here one finds dissen
sion: the position I am describing is that of Suarez.)
1. Uniqueness. Fonseca and Suarez emphasize that every kind of sublunary

substance can change into every other kind. Substantial change is in princi
ple ubiquitous. Experience makes it evident that "the elements act mutually
among themselves, and one is converted into another, either mediately or
immediately, and mixtures also are generated from them, and consequently
are resolved into them, so that all sublunary things, according to the virtue
[vis] of their nature and composition, are mutually transmutable. I say
'according to the virtue of their nature' because it can happen that some
parts of the elements are never transmuted, on account of their being in
hidden and very remote places, which the actions of contrary agents do not
reach" (Disp. 13§115, opera 25:396-397). 4 Substances could have been
arranged into disjoint genera, within which transmutation was possible but
4· "And if there is given a common subject of all transmutations, it itself cannot be made
out of yet another subject, for then it would not underlie every transmutation; therefore if in
every transmutation there is given a common subject, and all things (according to us) are
transmuted mutually into one another, it follows necessarily that there should be given a first
subject not made out of another," and thus a first matter (Fonseca In meta. rqq2§2; r :330F
331A). On the mutual transmutability of the elements, see De gen. 2q, 33oa61T; De cado 3c6,
304b23ff·

Authenticated
[86] Vicaria Dei
not otherwise. 5 But the elements can be changed each into the others, as
Aristotle argues in De generatione et corruptione. Provided that all other sub
stances, whether mixtures like wine or animate beings like humans, can be
resolved into their elemental components, the mutual transmutability of
the elements entails that all other sublunary substances are in principle also
mutually transmutable.
Mutual transmutability therefore renders idle any suggestion of specific
differences in matter. It is "of itself indifferent to all forms of corruptible
things, and to their dispositions; it requires therefore in itself no distinction
or multiplication of kinds" (Suarez Disp. 13§2'18, opera 25:401). If it must
have a form, that form will have to be universal.
Descartes too holds that all material substances can be changed into one
another (Monde 5; AT 11:28), unlike at~mists like Gassendi, for whom the
ultimate constituents of matter are incorruptible.6 In Descartes's physics,
extension occupies the role of prime matter. It occupies all space, underlies
all corporeal things, cannot be destroyed or created except by God. Since
Aristotelian matter cannot naturally subsist without quantity, only one
point, it would seem, separates Descartes from his predecessors: the con
ceivability, as opposed to the physical possibility, of matter deprived of ex
tension. If, as usual, we put the question in terms of the absolute and
ordained power of God, then the difference is this: the Aristotelians be
lieved that God, according to his absolute power, could allow matter to
subsist without quantity, while Descartes did not (§4.2). Of course, once
Descartes identifies the substrate with extension, or nearly so, he excludes
from matter all properties not contained in extension. That cuts deeply
enough for the common ground to escape notice. Still he granted the
universality of change, and with it the consequence of a single substrate.
Instead of declaring prime matter absurd he simulated it.
2. Simplicity. The second goal is to show that prime matter, though not an
element, is not a unique substance underlying all others-that it is not itself
composite. There were reasons to believe othetwise. 7 One argument is that
in generation what receives substantial form cannot have been entirely
devoid of form, since then the form would have been produced from noth
ing. So even prime matter must have an "inchoative" form out of which
5· There are, in fact, two genera. Terrestrial and celestial matter are not mutually transmuta
ble (Suarez Disp. 13§1,11). Celestial matter is incorruptible (see Aristotle Meta. 12c2,
106ga3off; De ccelo 2c1, etc.). It undergoes no substantial change, and a fortiori no change into
terrestrial matter. The question therefore does arise whetller celestial and terrestrial materia
prima are of one species (see Suarez Disp. 13§11; Zabarella, De natura coeliin De rebus nat. 270
290).
6. See opera 1:256b, 266b, 273b.
7· I consider only two physical arguments. For oilier arguments, chiefly concerning the
human body, see Soncinas Q. meta. 8q8.

Authenticated
individual substantial forms are "educed."8 Since this argument concerns

the production of forms, I defer it until §5·3·
The second argument requires a few preliminaries about form that were
not needed earlier. We have seen that in the Categories Aristotle divides
substances into first and second. First substances are concrete individuals.
Second substances are genera and species. The analogy that warrants calling
them both 'substance' is that both are subjects of predication: one may
equally well say 'Sappho is wise' and 'The human is wise'. The Physics and
the Metaphysics introduce into first substances a structure unmentioned in
the Categories: each individual is a composite of matter and form (or forms).
Instead of being motivated logically, the new distinction is motivated, as we
have seen, by an analysis of change.
The question naturally arises how the two accounts are to be reconciled.9
Genera are divided into species according to differentia like 'winged', 'ra
tional'. In the biological works, the animal and plant worlds are classified in
terms of numerous crosscutting criteria like 'terrestrial', 'aquatic', 'two
footed', 'four-footed', and so forth. That system would sit happily with a
metaphysics in which the defining form of a thing was a set of accidental
forms. But we have seen that substantial forms are not merely collections of
accidental forms. So although a human is an animal and rational, the hu
man form is one, not several. We have, then, a problem: how are the obvious
relations between substances that are exploited in taxonomy to be reflected
in the metaphysics of matter and form?
The second argument in favor of a forma corporeitatis rests on one answer
to that question. Individuals are grouped under a genus or species by virtue
of a common nature, that if the generic term is univocal must be real rather
than imposed by reason. But there cannot be such a common nature except
if there is a form common to all those individuals. As Averroes says, "all the
parts of a definition are forms; the genus is a universal form, while the
differentia is a particular form. "10 Taking that claim to be correct for the
moment, we then have a straightforward general argument on behalf of a
forma corporeitatis: "Body, which is the highest genus in the category of sub
8. Although the forma inchoativa and the forma corporeitatis have different origins and
rationales, both terms are used in discussions of Avicenna. On forma inchoativa in Albert and
Thomas, see Nardi 1g6o, c. 2. There the doctrine is traced back to the semina or rationes seminales
ofAugustine (Gen. ad litt. 601, gc 17) and to Stoic-Neoplatonic doctrines (Nardi 1960:76).
g. Recent commentators usually explain the difference in terms of Aristotle's develop
ment (see Furth 1g88; on the developmental mode of criticism, see Owen 1978:xv-xvi, and
the references there cited).
1o. Zabarella, ib. "Definitions are composed of a universal form, which is the genus, and a
proper [form], which is the differentia" (Averroes In Phys. 2text28, Dpera4:5gK). "But you may
object that Aristotle[...] says that the parts of the definition correspond to parts of the thing;
but the parts of the definition are the genus and the differentia; hence to those parts there
correspond in the defined thing parts that are really diverse, and which are not matter and
form alone" (Suarez Disp. 15§1015, opera 25:537; cf. Aristotle Meta. 7c1o, 1035b2olf).

Authenticated
[88] VicariaDei
stance, is the univocal genus of all perishable bodies, and so there corre
sponds to it in things a certain form participated in by all such bodies; there
exists, therefore, a 'form of body' fforma corporis]" (Zabarella De rebus nat.
207E; cf. Soncinas Q. meta. 8q8, p1g2a; Coimbra In Phys. 1cgq3a2). The
"form of body," because it has no contrary form, can never perish; nor can it
be generated; and so it will be, as the expression goes, "coeternal" with
matter.
What was at stake in the question is best grasped by looking at the argu
ments against the forma corporeitatis. One rests on the by now familiar distinc
tion between substantial and accidental change. If the forma corporeitatis
suffices to make matter substance, and if there can be but one substantial
form in each individual, then all other change-from elemental mutation
upward-must be accidental. Indeed, since material substances are never
changed except into other material substances, all physical change must be
accidental. We have seen, however, that elemental mutation must be
distinguished from, say, alteration in degree of heat. There can be, there
fore, no forma corporeitatis.
But must substantial forms be unique? That is what the second argument
denies. The forma corporeitatis suffices to make matter a substance, but not to
make that substance animate or human. To be animate is to have a second
form, the soul. Some philosophers argued that humans have three souls
vegetative, sensible, and rational. The vegetative soul corresponds to the
genus living thing; the sensible to the genus animal; the rational to the
species human. The same, they believed, must hold of every individual,
animate or not: it will have a full range of distinct forms, at once "substan
tiating" and "specifying," from genus generalissima to infima species.
To argue against the forma corporeitatis, then, one must show that a single
individual cannot have several "substantiating" forms. Only the last of those
forms, it should be noted, will yield a fully specified individual. The others,
as Zabarella puts it, "indeed produce an actus [in matter], but an imperfect
one, mixed with potentia. "11 One cannot argue, then, simply that a plurality
of forms would-absurdly-yield a plurality of fully specified individuals
out of one matter. Suarez, recognizing that, rests his argument on the thesis
that "there can be no form that produces just generic being."
He offers two arguments, one inductive, one from the ratio of substantial
form. The inductive argument will likely strike readers now as strange, and is
11. 'They [the defenders of Avicenna] would say that substantial form is twofold, one
specific and[...] ultimate, which is the ultimate form of the species, like man, horse, cow; the
other common and general, by which is constituted not the species infima, but a genus[...] To
form of the second sort existence cannot be attributed; only with the specific form can this be
done, since the general form indeed produces an actus, but an imperfect one, mixed with
potentia [cum potestate commistum]; while the specific form produces a perfected actus, requiring
no further form in order to exist" (Zabarella De rebus nat. 208C-D).

Authenticated
Matter, Quantity, and Figure [8g]
worth noting for that very reason. It is an instance of the ordering of forms
(see §6.2). Accidental forms never fail to be ultimately specific. Nothing is
just colored; it must be a particular shade of red or blue or whatever. So too
for the entirely spiritual forms of angels. But material substantial forms are
"a kind of medium [quasi media!] "between the two, and there is no reason to
think them different, so they too must be ultimately specific. The point
angels aside-is that in both instances there is a genus-species distinction,
and yet no form of either kind fails to be as specific as it can be (Disp.
13§31 q, opera 25:407).
One could argue that the question is begged. Perhaps there are generic
accidental forms too. The argument must rest on an implicit understanding
that since the effects of quality can always be explained in terms of specific
forms, there is no reason to suppose any generic accidental form. What we
see, for example, is always an ultimately specific shade of color, not color in
general. Where accidental form is concerned, there is no other reason to
suppose generic forms. The forma corporeitatis, too, since it occurs in every
material substance and can neither be generated nor corrupted, has no
physical effects peculiar to it. Experience can yield no evidence on its behalf
(Disp. 13§3!14, opera 25:406). 12 We have no a posteriori argument for
generic form.
Indeed the very idea of substantial form precludes "substantiating" forms
that confer merely generic or incomplete being. "Every substantial form has
its proper mode of being [propriam entitatem distinctam] in the thing itself,
distinct from that of matter or other forms; it is therefore necessary to
understand in that mode of being an ultimate and proper difference
[differentiam], or a reason by which it is distinguished essentially from other
forms with diverse essences, with which it agrees with respect to the com
mon reason of being substantial form" (Disp. 13§31 17, opera 25:407). The
crucial point is that for each substantial-or "substantiating"-form, its
mode of being in its subject (i.e., as informing the subject) must be
distinguished from those of other substantiating forms by a difference that
amounts to the difference between species infima. It will not suffice for two
such forms to differ as genus and species. The reason, I take it, is that the
genus, being contained in each of its species, is not really distinct from
them. 1 3 The generic form of animality and the specific form of humanity in
12. Suarez notes that one might argue that quantity, which is accessible to the senses,
inheres in matter directly, rather than through an individual substantial form, and thus re
quires that matter have a generic form of its own prior to individual form. But the presence of
quantity in matter presupposes that of substantial form (cf., however, §5.3), and so the argu
ment fails.
13. See Disp. 6§g1[15ff, Opera 25:24off, which cites Fonseca In meta. 5c28q14§3, 2:1078 11•
The argument in brief is that a genus is such by virtue of being differentiable into species; so it
must first be differentiated specifically before it can serve as a principle of individuation. A

Authenticated
[go] Vicaria Dei
Socrates are not really distinct and share the same mode of being in him.
Only another specific form-that of dog, say-would be really distinct from
that of humanity. But clearly those two forms cannot coexist in one individ
ual at the same time.
To put the point another way: suppose for a moment that a substantial
form were just a list of essential properties, or rather, not just any list but a
complete (and ordered) list. A genus could not be a complete list, since it can
still be added to. Only a species infima could be complete, and any species
infima will contain the genera superior to it. None of them will be distinct
from it. Species infima of the same genus will be complete lists differing in
their last component. But the differentia of species under the same genus
are contraries. Hence no one individual could have two species infima. In
short, of the candidates for substantial form, the genera superior to a species
infima are all contained in it, while other species infima are incompossible
with it. Substantial form, therefore, is unique.
The argument might still seem to beg the question. It depends on our
agreeing that substantiation and individuation go hand in hand, so that the
only substances are (fully specified) individuals. Incompletely specified
substances-in the sense in which the forma corporeitatis, together with its
matter, was earlier considered to be an incomplete substance-are not
really substances at all, or if they are, they are only complete substances
incompletely described. To go further with this point would take me into
the dense and treacherous domain of universals. Suffice it to say that, al
though Aristotelianism is not celebrated for the austerity of its ontology, the
list of things to which its physics accords full-fledged individual existence is
in fact rather short. There are concrete individuals like Socrates and Fido,
and within each such individual its prime matter and substantial form
though matter and form are, as we have seen, regarded as "incomplete
substances," since each requires the other to exist naturally.I4 There are,
moreover, certain accidental forms that the Aristotelians held to be really
distinct from the substances they naturally inhere in. Everything else de
pends, either by inherence or by stronger relations of dependence, on
them.
The argument against the forma corporeitatis, then, exemplifies a more
comprehensive refusal to countenance a plurality of substantial forms in
species infima, on the other hand, includes "all specific, generic, or higher predicates that can
be abstracted or conceived in it" (240). To say ofsomething that it is animal is incomplete. One
hasn't, so to speak, yet enough information to pick out the thing as 'this so-and-so'. Only the
species infima will do. But if one has picked it out as 'this human', the predicates 'animal',
'living', 'corporeal', 'substance' are all implied and need not be separately asserted of it.
14· I am setting aside the doctrine of real accidents whose primary role was to explain how
transubstantiation was possible. I should also emphasize that the Aristotelians recognized an
infinite variety of spiritual substances: human souls, the angels, and God.

Authenticated
one individual. We will see that refusal again in arguments about the unity
of the soul. I will close this discussion with two remarks.
Zabarella's argument suggests that individuals might originate in a pro
cess that, from faceless prime matter, proceeds through ever more specific
genera to the species infima and thence to individuals. 15 Aristotelianism ac
knowledges such an order-an order of specification. But it refuses to
translate that order into an order of generation. In the Creation, what God
made was creatures like Adam and Eve-fully differentiated concrete indi
viduals. Any suggestion of a successive emanation from an original One or
Being, such as one finds in some Neoplatonist and Gnostic cosmologies, is
turned aside. The word 'emanation', as we will see, does occur in Aristo
telian metaphysics. But it does not denote a succession of ever more specific
substantiating forms within individuals, or of one individual from another.
It denotes the relation of proper accidental forms to the substantial forms
that unite them. Emanation in that sense is not a temporal relation; nor,
more important, is it a diversification of substance.
The second remark concerns prime matter. Some philosophers, among
them Scotus and Suarez, held that prime matter could exist without form
(§5.1). But it could do so only preternaturally, through the absolute power
of God, who then takes over the role of form as an extrinsic cause of matter.
Matter remains in that circumstance as devoid of proper activity, and as
dependent on something else for its existence, as it is in the natural order.
For Gnosticism, and especially for Manichaeism, matter was to some
degree independent of God and resistant to his will.16 By establishing that
matter depended entirely on God for its existence and activity, Christian
theology, as Hans Blumenberg has argued, attempted to answer the gnostic
challenge-in his view unsatisfactorily (Blumenberg 1988, pt.2, esp. 139
149). Thomas and his followers reduced matter to pure potentia, lacking
even a proper existence. That view leads to paradox; the central texts with
draw to varying degrees from Thomas's position. But they minimize, all the
same, the independence of matter, and they remain one with Thomas in
according it no active powers.
Descartes, as we will see, preserves that view. Indeed he strengthens it: not
only matter but material substance, including animals, is devoid of activity.
Yet by virtually identifying matter with extension, he restores to it at least the
possibility of independent existence. Though he himself held that created
15. Thus Peter ofTarantasia (Innocent V) writes that the ratio semina/is (see n.g) is "like a
beginning or seed of the complete form in matter, which by the action of a natural agent is led
out from potentia to actus. It flows or passes from one being to another, until it arrives at the
being of the last, and completing, form" (In Sent. distr8qra3, cited in Nardi rg6o:76).
r6. The Coimbrans write that "the Manicha:ans, according to St. Augustine [De natura boni
r8], affirmed that matter was the formative power (fonnatrix] of all bodies" (Coimbra In Phys.
rcgq3a2).

Authenticated
Vicaria Dei
substances must at every moment be consetved by God, his definition of

matter, by implicitly giving matter a "form of body," opens the way to con
sidering matter again in its own right as self-subsistent. What remained was
to restore to it some sort of activity. That was accomplished when Newton
ascribed to all matter qua matter a power of attraction.
3· Actuality. If prime matter has no form of its own, the paradox remains
unsolved. What remains is either to loosen the tie between form and actu
ality or to show that the actuality enjoyed by matter in composite substance
suffices to solve the paradox. I will first lay out the Coimbrans' position,
which insists that matter is pure potentia, and then summarize the position of
the other central texts, which hold that it is not.
The Coimbrans argue that matter "is neither an actus, nor something
composed of potentia and actus" (In Phys. 1cgq3a1, p156). The argument is
by cases. Clearly matter is not a composite, since that would only raise a
similar question about the "matter" of matter. If matter is an actus, it is either
an actus subsistens or an actus injormans. An actus subsistens is an actus that
requires no matter ("a materi:e societate omnino seiunctus"), like God.
Clearly matter is not God (though some, "fallen with the Manich:eans un
der the blows of madness" (158), have sunk so low as to affirm this). If it
were an actus subsistens it would have no need ofform. So if it is an actus it is
an actus injormans. But then it is the informing actus of some subject, which is
contrary to its definition as first subject. It is, therefore, no actus at all.
It is easy to sympathize with those who found the doctrine unintelligi
ble.l7 It is difficult to conceive how something whose essence is potentia
could so much as exist Uust as God, whose essence is actus, cannot be
conceived not to exist). Something of the flavor of the difficulty can be
gotten from an argument of Averroes (Toletus In Phys. 1c7q q, opera
4:35ra): ifthe essence of matter were potentia, then when form wasjoined to
it, it would be destroyed, since form is actus and the opposite of potentia.l 8
The trouble arises as soon as substantial form is defined as that which "gives
being simpliciter," as.the expression goes. The contribution of matter to the
composite consists solely in being that to which form gives being. Like the
receptacle or matrix of the Timt1!Us, prime matter setves only as the indefi
17. Charleton was by no means the first. Toletus writes that matter "has an actus in its own
right [secundum se], and is not pure potentia. This is indeed so certain, that the opposite cannot
be understood" (Toletus In Phys. 1c7q13, Opera 4:34rb). Zabarella, too, writes that "if others
are endowed with so much acumen, that they are [...) able to imagine this incorporate
matter, I for one (confessing my ignorance) cannot do so at all" ( Zabarella De relnts nat. 2 18A).
By 'incorporate matter' (materia incarporea) Zabarella means matter considered in abstraction
from its being extended in three dimensions, and in particular matter considered as pure
potentia.
18. "Si potentia esset in substantia eius [i.e., primi subjecti mutationis], tunc esse eius
destruereter ablatione potenti.e" (Averroes In Phys. 1text7o, Opera 4:41F). See also Coimbra In
Phys. 1cgq3a2, p158 (2d argument).

Authenticated
Matter, Quantity, and Figure [g3]
nite ground of existence, distinguishing those forms which are from those
which are not-a sort of reified 3. But if the office of matter is nothing more
than to receive the actus of form, and if that is nothing more than coming to
be, then it would seem that matter is dispensable in favor of the creative and
conseiVing acts of God. Put bluntly: matter is nothing other than God
himselfwith respect to those acts. 19
That way lies heresy, or Spinoza. But it is not easy to see why prime matter
is not superfluous. Part of the answer is that it does in fact have a proper
actus. In Toletus and Suarez, the argument for this has, broadly speaking,
three stages. 20 The first is to distinguish between an actus that consists in
reception of form and an actus that consists merely in existence. Next they
show that the actus that consists merely in existence need not be an actus
subsistens, or the existence of a self-subsistent thing. 2I Then, they distinguish
between the perfect or complete actus of composite substance and the
imperfect actus of its components:
The first is that actus which in the genus of being simpliciter or of substance
is complete, so that it is neither constituted by a physical actus distinct
from itself [i.e., it is not a material form] nor is actualized by one [i.e., it is
not matter], nor does it require such an actus to exist. [...] [Imperfect
actus] designates that being which has some actuality, insofar it is actually
apart from nothing [in quantum actu est extra nihi[J. But the actuality it has
is incomplete and imperfect, because it is not so sufficient as not to need
another actus, either to complete it with respect to being simpliciter or even
in order to exist. (Suarez Disp. 13§5~8, opera 25:416)
Since substantial form too is characterized as an incomplete actus, that of

matter is said to be "in potentia to any perfect actus whatsoever" (Toletus In
19. Thomas records an argument whose conclusion is that God and prime matter are
entirely identical. The argument is just that both, being simple, cannot differ, because
difference implies differentia and thus compositeness (ST 1q3a8; cf. Coimbra In Phys.
1cgq3a2, p158). Thomas's refutation of that argument, which may have come from David of
Dinant (see note 26 below), does not touch the issue raised here.
20. Toletus In Phys. Ic7q13, Opera 4:34rb-35va; Suarez Disp. 13§5, Opera 25:414rr. Eu
stachius puts the point in terms of a distinction between the actus physicus of form (which is the
actus that figures in the definition of motus) and an actus metaphysicus common to both matter
and form. The latter seems to consist in nothing more than that matter indeed exists, since it is
part of the composite (Eustachius Summa pt3, Physica nnq3, 3:121). Even the Coimbrans
admit as much. What they deny is that the existence of matter in the composite is its existence,
or in other words, that matter itself can be divided in any way into what is actual and what is
merely potential. Since, following Thomas, they make a real distinction between essence and
existence, they hold that "even if in the thing [matter] is an existing essence, still that existence
is potential, so much so that it is not capable ofexistence without the mediation ofform; and in
that sense they can call matter 'pure potentia', even in relation to the actus of existence [in ordine
ad actum entitativum]" (ib. 'll 7). That, if I understand it correctly, is the view that I have
suggested leads toward Spinoza.
21. Suarez Disp. 13§5t5, Opera 25:415.

Authenticated
Vicaria Dei
Phys. rc7qr], Dpera4:34va). But that is not the same as being in pura potentia.
Even though, as Augustine said, prime matter is "next to nothing" (prope
nihil: see Conf 12c6, p. 219), still it is "created, receives form, and is a
component of the composite, all of which include existence" (1)2, 417).22
The crucial point is that matter is, "metaphysically" though not "physi
cally," a composite of potentia and actus. Physically it is not, because it has no
form. Metaphysically it is, because like form it is a component of substance.
Even so, matter doesn't amount to much: "Matter is an entity of such a sort
that by itself it is not enough to exist without a substantial actus perfecting
and actualizing it; so that from the force of its being alone it includes no
formal adus either formally or eminently [...] Whatever sort of being is in
prime matter, it is entirely in service to the potentia for receiving substantial
form; for to that it is primarily and per se instituted. "23 The arguments on
behalf of an actus proper to matter show only that the underlying substrate,
the first subject, whatever it may be, must be at least metaphysically in actu.
There is no reason why God could not play the role. Yet to put God in the
place of matter was deepest heresy, to whose proponents the Coimbrans give
no name.
The passage from Albert to which they refer does name names. Epicurus
and Alexander of Aphrodisias said that "God is matter or does not exist
outside it and everything essentially is God, and Forms are imaginary acci
dents without true being, and thus they say that all things are the same. "24
More recently, David of Dinant, whose works were condemned in the Paris
decrees of 121 o and 12 15 which forbade the teaching of Aristotle, identi
fied God with the underlying matter of both bodies and souls. In the Con
clusio to his Q;tatemuli, he writes that "it is therefore manifest that there is
but one substance, not only of all bodies, but also of all souls, and that
[substance] is nothing other than God himself. The substance from which
all bodies are made is called 'Matter' (Hyle); the substance from which all
22. "Matter, insofar as it is presupposed by form, and is the subject of generation, is not
entirely nothing, since othetwise generation would arise from nothing; it is therefore a created
entity, and so an actual and existent entity, because creation does not terminate except with
actual being [entitatem] and existence" (Suarez Disp. 14§4'1)3, opera 25:413). Creation has a
natural terminus ad quem, which cannot be mere possibility or being in potentia. It must there
fore terminate in the actual existence of the thing created. The Coimbrans adduce similar
arguments to show that matter can subsist without form, all the while insisting that "although
matter has a proper existence, nevertheless it is pure potentia" (Coimbra In Phys. 1cgq6a2,
P17o).
23. SuarezDisp. 13§51 11, opera 25:417. A "formal" actusiswhatl have been calling an actus
informans. To include something "formally" is to have it as part of one's form; to include it
"eminently" is, in this context, to have it within one's power to produce it. "What eminently
contains another is commonly said to require two things: first, that the containing thing be of a
superior nature to the thing contained; second, that what is inferior be found in the superior"
in one of four ways, of which one is "as in its producing cause" (Chauvin Lexicon s.v.
'eminenter').
24. Albert In phys. 1tr3e13; opera (lnst.) 4pt1:64.

Authenticated
Matter, Quantity, and Figure [gsJ
souls are made is called reason or mind (ratio sive mens). There it is manifest
that God is the reason of all souls and the matter of all bodies" ( Quaternuli
p71; Kurdzialek 1966:410). The writings of David ofDinant were burned;
only recently have fragments of the Quaternuli been discovered and pub
lished (Kurdzialek 196$ cf. Birkenmajer 1933). But his views are thought
to have influenced Catharism in the thirteenth century and John Wyclif in
the latter part of the fourteenth. One can infer also from Bayle's article on
Spinoza that David ofDinant's views were known in the seventeenth century
and were likened to Spinoza's (Dictionnaire, s.v. 'Spinoza', Remark A, note
5 ).25
What, then, is distinctive about material form, and why should there be
such a thing as matter? It has been asserted that material form is in
complete. Just as matter is instituted to receive it, so too material form is
instituted so as to be the actus of matter. It cannot naturally be in actu except
as the actus of matter. But how does 'being the actus of matter' differ from
'being actual', or 'existing'?
Suarez's answer is that "nothing is more repugnant to [the nature of]
God than the task [munus] for which prime matter is posited, which is to
receive [form], and to be a passive potentia, to be actualized and perfected"
(Suarez Disp. 13§14,15, opera 25:413). Matter, unlike God, does not in
clude existence in its essence. But that fails to take the heretic entirely
seriously. The heretic may as well deny also that God is pure actus, so that the
development of the world is the unfolding of God's power. Or deny that
matter is in the relevant sense actualized or perfected, as an atomist might.
God could still be pure actus, as the theologians have it, and yet one with the
world. If, like Spinoza, the heretic is willing to assert that whatever is, is
necessary, not even the contingency of the world can be argued against him.
The Aristotelians, if they foresaw such a defense, would have found it in
credible no doubt that anyone should accept the consequences of identify
ing God and matter. But if Suarez hoped to show that the very idea of matter
precludes it, his argument underestimates the cogency of the heretic's view.
There are in Aristotelianism two paths toward answering the heretic. One
can show that the first subject of natural change must have properties
inconsistent with those of the first cause or ens peifectissimum. Chief among
those properties is quantity, or the divisibility characteristic of quantified
matter. Or one can show that material forms as a matter offact have proper
ties that entail that they are incomplete and that what completes them
cannot be God himself. Here the chief property of interest is corruptibility.
Experience shows that the forms of sensible things come and go: what was
2 5· A brief treatment of David ofDinant can be found in Steenberghen 1991 :83-89; fuller
treatments include Thery 1925, Grabmann 1941, Kurdzialek 1966; relations with Catharism,
Wyclif, and Spinozism are examined in Kalivoda 1966.

Authenticated
[g6] Vicaria Dei
wood is now fire, what was a person is now a cadaver. If there are, on the
other hand, forms not subject to corruption, like God and the rational soul,
there must be a reason for that difference.
The distinction between created incorruptible forms and corruptible
forms rests on a distinction between two modes of coming to be and ceasing
to be: creation and annihilation on one side, generation and corruption on
the other. We have seen that one argument for prime matter was precisely
that it would be absurd to suppose that in natural change one thing is
destroyed and another numerically distinct thing created. That argument
does not rest on the principle that nothing comes from nothing (though
the principle is often invoked in arguments on behalf of prime matter). It
rests on the observation that the generation of material forms is preceded by
changes that prepare the way for it, as heating readies wood to become fire.
If generation were creation, those changes would be gratuitous, and any
thing could come from anything (a point that did not escape the opponents
of occasionalism). It is experience that shows that generation is not cre
ation, or corruption annihilation, and experience too that concludes in
favor of material forms.
The objections against the definition of matter as pure potentia are not
allayed by the dissenting opinions of Suarez and the rest. What we have is
largely negative. Matter is indeed neither a complete substance, nor quality,
nor quantity, nor in any other category.2 6 But until we escape the via nega
tiva, there would seem to be no response to the heretic. Even aside from
that, experience shows that the substrate of natural change cannot be en
tirely indifferent to all forms. Matter is so partial to quantity that it cannot
naturally subsist without it. Celestial and terrestrial matter cannot receive
the same substantial forms. Terrestrial matter imposes an order on the
reception of forms, so that the human form cannot be received unless those
of flesh and blood are already present.2 7 It is to such features that we must
look in order to explain what is proper to material substance, and thereby
answer the heretic.
Descartes, we will see, answered in his usual drastic manner (see §g.2).
Supposing that material substance is coextensive with sensible substance,
and that we can be certain of sensible substance that it is extended, then
matter cannot be pure potentia. It is a complete substance. There is no
26. "I say matter is what 'of itself' (that is, considered according to its essence) in no way is
'what' (i.e., substance), 'nor quality nor of the other categories by which being is divided or
determined'" (Thomas In Meta. 7lect2, Cathala 11286, paraphrasing Meta. 102ga2o).
27. "Although prime matter is in potentia to all forms, still it takes them on in a certain order.
It is first in potentia to elemental forms, and by their mediation according to various propor
tions of mixture is in potentia to various [other] forms, and so notjust anything can be made out
of just anything, unless perhaps through resolution into prime matter" (Thomas In Meta.
12lect2, Cathala 'l12438).

Authenticated
paradox in asserting its existence. If by 'form', moreover, you mean substan

tial form, then to suppose that God fills the office of matter is inconsistent
with the idea of a perfect being. Since God is not a deceiver, extension-the
unique corporeal form-must underlie sensible accidents. If by 'form' you
mean the properties by which natural kinds are differentiated, then matter
supplies to such forms the complete substance on whose existence they
entirely depend. In Cartesian physics, form is figure, and figure cannot
subsist without quantity even with respect to the absolute power of God (see
§4.3). Matter supplies, in short, everything to form except particularity.
God's conserving power cannot take its place. The heretic is answered, but
for one small detail. Descartes made extension an attribute ofsubstance and
not itself substance. So one may still ask: of what is it an attribute? With that
question the problem of the substrate is revived. Spinoza, for one, did not
hesitate to revive it.
4.2. Quantity and Prime Matter

Experience creates a presumption in favor of a special relation between
matter and quantity. It reveals to us that quantity, like matter, is found in all
sensible substances. Only the eye offaith sees that in the Eucharist the body
and blood of Christ are present without their quantity. In natural change,
accidental or substantial, quantity seems never to be destroyed or created
but only augmented or diminished. Quantity seems also to be more funda
mental than other accidents. The whiteness and coldness of a snowball have
the quantity, extension, and figure of the matter they inhere in.28 All these
phenomena could be accounted for if quantity were essential to matter.
We have seen, moreover, that if matter is regarded as pure potentia, it
becomes difficult to understand either why it should be posited or why
certain forms should require it and not others. If, on the other hand, matter
were essentially endowed with quantity, it would bring to form more than
bare existence; and one could at least argue from experience that some
forms do seem naturally to exist only in extension, while others do not. A
final point in favor of a special relation between matter and quantity was the
sheer difficulty in conceiving of material substance without it. The
Coimbrans, who defend Thomas's view that matter and quantity are dis
tinct, note that "if form could be united immediately and per se with matter,
28. "Omnia [accidentia] insunt substanti::e, media quantitate, ut omnes philosophi docent;
et videtur probari experientia, nam albedo in superficie extenditur, et calor similiter, cum
diffunditur per corpus, in quantitate eius extenditur": "All [accidents] inhere in substance by
way of quantity, as all the philosophers teaches us; and experience seems to prove this, since
white is extended over the surface [of a thing]; similarly heat, when it spreads through a body,
is extended in the quantity [of the body]" (Suarez Disp. 14§41:7, Opera 25:495).

Authenticated
[g8] Vicaria Dei
then it could be united to it, at least by divine virtue, without the mediation
of quantity, so that from a rational soul and matter there would result a
composite that was one per se, and which could be nothing other than
human. There would then exist a man without head or heart or other
integral parts, and that, it seems, ought not to be conceded" (Coimbra In
Phys. 1cgq11a1, 1:182). Nevertheless the Coimbrans agree that form can be
united immediately with matter and accept the consequence (see §5.2 be
low). Others, including Descartes, find it absurd. 2 9
Physical arguments were overshadowed, however, by a theological
doctrine that seemed to preclude any straightforward identification of mat
ter and quantity. The Catholic church held firmly that in Holy Communion
the bread and wine ofthe host, when consecrated, are by God's miraculous
act entirely transmuted into the body and blood of Christ. Immense inge
nious effort had been expended to reconcile that doctrine with Aristotelian
physics. The problem was that sensible accidents are determined by, or
"emanate" from, substantial form (§5.4). Yet in the Eucharist the accidents
of the bread remain even while the form of bread is succeeded by that of
Christ's body.3° The solution offered by Thomas and others was that the
quantity of the bread is miraculously sustained by God in the absence of the
matter and form in which it naturally inheres, and the sensible accidents of
the bread attach themselves to it. Appearances are saved even while the
underlying substance perishes.
Quantity was thus central to the account made dogma by the Council of
Trent.3I Any thesis on the relation between quantity and matter which
29. See, however, the letter to Mesland (9 Feb. 1645) in which Descartes implies that a
human soul can be joined with any matter whatsoever, in such a way that the matter can be said
to be its body.
30. On Suarez's explication of the doctrine of "real accidents," which was often singled out
for ridicule by seventeenth-century philosophers, see §5.1.
31. The relevant canon reads: "Si quis dixerit, in sacrosancto Eucharisti:oe sacramento
remanere substantiam panis et vini una cum corpore et sanguine Domini nostri Iesu Christi,
negaveritque mirabilem illam et singularem conversion em totius substanti<e panis in corpus et
totius substanti:oe vini in sanguinem, manentibus dumtaxat speciebus panis et vini, quam
quidem conversionem catholica Ecclesia aptissime transsubstantionem appellat: anathema sit
[Ifanyone says that in the holy sacrament of the Eucharist the substance of the bread and wine
remains with the one body and blood of our LordJesus Christ, and denies the miraculous and
singular conversion of the whole substance of the bread into the body and of the whole
substance of the wine into the blood, even while the species of the bread and wine remain (... )
let him be anathema)" (Cone. Tridentinum, sessio 13 (1551), Canon 2 (cf. ib. 4); Denzinger
and Schonmetzer 1976, no. 1652, 1642. The doctrine of transubstantiation was reaffirmed as
recently as 1950: cf. no. 3891).
It should be noted that the Council did not favor any particular philosophical theory,
Thomist or otherwise, that purported to show how the species-the sensible qualities-of the
host could remain after transubstantiation. Descartes was thus not compelled to accept any
existing theory; but he was compelled, if he wished to respect the doctrines of the Church, to
show that it was not impossible that the species should remain. On the debates surrounding the

Authenticated
Matter, Quantity, and Figure [gg]
threatened to undermine the possibility of transubstantiation had for that

reason to be rejected or modified. If the quantity of the host, for example,
were destroyed along with its matter, then Thomas's solution would no
longer serve. It was therefore incumbent on the Nominalists in particular to
show that their thesis did not render transubstantiation unintelligible or
inconsistent with experience. Descartes, too, had to show that his physics,
though it identified matter with extension, could coexist with dogma.
A cursory look at arguments about prime matter and quantity reveals that
their grounds were quite various. But the arguments given the most weight
cluster around two topics: the relation of quantity to extension, and the role
of accidents in generation and corruption. I will defer the second topic until
§5. In treating the first I will start with a Nominalist argument purporting to
show that matter and quantity are not really distinct, and then examine the
responses of the central texts. In them the Eucharist plays the role of an
exemplary and inevitable instance.
Quantity was divided first into discrete quantity, or multitude, and con
tinuous quantity, or magnitude. The continuous quantity peculiar to matter,
or extensive quantity, was then distinguished from the quantity pertaining to
qualities alone, or intensive quantity. 32 Extensive quantity was generally
thought to confer the following characters on material substance:33
(i) Extension: quantified substance is spatially distributed (extensum), or

at least capable of being spatially distributed ( extensibile).
(ii) Divisibility: quantified substance can be divided into really distinct
parts. A thing is divided, strictly speaking, only "when the parts which
were united are conserved separately, once their union is dissolved, "3 4
and divisible if it can be divided, naturally or by divine power. 3 5
Eucharist, especially after the Reformation, see Armogathe 1977, §1; Andresen 1982, esp.
1:548fl, 2:483; Redondi 1987, c.7.
32. Intensive quantity is the mode by which degrees of the same quality differ. Qualities thus
exhibit two kinds of motus: increase or decrease of intensive quantity, increase or decrease of
extensive quantity (diffusion). Whether increase of intensity consists in the addition of parts of
the quality or in the destruction of the old and the production of a new quality was much
disputed. See Maier 1951, Murdoch and Sylla 1978:231 1r.
33· Another character mentioned by Aristotle-being measurable-plays no role in
debates about prime matter. See Fonseca In meta. 503q 1§ 1, 2:634r; Suarez Disp. 40§3, opera
26: 53 8.
34· "Dividitur enim proprie res, quando, ablata unione, partes qme erant unit<:e, separat<:e
conservantur" (Suarez Disp. 40§ 1111, opera 26:531). That character applies to anything that
has really distinct parts, including composite substance, whose matter and form are really
distinct. Substance is excluded by another part of the full definition, which entails that quantity
is an accident, not a substance (cf. 'l[2, 19; Aristotle Meta. 5c13, 102oa8).
35· Celestial matter, "although it cannot be divided by any natural agent, can be divided by
divine power. The same is to be said of natural minima." In any portion of matter, celestial or
not, one can at least by "mental designation" pick out disjoint parts (Suarez Disp. 40§1'l[12,
opera 26:532).

Authenticated
[ 1 oo] Vicaria Dei
(iii) Impenetrability: if two substances have quantity, it is contradictory to

suppose that they occupy the same place. By contrast, spiritual sub
stances, which cannot have quantity, may do so.
The position of the Nominalists36 was that extensive quantity, or "bulk

quantity" (quantitas molis), was "not a thing distinct from substance and
material qualities. Rather the being [entitas] of each of them by itself has
that bulk and extension of parts, which is in bodies; that being is called
'matter' insofar as it is a substantial subject, and quantity insofar as it has
extension and distinction of parts" (Suarez Disp. 40§2!2, opera 26:53$ cf.
Coimbra In Phys. 1 c2q 2, 1 :g4). Quantity is nothing other than matter itself,
designated with respect to the distinction among its parts and their conse
quent extension in space.37
The argument is that quantity has no effects beyond those that already
arise from matter itself.38 If there were such effects, then among them there
would surely be a "real distinction or situation of the parts of substance,
since by the fact alone that a thing is understood to have one part outside
another [ unam partem extra aliam], both in being and in place, quantity too is
understood" (Suarez ib. '13). It remains to be shown that matter is indeed
understood to have one part outside another. But clearly its parts are dis
tinct of themselves. They require no further cause to be outside one another
in being. Moreover, whatever entities are distinguished in being "can also be
constituted in diverse places." They require no further cause to be outside
one another in place. But those are the effects quantity would have if it had
any proper effects. So any supposed distinction between matter and quan
tity is idle. 39
Though Descartes treats extension as if it were entirely unproblematic,
the Aristotelians regarded it as comprising two features. A thing is extended
only if it has really distinct parts. Some Aristotelians held that distinctness in
being sufficed for quantity, 40 but the more common view was that distinct
36. The authoriti~s listed by Fonseca are Petrus Aureolus, Ockham, Gabriel Biel, and Adam
of Wodeham; to them Suarez adds John Major and Albert of Saxony.
37. Ockham himself affirms that "quantity is nothing but a thing's having part outside part
and one part distant in place from another" (De sacramento altaris; cf. Maier 1955:183).
38. See, for example, Ockham Summula 1c13, Opera philos. 6:192-193.
39· Chauvin, who takes the Nominalists' side, writes that matter is per se extended, because
extension is "nothing other than the several parts of matter set outside one another [nihil aliud
est quam plures materite partes extra se invicem positm]" (Chauvin Lexicon s.v. 'quantitas').
40. Suarez cites Capreolus (In Sent. 2d3a1,$ 2d 18q 1) and Soncinas ( Q. meta. 5q 19, P7 5b) 0
The argument was that quantity is by definition divisibility (an opinion disputed by both Suarez
and Fonseca, although it finds textual support in Aristotle: cf. Meta. 5c13, 1o2oa8), and that
distinctness of parts suffices for divisibility ("divisibilitatem [...] solum esse quod una pars non
sit alia"). Since matter cannot but have parts, it follows that matter cannot be conseiVed
without quantity even by God's absolute power (Soncinas Q. meta. 5q22, p8ob), a view Suarez
calls "very common among Thomists" (Disp. 40§4'l15, Opera 26:544), and which he denies.

Authenticated
Matter, Quantity, and Figure [101]
ness in place, and thus quantity, is not entailed by distinctness in being. 41

The parts must not only be "distinct in being"-which is to say, separable at
least by the absolute power of God-but also "distinct in place." The notion
of place, as Descartes well knew, was itself not simple. 42 For the moment it
will suffice to think of place as what was usually called "extrinsic place." The
extrinsic place of a body is, roughly, the surface defined by the bodies
circumscribing it, or what one might call its (outer) bounding surface. 43
Hence to be distinct in place is to have a different bounding surface; to be
extended is to have distinct parts that in turn have different bounding
surfaces.
If the Nominalist argument works, there must be some special reason
why distinctness of being among the parts of matter should entail distinct
ness of place. Ockham's reason is that "the parts of matter can never be in
the same place" (Summula 103, opera philos. 6:191). That comes close to
begging the question, at least from the point of view ofthose who hold that
neither matter nor quantity per se is impenetrable. It also runs counter to
the standard account of the Eucharist, according to which the matter of
Christ's body is present in the host, while the quantity of the bread or wine is
in the place of the host.
But perhaps there is another reason. Take a globe of water. It is homoge
neous, yet it has distinct parts. How are those parts different? The only
criterion that suggests itself is spatial separation: this part and that part are
two if the place occupied by one does not coincide with the place occupied
by the other. Here the very idea of part seems to presuppose distance in
place. Prime matter is, of course, as homogeneous as can be. Its parts cannot
be distinguished by substantial form. Nor can they be distinguished by
quality. Only place seems to distinguish them. Matter's distinct parts are
distanced parts, and matter is extended per se.
I know of no Aristotelian who mentions such an argument. But some
thing like it may be behind the unease exhibited at the thought that sub
stance deprived of quantity would collapse into a point. How would one
identify its parts? In any case, I think the argument can be answered. We
have seen that the criterion of real distinction is that two things are really
distinct just in case God could, according to his absolute power, preseiVe
one while annihilating the other. Since God's power was limited only by
noncontradiction, two things are really distinct just in case it is not con
tradictory to suppose that the first exists and the second does not, and
41. In addition to Suarez, see Fonseca In meta. 503q2§3, 2:65oE. Suarez and Fonseca
credit Paul of Venice (In meta. 5c 12) with the distinction.
42. Descartes embarks upon what for him are unusually lengthy and careful discussions of
place and space in the Rules (AT 10:426, 443ff) and in the Principles (2§10-15, AT 8 1:45-49).
See §g.2 below.
43· See Suarez Disp. 51§213. opera 25:g8o.

Authenticated
[102] Vicaria Dei
conversely. If Socrates' body were deprived of its quantity, his head and his
feet would not occupy distinct places (though they could still be present in
different places, as Suarez argues). But they would remain distinct. The
presence of a real distinction does not depend on our being able to recog
nize it. Nor does it require that we can represent clearly the two things as
distinct. Distinction, in short, is not subordinate to the discernment of the
intellect.
What quantity brings to matter, then, is distinction in place. Earlier I
characterized "extrinsic" place as the outer bounding surface of a body.
What is sometimes called "intrinsic" place, or Ubi ('Where'), is quite
different (SuarezDisp. 51§214. Opera 26:980). The intrinsic place ofa thing
is "a certain real mode intrinsic to a thing which is said to be somewhere,
and from which such a thing has [the property of] being here or there"
(§1i13, 975), or "locally present" (114, 976). Although it cannot be ex
plained except through relations of distance like nearness and farness or
above and below, it is not itself a relation. Nor is it identical to or caused by
extrinsic place (i 18-20, 977). Nor, most important, is it a region in space,
where 'space' is understood as an empty receptacle waiting to be filled with
bodies (1:9-12, 974-975; cf. Toletus In Phys. 4c5q3, Opera 4:115ff).
There is, accordingly, a difference between having intrinsic place, or
being present at a place, and having extrinsic place, or occupying a place.
Having extrinsic place implies having quantity. Spiritual things cannot have
extrinsic place. But they can have intrinsic place. God, for example, "is
present to the corporeal universe, not only by presence [per prcesentiam],
that is, by cognition, and by potentia or action, but also by essence or by his
substance" (§3'![8, 984). So too for angels, which, because their powers are
finite, are present only at finite regions in space (§3'![ 10, 985). It is not
necessary here to understand what being present by potentia or by substance
might consist in. All that matters is that a distinction between being present
at a place and being in or occupying a place be provisionally granted.
Suarez uses the distinction to dissolve the Nominalist argument. He con
cedes that the parts of matter are distinct in being and that they can be
separated in place, even when matter is bereft of quantity, just as two incor
poreal beings can be present at different places. 44 Nevertheless he denies
that matter must be quantified of itself merely because its parts are sepa
rated. Remember that Ockham's argument was that quantity has no effects
not already given in matter itself. Suarez's reply is that
44· "If, therefore, quantity is taken away, while the substance is conserved, and no local
motion occurs in the substance, the substance will remain with the same substantial presence,
and with the same distance or nearness to the center and the poles of the world; and so the
whole will remain present in the same space, and its parts in the same parts of space" (Disp.
40§4125, opera 26:s5o).

Authenticated
(i) it is characteristic of quantified matter that its parts naturally exist at

diverse places;
(ii) it is not sufficient, if one is to attribute that characteristic to matter
alone, to point out that the parts of matter are, or can be, present at
different places, since incorporeal things too can be, but need not
be, present at different places;
(iii) moreover, matter itself if deprived of quantity need not be present at
different places;
(iv) so the characteristic in (i) does not arise from matter alone, but from
quantity. 45
Occupying several places, in other words, is not proper to matter, but only to
quantified matter. Quantity does have an effect not already given in matter
itself.
It remains to be shown that quantity is really distinct from matter. Here
the doctrine of the Eucharist comes into play. First a few fundamentals.
When the host is consecrated, its entire substance, both form and matter, is
destroyed, and the entire substance of Christ's body replaces it. Neverthe
less the accidents of the original substance remain, as our senses show. The
host retains its color, smell, and, more significantly, its figure and quantity.
In the standard account, the quantity of the host, after consecration, serves
as a quasi matter for its other accidents in the absence of the original matter.
The quantity of Christ's body, on the other hand, does not accompany its
substance: it is present at, but does not occupy, the place of the host.
From that doctrine one may infer first of all that quantity can subsist
without matter or form: "in the holy Eucharist the substance of the bread
does not remain, yet its quantity does, sustaining and uniting qualities that
are really distinct, like brightness, taste, and so on. Quantity therefore is
really distinct not only from the substance of the bread, but also from its
qualities" (Fonseca In meta. 5c13q2§2, 2:648A). If quantity can persist with
out matter, it is at least modally distinct from matter. But since no one thinks
that matter is a mode of quantity, it must be really distinct (Disp. 40§2'18,
opera 26:535). 46
45· "Aliud est enim esse posse in diversis spatiis, quod duabus rebus etiam incorporeis
convenit; aliud vero est naturaliter esse non posse nisi in diversis spatiis, quod duobus Angelis
non inest; igitur illud prius non requirit quantitatem, et ideo convenire posset partibus mate
ria:, etiamsi quantitate privarentur; hoc vero posterius omnino requirit quantitatem. Unde si
partes materia: sine quantitate essent, indifferenter esse possent, vel in eodem Ubi, vel in
diversis. Quod ergo sint ita disposit<e, ut necessaria requirant ex natura rei situs diversos, id
provenit ex quantitate" (Suarez Disp. 40§2"12o, opera 26:538). Suarez adds that the reasons
why quantity must be really distinct from matter are those he has adduced earlier in discussing
the Eucharist.
46. From the definitions of modal and real distinctness, it follows that if X can exist (at least
by God's absolute power) without Y, then Yis either a mode of X or really distinct from X,
depending on whether Y cannot or can exist apart from X.

Authenticated
Vicaria Dei
Quantity, moreover, is not actual extension in place, but only potential

extension. The body of Christ, like all bodies, has quantity. But when pre
sent at the place of the host it cannot actually be extended, since that would
exclude the quantity of the now departed substance of the host. Still "it is
actually so extended and ordered within itself that if it were not super
naturally impeded, it would also have actual extension at that place" (§41:8,
547). Hence quantity is "a form giving things corporeal bulk [molem], or
extension," not actually but potentially, at least with respect to the absolute
power of God. It is even possible, contrary to what the Nominalists argued,
for a body to be deprived of quantity without any local motion of its parts. It
will then still be present at every place it was present at before, but neither it
nor any of its parts will occupy any place.
It is time to summarize the results so far. Remembering that God's ab.so
lute power is limited only by what is logically contradictory, we have the
following:
(i) It is not contradictory to suppose that matter can exist without quan
tity, or quantity without matter. But matter does not naturally exist
without quantity, nor quantity without matter.
(ii) It is not contradictory to suppose that matter having quantity is not
actually extended, or present at a place without occupying it. But
matter does naturally occupy every place at which it is present.
(iii) Quantity is a potentia of matter, whose actus can be impeded super
naturally but not naturally, and whose primary effect is to prevent
other quantified things from occupying or being present at the same
place. 47
Matter is naturally extended in place, spatially divisible, and impenetrable,

just as experience tells us. But with respect to the absolute power of God, it
can be deprived of all of those features.
It must be admitted that unextended, penetrable matter is hard to imag
ine. There are two difficulties. One is to conceive of matter apart from
impenetrability; the other is to conceive of matter apart from extension
apart from being present at all the points of a certain region in space. The
matter of Socrates is naturally present at all the points of a region that is
larger than the region occupied by the matter of his head or finger. How,
then, might it be shrunk to a single point, without depriving its parts of their
separate identities within the whole?
47· The essential definition [ratio] of quantity is to be "a form giving to things corporeal
bulk [molem], or extension [...]What 'having corporeal bulk' is, we cannot state, except in
relation to [in ordinem ad] that effect, which is to expel similar bulk from the same space, not
indeed actually, since the effect can be impeded by God's absolute power, while its formal
effect is conserved, but in aptitude [ aptitudine]" (Suarez Disp. 40§{116, opera 26:54 7).

Authenticated
Matter in itself is not impenetrable. It is, as we have seen, the substrate of

all forms, the "first subject." That definition does not preclude the presence
of distinct matters at one place. Quantity, on the other hand, does confer
impenetrability. Matter without quantity would continue to be a subject
informed by substantial form. Socrates sans quantity remains a rational
animal; but his body no longer has the power to prevent others from oc
cupying any region in space. Matter with quantity can, as we have seen, be
prevented by God from exercising that power; but power frustrated remains
power: it would if it could. Matter without quantity no longer has the power:
it would require a miracle to bring it about that other bodies should not
occupy the place at which it is present. 48 In both instances the result is the
same: we have a bit of matter that no longer hinders other bits from occupy
ing any region of space. Whether one would say, in a given instance, that
one or the other was true depends on whether there is reason otherwise to
believe that it still has quantity. In the theology of the Eucharist, there was
general agreement that Christ's body in heaven has all the properties of
matter, including quantity. That it does not occupy space in the host, then,
must be because God has prevented it from doing so. 4 9
Suppose, anachronistically, that every point of material substance were
endowed with a repulsive force. So long as a material point has that force it
will prevent other material points from being where it is. There seems to be
no contradiction in supposing that Socrates' body continues to serve as the
substrate of his soul even while the repulsive force is removed from the
matter of his body. Like a ghost, Socrates' body could then penetrate or be
penetrated by other bodies. Yet it would not have become a spiritual sub
stance: unlike any such substance, it would, if God ceased to act on it
miraculously, immediately occupy space.
Ghostly matter is difficult enough to conceive. To conceive of unex
tended matter was, for many philosophers, not merely difficult but impossi
ble. Imagine Socrates-or for that matter, the world-compressed into a
single point. Would not its parts become confused "and all immediately
joined together in substantial union," as some Thomists thought?50 One
might suppose that the parts at least of elemental substances, which are
entirely homogeneous, are distinguished only by their relative positions in
48. "Substantia autem quantitate carens nee posset impellere aliud corpus, nee ab eo
impelli, nee etiam posset a suo loco excludi ab alio corpore, quia non esset in loco, quantitativa
modo, nee esset impenetrabilis loco cum quolibet corpore" (Suarez Disp. 40§ 1'l[ 10, opera
26:534).
49· The other alternative is seen in the doctrine rather obscurely alluded to by Fonseca: the
bodies of saints can "have several parts in the same place, for instance, two hands or two feet,
and indeed however many [parts] as they wish, which can be at once in the same place as other
bodies" (Fonseca In meta. 5Cl3q1, 2:643D).
so. Suarez Disp. 40§4'l[28, opera 26:sso; cf. 119-20.

Authenticated
[106] Vicaria Dei
space. If such a substance were deprived of quantity, and therefore of exten

sion, those parts would coalesce into one.
For more complex substances, at least, the outline of an answer can be
found. In a passage quoted earlier, Suarez holds that a material substance is
"actually so extended and ordered within itself' that if not hindered by
divine intervention it will be present at, and occupy, a more or less extensive
region of space. It has, in other words, an intrinsic order prior to the order
ing imposed on it by its relations to other material substances. The nature of
that order is suggested by some remarks in Fonseca. He distinguishes the
"order of parts in the whole" from the "order of parts in place." Socrates'
head and feet, for example, are extremities of his body, while his chest is in
the middle. That holds wherever Socrates may be, and no matter what
relative position in space the parts of his body may occupy (Fonseca In meta.
5q1§4; 2:643). It is prior to, and determines, the ways his body may actually
be extended in space.
Imagine a crude picture of Socrates' body, in which his hands, feet, head,
and so forth, are designated by dots, and the relations of contiguity among
those parts by lines connecting the dots. Such an object would now be called
a graph. Certain properties of graphs hold independent of any particular
embedding in space. A graph is connected, for example, if any two points or
nodes in the graph can be connected by a path along the edges of the graph.
A connected graph is a tree if there is exactly one such path between any two
points. The graph of Socrates is a tree. There is, for example, just one path
from the head to the left foot, and no matter how the graph is drawn, that
path must encounter the torso and the left leg. An extremity of the graph is
a node that meets exactly one edge. Nodes are extremities or not indepen
dent of any particular embedding. With due allowance for anachronism,
Fonseca's remark can be taken to express that independence. Socrates'
head and feet are extremities, and the torso lies between them, no matter
what region of space he happens to occupy, and no matter how he is
oriented.
One cannot credit the Aristotelians with twentieth-century mathematical
ideas. But Aristotelianism encourages a view of substances and their parts
which is, in its way, less concrete than that of many seventeenth-century
philosophers. Medieval diagrams of the eye, for example, unlike those
found in Descartes's Dioptrics, make little attempt to convey the precise
spatial dimensions of its parts. The diagrams depict what one might call the
structural characters of the parts of the eye, defined, as we will see, by their
operations and their "topological" relations. 5 1 To argue, as Suarez does, that
51. See the examples in Lindberg 1970; Reisch Marg. phiL 10tr2cg, p421. The drawing in
Reisch (1517), unlike the medieval drawings reproduced by Lindberg, does not represent the
various "tunics" merely by concentric and intersecting circles. But it comes hardly any closer to

Authenticated
a material substance that is not actually extended still has a potentia to

occupy space, both as a whole and part by part, is, I take it, to recognize that
there are determinate relations among the parts independent of its actual
existence in space. Such relations cannot be predicated of spiritual sub
stances. There is no graph of the powers of the rational soul, even if these
are regarded as parts.
The predominant doctrine concerning condensation and rarefaction
also militated against identifying parts through actual spatial relations, or
quantity with actual extension. The same "quantity of mass," it was thought,
could occupy volumes of quite different size. A passage from the Sentences
commentary of Franciscus de Marchia expresses the common view: "Quan
tity of mass in one thing, and the variable extension of that quantity, even
while it itself does not vary, is another, as is apparent in the motus of conden
sation and rarefaction. A rarefied body, though no part of quantity is added
to it by rarefaction [...] occupies a greater place than before, which is not
on account of its having a greater quantity now than before, but only on
account of extension. "5 2 Since a vacuum was thought to be impossible,
rarefaction could not consist in the dispersion of particles into a larger,
otherwise empty space, or in the enlargement of vacuous pores in a
spongelike connected mass. It consisted instead in what one might call a
"thinning out" of the same continuous quantity of matter. 5 3 In watercolor
painting, a brushful of paint can be dabbed densely onto a small spot or
spread over half a sheet. The bulk of paint is the same; but its place varies in
size, and thus its extension in the sense that more places are occupied by
that it, and other bodies excluded. Where we would see, in the expansion of
gases when heated, a fixed number of particles, each itself unchanged,
dispersed into ever larger spaces, the Aristotelian sees a fluid being
smoothly and continuously redistributed. The parts of an expanding body
depicting actual spatial relations (cf. the modern drawing in Barlow and Mollon 1982:36 [fig.
2.1]). .
52. In sent. 4d13q1, quoted in Maier 1955:142, 200. Cf. Toletus In Phys. 4c9qu, opera
4:132rb, which resembles closely the definition of iEgidius (see n.56). Rarefaction and con
densation are distinct from augmentation and diminution. In augmentation and diminution,
quantity is added to or taken from a substance. Animals and plants do not grow by rarefying,
but by converting nutrients into their own matter. Air that expands upon being heated, on the
other hand, does not increase in quantity; experience shows that it expands even when heated
in a closed jar (Toletus In Phys. 4c9q11, opera 4:132va).
53· Needless to say, the interpretation of condensation and rarefaction was controversial.
The view I summarize, which seems to me to make the most sense, follows that of iEgidius
Romanus (see his In phys. 4lect1 7, p96va; quoted in Maier 1949:29-30). iEgidius argued that
there were "two kinds of quantity: one by which matter is so and so much, and which is great or
little, and one by which matter occupies so-and-so-much space as [being] bigger or smaller."
Call the first kind of quantity 'bulk', the second 'volume'. A substance is rare if its bulk occu
pies a large volume, dense if its bulk occupies a small volume. A given bulk of water, for
example, will occupy ten times the volume if transmuted into air. Only its volume changes,
however; its bulk remains the same throughout.

Authenticated
[108] Vicaria Dei
Fig. 3· Condensation
will of course expand with it; their shape need not be preserved, but their
graph-theoretic and topological relations will be.
Ockham's picture seems to have been that condensation and rarefaction
were. what would now be called 'dilations' of quantified matter (see Figure
3). Letybe a constant greater than o. Take a fixed point Oin space, and letP
be an arbitrarypointdistinctfrom 0. Let lbe the line from Othrough P. On l
lay off a segment OP* whose length is equal to y times the length of the
segment OPfrom 0 toP. The point P* is the dilation of Pwith respect to 0
and x. If the same transformation is applied to all points of a region R the
result will be a region R* whose points are in one-to-one correspondence
with those of R. Ockham writes that "from the very fact that the parts of a
substance or quality are more distant from each other and from the center
of the whole, without any other thing arriving [i.e. without the reception of
any new form], a body is rarefied; and if [the parts] were less distant before
it was then dense" (Ockham Q in phys. 97, opera philos. 6:658). So every part
into which the solid matter could be divided must be more distant from its
neighbors. The transformation of the region must be continuous and one
to-one. If it is, there will be no interpenetration or destruction of parts;
shape will even be preserved if the dilation is uniform (i.e., ify is constant).
Ockham concludes that quantity remains numerically identical through
condensation or rarefaction, and that there is no need to suppose, as the
atomists did (and- as Descartes later does), that condensation and rarefac
tion are caused by the contraction or expansion of invisible pores. Toletus
agrees with the second but not the first: "I judge [...] that there is no
change according to rarity or density unless something of quantity is lost or
newly acquired" (In Phys. 4cgq 11, Dpera4: 132vb). Toletus treats rarefaction
and condensation on a par with other alterations like heating and cooling.
Alteration in general consists in the addition or removal of degrees of
quality; the same holds for rarefaction and condensation. Quantity no more
remains identical in such changes than do qualities in alteration.
Whichever view one considers, the identity of parts within a body cannot
be tied too closely to their actual spatial relations. Neither view precludes
the possibility that a body could be shrunk to a single point. In that case
there would be no spatial relations by which to distinguish parts (which is
why philosophers like Capreolus believed that there would be no parts). But

Authenticated
Matter, Quantity, and Figure [lOg]
there can still be relations of order-connectedness, extremity, and so

forth. Such relations, I take it, may be what Suarez and Fonseca have in
mind when they insist that even under such conditions a material substance
may remain what it was. In that sense body can be conceived apart from
quantity, and without being confused with spiritual substances. As Suarez
notes, even when deprived of quantity altogether, a material substance,
unlike any spiritual substance, remains the sort of thing that could occupy
space, and indeed would if its natural state were restored to it.
4·3· Figure and Other Qualities

Many interpreters of Descartes have noted that the definition of matter as
extended substance and the restriction of its forms to the modes of exten
sion amount to a rejection of active powers in nature. But few to my knowl
edge have looked at the Aristotelian treatment of figure. Figure is indeed
not central to Aristotelian physics. Its role in physical explanation is, by
comparison with that of quantity or motion, more restricted and less con
troversial. The central texts agree that it is entirely passive and never an
essential property of any natural thing. Only in artifacts does it come into its
own; but artifacts have no natures, no internal principles of change, and are
therefore not among the objects of physics. Nevertheless figure recurs in
physical arguments as a kind of stand-in or likeness of form, especially
substantial form. We find Toletus writing that "the figures of things are
strongly analogous [ magnam habent proportionem] to substantial forms. For as
proper substantial forms are consequent upon singular things and species,
so too are their exterior figures; so that indeed many have mistakenly
judged that substantial form is nothing other than this exterior form"
(Toletus In Phys. 7c3q3, opera 4: 1g8va). Toletus adds thatjust as matter and
substantial form are composed into a thing that is unum per se, so too figure
and a "subject proportioned" to it are composed into something that is
"somehow" unum per se also. One indication of that unity is the use of the
same word to denote both people and images of people, as if the form of the
image conferred determinate existence on its "matter" as the human form
confers existence on prime matter. Figure, moreover, is not alterable. Un
like other qualities, it is not altered per se, but produced anew when some
other quality changes. 54
Figure thus enjoys a kind of double life within Aristotelian physics. It is,
on the one hand, a relatively unimportant part of the furniture of the world,
never initiating change, or even being changed independently of other
more fundamental properties. It is, on the other hand, the character by
which artifacts are defined, and through the recurrent analogy of artifacts
54· Aristotle Phys. 7c3, 245hgrr.

Authenticated
[110] Vicaria Dei
with natural objects it becomes a simulacrum, an "index," a representer of

substantial forms which, because their activity is always mediated by acciden
tal forms, cannot be immediately known to us. In this section I consider the
definition, mode of existence, and physical properties of form, deferring
the analogy of art and nature to §7.1.
The Categories, as I have mentioned, put figure among the species of
quality. It is, in the words of the Coimbrans, a "quality resulting from the
termination of visible quantity in a natural thing, like the external aspect of
man orlion" (Coimbra In Log. 1:489; Smirez Disp. 42§41)5, opera 26:615).
It is the external termination of quantity, "a certain agreement and harmony
of the parts of quantified substance [ quantz1," the internal termination
being what we might call the amount of a finite quantity-two feet, say.ss
With figure is included, significantly, form, not the form which is a principle
of change, but what was variously interpreted as a special case of figure
whether that of animate as opposed to inanimate things, or of beautiful
things, or of corporeal as opposed to mathematical things. 56
It may be surprising to find figure regarded as a quality. Certainly
Descartes, when he inveighs against qualities, did not intend to include
figure. Indeed the genus of quality seems a mixed bag, and it is not evident
at first sight what holds it together. The four species defined in the Categories
are:
(i) habitus or dispositio, a property ''which is per se and first of all ordered
to an operation" (unlike substantial form), not however as a primary
power of acting (unlike potentia), but as "assisting and facilitating"
the actualization of such a power (Suarez Disp. 44§116; on disposi
tions, see §5.3 below);
(ii) potentia (§2.3) and impotentia (the privation of a potentia, like blind
ness or lameness);
(iii) "passive" qualities, like heat and cold, rarity and density, colors, and
so forth;
(iv) figure and form.57
There were numerous attempts to explain why the four species were all
called 'qualities'. Aristotle has it that 'quality' is "that by which a thing is said
55· "Duplices autem terminos forma pr~stat, extemos, & intemos; externi sunt illi, qui non
sunt de genere quantitatis, ut figura, qu~ est in quarta specie qualitatis, & nil aliud est, nisi
quidam partium quanti inter se concentus, & harmonia; intemi vero illi sunt, qui sunt de
genere quanti, & certam quanti speciem constituunt, ut bicubitum, tricubitum, & huiusmodi"
(Zabarella De prima rerum materia 2cg, De rebus nat. 1g6C; cited in Goclenius Lexicon s.v.
'figura').
56. The first is suggested by Zabarella (opera logica, Tabulfl! 124); the second is attributed to
Simplicius and Boethius, and the third, which Toletus favors, to Iamblichus (Toletus In Log.,
Prfl!dicarrumta 8, opera 2:158).
57· On the classification of qualities, see Fonseca In meta. 5C14q1 and 2; Suarez Disp. 42§4;
Toletus In Log., Prfl!dicarrumta 8, opera 2:153ff; and the useful 1lzbulfl! of Zabarella, included in
his opera logica.

Authenticated
Matter, Quantity, and Figure [ lll]
to be quale" (8b25). That is uninformative, to say the least. Later he notes

that qualities have contraries, are capable of degrees and provide the
ground of similarity and dissimilarity. But the first two do not apply to all
four species, and in particular not to figure. The last, though shared by all
four species, is not shared by them alone. Aristotle's text leaves the problem
unsolved.
The most common answer was that a quality is an accident by which a
substance is perfected in being or in acting. Unlike substantial form, quality
presupposes a substance and cannot itself constitute a substance from
prime matter. It "perfects" substance only in the sense that any actual exist
ing thing must be fully specified in every way that things of its kind can be. In
Suarez's careful formulation: "Quality, therefore, seems to be a certain abso
lute accident, adjoined to created substance in order to complement its
perfection, both in existing and in acting" (Suarez Disp. 42§ 115, opera
26:607). How, then, is figure a quality? Every material substance naturally
has quantity. Its quantity must be of some definite amount, and in that sense
specified or "terminated." The same quantity may be terminated in many
different ways, may have many different shapes or figures. Any actual sub
stance must at each moment have one particular figure. Figure completes
finite quantity, makes it fully determinate and thus actually existent. It satis
fies, therefore, the definition of 'quality'.
Among qualities, however, it has several peculiarities. Suarez admits that
figure "would seem to follow upon [ consequz] quantity" rather than form. If,
moreover, quantity exists in matter prior to its union with substantial form
(§5.3), then figure too, as a mode of quantity, might precede substantial
form rather than complete it. Nevertheless it "follows upon form, insofar as
it pertains to the finishing [ ornamentum] of substance, and can, in a certain
way, be of seiVice in action or natural change" (opera 26:607). As Aristotle
says in De anima, each thing, especially each animate thing, has a figure
proper to it (De an. 407a23ff). Matter has no power to determine such a
figure: it must therefore come from form.
Indeed, figure is interchangeable with form in certain contexts. Ockham,
who came closer than most to identifying form and figure, agrees that "in
artificial things there is neither matter nor form" when those words are
taken in the strict sense. But if by 'matter' one means merely 'that by whose
transmutation some proposition now is true that was false before', and by
'form' 'something verifiable afterwards (i.e., after the change)', then figure
can be the form of an artifact. So, for example, the figure of a bronze statue
can be called the "form" of the bronze, since the bronze was not a statue
before and now is one. 58 The example recurs, as well it might, in the
attempt of Le Bossu to align Cartesian form with Aristotelian form (Paraltele
s8. Ockham Brev. sum. 1C3, Summula 2C2; opera philos. 6:23, 216-217.

Authenticated
[ 112] Vicaria Dei
12, p 133). Le Bossu simply substitutes figure for form. For him, as for
Descartes, the figure of a statue and the form of its bronze are forms in
precisely the same sense. With that none of the central texts could have
agreed. Even Ockham denies that 'form' can be applied unequivocally to
artificial and natural things. But the possibility was there to be exploited.
The second peculiarity of figure follows from the first. Unlike other
qualities, it is peculiarly dependent, not merely on substance, but on an
other accident-quantity-for its existence. In Aristotle's list, it is last be
cause least, "both in perfection and in its mode of being, so that to some it
seems scarcely to merit the univocal designation of 'quality'" (Disp. 42§5! g,
opera 26:625). Other qualities can, at least with respect to the absolute
power of God, exist apart from quantity (7§1119, 25:257). But figure can
no more exist apart from quantity than motus apart from the mobile. Neither
has "by way of its own concept sufficient being to be conseiVed; only by way
of a certain identity with that in which they inhere [do they have enough
being to be conseiVed]" (7§21)o, 25:265). Yet figure is not identical to
quantity: quantity can exist without any one of the particular figures it is
capable of, though not perhaps without any.
The concept of figure, as we have seen, is of a certain mode by which
finite quantity is terminated. Figure is, so to speak, an accident of an acci
dent, an adverb of quantum. The quality of heat, on the other hand, is not
thus related to any other accident of the substance it inheres in. Though a
lump of coal may be hot and square, its hotness is not a way of being
quantum or of being black. What stands to heat as figure to quantity is
intensity or degree: just as a thing cannot be just quanta, but must have a
definite figure, so too a thing cannot be just hot, but must have a definite
degree of heat.
Neither figure, then, nor intensity inheres in bodies independently of
other accidents. Since, as we have seen several times, independent existence
is the ground of distinction, figure, unlike other qualities, cannot be really
distinct from the quantity ofwhich it is a mode. It might be objected that by
the same reasoning other qualities too must be modes of quantity. The
crucial instance would be one in which matter and qualities remained, while
quantity was ablated. But we have seen that the Aristotelians generally
agreed that a corporeal substance can exist without its quantity. Its figure
will, of course, not continue to exist. But that is because of its peculiar
relation to quantity. There is no contradiction-so it was thought-in hold
ing, say, for heat that it exists in a body even while that body lacks quantity.
The heat will not occupy space, to be sure, but again there seems to be no
contradiction in that consequence.59
59· Color-to take an instance discussed in the seventeenth century-might well not be

Authenticated
Descartes, like Boyle, was inclined to ridicule the doctrine of real qualities
or real accidents. But it seems to me that the difference discerned by the
Aristotelians between "adverbial" accidents and "primary" accidents is gen
uine. That they expressed the difference in holding that primary accidents
are real and adverbial accidents only modes was unfortunate, although the
discomfort of later philosophers rests largely on misinterpreting reality as
substantial existence. Still the intuition was sound. A real accident is a
feature of the world that, though neither a complete individual in its own
right nor capable of conferring existence of itself (like substantial form),
serves as the basis for further specification, as quantity for figure, or heat in
general for intensities of heat. An adverbial accident is such a specification.
That is not to say that there is heat in general, or indeterminate quantity,
and then also particular intensities or figures. Every actual instance of quan
tity inheres in a completely specified individual and is thus itself completely
specified. But a thing must have quantity if it is to have a completely spec
ified quantity. Though the converse also holds, the same instance of quan
tity can, as I have already said, be specified in different ways. So although
each instance of quantity can exist without the figure it now has, its figure,
being the specification of that quantity, cannot exist without that instance of
quantity. Another instance of quantity could, of course, have an exactly
similar figure; but that figure would be numerically distinct from the first. It
is in that sense that figure is dependent on quantity, or intensity on heat.
The final peculiarity of figure is passivity. The Coimbrans, arguing that
figure has no active power whatsoever, begin with the commonplace asser
tion that "every natural composite comprises matter and form, of which
[matter] has the task of undergoing action, and [form] that of acting."
Quantity serves as an "assistant" [administrnm] to matter, as quality to form,
in carrying out those tasks-quantity by way of supplying extension to mat
ter, quality by way of supplying active powers to substantial form. Of the
qualities we find in matter, some, like the elemental qualities, generate their
like in other substances. Quantity, on the other hand, does not: it "is conse
quent upon matter, and thus imitates its nature, and exists to receive, not to
act" (Suarez Disp. 18§4i3. opera 25:624). Although it is true that an animal
in reproducing will generate a quantified substance, its quantity serves only
as the point d'appui for the active powers that do the real work of reproduc
tion. Hence there is some reason to hold that quantity is not itself active. But
if quantity is not active, then a fortiori figure, which is merely the specifica-
capable of existing without quantity. Color was thought to be a quality inhering in the surfaces
ofbodies. A body without quantity, since it does not occupy any place, has no surface, and so no
color either. Of sensible qualities, perhaps only hot and cold would persist in the absence of
quantity. Even wetness and dryness, because they are defined in terms of the limits of bodies
dry bodies having definite limits, wet bodies indefinite limits-depend on quantity.

Authenticated
Vicaria Dei
tion of quantity, is not active either (Disp. 18§4!8, opera 25:625; Coimbra In
Phys. 2c7q1g, 1:302).
Figure does affect the way in which active powers operate. A knife must be
sharp in order to cut. For that reason some philosophers had argued that
figure is a "principle of local motion. "60 Albert, for example, argued that
"the acute figure [of an axe, say] is part of the active principle by which it
cuts," along with its heaviness and hardness. Similarly Durandus argued that
"the figure of a seal is the principle per se of the similar figure impressed in
wax" (Suarez Disp. 18§41g, opera 25:626). But Suarez and the Coimbrans
both argue that figure contributes to the result only as a "disposition"-a
term I will return to (in §5.3): "But one may answer that such figures are
only dispositions on the part of, the instrument or body, so that it may move
or be moved more easily in such manner, either because it resists the move
ment less [...] or because it is less resisted, as in the motion of cutting; for
where an instrument is sharper, it has fewer parts, and thus less resistance
occurs" (Suarez Disp. 18§4'fg, opera 25:627; cf. Coimbra In Phys. 2C7q1g,
1:303). The pointedness of a knife consists in its having a small cross section
relative to its length and the gradual increase of the cross section as one
moves from the point inward. Since resistance varies directly with cross
section (see, e.g., Coimbra In Phys. 4c9q5, 2:75), a pointed object encoun
ters less resistance than a blunt one. But the mere arrangement of parts (so
the argument assumes) cannot of itself initiate the movement. Only the
heaviness can do that. Figure contributes not to the existence of the move
ment, but to its manner.
Figures change when other qualities are altered, and when substances are
generated or corrupted, but always as a by-product. In any natural change,
as we have seen (§3.1), there is a terminus ad quem, a state aimed at, which
consists in the reception of a form. That form is the proper effect of the
causes of the change: the material cause is its subject, the formal cause the
form itself, the efficient cause the agent initiating the change, and the final
cause again (with some qualifications) the form. Figure, as we have seen, is
never the terminus of any natural change. It comes about, to use Toletus's
word, with change, but not because of change (Toletus In Phys. 2c2q3, opera
4:52va).
Natural change is paradigmatically change between contrary species un
der the same genus, as from cold to hot. Toletus divides such changes into
"immediate" and "mediate" (Toletus In Phys. 7c3q3, opera 4: 1g8rb). Imme
diate changes are between qualities that "are contraries, and by themselves
act and undergo action." Only the elementary qualities satisfy that condi
6o. Suarez and the Coimbrans both mention Durand us (In sent. 1d45q2n8) and Albert (In
de ca!lo 4 ad finem); in support of the opposing view the Coimbrans cite Scotus (In sent. 4d1qs),
Thomas (Contra gent. 3c 105, ST 2pt2qg6a2), and the Thomist commentator Ferrari us.

Authenticated
tion. Mediate changes are between qualities that "do not act of themselves,
but by way of the first [qualities], from the mixture of which they arise." All
sensible qualities satisfy that condition; they are contrary to one another
derivatively, by virtue of the primitive contrariness of the elementary
qualities they are composed of. Figures, on the other hand, are neither
elementary qualities nor composed from elementary qualities. They are
contrary to one another neither primitively nor derivatively, but only in the
very weak sense that no body can have distinct figures at the same time.
Change of figure cannot be subsumed under the schema that was intro
duced in §3.1. 61
The natures of things depend primarily on their active powers (§7 .1); the
reception of the substantial forms from which those powers emanate de
pends ultimately on the elemental qualities of the underlying matter. Quan
tity is indeed a necessary condition for the reception of form (§5.3). But of
itself, it is entirely indifferent to the particular forms that inhere with it in
prime matter. There are indeed certain limits to the size of complex mate
rial substances, especially animals (§6.4). In that sense the determination of
quantity, and thus to some small degree figure, has a role in the reception of
form. But in general, figure, however important to our knowledge of sub
stances and to the practical arts, has virtually no role in explaining natural
change.
That is, nevertheless, not the whole story. Figure has a role in arguments
against Atomism, since their identification of figure and form must be
refuted. It is also prominent in questions about the motions of the heavens,
since Aristotle held that the proper motion of the simple substances of
which they are composed was circular. There are a few questions even in
terrestrial physics which refer to figure, notably those that concern the
shape of the earth. But the figures in question occur as effects of motions
whose end is other than the production of figure.
The central texts argue that figure is at most a "disposition" contributing
to the effects of active powers. The fourteenth-century Parisian philosopher
Nicole Oresme argues, however, that configurations of qualities can have
61. The point is notjust that changes in figure are derivative upon other changes. Changes
in color or taste are also derivative. But in those cases the quality attained by the change can
still be regarded as the terminus ad quem of the change, and the state of the thing with respect to
that quality before the change as a privation. But (so I read Toletus) the figure attained in
change is never that which is aimed at in the change; it comes about incidentally as something
else is attained. When bits of earth coalesce at the center of the universe, the figure of the
resulting mass is spherical; but sphericity is not part of the terminus ofthe local motion of any of
the bits. The case of animal parts is more difficult. It would seem, for example, that when
carnivores grow teeth, sharpness as well as hardness would be the terminus, since otheiWise the
teeth would be ineffectual. Nevertheless, it may be that the sharpness of the teeth can be
regarded as derivative, this time not upon the actions of the elements, but upon that of the
soul, whose end is not the sharpness of the teeth per se but their effectiveness in eating or
fighting (cf. the discussion of the shedding of horns by deer in Gotthelf 1987:220, n.32).

Authenticated
[116] Vicaria Dei
proper effects. Contrary to the Coimbrans, who are very much against at
tributing efficacy to the figures and characters used in divination (In Phys.
2e1q7a2, 1:219), Oresme, though skeptical, does not reject the possibility
out of hand (De config. 1c31, p231). His doubt concerns not the possibility
of efficacy, but whether it can be achieved by art, since "it is much more
probable that bodies have an efficacy or power [ virtutem] arising from a
natural figuration of active quality than from an artificial figuration of qual
ity." Natural figurations probably do have proper effects. Different kinds of
animal, for example, may well have different characteristic configurations of
heat. The natural friendship between certain species might result from the
"fitting accord between the configurations of the primary or other natural
qualities" in each (Oresme De config. 1c24,27; p233, 241f). Differences in
the expression of the active powers of things-the sympathies and antipa
thies, for example, of natural magic-can be correlated with differences in
their qualitative figures. In the soul, the seat of memory and imagination
can be "figured qualitatively in a variety of ways depending on the diversity
of the forms or species which it receives [in sensation]" (ib. IC3l, p24g).
For Descartes the suitability of spatial figure, in its infinite variety, to provide
the terms for a "coding" of physical properties into shapes is the key to his
theory of the senses (see Dioptrique 4, Regula! 12). And when he enthusiasti
cally affirms in the Regula! that all physical "dimensions," qualitative or quan
titative, can be represented by lines and plane figures, he repeats one of
Oresme's contentions.
Nevertheless in all such instances figure remains a means of representa
tion only. More significant to Descartes was the use of geometric reasoning
in music and optics. These were among the "middle mathematics," neither
wholly physical nor wholly mathematical; they were also, and not coinciden
tally, the domains in which Cartesian science came closest to fulfilling the
promises made for it. I want here to mention one aspect of the relation
between physics and mathematics: the application of the matter-form dis
tinction to mathematical objects.
Toletus, following St. Thomas, distinguishes sensible from intelligible
matter:
By sensible matter [St. Thomas] understands a substance endowed with

elementary qualities-heat, cold, wetness, dryness-and composed from
them; while by intelligible matter, he understands substance conceived
solely by way of quantity, and which is considered in mathematics; so it is
commonly said. But as for intelligible matter, it seems to me that one
should say that quantity alone is intelligible matter. Note, however, that
the Mathematician has as his object not quantity, but figures and forms,
which occur in quantity. For as the ironworker is concerned not with iron

Authenticated
but with the form that is made in iron, so too the Geometer [is con
cerned] not with quantity but with the figures made in it. (Toletus In Phys.
1C1q3, opera 4:8vb; cf. 2c2q3, 53va)
The quantity that serves as the matter of mathematics, Toletus adds, is not
quite that which is found in things. Points, lines, and surfaces must be
abstracted from the quantity we encounter in experience before they can be
treated in mathematics ( gra).
The intelligible form corresponding to intelligible matter is figure. The
mathematician "proceeds by form [i.e., external form or figure], that is, in
proving passions by their forms or subjects; for he considers primarily form,
and makes proofs by them" (ib. gra, 53va). Why not, one might ask, also
colors or heat? After all, these too have extension. The answer has been
anticipated earlier. Figure alone, unlike other qualities (which also inhere
in quantified matter), is a mode of quantity, and like quantity, it is neither an
effect of natural change nor a cause (In Phys. 2c2q3, opera 4:52va). Quan
tity, moreover, when abstracted from matter, retains its modes (its "pas
sions," Toletus says), including, of course, figure. Other qualities, even if
they inhere in matter, are not modes of quantity, and will be cast aside.
The analogy is this: figure is to quantity as form to matter. Quantity is
therefore said to be the intelligible matter of mathematics. But in middle
mathematics quantity cannot be abstracted altogether from sensible matter.
Music is the arithmetic of sounds, optics the geometry of light rays. The
Coimbrans, defending the thesis that middle mathematics is "more mathe
matical than physical," note first that optics, music, and astronomy are
commonly included among the mathematical arts, and that "those who call
themselves Mathematicians, customarily treat them." More importantly,
"the affections that they demonstrate concerning their subjects are usually
shown by means of Geometry and Arithmetic." Since the medium of
demonstration is a "kind of form" in the sciences, middle mathematics, as a
branch of knowledge, differs from physics in form if not in matter (In Phys.
2c2q1, 1:227). But the argument, it should be noted, ill suits the practice
even of the Schools themselves. Local motion, which is definitely part of the
subject matter of physics, was treated geometrically, and, as we have seen,
there were efforts to extend that treatment to alterations as well.
A more promising distinction is found in the Coimbrans' reply to one of
the arguments for the opposing view. They grant that neither astronomy nor
optics treats quantity in general. To that extent the two sciences differ from
arithmetic and geometry, and can be said to be concerned with substance as
well as quantity. But "they do not investigate [scrutentur] their substance."
Instead they touch on substance-light or the ether-only incidentally,

Authenticated
[118] Vicaria Dei
because they consider the quantity only of certain kinds of substance. Their
content is tailored to suit the nature of the substances they treat.
The mathematics of music illustrates well what the Coimbrans may have
had in mind. 62 Beeckman writes: "I suppose that the nature of the human
voice, of pipes, lyres, and any sort of musical instrument is the same as the
nature of strings, since experience shows that all voices can be consonant
with the voices of strings" (Joumal1:54). Beeckman no doubt means that
''voice" or sound is the same sort of thing no matter what the source. But the
succeeding investigation, after starting with an idealized divided string,
consists simply in the comparison of ratios. Not only do the sources drop
out, but sound itself, except incidentally as that which motivates the mathe
matics. When, on the other hand, Beeckman examines properly physical
questions about sound-why, for example, two strings tuned to the same
pitch resonate with one another, or why a single voice may be pleasant or
unpleasant (ib. 166, 177) -mathematics is conspicuously absent.
Mathematically, music consists in the computation and comparison of
ratios of whole numbers (see Figure 4). Neither geometry nor algebra is
used. Nor are just any ratios of interest: only those generated from small
whole numbers are studied, since only such ratios are heard as consonant.
The mathematical content of music is thus dictated by the long-known
relation of string lengths to perceived pitch and by the felt pleasantness or
unpleasantness of sounds. Though in principle it could treat ratios with
arbitrarily large prime denominators, or even irrational ratios, it does not,
because such ratios are regarded as dissonant in every context, and are thus
never used. 63 To that extent the content of music depends on the sub
stances it treats. But since it never concerns itself with either the causes of
sound or the physiology of hearing, it can reasonably be called a mathemati
cal branch of knowledge.
Similarly in optics, the nature and causes of light are disregarded; quan
tity, in the form of "rays" or line segments, surfaces (of reflective objects),
and volumes (of refractive media), is again all that remains. Certain physical
features of light-that it travels in straight lines, that the sun radiates light
along all the radii extending outward from its center-guide the choice of
quantities. But once the choice is made, optics no longer concerns itself
with physical change as such, or with its causes and principles. When
Descartes declares, near the beginning of the Dioptrics, that he will not
concern himself with the nature oflight, he is doing no more than making
62. On music theory in the period of Descartes, see Cohen 1984 and Dear 1g88, c.6.
63. Beeckman, like Descartes, is somewhat conservative in this respect. Simon Stevin had
already constructed the "tem2ered" scale, in which the ratio between every pair of notes
separated by a semitone is 12-v2. Beeckman rejected that innovation (To Mersenne 30 Apr.
I63o,joumal1:I8o).

Authenticated
Matter, Quantity, and Figure [ug]
1
3. Oclava.
1
.
Ouotl«.iiiUJ 2
~ -s- t'Jitll~lir...
1
Det,m.~~.5~ .!. Otfava.. :s
4 4 4 r;ua,./t~,
1
Dtllin«J."'
·~ .... '3 '":,aj.
10 .,,. 4 JJi/lntl.l .
5 5 .$ 5
'
OdavtL. 4 lJuinle.. 5
1 D«. "
11114.J· 1i f fl.". ~ 7',.A«.m,~
6 6 6 6 6
Fig. 4· Intervals (from Descartes, Compendium musicce; AT 1o:g8; note that the largest
prime denominator is 5; intervals with prime denominators of 7 or more were
regarded as disonant when played both together and in succession)
explicit what anyone reading his work would have taken for granted. It is not
that declaration but the subsequent violation of it that evoked the criticisms
of Morin.
What distinguishes middle mathematics from pure mathematics is not its
objects but the principle by which those objects, and the propositions to be
proved about them, are selected. That principle comes not from the objects
themselves (if that is possible), or from simple curiosity, but from certain
physical or phenomenal properties of the particular kinds of bodies to
which a branch of middle mathematics is to be applied. Music deals in ratios
because it is the science of consonant sounds, optics in rays and surfaces
because it is the science of light. Middle mathematics, because it concerns
itself with quantity and its modes, is, from the Aristotelian point of view,
inept to determine the natures of things. Quantity and its modes are neither
the causes nor the effects of change. Though essential to every physical
thing, they are among the least physically significant of its properties. To
study them alone, or to postulate an object for physics that had only such
properties, would be, for an Aristotelian, to give up the project of natural
philosophy.
Indeed, a comparison of Descartes with Oresme and his imitators raises
an issue that I will return to in the next part. It is clear that for Oresme the
configurations of qualities, though real enough, and representable by geo
metric figures, are not-and could not be-themselves such figures.

Authenticated
[120] Vicaria Dei
Qualities, though by virtue of their intensities analogous to quantities, are in

no way reduced to quantities. The issue, then, is not so much why Descartes
thought that physics had to employ magnitude and figure in representa
tion. It is why he thought it had to conceive of the things represented as
nothing other than magnitudes and figures. The methodological virtues of
a geometric treatment do not entail that one regard the objects treated as
exclusively geometric, or even-as the example of Cusa shows-as quan
tities. So even though Descartes does often enough tout the methodological
soundness of what he and Beeckman called "physico-mathematics," I think
one must look elsewhere too. The profounder motives behind the identi
fication of matter with res extensa are to be found not in method but in the
philosophy of nature: Cartesian physics is, from the Aristotelian standpoint,
an anti-physics.
In prime matter we find a philosophical concept that, if it does its job,

threatens to put itself out of business. The hidden paradox in the maxim
that forma substantialis dat esse simpliciter is that what gives being simpliciter
must, on a plain understanding of the maxim, give it to nothing. The
solution of Suarez and Toletus is to separate the metaphysical actus of mere
existence from the physical actus of substantial existence. This does not
quite beg the question, but it leaves open the question of what it is that is
said to exist. The 'what', in Aristotelianism, is standardly given by designat
ing a form. But here there is no form to designate. None, that is, unless one
follows the way of Scotus and assigns to matter a form of corporeality; or
takes, following Averroes and Zabarella, or Ockham, quantity to be insepa
rable from matter even in conception.
That was the way of Descartes also. What may be surprising is that, con
trary to Descartes's confident supposition, the coincidence of quantity and
extension is not a matter of course. In the Aristotelian treatment, two quite
different reasons for separating them are given. From the doctrine of the
Eucharist comes the Su:irezian distinction between being present at and
occupying a place. That distinction will again have a role in the theory of the
rational soul. For now it suffices to retain the thought that the occupation of
place-whose primary effect is the passive power to resist occupation of the
same place by others-is not a necessary condition for having physical
effects at a place. The second reason for separating quantity from extension
is that 'quantity' denoted what came to be regarded as two quite different
properties of bodies: volume and specific density. Even in Aristotle there are
hints of such a distinction, hints made explicit by .1Egidius. 64 Cartesian
physics never quite met the challenge of explaining it.
64. See Hussey 1991:223, which cites De uelo4c4.

Authenticated
Matter, Quantity, and Figure [ 121]
Descartes's ideal was to displace Aristotelian physics from the center of

natural philosophy and to put middle mathematics there instead. He saw
that to do so required not merely a new method-his own, after all, only
canonized the methods of existing middle mathematics-but new objects
of study coextensive with those of the old physics. Hence the identification
ofjrmna corporeitatiswith quantity, and ofform with figure. The surreptitious
substitution of extension for Aristotelian quantity had the salutary effect of
sweeping aside superannuated puzzles, and kept, for a time, the specter of
Deus sive Natura at bay. It also ignored the real problems I have just men
tioned. But when those problems were resurrected, the context had
changed: solidity and density were now superadded to res extensa.

Authenticated
The Structure of Physical Substance
S o far I have examined matter and form separately. In the ordinary

course of nature, neither ever exists without the other. Each has its
proper office: that of substantial form is to give specific existence, to
cause active powers, to individuate; that of prime matter is to sub
tend quantity, to be the subject of the dispositions that enable the active
powers of substantial form to operate. The two thus complete one another
to such a degree that, as the Aristotelians argued, to suppose that either
should exist separately except by a miracle is to suppose that nature has
produced it in vain.
The intimate association of matter and form raises the question of their
distinctness (§5.1). Though the question becomes especially pressing in
psychology, it is asked and answered first in physics. The central texts have
no trouble with the separability of form from matter; on the separability of
matter from form, they record controversy. The more common view was that
matter could exist apart from form by divine power. One consequence is
that matter and form are really distinct. The real distinction of body and
soul is merely a special case. There were other more specific reasons to hold
that rational souls can exist without matter; but the availability of the gen
eral argument created a presumption in its favor.
The stronger the distinction between form and matter, the more urgent
the need to explain the manner in which they are united (§5.2). The
predominant view was that substantial form and prime matter, when united
in a complete substance, are unum per se, not per accidens. Yet it is not obvious
how that could be so, if they are really distinct. The difficulty will be familiar
to readers of Descartes's letters to Elizabeth. To one side, the danger was the
reduction of form to a mere mode of matter, with the consequence that the
[122]
Authenticated
soul is corruptible. To the other side, the danger was Platonism-a world in
which the soul, like every incorruptible essence, stands apart from matter.
The doctrine of substantial union avoids both dangers, but not, as we will
see, without raising new difficulties.
At the end of §3.2 I suggested that substantial form is known not by
uncovering a hitherto concealed level of structure but by its effects and
causes. Among them are the accidents required for matter to receive form,
which are both part of the material cause of form and, in generation,
efficiently caused by it. The most fundamental is quantity. In §5.3 I look at
the relation of quantity to form and the more general question of the
"dispositions" required for the reception of form. Unlike the Thomists,
Suarez and Toletus hold that quantity and other accidents common to
material substances precede form both in generation and in the constitu
tion of substance. In the resulting structure the quantity of a substance and
the qualities that "accommodate" prime matter to particular forms inhere
immediately in matter, while the active powers associated with the form
inhere in matter only by way of the form. That picture is already suggestive
of the division of powers in Descartes's physics.
The term dispositio, though never the subject of sustained scrutiny, recurs
frequently in physical arguments. It denotes, roughly speaking, the arrange
ment of parts-spatial or other-to some end. As we will see, it figures
prominently in Descartes's physics also, conveniently ambiguous as between
an innocuous sense of 'spatial arrangement' and a charged sense of 'pur
poseful arrangement'. Such ambiguities allow Descartes nominally to agree
with his opponents when in fact something quite different, and often anti
thetical to the Aristotelian sense, is intended. After examining the defini
tion of dispositio, I will look at some examples of its use in physical argument
(§5.3). The full account of a substantial form ought in principle to spell out
the dispositions necessary to its reception in matter; its active powers might
then be exhibited as efficient causes of those dispositions. Yet the texts here
examined rarely go past showing that such and such a power or disposition
must exist. At the end of §5.3 I will briefly consider why that is so.
The arguments in §3.2 on behalf of substantial form frequently referred
to the active powers united by it. It is, as I have noted, those active powers
that chiefly differentiate substantial forms. In §5.4 I examine the "emana
tion" of active powers from substantial form and the role of such powers in
the generation of substances. Although the substantial forms of terrestrial
substances are, by general agreement, incapable of generating their like
without the aid of celestial powers, they are capable of preparing matter to
receive copies of themselves; their active powers give them the means to do
so. In §6 I will follow out the teleology implicit in that account: here I
concentrate on the process.

Authenticated
Vicaria Dei
5.1. Matter and Form Distinguished

Substantial form and prime matter are really distinct only if each is capable,
at least by the absolute power of God, of existing without the other. There
seems to have been little doubt among Catholic philosophers that form
could exist without matter. 1 The existence of matter without form was an
other story. Those who, like Thomas, denied that matter had a proper
existence were inclined to deny also that it could exist without form even by
divine power. The central texts all diverge more or less greatly from that
position. Even the Coimbrans, who hold that prime matter is pure potentia,
distance themselves from the Thomista!. Suarez and Fonseca, since they
distinguish the existence of matter from its actualization by form, have little
trouble arguing that God can conserve matter in actu even without form.
Explicitly the question turns on the actuality of matter. But a more tren
chant issue is in the wings. In Genesis we read that "the earth was empty and
void." Augustine and other church fathers read 'earth' as matter, and
'empty' as 'formless'. 2 God first created matter and then supplied it with "all
the [species] of which this mutable world consists" (Augustine Conf. 12c8,
p220). So matter existed naturally without form for some time. Augustine
did not, of course, deny that God created form and matter from nothing.
But the thought that a formless, tenebrous matter preceded all creation
could not be far away. In the Tima!Us, matter is eternal; the demiurge im
prints it with simulacra of the Forms, and orders it toward himself and the
Good. 3 According to a certain Heraclides, disciple of Plato, there was "long
ago a formless empty world," in which everything "lay weary and burdened
1. Suarez writes that "Aristotle and [other) philosophers perhaps could deny that [material
forms) can in any way subsist separately from matter, because they judged that dependence
and actual inherence in matter are essential to them. But we Catholics, who believe that God
conseiVes accidents without their subjects, cannot doubt (although certain moderns do) that
God could also conseiVe substantial material forms without matter, since the dependence ofan
accident on its subject [...) is greater" than that of substantial form on matter (Suarez Disp.
15§9'l11, opera 25:532). Zabarella does not hesitate to ascribe to Aristotle the thesis that "being
in matter is a necessary and essential condition for [material forms)" (Zabarella De anima
2text3, p120A).
2. "Nonnulli enim [...) opinati sunt in ipsa nascenti mundi origine fuisse a Deo pro
creatam materiam absque ulla forma, idque significari cap. 1. Geneseos illis verbis, Terra autem
erat inanis & vacua, ubi, ut D. Augustin us cap. 4· libri imperfecti de Genesi ad literam ait, per
terram, qme omnibus elementis subsidet, materia, per inanitatem, eius informitas designatur"
(Coimbra In Phys. 1c9q6q3, 1:171).
3· See, e.g., for example Calcidius In Tima'Um '1306, 321, 329; pp307, 316, 323; Blumen
berg 1988:139. Weinberg has brief expositions of Platonist, Neoplatonist and Avicennian
cosmologies (Weinberg 1964:11, 21 rr, 116fT). Mersenne in a letter of 1630 reports that Robert
Fludd takes the Biblical creatio ex nihilo to be creation from formless matter: "Materiam (quam
s.epissime teneiffas vocat) esse id, quod proprie appelletur Nihil; ac proinde cum Deus dicitur
creare aut facere aliquid ex nihilo, intelligi creare aut facere ex materia." After which he asks,
"if these [opinions) are not impious, what can be?" (Mersenne A de Baugy 26 Apr. 1630, Corr.
2:442).

Authenticated
in sad silence," until the demiurge set the heavens apart from the earth, the
land from the sea, and gave the elements their proper forms (Coimbra In
Phys. 8c2q2a1, 2:164). It is not surprising that those who propounded this
theory bolstered their argument by appealing to Genesis. There is, after all,
nothing in the opening verses to contradict them.
If, on the other hand, matter cannot naturally exist apart from form, then
matter and form are either both eternal or both created. One could there
fore refute the eternity of matter by refuting that offorms, which was easier. 4
In showing, moreover, that matter is created along with form, one shows
that matter is not, as the Manichaeans thought, evil in itself. The Church
insisted as a matter of faith that all things, material and spiritual, "are
produced by God from nothing according to their entire substance,"5 and
that "every creature of God is good. "6 Only an uncreated matter, therefore,
could be evil; but whether reason alone could prove what faith affirmed was
another question.
Aristotelians, especially Catholics, had to navigate carefully among funda
menta that on their face were not easily reconciled. The world had to be, on
the one hand, utterly mundane, purged of divine attributes. No such
doctrine as that of Stoicism, with its world soul, could be admitted; nor,
certainly, could God be corporeal. 7 Nor could the existence of the world be
implied in that of its Creator: creation must be a free, hence contingent, act
of God. One route to enforcing the separation was to make matter and God
as distant as possible. God is actus purus, necessarily existing; symmetry
would have it that matter is potentia pura, having no proper existence at all. 8
Or, if it does, at least it exists independently only through miracle and is thus
entirely subordinate.
4· Of the twelve arguments used by Toletus to prove that the world has not existed for an
infinite time, eight presuppose that the creation was of complete substances, and not of matter
alone (Toletus In Phys. 8c2q2, opera 4: 214fT). The Coimbrans, who devote their energies mostly
to refuting the contrary view, offer four arguments, of which two presuppose that if the world
has existed for an infinite time so have humans (Coimbra 8c2q3a3, In Phys. 2:267lf).
5· Denzinger 1976, no. 3025 (Concilium Vaticanum I, 187o); cf. no. 8oo (Cc. Lateranense
IV, 1215, against the Cathares), and no. 1333 (Cc. Florentinum, 1442).
6. Denzinger 1976, no. 1350 (Concilium Florentinum, 1442). This follows an omnibus
anathema against heretics of all stripes, including Gnostics and Manichaeans.
7· See the discussion of God's body in Funkenstein 1986:42-47; on the nondivinity of the
world, see Milton 1981:191lf. On God as matter, see §4 n.27. Certain philosophers in the
twelfth-century School of Chartres identified the anima mundi of the Tim<EUS (see Calcidius In
Timamm 126, 54• etc.) with the Holy Spirit, the third person of the Trinity. One obvious
objection, noted already by William of Conches, is that the Holy Spirit is "consubstantial,
coequal, and coeternal with the Father"; the world soul, if it exists, must be a created thing
(Gregory 1955:134lf, 142, 149, 148n.1; cf. William's Dialectica [ed. Cousin], pp475-476).
Giordano Bruno, according to a scandalous report written the day of his death and published
anonymously in 1621, revived the identification (Mersenne Con: 1:137; cf. Michel 1962:118,
which cites Bruno Lampas triginta statuamm [1587], opera lat. 3:54).
8. See Coimbra In Phys. 1c9q6, 1:167; Suarez Disp. 13§4'll). opera 25:411.

Authenticated
Vicaria Dei
But since God's power is absolute, if it is not contradictory to suppose that

matter should exist without form, then God can bring about that it should
do so. The independence of matter from form is in keeping with the ten
dency in the central texts to mitigate the debasement of matter one finds in
Neoplatonism and in Augustine. Fonseca argues that if, as Dionysius af
firms, existence is the principal way for created things to participate in
divinity, then matter too must exist; for "otherwise the derivation of divine
esse would not descend through all ranks of created things" (Fonseca In
meta. 8Ciq I, 3:444aC). The readiness of Smirez and others to ascribe quan
tity and other accidents to matter and the increasingly elaborate theory of
mixtures are other indications that matter was being granted a more inde
pendent role in natural philosophy.
We find, then, a twofold claim: matter can exist without form according to
God's absolute power; matter cannot naturally exist without form.
I. The Thomists argued that matter without form would be actu sine actu,
which is openly contradictory. 9 Soncinas, citing Averroes, writes that "matter
differs from form but is never denuded of form, and indeed when it is
separated from one form, it takes on another, since if it were denuded of all
forms, what is not in actu would be in actu" (Soncinas Q. meta. 8q uesp,
pi8ob; Averroes In Phys. 2c2commi2, Dpera4:52G). If, moreover, the exis
tence of matter is a formal effect of substantial form, then since not even
God can produce a formal effect in the absence of form, not even God can
bring it about that matter should exist without form. 10
Those were the primary metaphysical arguments. Against the Thomists,
Suarez and Fonseca argue that the actus of existence is distinct from that of
informing. Averroes therefore equivocates: to hold that matter denuded of
form exists while lacking actus in the sense of being informed or composing
a complete substance is no contradiction (see Coimbra In Phys. Icgq6a{,
1:172). For similar reasons the second argument is rejected. It is false that
all existence is from form, "as from an intrinsic and as it were essential cause
(for this is true of complete substantial existence, but not of each partial
[existence]).'Though the existence of matter is naturally dependent on
that of the complete substance it is part of, and thus on the formal causality
of the form (which is, as we learn elsewhere, the union of matter and form),
still God may intervene and preserve matter even when the other is absent
and can have no effect (Suarez Disp. IS§gtg, opera 25:535).
g. Fonseca In meta. 8CJq1, 3:442aC; Coimbra In Phys. 1cgq6, 1:167; Suarez Disp. 15§9i2,
opera 25:532. Cited as arguing against the separate existence of matter are Thomas (ST
1q66a1, Contra gent. 2c54-55), Capreolus, Caietanus, Ferrarius, and Herv<eus-the usual
list-together with Bonaventura and Durandus.
10. Suarez Disp. 15§912, opera 25:532. Since a form can be an efficient and final cause as
well as a formal cause, the effect must be specified to be formal.

Authenticated
Since form can likewise exist without matter, the two are really distinct. 11
The manner of argument will be familiar to anyone who has read
Descartes's proofs of the real distinction between soul and body. His and his
opponents' reasonings, whatever their differences, share the fundamental
principle that whenever God can conserve each of two things without the
other, then they are indeed two things in the strongest sense. But where
Descartes admits such a distinction only between soul and body, and
demotes all material forms to modes of matter, the Aristotelians admit it for
all substantial forms, material and spiritual. The difference between the soul
and other material forms is established on other grounds.
In making a real distinction between matter and form, the Aristotelian
version of Aristotle stands apart from many recent versions. Many commen
tators now would, like the ethnici rather condescendingly mentioned by
Fonseca, deny that matter can exist without form or form without matter.
Form identified as organization or disposition, or as activity or power, must
be realized in a material subject. 1 2 In Aristotelian terms, that is to make form
a mode of matter, as Descartes did. One mark of how successfully he and his
like integrated their view ofform into the new science is that few commenta
tors other than those who continue the Scholastic tradition even notice the
shift.
Descartes is noticeably reticent about the means by which God may con
serve things that are incapable of existing apart naturally. Suarez, thorough
as ever, takes up the problem. How, he asks, can God supply the effect of the
form? His response is relevant not just here but wherever the Aristotelians
hold that two things not naturally separable can exist by divine power with
out one another. The opprobrium heaped by Descartes and Boyle on the
doctrine of real accidents is directed precisely against the notion that some
thing whose raison d'etre is to exist in another should nevertheless be held
capable of existing when that other does not. That they take to be not
11. Eustachius, exceptionally, seems to hold that matter can exist without form but not
form without matter: "Quod si materia virtute Divina omni forma exspoliaretur, posset sua vi
subsistere; quia prius natura est materiam esse quam formam: at vero forma, si virtute Divina
separetur amateria, non posset vi sua cohrerere, sed statim (nisi virtute Divina sustentaretur) in
nihilum rueret; quia ab eo subjecto pendet ut conseiVetur ex quo educitur [...] " (Eustachius
Summa pt3, Physica 1tnd2a9; 3:127). The difference is, however, smaller than it appears.
Matter can exist without form by its own power (sua m), but can only come to be without form
through God's power; form can neither exist nor come to be without matter unless sustained
by God.
12. Charleton, for example, writes: "how can there be a formal aspect which is not an aspect
of anything, a form which is not the form of anything?" (1987:422). On form as disposition,
see, e.g., Frede 1985:21; as structure and activity, Moravcsik 1991:46. As Charlton notes,
Aristotle's nondualist theory of the soul has newly become attractive to antidualist philoso
phers of mind, who tend to set aside as unfortunate lapses passages that intimate another view
(Charlton 1987:408).

Authenticated
[ 128] Vicaria Dei
merely implausible, but unintelligible: not even by a miracle could God

bring it about that quantity, say, should exist without inhering in matter.
It is worth examining, then, how an Aristotelian conceived of God's oper
ation in such instances. I will take two: the preservation of matter without
form; the preservation of accidents without a subject.
In Suarez's account, the first point is that the "dependence" of matter on
form is "extrinsic. "13 Matter depends on form either as "a condition or
actual disposition naturally required," or as "an informing cause [...] con
curring and assisting per se in the existence of matter." The first is "quite
extrinsic and a posteriori," amounting only to the fact that a thing lacking
such a disposition cannot naturally exist. The second, too, is extrinsic:
Although according to this mode [of dependence] form concurs in a

certain way as a cause to the esse ofmatter, it does not concur as an intrinsic
cause or component of that esse, or as its proper subject, but only as an
informing or actuating cause, and as [an] extrinsic [cause], in the sense
that it is entirely distinct from the effect; and God can make up for
[supplere] a cause of this sort as an efficient cause [efficiendo], even if it
causes by informing. The material cause, on the contrary, is intrinsic like
the formal [cause]. God cannot make up for such a cause with respect to
its effect[...] But he can make good its causality with respect to the other
component-form-even if matter, according to its genus [i.e, as the
material cause] contributes to [ injluat] the existence of its form per se and
as a true cause. And in that way God may make up for the dependence of
an accident on a subject. (Suarez Disp. 15§g'l[6, opera 25:534)
Of the four Aristotelian causes, two-the material and formal-are intrin

sic, since they are parts of the thing ofwhich they are causes, while the other
two-the efficient and final-are extrinsic. Not even God can make up for
the absence of an intrinsic cause, since (as Suarez shows elsewhere) actual
existence is necessary for an intrinsic cause to have its proper effect. But an
intrinsic cause may have other, incidental, effects that in the order of nature
invariably accompany its proper effect, but in relation to which it is not
intrinsic. The existence of matter invariably accompanies the intrinsic effect
of form (the complete substance). But matter itself is not intrinsically
caused by form, nor form by matter. In a human being, which is a complete
substance, God cannot make up for the absence of either matter or form.
But it is possible that God should conserve the body without the soul, since
the soul "causes" the body only by way of intrinsically causing the human
13. Dependentia and related words, used as technical terms, typically denote causal rather
than, say, logical dependence; causal dependence can, of course, be understood in terms of
any of the four Aristotelian causes. To say that the dependence of matter on form is extrinsic is
to hold that form is an extrinsic (efficient or final) cause of matter.

Authenticated
being of which both are parts: "God can make up for the causality of form,
not in the composite, but in the matter, not by informing it but by effecting
it [ neque informando, sed efficiendo]" (ib. ,6, 25:534).
Suarez's position therefore is that God can, extraordinarily, bring matter
into existence without the extrinsic contribution ordinarily provided by
form.l4 There is no difference in God's action, but only in its circumstances.
Whoever believes in the "concreation" of matter and form in complete
substances must admit the possibility of the creation of matter alone. Sim
ilarly, if God allows the form of a substance to perish while preserving its
matter, the miracle consists. only in his having continued the action of
conserving matter "in that state in which it should not be, and without the
conditions that are necessary to the natural mode of existence" (ib. , 7,
25:535). Whatever mystery resides in God's mode of operation in such
instances resides also in the most ordinary instances.
My second instance, the conservation of quantity and other accidents
without their subject, requires an action quite different from the ordinary.
In his lengthy treatment of the Eucharist, Suarez argues thus:
(i) When the subject of quantity is ablated in transubstantiation, quan

tity is deprived of a "positive, real, and intrinsic mode," namely its
actual inhesion or union with the subject (Suarez De sacrif myst.
d56§2, opera 21:280). Inherence is not a mere denomination, nor is
it simply presence. Hence "quantity, when it is separated from its
subject, is deprived of this intrinsic mode." So it will not suffice to
hold that God simply conseiVes quantity without its subject; some
new action is needed.
(ii) When quantity is conseiVed separately, it is altered by receiving a new
mode of being that is in fact repugnant to its former mode of inher
ence in a subject. Suarez admits that this conclusion is only probable,
and that the contrary position (held by Scotus and Domingo Soto,
among others), can be defended (1g, 281). The advantage of his
own position, which is that of Thomas, is that if the miracle by which
quantity receives the new mode of being is granted, no further mira
cles are needed for the persistence of other accidents without matter,
since they can be taken to inhere in quantity as a proxy-subject (see
28 2, and the passage from Thomas's Sentences commentary quoted at
274).
14. "Si a principio crearet [Deus] materiam solam sine forma, actio eadem omnino esset
cum ilia, qua Deus actu creavit materiam sub formis solumque constitisset miraculum, vel
pr<eternaturale opus in hoc, quod Deus faceret illam actionem sine concomitantia alterius, per
quam induceret formam in talem materiam": "If at the beginning [God] had created matter
alone without form, his action would have been the same as that by which God actually created
matter with its forms, and in doing so would have done something miraculous or preternatural
only insofar as God performed this action without the accompaniment of the other action by
which he induced form in such matter" (Suarez Disp. 15§9!7, opera 25:534).

Authenticated
Vicaria Dei
(iii) The new mode of being entails a new action of conservation by God,
rather than a continuation of the old action. In the natural order,
God conserves a subject, together with all its forms, in a single act.
Suarez is careful to note that in this act God does not act in the role
of material cause (which, being intrinsic, cannot be substituted for)
(114, 284).
What does the new mode of being consist in? God, Suarez writes, "can
bring it about that an accident should in a certain manner participate in the
mode of existence of substance, by a certain mode that has formally (so to
speak) a resemblance [similitudinem] with the substantial mode [of exis
tence] but materially and in being [entitative] is accidental, to which it
suffices that it should affect and terminate accidental being, although it is
not related to another subject" (, 13,283). The argument is largely
defensive-against the very complaint that would be voiced so many times
in the seventeenth century. It proposes to show that to call something an
accident and to suppose that it does not actually inhere in a subject is not
contradictory. Fonseca, citing the common view of Catholic theologians,
argues that the inhcerentia essential to being an accident is not actualis but
aptitudinalis. An accident, in other words, is not the sort of thing that must
inhere in a subject if it is to exist at all. It is the sort of thing to which it is
essential that ordinarily it would so inhere (Fonseca In meta. 7Ciqt,
3:tg8).15
The strategy should be familiar. We have seen it at work in Suarez's
account of quantity. There quantity was construed as not actual but poten
tial extension, here the inherence essential to accidents is construed as not
actual but potential being in a subject. Are such distinctions "groundless," as
Boyle says (Papers 22)? They are, first of all, entirely in keeping with the
spirit ofAristotelian physics, which defines, as we have seen, change itself in
terms of the passage from potentia to actus. Reconstructions of notions like
quantity as potentice are not, it seems to me, illicit extensions of the basic
scheme. They are indeed extensions: the occupation of space by quantity is
a "change" only for someone who is inclined to regard every actuality as the
15. "Inherence is twofold, actual or aptitudinal [ ...] Actual [inherence] is that by which
an accident inheres actually [actu] in subjects; aptitudinal [inherence] is that by which it has a
propensity to inhere [i.e., actually]. No Philosopher either discovered or affirmed this distinc
tion until after it was shown by faith that the accidents of the Holy Eucharist did not exist in
substance [...] All judged, before the mystery was divinely revealed, that there was no other
inherence than actual [inherence], and indeed that it was contradictory for accidents to be
separated from the subject substances in which they inhered" (Fonseca In meta. 7c1q1§2,
3:1g8aF). In the commentary on 7c1, Aristotle's view is said to be that "non posse ullum
accidens existere sine substantia, aut ex subjecta substantia in aliam substantiam migrare."
Both conditions are violated in the Eucharist, the first since the quantity of the host no longer
inheres in matter, the second because the other accidents, which used to have the matter of the
host as their subject, now have its unmoored quantity as their subject.

Authenticated
coming-to-be of something that was, or at least could have been, merely

potential. Inherence, although it is not, properly speaking, a motus, is an
actuality, an existing state of affairs not only for the subject but for its
accidents.
There is no doubt that Suarez's argument was intended to resolve an
apparent contradiction between reason and revealed truth. Since doctrines
of double truth, granting to each source of knowledge its own kind of truth,
had long since been ruled out, the contradiction had to be resolved, and in
favor of revelation. 16 The question is not whether the shift was motivated by
the desire to avoid contradiction-philosophers habitually make distinc
tions to that end-but whether it was entirely ad hoc. I have already sug
gested one way in which it could be regarded as a plausible extension of
methods already proven. Another way is this. We have seen, and will see
again, that the essences of things (other than God) are to be found not so
much in what they actually are, or have been at a given time, as in what they
potentially are at all times. Human nature is not well captured in a descrip
tion of Sappho as she is now: if she is sleeping, she neither senses nor
perhaps thinks; if she sits, she does not run. Yet she has the power to do all
those things. Actualities come and go; only potentialities persist. Perhaps
the same holds for accidents: what matters is not actual but only potential
inherence.
One could argue, of course, that quantity and inherence are exceptional.
But once the boundaries of the possible are marked by logical contradic
tion, or-what comes to the same-by God's absolute power, it becomes
quite difficult to argue that the actual possession of any accident or mode is
essential to a thing. Is it indeed contradictory to speak of 'matter without
quantity'? Only if prime matter cannot but have the power to occupy space.
But prime matter is that which, when joined with substantial form, yields a
complete substance, and nothing more.
The case of inherence is harder. Boyle, after quoting a standard defini
tion of inesse, and of 'accident' as id cuius esse est inesse (a formula cited by
Fonseca), writes that if the Scholastics ''will not allow these accidents to be
modes of matter, but entities really distinct from it and in some cases separ
able from all matter, they make them indeed accidents in name, but repre
16. The term "double truth" refers to two versions of the relation between philosophy and
revelation: (i) what can be demonstrated solely by reason or experience and what is revealed
are different-but all the same only one of the two, namely revelation, is true; (ii) the truths of
philosophy contradict, and therefore compete with, revealed truth. The first is found in a
number of philosophers in the thirteenth and fourteenth centuries, including Boethius of
Dacia. It was partly in response to his De O!temitate mundi that Stephen Templier, the Bishop of
Paris, issued the condemnations of 1277. The second version, though condemned by Tem
plier, was not, it would seem, unambiguously affirmed until the end of the fourteenth century
by Blasius of Parma. See MacClintock 1956, c.4; Hissette 1977=13; Steenberghen 1991:348ff
(on the condemnation); Maier 1955, c.1 (on Blasius).

Authenticated
Vicaria Dei
sent them under such a notion as belongs only to substances [...]," a

doctrine that is either "unintelligible or manifestly contradictious" (Boyle
Papers 22). He takes it to be part of the definition of 'accident' that an
accident "cannot exist separately from the thing or subject wherein it is" (Pa
pers 2 1). The only lexicon I have found that definition in is Chauvin's,
published after the heyday of Aristotelianism. Chauvin, in keeping with
Boyle's (and Descartes's) conception, defines accidents as modes: an acci
dent is "a mode affecting a created substance, and so dependent on it that it
cannot exist outside [that substance]." 17 But to define accidents as modes
begs the question against real accidents-unless, of course, you have al
ready accepted Cartesian physics.
The Aristotelian would insist that not all accidents are modes. Some, like
figure, are modes because they depend on other accidents (§4.3). To them
actual inherence is essential. Other accidents, including not only passive
qualities but quantity, depend naturally only on the subject. Since actual
inherence is inescapable for modes, one might conjecture that other acci
dents might be distinguishable from modes in that respect. Yet clearly
inherence-being in a subject-is, so to speak, diagnostic of accidents:
hence the move to potential inherence, which is therefore essential to all
accidents; for modes the stronger condition of actual inherence is added.
In an earlier section I suggested that in Descartes and his successors self
subsistence became the primary defining feature of substance, with the
consequence that the "second substances" of the Categories dropped out of
sight, along with 'being of' (rather than 'in') a subject. A parallel simplifica
tion occurs when 'accident' and 'mode' are taken to be synonymous. Actual
inherence and potential inherence coalesce; the notion of real accident,
meaning that which is both an accident and capable of being consetved
apart from substance, becomes, as Boyle quite rightly affirms, unintelligible.
But that unintelligibility is not a datum of human experience, nor is it an
inevitable consequent of the notion of accident. It is a by-product of the
struggles of the new science against the Schools.
All the same, to conceive whiteness apart from the wall it modifies does
strain the imagination. Suarez admits as much. His positive account of the
existence of accidents apart from substance rests on a rather suspicious
17. "Accidens pnedicamentale [...] est modus substantiam creatam afficiens, & ab eadem
ita dependens, ut extra earn existere non possit" (Chauvin, s.v. "accidens pr<edicamentale").
Chauvin's Lexicon is, however, a tainted witness on Aristotelian matters (Gilson 1913:v).
Goclenius, a better witness, mentions no such condition; he says only that being ordered
toward a subject is of the essence of accidents (Goclenius Lexicon 29). Micraelius states both
that accidents cannot be separated from substance (accidens non separatura substantiis), because
its esse is in esse, and that an accident is called "separable" if "vel reipsa, vel mentis abstractione a
subjecto separari potest." Accidens reale he defines simply as forma inhamms (Micraelius Lexicon,
s.v. "accidens"). The Port-Royal authors, like Boyle and Chauvin, take 'mode', 'accident', and
'quality' to be virtually synonymous (Arnauld and Nicole Logiquq7).

Authenticated
analogy with substances. One might regard it as what philosophers of sci

ence call a 'how possible' explanation: it defeats the claim of incoherence
by showing one way in which accidents could exist without their subjects,
but it does not assert that in fact they do in that way. The existence of matter
apart from form is on firmer footing. That God can create matter conjointly
with form will readily be granted by anyone who assents to the orthodox
doctrine. If form setves merely as a condition or assisting cause in the
creation of matter, then God, who can do whatever he does without as
sistance, can create matter apart from form, or consetve it when form per
ishes. It is thus rather easy to shift the burden onto those who deny that
matter can exist without form.
2. The absolute possibility of prime matter existing apart from substan
tial form is thus established. 18 Since form can, by similar arguments, be
shown to be able to exist apart from matter, the two are really distinct. What
remains to be shown is that the two cannot naturally exist separately.
On that issue little need be said. There was no great controversy among
the Aristotelians themselves. 19 The use of finality, however, in the two argu
ments given by Fonseca will be of interest later. The first shows that matter
cannot be deprived of one substantial form without immediately taking on
another. That is a consequence of a more general principle: "agents cannot
intend mere privation, but rather some actus and perfection, to which a
privation is joined." So the natures of things "are not instituted to despoil
matter of any form, unless it is in order to induce another" (Fonseca In meta.
8oq1§4, 3:447bC; cf. Coimbra InPhys. 1cga6a1, 1:167). It follows that the
corruption of a body is only the incidental effect of a cause aiming to
introduce into its matter a new substantial form. Since no natural cause aims
at producing matter devoid of form, there is no reason for it to come about
naturally.
No efficient cause will bring about bare matter; neither will any final
cause. Nature allows nothing otiose; but matter sans form would be, since "it
can by itself sustain no operation." Its passivity is recompensed when it is
joined with form, but not otheiWise. Matter without form is matter without
purpose. One might argue that it could exist for the sake of other things
that do have form and are capable of acting-humans, say. But it is charac
teristic of Aristotelianism not to construe natural purposes entirely in rela
tion to human ends (see §6.2). Indeed, the natural impossibility of matter
18. I will use the phrases 'absolute possibility' and 'natural possibility' occasionally as short
hand for ',possibility relative to the absolute (or ordained) power of God'. Absolute possibility
thus coincides with logical possibility, but its rationale is different.
19. The authorities cited by Fonseca and the Coimbrans are patristic (see n.2 above, and
Fonseca In meta. SCI q 1§4, 3:448bC). The only exceptions are Marsili us oflnghen and Gabriel
Biel, who held that Christ's body existed without a substantial form after his death.

Authenticated
Vicaria Dei
without form is based on the contrary claim: nothing can exist naturally that
does not have its own ends.
It is instructive to contrast these arguments with those of Boyle and
Descartes. Although one may show rather easily in Aristotelian physics that
matter without form has no place in the natural order, it is very difficult to
show that separate existence leads to absurdity. That, and not the mere task
of showing that matter without form is unlikely or contrary to experience, is
what Descartes and Boyle had to accomplish. It is not surprising that the job
was done by changing the definition of 'accident'; little short of that could
have worked. The change was not unmotivated. It was harmonious with the
new physics. Rejecting potential in favor of actual inherence in the defini
tion agreed with the general policy of banishing powers and potencies of all
sorts. But from the viewpoint of the eventual losers in the debate, the
arguments of the victors could not but have appeared to beg the question.
5.2. Substantial Union

Prime matter and substantial form are really distinct, a plurality indepen
dent of the manner in which they are conceived. Yet the complete substance
of which they are parts was held to be not two things but one. The question
was not whether but how matter and form make one substance. Since hu
man beings are an instance of substantial union, the question bears imme
diately on psychology and on theological doctrines about the immortality of
the soul. Among Descartes scholars, substantial union-long overshadowed
by the interaction of soul and body-has recently been studied with some
attention to its historical context (Hoffman 1g86). But it is still unusual for
the union of body and soul to be seen as what it was: a question whose terms
were set not in psychology but in physics. There are, of course, particular
difficulties concerning the rational soul. It alone among material forms is
not "educed" from matter; it alone can exist apart from matter even without
divine intervention. Nevertheless what was proved about the distinctness of
matter and form and about their union could be, and was, applied mutatis
mutandis to the soul. Indeed the insistence of the Church on that point
hindered attempts to show by reason alone that the human soul is immortal.
Since the controversy about substantial union concerns the claim that
matter and form are unum per se, I will examine that notion first. The
question then will be whether prime matter and substantial form, when they
coexist in complete substances, enjoy the highest form of unity that a com
posite thing can have-whether, in other words, a complete substance is
unum perse.

Authenticated
1. Unum per se, unum per accidens. The Coimbrans distinguish several
degrees of unity, from the "unity of aggregation" exemplified by a heap of
stones to that of simple substances "free from commixture with matter"
angelic souls, God. In between are the two kinds of unity most relevant to
the question: the unitas per accidens of accidents with their subjects, the
unitas per se that results from the "composition of parts coming together in
some third nature." Such unity we find in quantity, when two parts share a
boundary, and-the Coimbrans add without further argument-in the
composite of matter and form.
As an account of the unity of composite substances, that amounts to little
more than labeling what one hopes to explain. Nor does it help much to be
told that "all that arises out of distinct things without any physical and real
union between them" is only an ens per accidens (Suarez Disp. 4§3! 13, opera
25:129).20 After all, there seems on the face of it to be little reason to deny
that whiteness and the wall are physically united. Yet accidents, apparently
without exception, are supposed to be only unum per accidens with their
subjects.
In the Coimbrans' further remarks, however, we find two more compell
ing reasons to set substantial unity apart. The first is that matter, as pure
potentia, and form, as the actus substantialis of matter, are "ordered first of all
to each other." Alone, each is, as we have seen, an incomplete substance,
and each is completed just by the other (In Phys. Icgq11a2, 1:184). The
second is that matter and form are united without intermediaries. Though
whiteness and the wall are indeed one in a certain sense, the whiteness
inheres in its subject, the complete substance, only by virtue of the substan
tial form that has actualized the matter of that substance. Setting aside the
difficulties of the Eucharist, accidents always presuppose the presence of
both matter and substantial form.
Suarez sets up the general principle that whatever has one essence or
being [entitas] is unum per se. It will have one essence just in case it has
''whatever bears on its intrinsic reason or consummation." Simple sub
stances obviously satisfY that criterion. As for composite substances, "since
matter and form per se are not complete, whole beings in their kind [entia
completa et integra in suo genere], but are instituted by their nature to be
composed into [a substantial nature], that [nature], which is composed
immediately from them, is deservedly called, and is, a nature and essence
which is one per se" (Suarez Disp. 4§318, opera 25:127-128). Matter cannot
naturally exist without form or form without matter; in that sense, each of
20. Unum per se (per accidens) and ens per se (per accidens) are taken to be more or less
interchangeable: "that which is unum per seis called [...] an ens per se, while that which is only
unum per accidens is called [ens] per accidens" (Suarez Disp. 4§3i3, opera 25: 126).

Authenticated
Vicaria Dei
them, considered by itself, does not have all that is needed to fulfill its end,
which is precisely to exist as part of a composite.
Yet much the same could be said of accidents and substances. As we have
seen, potential if not actual inherence is essential to being an accident; and
substances are said to be completed or perfected by accidents, and are thus
"instituted" by nature to receive them. Why then are accident and substance
not unum per se? The union of accident and subject "is a third genus of entia
per accidens, which seems to depart more [than the unity of mixtures] from
the first and least [genus] ens per aggregationem, and to approach more to
unity per se, because the things from which it arises are not distinct with
respect to their suppositum [...] , and have a greater physical union; the
[substance] is truly in potentia to the [accident] [...] , and the [accident] is
by its nature ordered to the [substance], and in it has its connatural perfec
tion; in all [features] ens of this sort imitates that which is properly and per se
one, although it is simpliciter and absolutely per accidens. "2 1 In view of those
resemblances, it is difficult to understand why accident and substance are
nevertheless only unum per accidens. The reason we were offered in the case
of matter and form was that each completes the other. But, as Suarez notes,
much the same could be said of substance and accident. Accidents, as we
have seen, must at least potentially inhere in substances, and naturally must
do so; no substance can exist naturally without the accidents appropriate to
its form, either as necessary conditions or as effects.
Two more promising differentia emerge from questions on unity. But nei
ther looks decisive. The first is that matter and form united constitute a
single essence, although substance and accident do not. Without examining
the notion of essence, one cannot evaluate the point, but at first sight it
would appear that quantity at least is both an accident and part of the
essence of corporeal substances. Suarez writes, "But because we hold that
quantity inheres in prime matter, with which it does not make an unum per
se, but per accidens, it thus should be added that accidental form, by the very
fact that its nature is in relation to [ respicit] a being [esse] of another order
and category, and is not ordered so as to constitute or complete it in that
order and category, does not constitute an unum per sewith it, but an unum
per accidens, because what is unum per se must be of one order and category"
(Disp. 16§1,13, opera 25:570). The composite is a substance made up of
substances. Its components are of the same category as each other and as it
21. Suarez Disp. 4§3, 14, opera 25:130; cf. the similar discussion at 25:570. The term
suppositum, applied to created things, can be interpreted as 'first substance' in the sense of the
Categories: a "singular individual substance subsisting per se" (Disp. 34§ 11 7-g, opera 26:350).
'White' and 'Socrates' are said of the same individual, whereas the citizens of a republic,
though united, are distinct individuals.

Authenticated
itself. The general principle is that a thing can be unum per se only with
another thing of the same category. It is likely to seem ad hoc, since the only
case that matters is precisely the one to be proved. Nor is it clear-especially
to a modern philosopher inclined to reduce essences to essential
properties-why only a substance-substance union, and not a substance
accident union, should yield a single essence. The question of the unity of
complete substance has in effect been deferred to that of the unity of
essence.
The second differentia is that accidents are always united to an entity that is
itself already a composite of matter and substantial form. They are second
actus of matter, and presuppose a first actus, namely, substantial form. But
substantial form itself seems to presuppose certain accidents, notably quan
tity. In what sense, then, is it "first"? A Thomist can straightforwardly answer
that prime matter doesn't even exist unless joined with substantial form;
evidently no other form can precede it in union with matter. The central
texts, on the other hand, mostly hold that matter has a proper existence,
and some, like Suarez himself, hold that quantity sometimes inheres directly
in matter (see §5.3). The priority of substantial form remains to be ex
plained.
2. The unity ofform and matter defended. The first and most important of the
arguments refuted by the Coimbrans takes up that point generally. Either
prime matter and substantial form are united by way of some connecting
medium, or not. If there is a medium, then it must be an accident; but if
matter and substantial form are united by means ofan accident, they are not
unum perse.
If there is no medium, then since substantial form is supposed to unite
immediately with prime matter, it could, "at least by divine power, be united
without the intermediation of quantity, so that from the rational soul and
matter there would result a composite which was unum per se," and which
could only be a human being (rather than some other kind of thing). Thus
we could have a human without a head or heart, which "should not, it
seems, be conceded." Experience, moreover, shows thatjust as the genera
tion of a substance requires a particular kind of agent, so too it requires a
certain kind of matter. If form united immediately with matter, there would
be no reason for this: even without a miracle the human body might take on
the soul of a horse. Since that does not naturally occur, it must be that
substantial form can only unite with a matter already endowed with suitable
accidental forms (Coimbra In Phys. 1cgq11a1, 1:182-183). We are forced
back to the first horn of the dilemma.
The Coimbrans deny that between two things that are mutually accom
modated and proportioned, as are matter and form, there need be any

Authenticated
Vicaria Dei
medium by which they are united (<14, 1:186). 22 They choose instead the
second horn of the dilemma, and the consequences of holding that matter
is united immediately to form. Following Scotus they argue that, however
unnatural the result might seem, Socrates' soul could indeed be joined to
matter deprived of quantity. In fact, "any sublunary form can be conjoined
to any portion of inferior [i.e., sublunary]. matter by divine power" (Fonseca
In meta. 5c2q14§2, 2:1g6bF, cited by Coimbra).
It remains to be explained why not any form can be naturally joined with
any matter. Socrates' soul requires a body endowed with suitably arranged
parts, which themselves must be composed of bone and flesh so as to sub
serve the various powers of the soul. It is not enough to observe that such a
form could be joined to matter lacking suitable accidents, so that the two
would then be unum per se. The presence of the accidents in a naturally
generated substance may still be grounds for holding that as it actually is it is
not unum per se. A ship and dock joined by a gangway are not unum per se,
even if by welding ship and dock together we can make a thing that is unum
per se. We find again the question raised a moment ago: are matter and form
immediately united, even in the presence of accidents that are admitted to be
naturally necessary to their union?
5·3· Conditions for the Reception of Form: Dispositions

That certain accidents were naturally necessary to the reception of even the
simplest forms was universally acknowledged. Quantity at least is common
to all. The Aristotelians made a standard distinction between "proximate"
matter, the already formed stuff required by a particular substantial form,
and "remote" matter, that which supports all the forms of a thing, substan
tial and accidental. We have seen that there were various opinions about the
essence of prime matter. But all agreed that, however the fact was to be
explained, substantial form is always naturally accompanied by a characteris
tic set of accidents when joined with matter.
The explication of the structure of material substances was bound up with
that of generation and corruption, and of life. The active powers of a living
thing have as their collective end its continued existence and the generation
of its like in new matter; even sensation and intellection are subordinate to
those ends for people ifwe consider them only as living (if we consider them
22. Similarly Suarez writes that the union of matter and form "depends essentially both on
form and on matter [...] because it is their actual bond, and no other bond inserted between
them is needed [...], since otheiWise it would proceed to infinity" (Disp. 13§919, Qplffa
25:431). The union, which is nothing other than their mutual dependence in the complete
substance, is a joint mode of each, just as motus is a joint mode of agent and patient.

Authenticated
as rational, their ends are different). Experience shows that the effect of the
generator consists largely, perhaps entirely, in preparing the matter of the
new substance to receive its form. Since active powers are defined by their
(completed) actions, and since actions are defined by their termini, the
scientific understanding of substantial forms depends on that of the disposi
tions that it is their purpose to sustain and reproduce.
The simple formed matter of Physics lis thus unfolded into several layers.
The first is prime matter. The second is prime matter together with quantity
and dispositions, the proximate material cause of substance, and thus of
substantial form. That relation will be studied here. The third layer consists
in the active powers of which substantial form is the efficient cause. Those
powers are in turn the proximate efficient cause of the dispositions of the
new substance. I will examine that relation in the next subsection.
The role of accidents in the reception and sustenance of substantial form
is touched on in a variety of contexts. I will divide the discussion into three
parts:
(i) The rejection of preexisting substantial forms or "seeds" in matter;

(ii) the inherence of accidents, especially quantity, in prime matter inde
pendently of, and prior to, substantial form;
(iii) the notion of dispositio and its use in the theory of generation and
corruption.
Although the first two questions become largely moot for the new science,
the third retains its importance. Mechanistic hypotheses on the phenomena
of life were for some time not markedly better or less ad hoc than those of
the Aristotelians. 2 3 Their advantage consisted mostly in being mechanistic.
Descartes acknowledged that his physiology was incomplete without a the
ory of generation; yet his attempts to devise one were at best mere prom
issory notes, at worst question-begging. Biology has since redeemed those
notes. But we should not overlook the fact that, as Leibniz soon recognized,
the arguments Aristotle raised against the reductive theories of his pre
decessors had lost but little of their force in Descartes's day.
1. Form educedfrom matter. Doctrines of preexisting form typically came up
in questions about the maxim forma educitur a materia (form is educed from
matter). The maxim was as usual not questioned; only its interpretation was
subject to debate. One obvious reading is that form, if it is educed from
matter, somehow already exists in it, but occultly or imperfectly. Anaxagoras
"contended that everything is in everything, [...] so that the generation of
23. Two notable successes in this period-Kepler's theory of vision and HaiVey's theory of
the circulation of the blood-were arrived at independently of mechanistic conceptions.
Descartes's theory of the action of the heart, on the other hand, and the animal spirits invoked
in his psychology of perception were soon recognized to be mistaken.

Authenticated
[140] Vicaria Dei
a thing is only of the form," which exists in its entirety in matter beforehand.
Until they are generated, such forms "are not apparent from outside on
account of the indistinction which they possess in matter." Generation is
simply the manifestation of a form that is otherwise not different before
than after (Nifo Disp. 7disp 11 c3; tg6a). Similarly Augustine argues that "for
all things that are brought before our eyes, there are certain seeds con
cealed in the elements of the world, through which those [forms] that were
hidden come out into the open" (Coimbra In Phys. tcgq12a1, I:t8g).24
The primary argument on behalf of latent forms was quite simple. All
philosophers accept the principle that nothing comes from nothing. Form,
as we have seen, is a thing distinct from matter. Hence "either it is some
thing prior to generation, or it is nothing" (Suarez Disp. 15§2'l[ 1, opera
25:505; cf. Coimbra In Phys. tcgq12at, t:t8g). If it is nothing prior to
generation, then the principle is violated. If it is something, it cannot be
matter itself. Indeed it cannot be other than form. But that is impossible,
''whether because otherwise there would be no substantial generation"
since in effect the substance to be generated already exists-"or because
otheiWise contrary forms would exist together in matter. "25
One could avert the last objection by supposing that the forms latent in
matter are "inchoate" or "imperfect." Averroes, according to Nifo, held that
in matter, privation was something more than the mere potentia passiva to
form. But it cannot quite be form, since if matter already had a perfected
form, it would not be fit to receive all forms. The forms that preexist in
matter are therefore not merely hidden, as Anaxagoras and Augustine
thought, but "imperfect" (Nifo Disp. 7disp11c3; 197b; cf. Averroes In Phys.
1com7g,66,62; opera 4=4sC,4oi,37H).26
24. Cf. Augustine De trinitate 3c7, p14o: "Omnium quippe rerum qme corporaliter
uisibiliterque nascu.ntur occulta qmedam semina in istis corporeis mundi huius elementis
latent." Boyle found the maxim "not comprehensible" except if understood in some such way
(Papers 55). But he does not therefore agree that there must be preexisting forms. Instead he
argues that what the Aristotelians call "generation" is only alteration or the production of
subtle from gross matter. Maignan, in his question on the eduction of form, concludes that
matter, understood as pure potentia, is a "mere figment" in physics, and that the "matter" of
generation consists in already formed elements (Cursus, Philosophia: natura: 1pr5, p142).
25. So Zabarella, after Avicenna: "Unless this form [of matter as such] is admitted, there
will be a creation ex nihilo, namely, what is usually called generation will be creation [...] The
form which is produced is produced as a whole [tota producitur], without any part existing
beforehand [nulla sui parte prcesupposita]. But what is produced as a whole, without any part
existing beforehand, is said to be created. Therefore form is created" (De rebus nat. 207B). See
also Toletus In Phys. u8qr 5, opera 4:36ra.
26. The same view was ascribed to Albert (In phys. 1tr3c3&16; opera [Inst.] 42b,7oa). For
an extensive discussion of his views and the responses to them, see Nardi 1g6o, c. 2. Soncinas
(Q. meta. 7q28, 15gb) and Smirez (25:506) deny that by "inchoatio forma:" and similar phrases
he meant anything other than that matter is in potentia to form (cf. Nardi 1g6o:87-93). One
must admit, however, that Albert's manner of speaking is suggestive. Nardi argues that Albert
did believe in an inchoatio formce distinct from the potentia of matter (g2IT).

Authenticated
The problem such arguments raise is not entirely obsolete. A similar

dynamic operates in Thomas Nagel's argument for panpsychism. Call an
emergent property any property of a composite whole which is not
reducible-by whatever means count as reductive-to the properties of its
parts. Nagel argues that mental properties are not emergent. On the other
hand, they cannot come from nothing. Something like them must be in
choately present in the building blocks of matter: Nagel speaks of "proto
mental" properties. The Aristotelians would agree that souls cannot come
from nothing. But for them having a soul is an emergent property, distinct
from the disposition of the underlying matter. Their conclusion, as we will
see, is not that matter is endowed with a protosoul, but that the rational soul
is not educed from matter. The mental does not seep up from below; it slips
down from above.
Any solution to the dilemma had to meet three conditions. First, it must
show that there are no actual preexisting forms in matter. Second, it had to
somehow interpret the maxim forma educi a materia so as to make it come out
true. Third, it had to take account of those forms that were believed not to be
educed from matter: the rational soul, as I have just mentioned, artificial
forms, and grace (which comes to us not out of our nature but by a super
natural act of God).
To the first point Suarez argues that on pain of supposing an infinite
number of mutually repugnant forms in prime matter, one cannot take the
supposed preexisting forms to be just as they are in generated substances. If
they are there, they must be, as Averroes thought, somehow imperfect.
Generation, Suarez supposes, would then be something like an increase in
intensity. But that will not work. Substantial forms do not admit of intension
or remission. There would, moreover, be an "infinity of actual entities" in
matter, for though they were imperfect, the "remitted" forms would not be
merely possible. The higher grade of intensity of the generated form would
have to be added by the agent, and so the same question arises for it (Suarez
Disp. 15§216, opera 25:507). The forms educed from matter, therefore, do
not exist in it in any way.
The problem now is to reconcile the fact of generation with the principle
that nothing comes from nothing. 'Eduction' is defined thus: ''We say that
for a form to be said to be educed from the potentia of a subject, two
conditions are required and suffice: the first is that there should be in the
subject a natural potentia to it, since otherwise the eduction will not be
natural; the other [condition] is that what is educed can neither be brought
about nor endure without the aid [sine adminiculo] of such a subject; which
[condition] they call 'depending on matter in becoming and being con
seiVed [pendere in fieri & conseroari amateria]'" (Coimbra In Phys. 1cgq 12a5;
1:1gs; cf.John of St. Thomas Nat. phil. pt1q4a1, Cursus 2:85; Suarez Disp.

Authenticated
[142] Vicaria Dei
15§2,15, opera 25:510). Grace and artificial forms fail the first condition,
since matter has no natural potentia to them; the rational soul fails the
second, since it can be produced and endure without the body. As for the
principle, if it is taken strictly, so that to come from nothing is "to come as a
whole, that is without presupposing any of the parts," then substantial form
does not come from nothing. It comes from matter, or rather from "a matter
disposed and determined [ dispositam determinatamque] toward forms of its
sort" (Fonseca In meta. 5c2q4§2, 2:goF). It is, in short, generated, not
created.27
Even if the solution to the dilemma is sound, it is incomplete. Substantial
form is really distinct from matter; it is likewise really distinct from whatever
accidental forms may naturally accompany it in a complete substance. Since
the progenitor need not perish in reproducing itself, the form of the prog
eny cannot be numerically identical to that of the progenitor. That new
existence, then, is left unaccounted for, even if we distinguish generation
from creation.
The question 'Where does the progeny's form come from?' can be under
stood in several ways. One concerns the existence of the individual: 'How
did the matter of this individual come to be determined (in some way or
other)?' The appropriate answer is that such and such an agent acted on the
matter. Another concerns the kind of thing the individual has turned out to
be: 'How did the matter come to be determined thus?' Once the individual
exists, the form itself is the answer to that question. But I am asking about
the cause of the form, which is, as I have noted, really distinct from, and not
already in, the matter of that individual.
One answer, favored by Boyle, which would still be favored, is that the
disposition is the cause of the form, or rather that the form is nothing other
than the matter so disposed. The Aristotelians, however, denied that sub
stantial form could be thus reduced to accidental forms. A second answer is
that of Avicenna. There is a celestial intelligence, "by which the forms of
natural things are induced in matter disposed beforehand by a physical
agent," and which is therefore called the dator or datrix Jormarum. 28 That
27. "Generatio enim est ex aliquo pnesupposito, unde non facit totum ens, quod gener
atur, sed materiam perficit: at vero creatio est productio ipsius rei totius: creare ergo equum,
est totam equi substantiam de novo facere: "Generation is from something given beforehand,
and so it does not bring about the entire being of the thing generated, but perfects its matter.
Creation, on the other hand, is the production of the whole thing; to create a horse, therefore,
is to bring about for the first time the entire substance of the horse" (Toletus In Phys. tcgqtg,
opera 4:41vb). As a motus, generation presupposes not only a terminus ad quem-the thing
generated-but a terminus a quo, that from which the thing is generated. As an action, it
presupposes an object acted on, namely the matter of the terminus a quo. Creation has only a
terminus ad quem and requires no object.
28. For a,_ sketch of Avicenna's cosmology, see Fonseca In meta. 5c2q8§ 1, 2: 122crr. There
the datrix omnium formarum substantialium is identified with the motrix intelligentia of the

Authenticated
answer too was rejected. Since "every agent produces what is similar to
itself," only a composite substance can cause another composite substance.
God is the sole exception; in him alone all form and matter exist virtually.
Because "natural causes are determined by nature to certain actions," and
those actions to effects peculiar to them, even angels and demons can
introduce forms into matter only by bringing about the conditions under
which natural causes will do what comes naturally (Coimbra In Phys.
1cgq12a3, 1:Ig0-191; cf. Fonseca In meta. 7c7q2§2, 3:251E).
What is primarily generated, the Aristotelians emphasize, is not form or
matter alone but the composite. Toletus writes that "form is not produced,
or made, but the composite, in whose production form is co-produced and
co-made: just as when someone bends a staff, he does not produce cmved
ness, but makes [something] curved, that is, when he makes the wood such,
curvedness is co-produced. "29 The human form, say, is not the primary
terminus of human reproduction: what is reproduced per se is not that form
but a human being, composed of soul and body. The form is indeed made,
but per accidens, as a by-product so to speak of the making of the composite.
But the composite too is a new existence. To that objection Toletus responds
that one cannot deny genuine activity to particular causes3o, and so one
must ascribe to them a ''wonderful virtue," by which they participate in the
creative power of God. That, like the similarly wonderful virtue of matter to
have such forms drawn out of it, "are not to be denied because they are
WOnderful" (/n de Gen. 1c3q2, opera 5:254vb).
We are left-as we should be, no doubt-with material substances them
selves as the datrices Jormarum. But though Abel's form came from Adam,
Adam's form, like his matter, must have come from God. In the beginning
there was a datorJormarum, that was also and at the same time a dator materia!.
It has often been remarked that Christian Aristotelianism incorporated
elements of Plato's natural philosophy. One need not look far for them:
Avicenna's intelligences are not rejected entirely, for example, though their
role is restricted. But in the Aristotelian world there is, barring miracles,
lunar sphere; "natural things (even the celestial orbs), since they act by motus, can do nothing
other than dispose matter in order that this last inteUigentia [i.e., the lunar] should introduce
substantial forms into it" (122F).
29. "Magis placet solutio A':gidii [...] quod non producitur forma, nee fit, sed com
positum, ad cuius productionem comproducitur, & confit ipsa forma: sicut qui virgam incur
vat, non facit cmvitatem, sed facit curvum, id est, dum lignum tale facit, cmvitatem
compr~ducit" (Toletus In de Gen. 1C3q2, opera s:254Vb; cf. Suarez Disp. 15§21 15, opera
25:511; Thomas ST 1qgoa2). That what is made is not the form but the composite is part of
Aristotle's argument against Platonic ideas (Aristotle Meta. Sq, 1034a25rr).
30. Particular causes, which include all sublunary agents, are contrasted with universal
causes, the heavenly substances that concur in earthly generation and that introduce substantial
forms in cases where the sublunary agent is absent, as in the equivocalReneration offrogs and the
like from putrefying mud (see, e.g., Fonseca In meta. 7c7q2, 3:250 , or virtually any De anima
commentary).

Authenticated
Vicaria Dei
exactly one "donation offorms" to the world. From that moment on, nature
acts pro deo, by its own power though in accordance with God's wi11.31
We see here an exacting economy of the world's dependence on God.
The Coimbrans write that Aristotle occupies a middle place between two
extremes. Some, like Anaxagoras and the atomists, put preexisting forms in
matter; some, like Avicenna, deny to sublunary things the wonderful virtue
of making forms and would have them come always from above. One side
threatens to make God superfluous by attributing to matter all it needs to
exist; the other robs Nature, as the Coimbrans put it, of "her most noble
operation"-the generation of substances (Coimbra In Phys. Icgq12a4,
1: 192). In Aristotelianism the world is neither so independent as to do
without its Creator, nor so destitute as to require his intervention in the
operations proper to it.
2. Accidents prior to form. Not all the accidents that invariably accompany a
substantial form are conditions for its reception. Some truly "prepare and
adapt" matter to form, some merely "embellish and perfect" the form
(Suarez Disp. 14§3'128, opera 25:480). Among the latter are the active
powers that, as we will see in §5.4, are said to "emanate" from form. It is
reasonable to believe that such accidents inhere in matter only by way of
form. Heat and lightness ( levitas) are both invariable concomitants of fire.
Yet although heat is necessary to its production, lightness might well be a
mere consequence of the form. Heat is both an active power of fire and a
preparatory accident in its generation; lightness is not. In more complex
substances, and especially in animals, the active powers that operate in
generation cannot be reduced to elemental qualities. There the forms that
follow upon substantial form, and those that prepare the way to it, become
increasingly dissimilar.
Embellishing accidents, then, inhere in matter by way of form; but what
about those that prepare matter for it? The Thomists argued that they too
inhere in matter only by way of form. Since in generation the terminus ad
quem is a new substantial form, and since there cannot be more than one
form in each substance, the old one must yield to the new. As the slogan has
it, the generation of one is the corruption of another (Aristotle De gen. 1c3,
31ga6). But then-Thomas and his followers argued-none of the acci
dents of the old substance, including its quantity, can be numerically identi
cal with any accident of the new substance. Substance must be "resolved"
into prime matter before it is resurrected.
31. Nature depends on God in esse, since God's conseiVing power is necessary to the
existence ofall created beings. Nature depends on God in fieri only insofar as God must concur
in the actions of natural agents. On nature as prrrdea, see Alain de Lille Anticlaudianus 2:69-74;
Speer 1991: 11 1 ff; and §7 .1 below.

Authenticated
The Thomists and their opponents32 agree that the persistence of acci
dents through substantial change would be a good reason to believe that
they inhere in matter alone. They also agree, by and large, on the experi
menta that I will turn to in a moment. Indeed Fonseca rests his defense of the
Thomist view entirely on metaphysical arguments and finds himself obliged
to explain away, or dismiss, the appearances. The real issue must lie else
where.33 The positive arguments for their position reveal that the Jundamen
tum is a sheaf of mutually supporting propositions already encountered in
arguments about the essence of matter (§4.1):
(i) The subject of inhesion of all accidents is the composite of matter

and substantial form (see John of St. Thomas Nat. phil. 3q1a6,
3qga1, Cursus 2:582,755).
(ii) Matter is first of all and per se completed by and united with substan
tial form (ib. 583).
(iii) Matter is pura potentia, having no proper existence or actuality.
All existence, in short, comes from substantial form, and only that which
exists can have accidents. Matter, "since it is pura potentia in every genus of
beings, cannot of itself alone give birth, as it were, to quantity, which is an
actual being, except when at the same time it is joined with substantial
form."34
We have seen that the doctrine is troublesome, even paradoxical, where it
concerns existence. Now we will see that it is troublesome in the theory of
generation and corruption. Suarez, in a section of unusual length even by
his standards, establishes first of all that matter has indeed "sufficient being
[entitatem] so as to sustain this or that accident" (Suarez Disp. 14§3, 12,
opera 25:474; cf. §4-2). Even if its subsistence in a complete substance is
only incomplete or partial, that subsistence will do for "accidents propor
32. Thomas (ST1q76a6, In de Gen. 1lect10), Albert, Durandus, and the usual Thomists are
lined up in favor of resolution into prime matter (Fonseca In meta. 8e1q1§3, 3:453bF); Sim
plicius, Philoponus, Averroes, Avicenna, Henry of Ghent, Ockham, and "most of the nominal
ists," Gregory ofRimini, Gabriel Biel, and Augustino Nifo against it (Fonseca 3:453aE, Suarez
Disp. 14§3,10, opera 25:474).
33· Theological issues weigh lightly on the dispute. Suarez mentions in passing that his
position is "favorabiliorem (ut sic Ioquar)" to explaining the Eucharist and justifYing the
veneration of relics. That may have been a touchy point. If no accident of the living person is
identical to any accident of her remains, the bones of a saint and the saint herself share only
their prime matter; but that seems unworthy of veneration (cf. Fonseca In meta. 8e1q1§2,
3:451aC). If mere similarity of accidents suffices, an exact replica ofa saint's scapula, though it
belong to Aretino, is as venerable as the saint's.
34· "Materia enim, cum sit pura potentia in toto genere entium, non potest ex se sola quasi
parturire quantitatem, qme est ens actuale, sed simul copulata cum forma substantiali,
quatenus ea forma dat esse corporei in genere substanti<e" (Fonseca In meta. 8e1q2§3,
3:4s6bB).

Authenticated
Vicaria Dei
tioned to it," just as the subsistence, though partial, of the rational soul in a
human being will do for its volitions and habits.
That much establishes the possibility of accidents inhering in matter
independent ofform. The more interesting part of the argument consists in
showing, by appeal to the experimenta I mentioned a moment ago, that in
fact they do.
Start with what is common property. Quantity answers to matter as quality
to form. It is, as the Coimbrans put it, "the first disposition of matter,"
necessary to the reception both of substantial and other accidental forms.
For the latter it seJVes as a kind of secondary subject, so that not only the
substance itself but its color and heat are said to have size and figure.
Though the central texts deny that quantity is absolutely necessary to matter
(§4.2), they agree that it is naturally necessary (Coimbra In Phys. tqqt,
1:105 ).
All this suggests a structure for material substance in which prime matter
together with quantity, and perhaps certain other dispositions, make up a
proximate matter ordered to and receptive of substantial form. Toletus,
concluding that in corruption "there does not always occur a resolution
immediately into prime matter," argues that in many instances we can
distinguish two sorts of accidents: those that are common to the old and new
substances, and those that are proper to one. Clearly what is proper to the
corrupted substance alone will not suiVive the change. But for those which
are "common in some way, and which have no contrary in the generated
substance," nothing opposes their remaining numerically the same in both
(Toletus In de Gen. tqq7, Opera 5:262va). The clear implication is that such
accidents must have matter alone as their subject, since (as all Aristotelians
agree) accidents cannot "migrate" from one subject to another. Of the two
incomplete substances that constitute the corrupted complete substance,
only the matter persists; it alone could be the subject of accidents common
to both the old and the new substance.
Experience shows that certain accidents of the old substance have exactly
similar counterparts an instant later in the new one: "We experience that in
the cadaver of a human just after its death the same accidents remain which
were in the living human, except those faculties which are proper to living
things, and which therefore effectively depend on the soul" (Suarez Disp.
14§3~20, Opera 25:477). One may reply, as Fonseca does, that exact sim
ilarity does not prove identity, and that where reason shows that something
cannot be so, the evidence of the senses is of no avail. 3 5 Suarez answers that
35· "Not by sense but by reason should one examine the numerical identity and diversity of
things; as when someone sees two exactly similar eggs from the same hen: he would not be able
to distinguish them numerically from each other, but someone who knows that the hen laid
only two eggs, and has eaten the one that the other person saw first, will demonstrate to him

Authenticated
the reply "could indeed be admitted, if some reason intervened that would
effectively compel us to correct our senses" and also exhibit a cause for the
new accidents in the generated substance. But the metaphysical arguments
are not compelling, and in many instances no efficient cause offers itself.
The heat, for example, that remains in a cadaver certainly does not result
from the form of the cadaver, since it quickly disappears. Nor could any
external agent induce such heat, "especially when death comes about vio
lently by suffocation, strangulation, etc." The particular cause in such in
stances certainly cannot, and the universal causes that are said to educe the
forma cadaveris from the matter of the victim have no reason to do so (Suarez
Disp. 14§3!21, opera 25:478).
The forma cadaveris with its scars and dwindling heat would occupy a
useful though morbid chapter in the history of Aristotelianism. Here the
point is that the Thomist position requires that many accidents of the old
substance be exactly duplicated in the new. Some authors argue that this
already offends against Ockham's razor. Worse are cases where from the new
form accidents contrary to the duplicates would follow, since it cannot then
be their cause. A cadaver is naturally cold; its form would not of itself
produce heat. The Thomist position requires also that the quantity of the
old substance be produced anew. Typically they say it is introduced at the
exact instant in which the new form is introduced and the old quantity
perishes. Though Suarez admits that the claim cannot be refuted, he con
cludes that it is "more philosophical and more in accordance with corporeal
agents" to suppose that every action supposes "a quantified and divisible
subject, distinct from others not only in being but in quantity [non solum
entitative, sed etiam quantitative]" (i 25, opera 25:480). More generally, since
prime matter is in potentia indifferent to all forms, its taking on the forma
cadaveris, say, rather than some other "requires that it should be accommo
dated and disposed by the agent to such a form; but this accommodation
and disposition occurs only through previously existing accidental disposi
tions," including the quantity that supplies extension to all the rest.36
that they are diverse." Likewise the sun and moon appear to have the same size, yet reason
shows that the sun is "five hundred" times larger. Hence even if the scars of a corpse appear to
be numerically identical to those of the living person, "reason, which demonstrates the op
posite, is to be believed, since to its inward tribunal belong that [question]. and [all] the
insensible numerical differences of things that are very similar" (Fonseca In meta. 8e1q2§6,
3:458bC). This passage, rather unusually, humbles the senses, thereby to exalt reason-a
Cartesian strategy. But it is one of the few instances where a thesis manifestly contrary to sense is
defended.
36. Although there are Thomist replies to these points (see John of St. Thomas Nat. phil.
3q 1a6 for explicit replies to Suarez), it would be tedious to examine them. Toletus, I should
note, attempts a rather uneasy accommodation of the two views (one that Fonseca rejects). He
also excepts quantity from among the accidents that survive through transmutation, though his
argument othelwise resembles that of Suarez. Cf. Toletus In de Gen. 1qq7, opera 5:263-264;
Fonseca In meta. 8e1q2§4, 3:454.

Authenticated
[q8] Vicaria Dei
Since quantity, at least, persists through all substantial change, it is, for
reasons already mentioned, reasonable to believe that it inheres imme
diately in matter rather than in the composite. A leading special case is that
of the human being, since the soul is a spiritual thing. Like all spirits, it is
incapable of receiving quantity. But it is also incapable of being naturally
joined with matter except if that matter has quantity. The quantity of a
human being must inhere in the matter alone.37
The conclusion can be generalized. The union of a particular quantity
with its matter can persist even when whatever union it has with substantial
form perishes. The quantity of Caesar's body is united, we may suppose,
both with his matter and with his soul. When Caesar was murdered, that
quantity was still united to the same matter, but its union with the soul
departs when the soul does, and it is then united with the forma cadaveris.
Hence the union of quantity with matter and its union with form are dis
tinct. Suarez argues that in all material substances other than the human
"there intervenes no special partial union between form and quantity"
(146- 48, opera 2 5:488ff). Indeed in all material substances, not just hu
man, quantity and form are united only by way of the union of form and
matter, on the one side, and that of matter and quantity, on the other. Form
and quantity are not unum per se, but per accidens: a conclusion that will have
its counterpart in Regius's heterodox Cartesianism.
The differendbetween the Thomists and their opponents can be traced to
a dissension within Aristotle's texts themselves. The Categories treat individ
ual substances as unanalyzable; accidents cannot but inhere in complete
substances. The Metaphysics and the Physics add a new level, matter and form
(see Furth tg88, §g). In Scholasticism that distinction was combined with
the Categories account to yield prime matter, substantial form, and acciden
tal forms.
The subject of inherence in the Categories is a concrete existing individual:
Socrates is white, not his soul or body. In the Metaphysics and the Physics, on
the other hand, it is at least an open question whether forms other than
substantial form inhere in the composite. The Thomist position sticks with
the Categories. From the Metaphysics it takes the distinction of matter and form,
interpreting the pair as prime matter and substantial form. Only the union
of the two yields a proper subject of inherence. Thereafter everything pro
ceeds as in the Categories. The structure is shown in Figure 5· In generation,
the complete substance perishes, and thus the subject of inherence of quan
tity, and so too of all other accidents (I omit those accidents that reside in
37· Suarez Disp. 14§31 1611, opera 25:476«. Descartes expected no trouble from Aristo
telians when he held that the immaterial mind is joined with the material body: far from being
a reason for believing that the mind is extended, the fact that it cannot be is reason for holding
that quantity inheres in the body alone.

Authenticated
Formal
cause
Other Inherence
Substantial Complete accidents
form substance
:nformation Material
cause
matter
Fig. 5· Structure of physical substance (Fonseca)
the rational soul alone). Since quantity cannot naturally exist unless it in
heres in a subject, it too perishes. Everything above the dotted line in Fig. 5
is replaced when the new form is introduced.
In the theory of Suarez and Toletus, a compromise between nominalism
and orthodox Thomism, the unanalyzable first substances of the Categories
are almost entirely abandoned. Preparatory accidents inhere immediately
in matter; those that are common to the old and new substances may
therefore persist, along with those accidents that, although not "con
natural" with the new form, nevertheless are shown to persist by experience.
In particular quantity always inheres immediately in matter. The structure
according to Suarez is shown in Figure 6. Below the line now is all that
pertains to proximate matter. Much of it, and certainly quantity, is preserved
through substantial change. The bones and scars of a dead saint are,just as
the senses would have it, the very same bones and scars as those of the living
saint. The role, evident to experience, of dispositions in promoting the
eduction of the new form, can be explained without difficulty, since the very
same accidents that "proportion" matter to its form in an existing substance
existed in its predecessor.
Suarez's solution (which is, I should note, not unique to him) has the
advantage of not requiring us to deny the evidence of the senses. It has the
disadvantage of raising more acutely the question of unity of complete
substance. We have seen, however, that Smirez believes that the union of
matter and quantity is distinct from the union of matter and form. Hence
matter and form are not in fact united f7y means ofquantity. Indeed form can
be received by a matter lacking quantity altogether (§4.2). So they can still
be unum per se. But matter endowed with quantity can likewise exist

Authenticated
[150] Vicaria Dei
Formal
Complete
substance
cause
Information
Material Preparatory Inherence

cause accidents
Fig. 6. Structure of physical substance (Smirez)
without form. The union of matter with form and its union with quantity are
on an equal footing, a result closer to Ockham than to Thomas.
Matter has in Aristotelian physics three functions. It is the subjectum of
substantial form, the indeterminate being-somehow-or-other that is specified by
a particular somehow. It is the substrate of transmutation. It is that which all
material substances have in common. In the Thomist account, as we see it in
Fonseca andJohn ofSt. Thomas, the subjectumand the substrate are one and
the same: both pura potentia. In fact the identification of the subjectum with
pura potentia virtually forces a like identification of the substrate. Once that is
arrived at, matter cannot actually be said to have those accidents that are
proper to material substances; it can only be said to be in potentia to them.
In Suarez's account the subjectum is accorded a proper existence. It is not
pura potentia. But in agreement with the Thomists he does not take any of
the accidents common to material substances to be essential to matter.
Since God, according to his absolute power, can deprive matter of all those
accidents, what is essential to matter is only that it be in potentia to quantity,
and thus to the occupation of space, to mobility, and so forth. The substrate
of change, on the other hand, does essentially have quantity. In particular
instances it may have many other accidents as well. It can be as complicated
as the human body. Yet Suarez rejects the Scotist plurality of substantial
forms: the proximate matter of a given form is not itself a substance. So we
have not yet arrived at Cartesian body and soul.
Suarez's Disputationes are very nearly contemporary with the Qur.estiones of

Authenticated
Fonseca; Toletus's Physics commentary predates them both. One cannot

speak unproblematically of "development." But in light of what is to come,
two tendencies may be discerned in Suarez's theory of matter. The first is
that accidents, divided into the dispositiones that precede form and the
powers that follow it, are being allotted different subjects of inhesion.
Though the powers are said to inhere in the composite, the parallel with
dispositiones, which inhere in matter alone, might incline one to regard them
as inhering in the form alone. We would then have a substance with two sets
of properties, one attached to matter, one to form-in animate things to the
soul. The parallel with Cartesian dualism need hardly be emphasized; what
will pave the way to it is the reduction of powers to dispositiones, and of
dispositiones to the modes of quantity, and the elevation of all powers of the
soul above the capacities of matter.
The second tendency is toward the reduction not only of powers but of
forms to the dispositiones of their matter. Indeed if one could show that their
so-called powers follow immediately from their dispositiones, the place of
substantial form would become exceedingly precarious. The arguments
outlined in §3.2 would still have to be addressed. But once matter is granted
not only actual existence but a characteristic set of properties, an important
motive behind the introduction of substantial form disappears. So the Tho
mists may well have thought. The dangers were not hard to discern. Matter,
having been given title to the common properties of bodies, might now be
free to declare its independence not only from form, but from God.
3· The notion and uses of 'dispositio'. The semantic field that contains
dispositio merits attention, not least because, as I have said, Descartes takes
over the term to his own ends. In the shortest chapter of the Metaphysics
Aristotle defines dispositio as "the order of that which has parts, either ac
cording to place or potentia or form" (5C19, 1022b1), deriving 8tci9Ems
from 8€ats, meaning 'position'. 3 8 Disposition in place is that of a house
the roof highest, and so forth. Disposition in potentia is illustrated by com
plex virtues like prudence, ''whose first part pertains to deliberation, the
second to judgment, the third to command. "39 Most important for my
38. The Categories define 'dispositio' as a transient, easily mutable habitus or state ( 1c8,
8b25ff). In that sense heat and cold, sickness and health are dispositions, knowledge and
virtue are habitus. Aristotle seems more to be defining a certain respect under which qualities
of various sorts can be regarded than a class of qualities. The Aristotelians tend, on the other
hand, to take 'quality' as a genus of which hahitus/dispositio, potentia/impotentia, and so forth
(§4.3), are distinct species, as is indicated by standard questions on whether the classification is
exhaustive and di~oint. Zabarella restricts the term to animate things (Zabarella opera logica,
Tahul01 logic01 120); Chauvin, though he recognizes the broader sense of the Metaphysics, holds
that the term is "properly" used in the Categories sense.
39· "Formally, if we suppose that the habits of the soul are not simple qualities, but emerge
out of several habits duly coordinated, so may the whole science be called a dispositio, because it
is like a certain order of that which has parts secundum virtutem" (Suarez Disp. 42§3'l[6, opera

Authenticated
Vicaria Dei
purposes is the last, disposition according to form, which Fonseca takes in its
Categories sense of figura. He identifies it with what Aristotle calls the "place"
or "order" of "parts in a whole," which is to say with quantity itself (see
sc 13q1§4, 2:642F).
One might well ask why disposition is not distributed among other
categories-place, quality, quantity, and relation. The answer is that the
"order" of the parts in question is with respect to an end. It can therefore be
applied even to things which have no parts but which "affect the subject for
the worse or the better." Even in those that do have parts, what is foremost is
that they affect the subject for the worse or the better, since only when so
regarded are they qualities.40
Suarez's more extended treatment, well attuned to the range ofthe term
dispositio, emphasizes the orientation to an end. Whatever exhibits order,
ultimately, or is understood as ordered, spatially, temporally, or secundum
vinutem, can be called dispositio, whether it is a habitus of the soul, a passive
quality of the body, or health and beauty. Dispositio thus falls within the
broad ambitus of ordo, which in its highest use denotes the "unifying struc
ture of the multiplicity of being in the Universe," the unitas universi
(Michaud-Quantin 1971 :gs).
In use dispositio is often collocated with proponio, a word the lexica all
define quantitatively. Arguing that prime matter has a natural potentia to
ward the rational soul, Suarez writes that "matter is naturally disposed
[disponitur] ultimately to receive the rational form," and again that "the soul
is an natural actus proportioned [proponionatus] to matter" (Suarez Disp.
15§2t 12, opera 25:509). The Coimbranswrite that form is educed from the
potentia of matter only if "its privation and dispositions preexist in matter,"
and later that "since a potentia necessarily corresponds in proportion [propor
tione] to its actus, matter has a natural potentia only to those forms that are
natural to it." Concerning mixtures, Toletus writes that all parties agree that
"a certain qualitative dispositio is necessary and proximate to the form of the
mixture" (Toletus In de Gen. 1Cl0q17, opera s:303Va). That dispositio, or
complexio, consists in a fixed proponio or temperament of all four elements.
Gathering such contexts together, one sees that like ordo, dispositio is
distributed over a spectrum extending from the de facto arrangement of
parts to states or features unspecified except in relation to an end. To what
extent, then, did an Aristotelian expect, or feel obliged, to produce a
definite characterization of the dispositiones he proposed? Suarez writes that
25:612). Capacities analyzed into other, simpler, capacities are said to have the latter as parts
secundum virtutem.
40. Fonseca In meta. ad 5c1g (text24): "Cum partes aliter ac aliter ordinari possunt, si certo
quodam modo ordinentur, bene se habet res, si opposito, male: ut patet in exercitu, domo,
complexione humorum, in artibus, scientiis, & similibus" (2:921).

Authenticated
any form necessarily requires an "accommodating dispositio from the side of

matter," the necessity being as usual natural rather than absolute (Disp.
15§6ts, opera 25:519). The statement is plausible enough. But what is
being asserted here, other than that the association offorms with proximate
matters is nonarbitrary?
In the metaphysical context, the assertion is intended to remind the
reader of Suarez's earlier arguments against the Thomists. Dispositiones are
necessary conditions for forms to be causes not merely in esse but in fieri. Not
only must they be present in order for a form to exist, they must also
temporally precede its introduction (see Suarez Disp. 14§3t28ff, opera
25:48off). Since dispositiones are accidents, the term also setves as a re
minder that proximate matter does not have a substantial form. The Scotist
would have it that the proximate matter of the rational soul already has a
substantial form, the forma corporeitatis, which persists after death. Suarez
and the other central texts hold instead that what persists are the disposi
tiones, and that the forma cadaveris is introduced at the moment of death.
But a philosopher now would rather know what those dispositions are. It
is one thing to say, for example, that conscious thought must supetvene on
certain physical processes, quite another to say what those processes are.
One may certainly attempt to establish the first without having any idea
about the second. But even committed physicalists, I'm sure, would like to
be able to point a finger at the appropriate part of the brain. The inability to
do so has no doubt been one reason why physicalist arguments, however
cogent in the abstract, have never quite succeeded in carrying the day.
To some extent the impression that appeals to powers or dispositiones are
vacuous rests on ignorance of other works in the Aristotelian canon. 41
41. That ignorance is not without basis. By far the greater number of commentaries were
devoted to the Organon, the Metaphysics, the Physics, De generatione, and De anima. In the statistics
gathered by Blum from Lohr's bibliography of commentaries published between 1500 and
1650 (Blum tg88), the works rank as follows (the percentages indicate the proportion written
by members of monastic orders): Organon: 1165 (42%); Physics: 1103 (4o%); De anima: (603
(43%); Metaphysics: (463) 51%); De gen. et corr.: 286 (s8%); De ca!lo: 247 (48%); Meteorologia:
241 (42%); Paroa naturalia: (113 (25%); Gen. anim., Hist. anim., Intr. anim., Part. anim.: 25
(4%).
Though the curriculum undoubtedly had the major part in determining this distribution,
the relatively small numbers of commentaries even on De gen. et corr. and De ca!lo, together with
the near absence of commentaries on the animal books, which exhibit Aristotle at his most
empirical, indicate a certain lack of interest and (no doubt) aptitude for their subjects. (I
should note that members of medical faculties, who did not write Aristotle commentaries, also
studied plants and animals, and that there were commentaries on Theophrastus's De plantis
and Pliny's Naturalia.)
The cursus tell a more favorable story: in them, natural philosophy tended to grow at the
expense especially of metaphysics, and to incorporate varying amounts of new information
(see Schmitt in Schmitt and Skinner tg88:8o2 1f). Nevertheless, it must be said that despite
recent attempts to rehabilitate the Aristotelians, the impression given by the corpus en masse is
that the character attributed to the textbooks by their enemies-the absence ofexperiments as
opposed to authorities or commonsense generalities, the continued reliance on ancient and

Authenticated
[1541 Vicaria Dei
Physics and Metaphysics commentaries are sparing in details about particu

lar substances. But it is not their task to study them. The first descent to
particulars comes with the theory of the elements in De generatione. There we
find that among the necessary conditions for receiving the form of fire is
heat to the highest degree (summa calor), and similarly for the other ele
ments (Toletus In de Gen. 3q5, opera 5:315ff). In De anima commentaries we
are told, for example, that the brain must be humid enough to receive
impressions, and yet dry enough to retain them. Such claims may be mis
taken or primitive, but they are attempts to describe more precisely the
dispositions underlying various powers or forms.
An Aristotelian will not accept offhand just any inference to a disposition
or power. Sometimes Ockham's razor is invoked, sometimes the absence of
any conceivable efficient cause. What is typically not done is to demand an
analysis of the power in question, and, if an analysis is not forthcoming, to
reject it. There seems to be, in other words, no expectation that the conclu
sion that such and such disposition must be supposed will be followed up by
a research program that would lay bare its underlying conditions or break it
down into simpler component dispositions. Even when experience is
brought to bear on such problems, the facts adduced are rarely new. Almost
never can one infer that a fact was witnessed with the particular intention of
resolving a disputed question, or that the person adducing them witnessed
them. It is far better, in fact, for the generalization to be taken from an
authority, from a text that, by standing the test of time and the acid of
dialectic, has become an auxiliary to common sense.42
Take the question of minima naturalia. Among the dispositiones required
for a particular substantial form was the quantity of its matter, or rather the
upper and lower bounds of the quantity it was naturally compatible with.43
Scholastic treatments-is not unjustified. The Aristotelians who were progressive in relation to
the new science were typically not those who wrote commentaries or cursus. Though much
more work would be needed to bear out these impressions, they do suggest that-as I will
argue shordy-the barriers to innovation were not conceptual but institutional.
42. See Dear 1985:148ff, and Dear 1987. The second essay, on Jesuit astronomy and optics,
argues that the Jesuits, like the Royal Society later, began to stress singular reports whose
veracity was attested by obseiVers who had themselves witnessed them; the "objectivity" of the
obseiVers was assured by their institutional position as researchers. Of the works I have been
citing, the central texts largely antedate the developments brought out by Dear. John of St.
Thomas repudiates them (ignoring Galileo, he cites Thomas's arguments against mountains
on the moon; he maintains that since the heavens are immutable the new stars of 1572 and
1603 were created supernaturally; see Cursus 2:850, 851). Arriaga, aJesuit, and Maignan, a
Minim, were to varying degrees conversant with them (Schmitt 1989, XI:223ff).
43· The question of minima concerned especially the minimum quantity possible for homo
geneous substances, defined as those whose parts (or whose at-least-minimal parts) were all of
the same form as the whole. Emerton argues that developments within Aristotelian theories of
minima and mixtures laid the way for seventeenth-century corpuscularism (Emerton 1984, c.3
and 4). The central texts, it must be noted, though they affirm the existence of minima, show
no movement in that direction.

Authenticated
That would seem to be a question susceptible to empirical test, even if, as it

happens, the means turned out to be beyond those available to seventeenth
century science. Yet the reasons for and against minima, though they do
draw on experience, do not arise from any investigation into that question.
Here are some arguments against:
In these [substances] there is no limit to bigness[...]; therefore [there is

no limit] to smallness, since the reason seems to be the same.
Such a minimum, if there is one, is a quantity, and is therefore divisible: it

can therefore be actually divided, and then would be smaller than the
minimum.
An agent acts successively, and therefore produces the lesser before [it
produces] the greater, and there will never be a minimum without some
thing less being produced before it: therefore there is no limit in these
things.
Here are some arguments for minima:
The actus of active [powers] occurs in a [suitably] disposed patient, since

form is not received except in disposed matter; but of such dispositions
the principal one is quantity; therefore just as the rest are determinate, so
too is quantity.
To the action of an agent a certain quantity is necessary, in the absence of

which the action will not occur; an axe can be so small that it will not cut.
[...] A thing cannot be of a quantity such it cannot operate; for there is
no being without operations. But for operation a certain quantity is neces
sary, with respect to smallness, and so also to being itself. (Toletus In Phys.
1qq 10, opera 4:25rb-va)
The premises of these arguments are general; all are repetitions of state
ments to be found in earlier authorities; none but that of the second argu
ment is a priori. What interests me here is that none of the premises bears
immediately on the question at hand in the direct way that, say, dropping
weights from a tower bears immediately on the question of whether the
velocity of fall is proportional to weight. All of them could well have oc
curred to someone who had never heard of minima.
What could have borne directly on the question? The discovery of Brow
nian motion lay some two centuries ahead; instruments capable of imaging
individual atoms were not available until the 1950s (Crewe 1993). On the
other hand, one consequence of the existence of minima is that mere me
chanical cutting ought sometimes to produce a new substance. Minima are

Authenticated
Vicaria Dei
not atoms; they could, presumably, be deprived of the requisite quantity and
corrupted. Does grinding ever result in transmutation? Could flesh be cut so
finely as to cease to exist? Just how small can an axe be and still work? Such
experiments might not decide the question, but that would itself be
informative.
Aristotelian qucestiones did not give birth to new "experimental" questions,
answers to which would "test" answers to the original question. That particu
lar barrenness distinguishes the new science from the old. The presumption
behind my queries is that one should do things, make things happen with
the intention specifically of resolving a dispute. The textbook authors did
not take it to be part of their brief that they should, in the experimental way,
do anything in particular. This is not to say that new questions never
emerged. John of St. Thomas, having insisted that in generation the acci
dents of the new substance are all numerically distinct from those of the old,
adds a question, not found in other authors, on "how the ultimate disposi
tion is caused, or causes substantial generation" (Nat. phil. 3q1a7, Cursus
2:588). But his new question is simply a continuation of the argument
against Suarez by the same means. It brings no new phenomena to light; it
only shows that familiar generalities can be saved.
The fruitfulness of Aristotelian or Cartesian hypotheses cannot be mea
sured by reference only to the concepts used in them. Descartes's own
postulated mechanisms and configurations met with criticisms not unlike
those raised against powers and qualities. It was as easy, and as informative,
to suppose some mechanism or configuration that causes the phenomena
we wish to explain, as to postulate a new power or quality.44 Aristotelianism,
for its part, need not have been an insuperable obstacle. There were times
when Aristotelians did actively seek new experientia. This was true in the
thirteenth century of Albertus Magnus and Roger Bacon. In Descartes's
time the Jesuit Scheiner made extensive observations of sunspots and par
helic phenomena; Galileo's opponent Grassi emphasized his having wit
nessed himself the experimenta he published; Mersenne's appetite for facts
and his devotion to' experiment preceded his conversion to mechanism. 4 5
Even when their philosophy remained Aristotelian, their questions were no
longer to be answered only through the interpretation and reconciliation of
received experientia.
The problem with Moliere's doctor is not that he appealed to a virtus
dormitiva-which, after all, opium does have-but that the inquiry was sup
posed to end there. At its best, in Aristotle's theory of the soul or the theory
44· See Gabbey 1990, esp. 279-282.

45· See Schmitt 1981, essays 7 and 8; Dear 1987 (Scheiner and Blancan us); on Mersenne's
relation to Aristotelianism, see Lenoble 1943:211-222. On jesuit instruction and research in
mathematics and astronomy, see Dainville 1978, esp. 311-354.

Authenticated
of mixtures, Aristotelian ism did not rest with powers so obviously complex.
That it did not penetrate very deeply had to do, not with the notions of
virtus, dispositio, and potentia themselves, but with the techniques by which
new questions were generated out of old ones.
5-4- Substantial Form and Active Powers

The arguments just laid out, and those of §3.2, suggest that in individual
substances three stages of specification are built upon prime matter. The
first consists in quantity together with preparatory accidents; these inhere
immediately in matter, though not of course without an accompanying
substantial form. The second consists in the form itself, and the third in the
active powers and embellishing accidents of which form is said to be the
unifYing ground. I will here finish the description of the structure of mate
rial substances by examining the causal relation of form to active powers,
and of active powers to dispositions.
Active powers introduce into the matter of the thing generated the
dispositions required by its new form. The new form will in turn effect the
powers and embellishing accidents proper to it, so that the cycle can begin
anew. Thus briefly stated, the account raises a certain doubt. Actions can be
defined in terms of their termini and active powers in terms of the corre
sponding actions. The active powers associated with form, if they pertained
only to generation, would be determined by the dispositions proper to that
form. 46 The form itself has no effects but by means of them and can be
known only by their actions. It too, therefore, would be entirely determined
by its proper dispositions, and so superfluous.
But in fact even the elements have powers with no role in generation.
Though heat alone suffices to generate fire, fire is not only hot but lumi
nous and light (levis). Mixtures and animate things typically have many
powers with no immediate relation to generation: the medicinal virtues of
herbs, the senses of animals, the human intellect. It could be argued that
even such powers are instrumental, if not to generation itself, then to sur
vival. But because they are at best among the remote causes of the disposi
tiones necessary to the reception of form, they cannot be said to be deter
mined by those dispositiones in any obvious way.
In the new science the banishment of substantial forms rested on show
ing, or asserting, that the "actions" ascribed to powers could all be explained
by appealing to dispositiones alone. The word 'disposition' has, in the vocabu
46. On the centrality of the generative among the powers of substance, see Suarez Disp.
18§31! 19, Opera 25:621, quoted below. In De anima Aristotle calls generation the "most natural"
of the soul's powers ( 2c4, 415a26), because it is common to all material forms, animate or not.

Authenticated
Vicaria Dei
lary ofa philosopher like Quine, usurped the role that was played by powers;
dispositions he admits only as stand-ins for eventual reduction to micro
structure. That reduction was already envisioned by Descartes and Boyle, as
we will see. Descartes's mechanisms, Boyle's analogy of key and lock, were
intended not only to show that the phenomena can be saved without put
ting any properties in matter except those it has by virtue of being extended,
but also that nothing mediates between dispositions-or mere arrange
ments of parts-and so-called powers. Form thereby vanishes.
Even within Aristotelianism, the mediating role of substantial form called
for clarification. Active powers were supposed to be grounded in it, disposi
tions somehow caused by it. The first relation, which was called emanatio or
resultatio, did not, oddly enough, attract much attention. 4 7 Of the authors
studied here, Suarez and John of St. Thomas alone discuss it explicitly. The
neglect is odd in part because substantial forms were held to act only
through their active powers. Indeed, as we have seen, demonstrations of
their existence relied not on showing that they had immediate sensible
effects, but on the necessity of a unifying ground of the powers that did have
such effects. The indirectness of their relation to the phenomena proved to
be a point d'appui for skeptics, who were happy to declare them inscrutable.
I will look at the Thomist position against which Suarez argues, and Suarez's
arguments for his own.
The second point touches on the activity of created substances. The
central texts agree that they are effective, but only within definite limits.
Unlike the first cause, God, they can in no way create complete substances,
matter and form. They can, on the other hand, generate their like in exist
ing matter; on that both common sense and the authorities agree. But
substantial forms were believed to act only through their powers. Active
powers are accidents and thus inferior to forms. Since in general the less
perfect cannot bring about the more perfect, accidents should not be able
to produce substantial forms, nor should substantial forms themselves be
able to. Experience and theory thus seem at odds. In looking at the role of
substantial forms in generation, I will examine how the clash was mitigated.
1. The emanation ofpowers. Thomas, having proved that the powers of the
soul are distinct from its essence, asks whether they "flow" from its essence.
His answer is that they do, since each subject is "the cause of its proper
accidents." The powers of the soul are actus of its potentia, and can therefore
be called its accidents, and so are caused by it. They are not, indeed, the
4 7. In De anima commentaries the relation of the soul to its powers is a standard question.
See for example the questions from Toletus and Suarez cited below. The presuppositions of
the general question, and in particular the distinction between form and powers, were not
shared by all Aristotelians. The controversy over dimanatio occurred largely among Thomists
and their closest opponents.

Authenticated
effects of any motus in the soul, properly speaking, since that would imply
the soul's acting on itself. But they flow from it "by a certain natural result
[resultatio], as color from light" (Thomas ST 1q77a6, Parma 1:302).
It is easy to see how form could, by analogy with its own relation to matter,
be called the material cause of active powers. Some philosophers believed it
was that and nothing more. But Toletus, like Thomas, holds that the soul is
not only the material cause of its powers, but also "quasi efficiens, & quasi
formalis," not, certainly, by effecting them as it itself is by its generator, but as
a subjectum determines its own accidents, which depend on it and are or
dered by it (Toletus In de An. 2qq10, opera 4:7ovb). (Prime matter, of
course, is a subjectum too. But unlike form, it is, as we have seen, in potentia
indifferent to its accidents. Though they "depend" on it as a subject of
inherence, they are not "ordered" by it. Ordering is the job of form.)
The vagueness of the language-Thomas's resultatio, Toletus's quasi
efficiens-points to a puzzle that the mere coining of terms cannot solve.
Either resultatio is an action ofwhich the soul is an efficient cause or it is not,
and the soul is at most a material cause. John of St. Thomas argues that
emanation is an action, not of the form, but of the generator. 4 8 It is indis
tinguishable from the action by which the form itself is educed from matter.
What gives form, gives its consequences. 49 The terminus of generation is not
a "bare substance," but a substance equipped with whatever it needs for its
proper operations. Every genuine action, moreover, "requires an operative
virtue," that must be distinct from substantial form, "since othelWise the
whole ground for distinguishing operative powers from substance is over
turned." Form, properly speaking, has no actions at all, whether on itself or
on other substances.
Emanation is an action ofform only in the way that unlocking a door is an
action of the key that turns the lock. When an agent produces one thing by
way of producing another thing of which the first is a consequence, the
second "is said to behave actively [active se habere]" (John of St. Thomas
Cursus 26gb, citing Thomas Summa logica 1c6, Parma 17:5gb). A generator
produces the powers of its offspring by producing its form, and only in that
sense does the form "act" to produce its own powers.
Analogy has its perils. It cannot be that no intermediate link in any causal
chain genuinely acts. OthelWise parents wouldn't genuinely act in produc
ing their parents' grandchildren. What is peculiar to the instance at hand is
that form, in effecting the accidents proper to it, would be acting on itself.
48. Cited in support of the view are Caietanus, Ferrarius, Soto, and the Carmelite Cursus
Complutensis. The Scotists hold that "those passions which are identified with substance, like
active powers, do not emanate from substance, but come to be through the act of generation
itself. But those that are distinct from substance, like quantity[ ... ] emanate from substance by
a genuine action" (John of St. Thomas Nat. phil. 1q12, Appendix,Cursus 2:268a).
49· Cursus 2:268b; cf. Thomas In de Ca!lo 3lect7, Parma 19:16ob.

Authenticated
[160] Vicaria Dei
Though John of St. Thomas grants that it "aids and guides" the production
of those accidents, the initiator must be something else.
Suarez takes the second way. Accidents that are "immediately connected"
with form, as the intellect to the soul, "are caused by substance, not only
materially and formally, but also efficiently by natural result [effective per
naturalem resultantiam]" ( 18§3'l[6, 25:617). Resultantia, he admits, cannot be
a full-fledged action: actions are accidents too, and so there would be a
regress of accidents. By way of making the notion more precise, Suarez
argues that it applies only to those accidents that are immediately con
nected with form. The intellect is connected thus with the soul, but the
figure of the body results by way of quantity. More generally, all those
dispositions which precede form in generation, and which, as we have seen,
may persist through substantial change, are caused by form only by way of its
active powers.
The strongest argument for supposing that resultantia or emanation is
distinct from any action of the generator is that emanation could be "sus
pended" by God, so that, for example, a soul could be generated that lacked
an intellect. Nor is resultantia the mere copresence ofform and power in one
substance. God could not only prevent the intellect from proceeding out of
a soul, he could also at the same time immediately produce an intellect in it
('l[13, 25:61g). Only if having an intellect were a logical consequence of
having a soul would God be unable to do so. 50 The making of a fully
equipped soul, Suarez concludes, requires not one action, but two: creation
properly speaking, whose terminus is substance, and resultantia, whose ter
minus is the accidents that complete the substance (ib., see also De anima
2c3'l[10, opera 3:582).
Even in less exotic circumstances, emanation may occur later than gener
ation. Water condensed from hot vapor is only gradually restored to its
pristine coldness; a stone may be prevented from falling until long after its
maker has vanished ('l[ 7' 25:617). John of St. Thomas replies that "in the
form produced by generation [the action of] generation remains virtually,
and when the impediment is removed it is again expressed in a passion and
emanation" ( Cursus 2:2 70). This is part of his namesake's explication of the
slogan 'Everything that is moved is moved by another': the falling stone does
not move itself, but is moved by the generator, whose action may continue to
exist even when it itself is destroyed (see Cursus 2:465).5 1 When impedi
ments are present, part of that action may be delayed, just as the action of a
50. "Nam finge Deum creare substantiam anim<e impediendo emanationem potenti<e ab
ilia [...]; maneret ergo tunc substantia anim<e absque intellectu et voluntate auferat vero
postea Deus impedimentum, reliquatque animem su<e natur<e, certe dimanarent ab ilia intel
lectus, atque voluntas, sicut in primo instanti generationis, aut creationis" (Suarez De anima
2C3,10, opera 3:582; Cr. also 2Cl,7, 574).
51. On Thomas's understanding of the slogan, see Weisheipl 1985, c.4, esp. gorr.

Authenticated
testator may be delayed by inheritors' lawsuits. Where Suarez sees two

actions-generation and emanation-John of St. Thomas sees one.52
If the issue seems abstruse, it is because we moderns, like the nominalists,
don't believe that form and powers are really distinct. 53 The conclusion can
be generalized. So long as powers are understood in terms of the physically
necessary consequences of structure, they will "emanate" from it as a matter
of course. Descartes's machines, as we will see, illustrate just such an under
standing. In his world the issue that set Suarez and the Thomists at odds
cannot arise.
2. Forms and powers in generation. Generation begins with corruption: since
nothing can have two substantial forms at once, the matter of the new thing
must be released from its old form. When water is changed to air its heat
must first be raised above the latitude permitted by the form of water. That
alteration at once destroys the form of water and makes its matter ready for
the form of air. The moment that the work of the active powers is complete,
the new substantial form emerges. Generation is completed by the emana
tion in the new substance of its active powers and other embellishing
accidents.
Such is the gross anatomy of generation. Both substantial form and active
powers appear to have some part in the process. But how is their labor to be
divided? Can accidents by themselves "attain to the production of sub
stance"?54 In their version of the standard question, the Coimbrans first lay
out what they call the two extreme views. One, which they attribute to
"certain recent Philosophers," is that "the substantial form of the generator
lends no active influx to generation or to other transitive actions; accidents
alone, by their own force [virtute], and as principal causes, generate sub
stance" (Coimbra InPhys. 2c7q 18a1, 1:293) .55 The other, that ofScotus and
the Nominalists, is that "accidents attain to the production of substance
52. Elsewhere Suarez grants that emanation continues and completes the act of genera
tion, and that the soul, as the efficient cause of powers, is the instrument of the generator
(Suarez Disp. 18§7, 1o, opera 25:633). But since he believes that instruments can have genuine
efficacy of their own (§7.3 below), there is no contradition.
53· The (real) identity of the mind and its powers follows, for example, if the mind is
conceived to be a computer program. Hilary Putnam and Martha Nussbaum have argued that
Aristotle's conception of the soul is functionalist, and so-according to the predominant
brand offunctionalism-analogous to a program. Mary Louise Gill compares substantial form
to a "list of instructions that determines the materials and tools needed to realize a particular
end" (Gill 1991:251).
54· Even though the question is stated generally in terms of substance and accidents, the
only relevant cases involve substantial form and active powers. Matter and quantity cannot
produce substance because they are entirely passive (Fonseca In meta. 5c2q6§ 1, 2:98C; Suarez
Disp. 18§2,3, Opera 25:599). Matter considered as substance can neither be generated nor
corrupted.
55· The recentiores include Albert of Saxony and John Mair, a Parisian philosopher who
flourished in the early sixteenth century, as well as the earlier Richardus de Mediavilla and
Thomas Argentinus (Suarez Disp. 18§21 14, Opera 25:602).

Authenticated
Vicaria Dei
neither by their own power [vi propria], nor as the instruments of substance;
they merely make matter apt and fit, so that in it the substance of the
generator can immediately produce substantial form" (ib. 293-294). In the
technical terms used to answer the question, accidents are, according to the
second view, not even "instrumental" but only "dispositive" causes in genera
tion. An instrumental cause has the same effect as the corresponding princi
pal cause. A dispositive cause has a different effect that promotes the opera
tion of the principal cause. 56 The utterance Hie est calix sanguinis mei is
instrumental to the sacrament, the freshness of the wine merely dispositive.
To deny that accidents are instrumental causes in generation is to deny that
they are causes of substantial form at all.
The two extremes, then, deny that accidents are instrumental causes. The
first makes them principal causes, at least sometimes; the other, merely
dispositive always. The central texts take that favorite path of Aristotelia
nism, the middle way: accidents cannot of themselves bring substantial
forms into existence, but substantial forms can only generate other substan
tial forms by way of accidents. Accidents, especially active powers, are neces
sary instruments in the production of substance.
That answer confers upon created substantial forms and their powers a
well-defined, and narrow, rank in the hierarchy of causes. Created forms do,
on the one hand, have the power to generate other forms. As the Coimbrans
say, activity is "one [... ] of the principal affections of natural things"
(Coimbra In Phys. 2c7q 18a2, 1:294), and generation is the noblest of natu
ral actions. Forms, the noblest part of substance, ought not only to be
endowed with activity but also to be the cause of the noblest action. Only
those who deny efficacy to second causes altogether would go so far as to
deprive form of activity.
Accidents, on the other hand, since they are less noble than forms, are
incapable of producing them without assistance: "Whatever is in the effect,
exists beforehand in the efficient cause, [which is] equally perfect if univo
cal, more perfect if equivocal: for nothing can give what it does not have
either formally or eminently, as they say, that is, more excellently: but an
accident, if it is compared with substance, especially with the form which is
introduced [in generation] or with the composite which is generated, is
much less perfect: substance is therefore not produced by the virtue of
accidents [virtute accidentis]" (Fonseca In meta. 5c2q6§2, 2:9gbE; cf.
Coimbra In Phys. 1 :295). That is enough to rule out the first extreme for the
56. Suarez, after an analysis well worth reading, concludes that "in the [...] most proper
way [of speaking], 'instrumental cause' is said of that which concurs or is raised up to bring
about what is nobler than itself, or exceeds the measure of its proper perfection and action,
like heat, insofar as it concurs in the producing of flesh, and in general accidents whenever
they concur in producing substance" (Disp. 'l[ 17, Opera 25:590).

Authenticated
time being. But the Scotists and nominalists conclude further that accidents
can in no way be said to have substances among their effects and that
substances are the immediate principles of generation (Coimbra In Phys.
1:293, Abrade Raconis 54).
Against that conclusion the central texts urge experience, reason, and
authority. 57 The elements, the heavens, mixtures, all are seen to exert their
actions through accidents. For the soul, too, "the principles of accidental
operations are accidents also, as is clear from vital or nutrimental heat, and
from the power of gathering together and expelling excrement, and the
like" (Suarez Disp. 18§31[ 16, opera 25:620). Some substances, including the
heavens and the soul, appear to act at a distance. Since every immediate
cause must be present wherever it acts, the heavens themselves cannot be
the immediate cause of the generation of minerals (Suarez 620, Fonseca
2:102E-F). In the reproduction of animals and plants, the seed "gradually
fashions through its formative power" the limbs, and so forth, even if the
parent has meanwhile died (Coimbra 1:2g6). Although God could supply
the missing efficacy of the parent (see Abra de Raconis 55), so widespread
an appeal to universal causes is to be avoided, "since the right disposition of
the world would have it that whatever can be done suitably and naturally by
second causes is done by them" (Suarez Disp. 18§21[15, opera 25:603). 58
Accidents, then, are necessary instruments in generation. But only instru
ments, never principals. John Major noted, though he was no doubt not the
first, that hot iron can set wood afire; and we all know that rubbing two sticks
together can accomplish the same end. The Aristotelians knew that a drop
ofwater in a vat of consecrated wine will be corrupted by the accidents of the
wine, even though the substance of the wine has departed. In all such cases,
it seems, an accident brings about the production ofa substance whose form
is not present.
But we have seen that all such claims would have the less noble produce
the more noble. When heat alone appears to generate fire, it must receive
the aid of a superior form. If there was disagreement on that point, it
concerned only the nature of the superior form: was it material? was it the
celestial intelligences? Scotus and Suarez did not forego appeal even to the
First Cause. Other philosophers referred the equivocal generation of frogs
and worms from decaying matter to the heavens; some held that metals and
57· The authorities include Aristotle De sensu 4C4• 441b13; Averroes In Meta. 7com21, 31
and 12com18; along with Thomas and many of his followers.
58. Suarez adds (25:6o2) that the Scotist view "smacks of Plato [sapit Platonem] ,"and is
similar to that ofPhiloponus (Phys. 1 in fine), Themistius (who writes in De an. 3c52 that forms
"provenire ab anima mundi"), and Avicenna (see §5.3). To deprive second causes of genuine
efficacy was contrary not only to the opinion of all the Scholastics but also to faith (Disp. I 8§ rf5
& r2, opera 25:594, 597; cf. Concilium Tridentinum, sessio6, canon4, in Denzinger 1976, no.
1 554).

Authenticated
Vicaria Dei
precious stones were produced in the earth by the sun; and many believed
that the forma cadaveris must be introduced by a superior cause.
Take a case where, by our lights, superior causes are obviously not
needed: the generation of fire by radiation or friction. Some philosophers
explained the combustion caused by a hot iron by supposing that "some
true fire has seeped into the pores of the iron, from which the [new] fire is
generated" (Suarez Disp. 18§2t3o, opera 25:6og, cf. 67g). Fonseca holds
that the iron or the source of its heat is either fire or nobler than fire-the
iron itself, being a mixture, contains all four elements virtually-and is thus
suited to produce its form. The same goes for friction: ''what rubs [...] is a
mixture, in whose virtue fire is contained" (In meta. 5c2q6§3, 2:1o4bF).
Such explanations avoid an appeal to celestial causes. In other instances,
according to the careful analysis of Suarez, they cannot be avoided. His
account of the generation of fire and minerals starts with the observation
that the heat in an "extraneous" subject "never has the intensity necessary
for the form of fire." 59 Of all the elements fire has the highest degree of
heat, and in any mixture the elemental qualities are always tempered. But it
may happen that "although the proximate agent introduces heat only to a
certain degree, there results per accidens the disposition of a quite different
form, as when wet matter is heated by fire, and a warm wet temperament
readily results, and [thus] some form that requires such a disposition" (4l[3o,
61 o). We might take this to justify not the introduction of a superior cause
but the reduction of those forms to their dispositions. Suarez instead con
cludes that "it is not difficult to believe that the virtue of a superior agent is
required [...] , not only because the proximate agent does not have an
equal or similar substantial form, but also because neither its quality nor its
temperament suffices per se to dispose and prepare the matter" (ib.). The
superior agents turn out on the common view to be celestial bodies, espe
cially the sun.
Animal souls are the noblest of nonhuman forms. But in generation,
because they reproduce by seeds, they turn out to require even more as
sistance than inanimate forms. The form of the seed is less perfect than that
of the parent, and the mother's form cannot make up for its deficiencies. A
higher cause must not only concur with the action of the seed but supple
ment it: "insofar as it attains to the production of the sensitive soul, not only
does the seed lack power with respect to the mode of action, but also with
respect to its substance (so to speak), and thus with respect to that action
needs much more the actual help, influx, and concurrence of a superior
cause" ('li33· 611 ). The slightly obscure language refers to an argumentjust
59· 'Extraneous' means 'not belonging to a thing by nature'. Iron is naturally cold. But
there are other substances-air, of course, and blood-which are hot by nature. So it is not
quite clear what Suarez has in mind, unless that fire is the hot thing par excellence.

Authenticated
made that the seed probably does not suffice even to organize the matter of
the foetus. Still less, then, will it be able to educe the sensitive soul that
requires such organization. Many philosophers held that the higher cause is
either the celestial intelligences or the heavens working as their instru
ments. But they too must act through their accidents, and their accidents do
not differ in kind from those of earthly things. So they are no more able to
produce souls than are animals themselves. Suarez, despite the words of
caution quoted above, is led to conclude that God alone is the coauthor of
souls, human or not.
The oddity is that lower forms-those of elements and many mixtures
are capable of reproducing themselves without assistance. That looks to be
"both an ill-ordered [inordinata] institution of nature and a great imperfec
tion in the causes that generate in this way" (1 34, 61 1). Suarez replies that,
just as it is no imperfection in man that two of his species should be needed
in reproduction, so too propagation by seed, which requires the assistance
of God, is "a natural condition or need" that animals "postulate by their
nature." Indeed it is because living forms are so much more perfect than
nonliving that they require an "exquisite and peculiar mode of generation."
For that reason also they require assistance; but it remains true that nature
has given to them, as to every corruptible kind, all they need to propagate
themselves (141, 614).
Nevertheless it seems odd that the more noble and more powerful a
material form is, the more fragile its mode of propagation should be, and
the more dependent on assistance from God. It is as if the presumption of
matter to assimilate itself to the divine had to be compensated for by a
proportionate vulnerability. There was, indeed, a standard view that the
more noble a substance was, the more difficult it would be to bring it into
existence. Higher animals must reproduce sexually: no one individual can
reproduce itself. The human soul, and all spiritual forms, could not be
brought into existence at all by material forms alone. Infinite being would
require infinite power to produce. Even the attenuated infiniteness of im
mortality, shared by the soul, could be communicated only by God himself.
The dustbin of history has long since swallowed up the Aristotelian theory
of generation. Already Petrus Aureolus had complained that the use of
celestial causes "is the refuge of the wretched in philosophy, as God is the
refuge of the wretched in theology. "60 Their reputation would only get
worse. Rather than flog a dead theory, it is more fruitful to ask what obstacle,
6o. In Sent. 4d1q1a3; cf. Maier 1951:I82n.2g, Maier 1952:13n.IO. For a brief suiVey of
similarly unfavorable sentiments about appeals to celestial causes or God, see Hansen's intro
duction to Oresme Mira., psorr. Oresme too calls the use of celestial causes, demons, or God to
explain maiVels the "ultimum et miserorum refugium" (Prologue, p136). On the naturaliza
tion of maiVels, see §6.4 below.

Authenticated
[166] Vicaria Dei
in Bachelard's sense, blocked a further understanding of generation. Less

tendentiously put: which elements of the Aristotelians' conception induced
them to bring in universal causes?
Most obvious is the principle that the less noble cannot bring into exis
tence the more noble. Accidents alone cannot beget substances. That is
bedrock: what remains is to explain apparent counterinstances. More bed~
rock: forms cannot act at a distance. Hence the push to the one cause that
can act immediately anywhere because it is present everywhere. Scotus gave
up the second principle to preserve the first. Suarez thought that doing so
smacked of Plato and Avicenna. 61 Yet his own considered view of the repro
duction of animals is similar. Perhaps it was inescapable: so long as forms
were considered more noble than accidents, while acting only through
them, they will always need a supplement; if that supplement too is a finite
form, it will want a further supplement, and so forth. Since it was already
firmly part of theology to hold that God concurs in every action of created
things, his assistance in generation could be seen to be merely a specifica
tion of that concurrence in certain instances.
Indeed the relation of forms to their powers is not only comprised in the
question of the efficacy of second causes; it is an almost exact parallel. God
could act immediately to bring about change in the world; finite substances
could, in principle, have been given the power to bring about accidental
change (Disp. 18§3, opera 25:621). Instead each has its proxies. The
difference is that God chooses to operate through second causes when he
does, and that some second causes are free; created substances, even those
with free will, cannot choose but to act through their accidents, which in
turn cannot but fulfill the ends of form. That difference aside, within each
substance the hierarchy of divine and created agents is repeated in the
hierarchy of substantial and accidental forms. Every substance is an image of
the universe.
If the Aristotelians had countenanced a reduction of substantial forms to
accidents, the ordering of forms according to nobility would have lost its
force. But reduction was refused even for elemental forms. Though some
philosophers, notably Alexander of Aphrodisias and Nicholas of Au
trecourt, were said to have affirmed that the form of each element is its
61. "Posterior vera pars [the claim that inanimate forms are educed immediately by the
first cause] sapit opinionem Platonis, qui dicit formas substantiales induci ab ideis separatis;
nam si verum est Platonem non posuisse ideas nisi in mente divina, perinde fuit ac dicere
formas substantiales effici a prima causa": The latter part smacks of Plato's view, when he said
that substantial forms are induced by ideas which exist apart [from the world]; for if indeed
Plato did not posit ideas except in the divine mind, it is as if he had said that substantial forms
are brought about by the first cause" (Disp. 25:602).

Authenticated
dominant quality, that view was rejected as "absurd" by the central texts. 62
What is substance, Averroes argues, in one thing cannot be an accident in
another. But heat, say, is found in animals too, and so it cannot be the
substantial form of fire. Not even an essential quality can be a substantial
form: if the ablation of heat destroys fire, it is not because heat is the form of
fire, but because the ablation of heat removes the form, just as cutting off a
man's head removes his soul (Toletus In de Gen. 2c2q2, opera s:3ogv-310r).
Since even the forms of the elements cannot be reduced to their disposi
tions, neither can those of higher forms. The reduction envisioned by
Descartes of all material forms to modes of extension could not be admit
ted. Confronted with similar proposals in ancient atomism, the Aristotelians
hardly felt the need for extended refutation. It sufficed to note that atom
ism does away with the distinction between substantial and accidental forms,
and thus with the distinction between generation and alteration. The irre
ducibility of form was not the only reason the Aristotelians had for holding
that substantial form is nobler than accidental. But it was perhaps the best
physical reason.
Descartes's physics, on the other hand, by doing away with the distinction
(except in the case of the rational soul), wiped away with one stroke the
principal reason that drove the Aristotelians to assign celestial agents and
the First Cause an essential role in generation. Or at least it promised to: he
himself was unable to explain the reproduction of animals except by invok
ing a seed endowed with powers of organization, whose effects he could
attempt to describe but whose explanation eluded him.
62. "Others would have it that elemental forms are the four primary qualities themselves
[...] Theyjudge Alexander and other Greek interpreters of Aristotle to believe the same. Nor
does it count against [the view], that the forms of the elements should also be substances: they
are referred (they say) to the category of substance, and for that reason they are substantial
qualities, just as the qualities that belong to the category of quality are accidental" (Fonseca In
meta. 8c3q3§1, 482aB). Toletus adds Nicholas to the list, and notes that John of Jandun
contests the ascription of the view to Alexander (Toletus In de Gen. 2q1, Opera 5:3ogva).
'Absurd' is his word.

Authenticated
[6]
Finality and Final Causes
D 'Alembert wrote that whatever Descartes's errors in physics, he

at least showed "those with good minds how to throw off the
yoke of Scholasticism" (Disc. prel. xxvi). Not the least of hisser
vices was to proscribe the use of final causes in physics. Final
causes are not merely superfluous, not merely the "vestal virgins" of philoso
phy, as Bacon said, but "dangerous." 1 The appeal to ends leads to absurdity,
and invites us to put ourselves in the place of the Creator, presuming to
know his ends. 2
Gratuitous examples of the sort that later critics probably had in mind are
not hard to find. Abra de Raconis, for example, writes that "the two most
important ends to which the sea is instituted are, first, that it should be the
common domicile of fish, and second that in it there should be navigation
to provide commerce and necessary goods; but to both ends its saltiness is
most fit, since saltiness keeps the sea from putrefying and makes it stronger
and denser so as to hold the greater weight of ships. "3 The Coimbrans write
of the ocean that "when by its flooding or daily ebb and flow or the force of
storms its [waters] are borne toward the shore, they contain themselves
within preordained limits as if terrified of some law written to them on the
sand" (Coimbra In Phys. 2cgq1a1, 1:224; cf. Suarez Disp. 23§10110, opera
1. See Bacon De augm. sci. 3c5, Works 1:571. It should be noted that Bacon'sjibe is directed
against the use of final causes in physics-where they were in fact usually treated-and not
against their use altogether. In metaphysics, ends, along with forms, are proper objects of
inquiry (cf. 570).
2. Encyclapedies.v. "Causes finales," 2:789. Cf. Descartes PP1§28, AT 8/1:15.
3· Phys. 384. He adds that since "God rarely effects himself what he can produce by way of
second causes," the efficient cause of the sea's saltiness is probably "fiery exhalations" and solar
coction (385; cf. Dupleix PhysiquqCtg, 485, which cites Aristotle, Pliny, and Plutarch). Cf. also
John of St. Thomas Nat. phil., Tract. de meteoris 7c3, Cursus 2:877a-b.
[168]
Authenticated
Finality and Final Causes [16g]
25:888). The sentiment is fine, but the explanation is superfluous and

based on a misapprehension of the facts. The ocean's boundaries have
changed enormously over the aeons; its transient containment can be ex
plained entirely by efficient causes. So too with saltwater fish: the sea was not
made for them, we now believe; they, in a certain sense, were made for the
sea.
But these are egregious instances. Descartes and those who subscribed to
his polemics exaggerated the sins of their opponents, ascribing to the Aris
totelians views the Aristotelians would have repudiated. Far from being
confused about which things have souls and which don't, as Descartes
thought, they carefully separated, as we will see, the case of rational agents,
which can recognize and judge their ends, from that of irrational agents,
which exhibit only the secondhand finality of instruments. While asserting
the primacy of the final cause, they did not, as d'Alembert insinuates, con
fuse final with efficient causation. The horrorvacui, which d'Alembert holds
up to ridicule, is an instance where the Aristotelians are quite clear about
the efficient causes. They also propose one or another final cause; they do
use intentional language in describing how natural agents act to prevent the
occurrence of a vacuum. But the use of such language does not of itself
imply confusion. In Abrade Raconis' s explanation of the saltiness of the sea,
the end is ascribed to God, not to the sea itself. Nor must one suppose that
final causes compete with efficient causes in explanation: the Aristotelians
recognize both, making each a separate subject of enquiry. When Descartes
and others declared that final causes were superfluous in the face of ade
quate efficient causal explanations, they forced the issue in a way that most
Aristotelians would not have regarded as necessary.
The Aristotelians had, moreover, come increasingly to restrict final causa
tion to the actions of rational agents. The beginnings of that development
can be found in Ockham, and its culmination in Jean Buridan. Though in
the period I am studying there was a return among many authors to the less
radical position of Thomas, at least one author-Hurtado de Mendoza,
whose Cursuswas published in 1624-omits the standard questions on final
causes from his Physica altogether, placing them instead among disputations
about the will. That was extreme, but it reflects an increasing reluctance to
take ends for "real and proper" physical causes except when cognized by
rational agents.
What remains, and what Descartes especially opposed, was finality.
Though the ends to which natural changes tend might not be proper causes
of those changes, it was inconceivable that natural change should not have
ends. The analogy implied in the use of intentional language to describe
natural change retained its force. Finality, and not the valetudinous
doctrine of final causes, was Descartes's real target.

Authenticated
[170] Vicaria Dei
The impression conveyed by Descartes of a vast distance between his

physics and that of his predecessors was, therefore, not unwarranted. The
very notion of natural change, defined as the imperfect actus of a potentia in
the thing changed (§2.1), presupposes that the basic and most comprehen
sible changes in the world are directed changes, which cease of themselves
after reaching their termini, and in which there is an intrinsic difference
between the patient, in which the change occurs, and the agent, which is
unchanged except per accidens. Other changes-corruption, say, or
monsters-must be redescribed to be understood scientifically. They are
not exceptions or counterexamples to the omnipresence of finality: they
are, rather, necessary outcomes when natural processes meet and interfere
with one another, or when a patient's matter is inadequate to receive the
form specified for it by the agent.
The prototypical event of Cartesian physics-the collision of two
bodies-is just the sort of event that must be redescribed. It has no terminus
ad quem. There is no intrinsic feature in the encounter thus described that
would allow us to pick out one as the agent and the other as the patient. To
bring it within the scope of science an Aristotelian must find, say, that one
body is a falling stone, the other a pilgrim leaving for Rome; only then can
the collision, now analyzed into two independent motus, be referred to their
ends and understood. The Cartesian condition of intelligibility is quite
different: paradoxically, at once more immediate in conception and more
remote in application. The falling of a stone is not a simple expression of
the heaviness imparted to the stone by its generator. It is the outcome of
innumerable collisions between the stone and the whirling vortex that sur
rounds the Earth. Each collision is subsumable under the laws of motion;
but the aggregate effect can only be understood, as we would now say,
statistically.
Nevertheless even Descartes, whose unbending opposition to finality was
maintained after him only by Cartesians of strict observance, did not rule
out reasoning from the perfection of the world as a whole. 4 Nor did he deny
that people and other ensouled creatures act according to ends. To under
stand why and how the explanatory burden in natural philosophy and meta
physics sometimes remained with the final cause, or was shifted to the
efficient cause, or even to no cause at all, it is a good idea to start with the
distribution of that burden in Aristotelianism.
4· Sylvain Regis sustained his opposition to final causes even in the face of Leibniz's jibe
that Descartes had "turned philosophers away from the search for final causes, or, what is the
same thing, from the consideration of divine wisdom" (Leibniz Ph. Schr. 2:562). Regis answered
that Descartes never intended to banish final causes from moral philosophy; as for natural
philosophy, he was right to do so, since "in Physics one does not ask why things are, but how
they occur" (ib. 4:334). For a discussion of the exchange between Leibniz and Regis, see
Tocanne 1978:70-71.

Authenticated
The first item of business is to reduce to order the welter of objects said to
have ends, at the same time classifying some uses of finality in physical
argument (§6.1). The second is to examine the existence of ends in nature,
both in the actions of individuals and in nature collectively (§6.2). Next I
look at the final cause, which was for a number of reasons not easily aligned
with the other three Aristotelian causes (§6.3). I conclude by studying the
use of teleological principles in explanation (§6.4).
6.1. Varieties of End

In human affairs, a single bodily act may be seen to serve many purposes at
once. I lick a stamp, intending thereby to signify that I have paid postage,
and so also intending to mail a letter, which in turn repays a debt and keeps
me in the good graces of the telephone company. No discussion of finality
in Aristotelian physics should neglect the complexity of the ends attaching
to natural change, or what I will call its teleological structure. Not merely the
ascription of ends, but the unification achieved through the subsumption of
one end to another, and of all to the ultimate end God, is what gave the
Aristotelian use of finality its force. When combined with a doctrine of
creation, moreover, it armed natural theology with evidence not only for the
existence and power but for the intelligence of the Creator.
I will classify the ends ascribed to natural changes into three categories.
The first, individual ends, includes the immediate ends of actions and of
particular things. The second, collective ends, includes all those ends that
cannot be achieved by one thing alone, like the preservation of the species.
Individual ends are typically in aid of collective ends; as we will see, a
collective end may be in conflict with, and override, individual ends. The
third category, cosmic ends, consists of those ends that may properly be
called the ends of nature as a whole, which for brevity's sake I will call
'Nature'. The most notable of those ends is beauty.
1. The most general definition of 'end' is 'that on account of which',
propter quid, something is done. Aristotle holds, and the Aristotelians agree,
that in any change there are two things that may reasonably be called the
immediate end of the change. There is a condition, the terminus ad quem, the
attaining ofwhich coincides with the cessation of the change. There is also a
thing for whose benefit the change occurs. That thing may or may not be
the subject of the terminus ad quem. In the standard example of an ill per
son's being restored, both health and the person herself can be called "that
on account of which," and so both can be called ends. 5 The condition was
most often designated the finis cuius, the subject the finis cui. The terminol

Authenticated
[172] Vicaria Dei
ogy is confusing, especially since some authors used other terms. Except in
quotations, I will avoid the Latin terms. I will call the condition the intended
state and the subject benefited the beneficiary.6
The intended state and the beneficiary have distinct roles in explanation.
It is, first of all, trivial that every natural change should have an intended
state. Natural change is defined as the imperfect actus of a potentia: it has,
therefore, a corresponding perfect actus, its natural stopping point. What is
not trivial is to decide which of the changes we see are in fact complete, and
if so, what their intended states are. Descartes denied that the changes of
bodies were natural in the Aristotelian sense; inquiry into their intended
states became moot. Nor did the Aristotelians hold that every occurrence in
nature is a natural change: monsters, which I will discuss in §6.4, are a case
in point. Designating the intended state of a change, moreover, enables one
to say that the change has been frustrated, or to explain why it has not yet
ceased. Aristotle, in fact, points to such judgments by way of arguing that
nature acts according to ends. It also, as we have seen in §2, serves to
distinguish one kind of potentia from another, and so, since the natures of
things depend on their potentice, provides the basis upon which to define
natural kinds.
It is not obvious, on the other hand, that every natural change has a
beneficiary. The hard case, as we will see, is generation, in which the in
tended state includes the existence of a new individual of a specified kind. 7
Here notjust a state, but the resulting thing, is the end of the action; but the
beneficiary is not that thing but something else-the owner, say. It was
customary to distinguish two ends in generation: the finis generationis, which
is the thing generated and thus the intended state of generation; the finis rei
genitce, or the end of the thing generated. Typically the finis rei genitce is an
end attainable by one or another operation of the thing generated, as
restoring health is the end of medicines. The beneficiary of medicine
6. "Finis Cujus dicitur, cujus adipiscendi gratia homo movetur, vel operatur, ut est sanitas in
curatione; finis Cui dicitur ille, cui alter finis procuratur ut est homo intentione sanitatis; nam,
licet homo curetur propter sanitatem, ipsam vero sanitatem sibi et in suum commodum
qu.erit" (Suarez Disp. 23§21[ 2, opera 25:847). In cuius gratia, cuius is the genitive of the relative
pronouns quis/quid, translating Aristotle's ou EVEKa or ou EVEKa TLVOS' 'on account of which/
something'. Cui, the dative, translates 4i EVEKa orou EVEKa TLVL 'for whose account'. To ou EVEKa,
id gratia cuius or id propter quid, is Aristotle's usual name for the final cause. Cf. Aristotle De an.
2q, 415b, Meta. 12q, 1072b; Zabarella De anima 2qtext37, 437B; Coimbra In Phys.
2qq2oa2, 1:305; Fonseca in n.7 below; Eustachius Physica pt1tr2d2q6, Summa 2:143.
7· Fonseca uses the term finis cuius to denote not only states like health, but also products of
action, such as houses, and immanent actions, such as contemplation, that issue in neither a
state like health nor a product. "Finis Cuius dicitur id, cui us consequendi aliquid fit, sive id sit
res aliqua, qme per actionem fiat, vel uti domus, cui us .edificand<e gratia artifex operatur, sive
actio immanens, per quam nihil aliud fiat, veluti contemplatio, sive usus aut fruitio rei, ut
habitatio domus, aut voluptas, qu<e ex contemplatione percipitur, cum quis propter volup
tatem contemplatur" (Fonseca In meta. 5c2qw§1, 2:153A).

Authenticated
Finality and Final Causes [1731
making is not the medicine itself but the person who takes it. In generation,
then, the intended state is the thing generated; but only sometimes is it
obvious that there is something for whose benefit the thing is generated.
Even if children help their parents and provide for them when they are old,
still it is not clear that children are born for the benefit of their parents.
The Aristotelians held, nevertheless, that to every intended state there
corresponds a beneficiary. 8 It made sense, for example, to ask not only
whether the saltiness of the sea was the intended state of the changes that
created it, but for whose benefit it is salty. The answer is in part that it is
useful to people. Though Aristotle himself seems to have recognized the
legitimacy of such questions, he does not bother much with them. Their
prominence in Aristotelian discussions of ends in nature rests, as we will see,
on the warrant provided by scriptural assertions of man's dominion over
nature.
Supposing, then, that all natural changes do have a beneficiary, naming
that beneficiary is essential to assessing intended states and the appropriate
ness of the means employed to reach them. Health in a human requires a
degree of heat that would kill an oyster. Even among animals for whom
being warm is part of being healthy, the means of heating might well be
different: a carnivore should be given spicy meat, a vegetarian hot peppers.
For that reason the Aristotelians said that the intended state is "subordi
nate" to the beneficiary (Fonseca In meta. 5c2q10§1, 2:154C).
Some authors thought that intended states could be assimilated to instru
ments or means (and thus that only the beneficiary is truly an end). Since an
instrument is an efficient cause,. that would reduce some final causes to
efficient causes. Yet there is, Sml.rez argues, "a great difference between a
means and a .finis cuius; the means[...] is desirable insofar as it is useful for
health; but health is desired on its own account, because it perfects per sethe
person for whom it is desired." He concludes that only the whole-a healthy
person-is the "entire and adequate end of this action.''9
That conclusion fails to do justice to instances where the intended state is
a newly existing individual. Although health and the person who is healthy
may be called a whole, a house and its owner may not. Though one could
evade that objection by talking instead about the finis rei genitre, the shelter
provided by the house, it is more illuminating to arrive at Suarez's conclu
sion by a different route. The end on account of which a patient is given
medicine is notjust the patient herself, nor even her continued existence. It
8. "Inter hos autern fines hoc interest, quod finis Cuius semper dirigitur in aliquem finem
Cui, finis autem Cui ex ratione sua non necessaria dirigitur in alium finem" (Fonseca In meta.
5C2q10§1, 2:153C).
9· Suarez Disp. 23§215. opera 25:848; Fonseca In meta. ib. See Maier 1955 for a discussion
of this controversy in Ockham and others.

Authenticated
Vicaria Dei
is a certain state ofthe patient. Other states require other actions, even if the
beneficiary is fixed.
In generation or in art, the teleological structure of actions has a com
plexity not exhausted in the distinction between intended states and benefi
ciaries. A veterinarian gives a horse mint to cool the blood. The intended
state of the operation is the cooling, the immediate beneficiary is the horse.
But the horse is treated so that the owner can plow his land. The owner is a
remote beneficiary, and the horse's power to work is also an intended state.
We will see a similar structure in the generation and preservation of species.
Recognizing that complexity is essential, because the two kinds of end
figure differently in Aristotelian physics. Questions about the causality of
ends and the objects of final causality concern primarily the intended state.
The intended state, rather than the beneficiary, is what "moves an efficient
cause to act." The beneficiary, on the other hand, comes into its own in
questions about the existence of ends in nature.
2. Collective ends. States of individuals and individual substances do not by
any means exhaust the list of possible ends. There are also collective ends.
For example, since no material individual can endure forever, a specific
form-the form or part of form common to all members of a kind-would
sooner or later cease to exist if it could not bring about the existence of its
like in new matter. Yet all forms strive, as we will see later, to assimilate
themselves to God, one of whose attributes is to exist at all times. God, for
his part, has given to each kind of form the powers necessary for each
instance of that kind to produce in new matter whatever is required for its
like to exist. It is therefore at least possible for every material form to exist
eternally, not indeed as an individual, but as a kind: "In all things there is an
inborn appetite for protecting and preserving themselves. Hence the care
with which they seek what is useful and healthy, and shun what is harmful
[...] Because in that way all perishable things strive to save themselves from
destruction, and if not numerically at least in species, if not per se at least
through the species to which they belong, strive to obtain immortality; so
that thus they approach, insofar as it is possible, a resemblance to the divine
nature" (Coimbra In Phys. 4cgq1a3, 2:62).
Although the individual form of the offspring, the terminus ad quem of
generation, is clearly the intended state of that action, neither the parent
nor the offspring can rightly be called the beneficiary. The parent, although
it may benefit by having offspring, need not have offspring for that reason;
the offspring, although it benefits, in a sense, from being generated, need
not have been the intended beneficiary of generation. What benefits in
every case is the specific form, regarded as a kind. 10 Yet no one individual
aims at the preservation of its kind. Preservation is a collective end.
10. Generation, writes John of St. Thomas, is the most natural of the soul's operations (cf.
Aristotle De an. 2Iect7, Thomas In de An. ad locum) because the nature of each living thing

Authenticated
Individual ends are typically appealed to in answering questions about

particular acts or types of act. Collective ends explain rather the existence of
powers in a form. The question one answers by referring to the preseiVation
of specific form is not 'Why did this individual reproduce?' but rather 'Why
do individuals with this form (or indeed any material form) have the power
of reproduction?' Similarly, the most obvious answer to the question 'Why
do souls have the power of nutrition?' is that a soul requires for its own
suiVival certain dispositions in its matter-notably heat-that can only be
maintained by eating and drinking.ll The soul, then, is not only the effi
cient cause by emanation of its self-sustaining powers (§5.4) but their
beneficiary.
That collective ends are distinct from, and occasionally opposed to, indi
vidual ends is well illustrated by the horror vacui. 12 Toletus, after affirming
that God could bring it about that a certain inteiVal was devoid of substance,
argues that nevertheless a vacuum cannot naturally occur. Experience
amply confirms that it cannot, and, as usual, the relevant phenomena are
taken to be actualizations of a sort of potentia, the horror vacui. But for the
Aristotelians as for their successors, to end the story there would have been
jejune. Toletus gives two explanations of the horror vacui: "I believe one
should say that contiguity among bodies is the most natural disposition, so
that just as a continuous part draws along by its movement another part
continuous with it, so too what is contiguous with another draws along what
is contiguous, since no other body can follow." The contiguity of bodies is,
moreover, "in accordance with the nature of the universe," since if a vacuum
inteiVened between sublunary bodies and the heavens, their contact with
the heavens would be interrupted, and the heavens could no longer guide
them (Toletus In Phys. 4cgq1o, opera 4: 13orb).
The Coimbrans argue more generally that in all bodies there resides a
"love of mutual conjunction and society." In drops ofwater and in the earth,
bodies form themselves into a sphere, "the greatest conciliator of union."
"inclines to it not on account ofits own good, but on account ofthe common good, namely the
perpetuity of the species, which is preseJVed by its being generated" Uohn of St. Thomas Nat.
phil. 4q3a1, Cursus 3:86b). Because the common good outweighs the private good, generation
more strongly attracts the soul than nutrition or self-preseJVation (ib. 87a). The reasoning
resembles that of the Coimbrans concerning the horror vacui, which I discuss below.
11. One may wonder, if all the powers of souls inferior to the human have, ultimately, the
preseiVation of the form in the face of corruption as their end, what end could be assigned to
the powers of angels, demons, and the human soul itself insofar as it is immortal? The answer
for angels, at least, is that there are other ends: the contemplation of God (which, being an
immanent action, is its own end), the administration of the· heavens, the revelation of divine
intentions to humans (see Suarez Deangelis 5 and 6, opera 2:565ff).
12. Aristotelian theories of the vacuum are discussed in Schmitt 1967 (Toletus is quoted at
PP355· 357), Grant 1973, c-4> Grant 1981. Schmitt discusses only empirical evidence for the
hmror vacui; he does not examine the causes proposed for it; Grant is likewise much more
interested in experimental evidence than in causes (but cf. n.14 below).

Authenticated
[176] Vicaria Dei
United and joined, they "operate much more strongly, and repel the inju
ries of their adversaries." The love of union brings even bodies of different
kinds together, since "there is no kind of thing that subsists by itself, torn
from the rest, or that if others are lacking can preserve its power or its
perpetuity." In particular, sublunary forms, as we have seen, require the aid
of celestial influxes to reproduce themselves. A vacuum, clearly, would
"dissolve this preservative union," and impede the reception of heavenly
force [vis], since "it cannot happen that this power should cross through an
empty intervening space."
The need for union is so great that it can overwhelm what one might call
the private tendencies of things.l 3 The Coimbrans note that "each natural
being strives resolutely on behalf of two goods, namely the common good of
nature as a whole, and its own peculiar good." But the common good is the
more "excellent and divine," so that sometimes a body will act to its own
detriment on behalf of it. Since union is a great good, and vacuum a "most
pernicious evil" to the world, we see heavy bodies move upward, light bodies
downward; "even contraries, which would otherwise flee one another[...]
come together as if they had given up their antipathy" (Coimbra In Phys.
4cgq1a3, 2:62-63). 14
3· Cosmic ends. A collective end, then, is an end which each agent can be
said to strive for but which cannot be accomplished by any one agent alone.
A cosmic end, on the other hand, is an end serving nature as a whole, an
intended state whose beneficiary is Nature. In their question on the ends of
nature, we find the Coimbrans responding to the argument that because
certain "minute animals" are "entirely useless," nature does not always act
according to ends: "Nothing superfluous or without an end has been
brought about by God, although to the ignorant it may at first glance seem
so, just as someone might judge the tools in some craftsman's workshop to
have been multiplied beyond necessity, because he is ignorant of their uses"
(Coimbra In Phys. 2cgq 1a3, 1:326). Each, being beautiful of its kind, at least
adds something comely to the whole: or rather God, by allowing such a
13. The Coimbrans contrast the "bonum privatum" of individuals with the "bonum com
mune" that sometimes conflicts with the bonum privatum. That the political uses ofsuch phrases
are not far off maybe gleaned from the references toAristotle'sEthiG!' (1c2), and to "those who
manage the public good" (Rempublicam tractantibus) earlier in the question.
14. Since I am here interested in the final causes of the horror vacui, I omit discussion of
Aristotelian hypotheses about its efficient causes. Scholastic authors had proposed a "univer
sal" nature that originates from the heavens; its action consists in impressing upon individual
substances the impetus that in tum moves them to fill up voids (Grant 1973:69-70). The
Coimbrans affirm that "frequently nature uses a certain motive force, which for this purpose
seems to be attributed, along with the peculiar virtues by which they seek their own places, to
all sublunary bodies" (ib. a3, 2:64). Neither they, Toletus, or John of St. Thomas mention
universal nature; Suarez, in his discussion of the horror vacui, refers to the "providence of
universal nature, or rather of its author" (Suarez Disp. 18§8114, opera 25:655), but in a
discussion devoted specifically to celestial causes he mentions only their role in generation
(22§517rr, 25:84off).

Authenticated
thing to exist, has done so. Yet adding to the beauty of the cosmos cannot be
the end of any individual's actions (I set aside angelic and human arts). Like
the brush strokes of Seurat, the gnats and mites of this world, unappealing
in themselves, have purpose in relation to the purpose of the whole in which
they are tiny but necessary parts.
The Coimbrans then answer the argument that some animals and plants
are not just useless but harmful: "Nor are venomous animals superfluous
[...] , because, as St. Augustine says [... ] , they should at least be praised for
reminding us to esteem that other better life [i.e., in God], in which there is
the greatest security [...] Or finally because, as Lactantius writes [...] , it
was useful to man that he should encounter some beneficial things, and
some harmful, so that in avoiding the one and striving after the other he
should exercise his power ofreason" (ib. 327). It cannot be any individual
animal's purpose, or even that of the specific form, to be useful to humanity.
They pursue their own ends. Since we do not produce them, or at least their
specific forms, it is not our purpose either to have made them useful. Only
in relation to the whole, in which there are both people to be reminded of a
better life and venomous beasts to remind them, can that purpose be made
out. Serving humanity is not, of course, itself a cosmic end. But the existence
of humans, too, perfects and ornaments God's work of creation, and so
whatever serves us indirectly serves that end.
Cosmic ends yield answers to very general questions about the constitu
tion of the world. We saw a moment ago that the preservation of specific
form is the end of generation, or rather of the power of generation. One
could take a further step back and ask why it should matter that specific
forms are preserved. The Coimbrans speak of the beauty of the world. They
do not mean merely that it is beautiful to us, but that it is a fitting illustration
of God's power. The multitude ofactual forms testifies to, though it does not
exhaust, the infiniteness of that power.
6.2. Existence of Ends

In Descartes's time anyone who set out to construct a physics without finality
would have found an immense variety of phenomena suddenly left out of
account. The range of items to which teleological reasoning was applied
gave rise to an equally broad range of arguments to support the legitimacy
of that reasoning.
Most fundamental is the regularity of natural change. In the second book
of the Physics, Aristotle sets before his opponents a dilemma: chance or
ends. What occurs fortuitously occurs irregularly; but natural changes occur
"always or for the most part" (aut semper aut plerunque). So they must occur
on account of some end. Worse yet, if there were not ends in nature, "all

Authenticated
Vicaria Dei
things would come from all promiscuously ... men would arise from the
sea, and scaly creatures from land, and winged ones burst from the air. "15
The argument is, on its face, unconvincing. Everyone agrees that efficient
causes necessitate their effects ("ifthe cause is given, so is the effect," writes
Eustachius with his usual brevity [Summa, Phys. 1tr2d2qs, 2:142]-posita
causa ponitur effectus). So people will not emerge from the sea ever if they do
not always: one does not need ends to account for that regularity. Given that
we have not seen any such occurrence, and that the sea remains constant in
composition, there is no reason to expect that the weird event will occur.
Likewise, if people have always given birth to people, and birds to birds, and
if they remain constant in composition, then there is no reason to expect
that people will bear birds or birds people. So if the regularity to be ex
plained is 'people give birth only to people, and no other kind of thing
does', then an appeal to the necessity of efficient causes seems to suffice.
The possibility of a world run like a great machine by efficient causes
alone did not escape the Aristotelians. They did not adopt final causes
through ignorance of such alternatives. Even if "God did not concur in the
actions of natural agents, but allowed them to perform their motus indepen
dently," still "the rock would descend, fire would generate its like, and so for
the rest" (Disp. 23§10!8, opera 25:888; Coimbra In Phys. 2cgqn3, 1:330).
Suarez acknowledges that regularity would result from the operation of
efficient causes alone; indeed such a world would even exhibit order. It
would appear to tend toward ends: "from this impossible hypothesis it fol
lows that nature operates in the most orderly way so as to tend toward an
end, without any direction or intention of the end." But the appearance of
order without the fact of being ordered is "absurd" (Disp. 1:g, 888). The
inference is familiar, as are the agnostic retorts: nature is not in fact well
ordered, and in any case "order," unless the term is used so as to beg the
question, does not entail ends.
What must be added to the mere observation of regularity or even of
order is either a list of phenomena that would not occur, or that would
occur differently, in a world of efficient causes alone; or else reasons to
believe that introducing ends, and speaking of them as causes, will yield a
genuine understanding of natural change which is not to be had in consult
ing efficient causes alone. In 16oo there were good arguments of both sorts.
Many phenomena seemed to require a coordination of efficient causes that
no known efficient cause could bring about. The generation ofliving things
was among the more striking. The sagacity of animals in some of their
15. Coimbra In Phys. 2cgq1a1, 1:323, quoting Lucretius De rerum nat. 1:161-162 (cf. also
Thomas Contra gent. 3c2, Suarez Disp. 23§10i3, Opera 25:886). Lucretius, as the Coimbrans
recognize, since they quote the same passage in their question on prime matter ( 1cgq3a1), is
arguing not that nature acts toward ends but that natural change never begins or ends with
nothing. The argument of Physics 2c8 is discussed and defended in Cooper 1987.

Authenticated
actions likewise outstripped the capacities of efficient causes, and indeed

seemed to entail the recognition of ends. To that the Aristotelians added
the long-term stability of the world, and the usefulness of its inhabitants to
humans.
The Aristotelians did not conceive of efficient causation in Humean fash
ion. Cause and effect are not related as antecedent to consequent in the
instantiation of a law. There are indeed regularities. But they consist in the
existence of kinds in which powers are united and the necessity with which
the actions of power are manifested under the appropriate conditions.
Efficient causation itself is blind and atomistic. Blind because it consists
merely in an agent's setting into motion the potentia of a patient; atomistic
because each agent, insofar as it is considered only an efficient cause, acts in
disregard of others. Yet in many natural processes the "triggering" of effi
cient causes proceeds in a regular order. Those causes seem, moreover, to
aim not only at their particular ends but to work in concert toward some
thing that none could effect alone. The Aristotelians believed that guidance
could not come from the efficient causes themselves, or from the matter
they act on. It could be accounted for only by supposing that they act toward
ends. Exactly how ends provide guidance will be seen in §6.s; here I will
examine the chief instances that seemed to call for that guidance.
1. Coordination ofpowers. In §3.2 I presented arguments to the conclusion
that substantial form is the ground of the unity of active powers, and in fact
their efficient cause. We saw, moreover, in §5.4 that the active powers that
emanate from a form seiVe to introduce into new matter the dispositiones
required for it to receive a similar form. That form is the terminus ad quem,
not immediately of the actions of the active powers, but remotely, as their
joint result. Generation is one action only with respect to its remote efficient
cause-the form of the generator-and its terminus-the form of the thing
generated. With respect to the active powers that a form must use as its
instruments, generation requires a series of alterations coordinated both
temporally and spatially. In higher animals, for example, the organization of
the fetus includes the "effecting ofa proportionate temperament of primary
qualities" and the "disposition of the members" in "maiVelous order"
(Suarez Disp. 18§2'll33, Opera 25:611). The particular efficient causes of
those changes have so to speak no cognizance of the rest, and no reason to
occur when and where they do. They will act whenever a suitable patient
presents itself. Nor will adducing further efficient causes, even celestial,
help unless it can be shown how they, considered merely as efficient causes,
could somehow regulate the causes already existing in the fetus.l6
16. A further argument to the same conclusion can be gleaned from the Coimbrans'
discussion of monsters. In general, an efficient cause and its effects are similar: heat causes
heat, and so forth. Each efficient cause, moreover, "strives to produce without error or devia
tion what is similar to itself." The instrumental powers of a seed, say, are nothing more than the

Authenticated
[180] Vicaria Dei
There was, as John Cooper notes, little reason in Aristotle's time to be
lieve that the "powers of matter," that is, the efficient causal powers of the
elements and their mixtures, could by themselves have the effects obseiVed
in generation (Cooper 1987:270). In the sixteenth century, the three prin
ciples of Paracelsus-salt, sulphur, and mercury-were no more promising
than the four elements of Aristotle had been. Celestial causes, the heavenly
intelligentia!, even God himself, are of use only insofar as they provide appro
priate repositories for the final causes that alone seiVe to explain
generation.
Just as the coordinated action of active powers in generation requires the
introduction of ends, so too does their coexistence in individuals. The same
holds for dispositions and organic parts, insofar as they provide the material
basis for powers. Suarez writes that "if matter is taken according to itself, it is
indifferent, and has no need of such dispositions or properties; but if it is
supposed already to be affected with such-and-such dispositions, already
those dispositions have been introduced on account of some end or form,
which requires them on account of its conseiVation or for some operation;
but that operation is in turn on account of the conseiVation of the species or
the individual itself [...] And so every connection and connate necessity
which exists per se in natural things, arises from being ordered to an end"
(Suarez Disp. 23§10!7, opera 25:887). In §5.3 it was argued that the acci
dents or dispositions that prepare the way for substantial form inhere imme
diately in matter. Matter has no reason to take on one set of dispositions
rather than another-it's just matter. It has them in order to receive the
form. But why does the form require those dispositions and not others? The
answer is that it requires whichever dispositions enable it to exercise the
powers ofwhich it is the efficient cause via emanation. But now one can ask:
why those powers? The form itself has ends, and in particular self
reproduction. Self~reproduction requires at least that the form be able to
produce in new matter the very dispositions that it itself needs in order to
act. Efficient causation suffices to show why, given the form, those powers
exist together in one individual. But it cannot, Suarez argues, show why
from that form those powers emanate. For that one must take the production
of the form to be not merely their common efficient cause but their com
mon end.
qualities arising out of the temperies or disposition of the elements contained in it. They
reproduce, without fail, their like in the matter of the offspring. But if, as the Aristotelians
thought, the form of the animal to be generated is not reducible to the disposition of its matter,
then the qualities, if they operated independently, could not themselves produce that form
except by accident. Something else must control their operation, and produce its like through
them. That something, of course, is the form, which sets the active powers in action to the end
of reproducing itself.

Authenticated
In Suarez's argument we proceed from dispositions to form, from form to

specific ends, and (by implication at least) from those to cosmic ends and
God. At each stage we ask: why do these items exist (or why are they compre
sent in one individual)? The answer looks not to their efficient cause (whose
existence indeed does explain theirs, in one sense) but to what will exist if
they are allowed to act. It looks, in other words, to their ends. The series
terminates when we reach God: provided that he is omnipotent and neces
sarily existent, and that an omnipotent being will create as rich a world as
possible consistent with its goodness, we need nothing more. 17
2. Sagacity and industry of animals. Among the most visible operations of
animals are raising their young, building webs or nests, looking for food,
and fighting or avoiding their enemies. For the Aristotelians, the thought
that such actions could be explained solely in terms of the dispositions of
animal bodies was a pipe dream. The resources of their science-or of
Descartes's-were not up to the task. The independence of animals' actions
from immediate stimuli, the coordination of their actions, the flexibility in
their means, all told against any explanation that did not take into account
the ends to which those actions are directed. The real question, which I will
return to, was whether animals were guided by reason or merely by instinct.
Here it suffices to note that in either case it was necessary to appeal to ends.
3· Stability. Under this heading I put observations not of the regularity of
particular causal relations but of the persistence of certain large-scale fea
tures of the natural world. We have seen that the ocean keeps mostly to its
place, and thus the proportion of ocean to land remains relatively constant.
So too for the elements:
The same is shown by the discordant amity of the four elements lying
spherically in four regions, and their alternating recirculation through
downfall and ruin. From earth water, from water air, from air fire, and
then in the other direction air from fire, water from air, and earth from
water are generated; so that in moderation, as if in equilibrium, the heavy
is made equal to the light, the cold to the hot, the lower to the higher, as
much as can be: to that end, namely, that they should last; while nature
teaches us in great things that impartiality [cequahilitatem] is the nurse of
concord, and concord the consetvator of all things (Coimbra In Phys.
17. Cooper writes that for Aristotle, the permanence of animal kinds "meant tlle perma
nence of a set of well-adapted, well-functioning life forms." If, moreover, it is "permanently true
that there are these given kinds of good, well-adapted plants and animals," then any physical
theory must take that into account. Aristotle, Cooper believes, chose to "accept as a fundamen
tal postulate of physical tlleory" tllat the world "so governs itselfas to preseiVe in existence the
species of well-adapted living things that it actually contains" (Cooper 1987:2 51). That sounds
rather like an anima mundi, as if the world itself were an organism. The alternative is to invoke
supernatural powers (271 )-which is, of course, what Christian Aristotelians did.

Authenticated
Vicaria Dei
2cgq 1a1, 1:224; cf. their question on the natural places of the elements,
4csq3al, 2:32ff).
Each particular instance of transmutation could be explained by appealing

to efficient causes only. What requires further explanation are certain
global features-that water and water vapor are in equilibrium, or close to
it, so that the air and water in the world do not eventually all get changed
into earth and fire. In the Aristotelian theory of generation, there is no way
to explain the rate at which one element is changed into another. So unless
one brings in a final cause-the sustenance of life, say-the balance of
reaction rates will be a brute fact.
Modern science has exhibited as much dissatisfaction with brute fact as
Aristotelianism did. The difference lies in the means of domestication.
Where efficient causes fall short, Aristotelianism turns to final causes and, as
we will see, to intelligent agents capable of cognizing ends. Modern science
finds ever more elaborate structures in which to combine efficient causes:
clockworks, feedback mechanisms, neural nets, chaotic systems. Though it
is true that very little of that machinery existed in Descartes's time, it is also
true that the will to devise it may well have required the proscription of ends
in natural philosophy. Although one may defend, on the grounds suggested
byJohn Cooper, the appeal to final causes by Aristotle and the Aristotelians,
one must, I think, also acknowledge that in assuaging the dissatisfaction
aroused by bruteness with the proferring of ends, Aristotelianism did pre
sent an obstacle to the invention of concepts adequate for explaining, in
terms of efficient and material causes alone, those otheiWise brute facts.
4· Usefulness. To our DaiWin-sharpened eyes the most glaring obstacle was
no doubt the explanation of natural phenomena by reference to their
convenience or agreeableness for humans. Yet here the Aristotelian would
have found a happy coincidence of experience and philosophy with re
vealed truth.
Start with the thought that the world is filled with an "inexhaustible
variety" of forms (Coimbra In Phys. 1:224). A world consisting only of the
four elements would be boring but not impossible. Consider also that those
forms are irreducible: the forms of elements cannot be reduced to their
qualities, or the forms of mixtures to the proportions of elements they
contain, or the forms of animals to the dispositions of their bodies. At every
level something new emerges. The four-element world would remain just
that, unless some power introduced into its matter the forms of higher
things. Hence each form bespeaks a distinct act of creation. Why, then,
would God have created so many of them?
For the Aristotelian, part of the answer is that God, given that he created
higher forms, would not have failed to provide them what they need to exist

Authenticated
and endure. In each individual the form is provided, as we have seen, with
the power to produce and maintain the dispositions it requires. Exercising
that power sometimes requires acting on other things, as in eating or
breathing. Those too must be provided. But they must not only exist, they
must be capable of being used: if our food, like Tantalus's, had the power to
evade our grasp every time we reached for it, we would starve. The things an
individual needs to use-where use entails alteration if not corruption of
the thing used-ought to be, therefore, no more powerful than the individ
ual itself. Conversely, what is less powerful is by that fact available. "It is," the
Coimbrans write, "established by the law of nature that things of inferior
grade are rendered to the more excellent, especially if they can sometimes
make use of them" (In Phys. 2cgq2a1 1:328).
That all thi?gs in the world are thus available to humans, and their
existence therefore explicable partly in terms of their utility, is thus only one
case of a truth that applies all the way down: "the form of an element, the
most contemptible of all, is ordained to [the use of] the form of a mixed
body; the form of a mixed body to the vegetative [i.e., the part of the soul
in plants the whole-devoted to nutrition, growth, and generation]; the
vegetative is possessed by the sensible; and this again by the rational soul,
which embraces all ranks and perfection offorms." So "man, by the inborn
right of his nobility, and the prerogative of the more eminent form, sum
mons the whole body of nature, and claims it for himself' (ib.).
Reason here only confirms what was revealed in Genesis. Suarez, in his
commentary on the six days, distinguishes three aspects of the "dominion"
given to men over the rest of creation. 18 Man has by nature the capacity for
dominion; by nature and by right he has power over and use of inferior
creatures. He has the capacity by virtue of having an intellect and a free will.
His physical power over animals consists in his being able to subject animals
to himself and use them either by orders or by force. Even if some animals
resist that power, "by reason, art, and industry, man conquers all [...] : some
he instructs, according to their capacities [...] , some he kills, either for use
or to prevent harm" (De opere 6 dierum 3ct6'l[8, opera 3:279). That physical
power is accompanied by a moral power, a "right" (jus) to exercise it: "Man
has this right, therefore, by virtue of his creation, and it is natural to him
with respect to inferior animals, both because they exist naturally on ac
count of man, and because no harm can be done to them in any use of
them, and so if man's nature alone is considered [i.e. disregarding Original
18. I use the word 'men' deliberately: in every species ofanimal, the male form is, according
to Aristotle, superior to the female; hence even in the state of innocence, Adam had power
over his wife, a power that is "a certain sort ofdominion, and is according to natural right, since
the man is the head of the woman" (Suarez De opere 6 diii1Um 3CI61)7. Opera 3:282; cf.
Ephesians 5).

Authenticated
Vicaria Dei
Sin], it is licit for him to use these animals in any worthy [ honestum] manner"
(De opere 6 dierum 3C16'117, opera 3:280; 'honestus' would exclude bestiality
and animal sacrifice). The argument is not that might makes right; that we
can, in fact, use animals is at best a sign that it is licit to do so. Not our
physical power alone but the possession of free will and the capacity to
deliberate about ends and means make us morally superior.l 9 The physical
power, since it has the exercise of the moral power as its end, is subordinate
to it.
The moral dimension of arguments from utility merits emphasis because
it illustrates the parallel between ends and goods that will be prominent in
the next subsection. The appeal to ends makes easy the transition from what
is to what ought to be, or from what a thing can do to what it may. In natural
philosophy, on the other hand, the explanatory value of such arguments is
slight. Earlier I quoted passages from the Coimbrans answering objections
to the effect that some natural things are useless or obnoxious. The utility
they manage to impute to gnats and vipers is tenuous; it could hardly be the
total final cause of those creatures. God could have found less roundabout
ways of commending himself to us, or of making us exercise our wits. If,
moreover, utility, as opposed to actual use, rests only on the inferiority of
their forms to ours, then everything in the world except angels and other
people has utility. So general an explanation hardly does justice to the
multiplicity of natural forms, or to the workmanship with which each was
fashioned.
I close this discussion with three remarks, one metaphysical, one meth
odological, one historical.
Consider an acorn: it has a form of its own qua seed, but it also is in
potentia an oak. The form of oak is the end to which it tends, and for which it
was produced. Whether oaks actually exist or not, if there are acorns, there
are actual things in which oakness is implied. Or consider the claim that in
order to conserve itself each form must have certain active powers. If that
form exists, those powers are implied in it even if God has miraculously
prevented them from emanating from it. Similarly the end to which those
powers act is given in them even if they never act.
Let us suppose that the end is really distinct from the thing whose end it
is. Since real distinctions hold not only between complete substances but
also between, say, a soul and its powers, the scope of the problem is not
reduced much by that supposition. How, then, can a thing have a property
19. When God said, "Let us make make man in our image and similitude," Soto writes, it
was as if he had said, "Let us make a man who, like us, exhibits judgment, and is capable of
choice, and since he has dominion over himself and his actions, let him rule in the same way
over the animals and other things that lack reason" (Soto De nat. & gratia 1c3; cf. Thomas ST
lptlqg6).

Authenticated
in which is implied the existence of another thing really distinct from it?
Colors, for example, have the power to produce in the eye what were called
"intentional species," which are really distinct from them. Is it then some
how intrinsic to colored things, somehow implied in being a color, that
there should be species and eyes? A. Mark Smith writes that "as visible [...]
every object is defined and restricted by its capacity to express its nature
though color. However, this same expression is equally restricted by what it
redounds on: the visible expressions of things are realizable only in eyes, not
in tongues or ears" (Smith tg81:575). Having the power to be colored in
the presence of eyes is indeed a property, real in the things that have it, but
dependent for its actualization on other things distinct from them, and on
unproblematic relations like propinquity and simultaneity with those
things. It does not quite imply the existence of those other things (save in
the presence of maxims to the effect that no power will forever go unex
ercised), but it does imply, I would say, that there are natures such that if
things with those natures exist, the powers will be manifested. In other
words: no power to be colored without vision and all that vision requires,
even if nothing actually has vision. Similarly, though no doubt less plausibly
by our lights, if lower forms exist on account of higher forms, then the
natures of the higher forms exist. Even if God had wound up creation on the
fifth day, with only the birds and fish, their existence, supposing that they
had the natures they now have, would have implied the natures of the
higher mammals, including humans. God, after all, would not have created
the means to an end without having that end in mind too.
That implication, that palimpsestic foreshadowing of the end in the
means, or of the perfected in the imperfect, is in part what led Descartes to
speak of the Aristotelian notion of potentia as 'confused'. If the natures of
things do imply other natures really distinct from them, then a complete
idea of a nature must include those other natures. Looking ahead to Leib
niz, one might argue that the inclusion in each individual essence of all the
others compossible with it is an implication of the same sort: because there
is but one end-the world as a whole-in God's creative act, each individual
has all the rest as its end. They are implied in it just as the oak's essence is
implied in that of the acorn.
My second methodological remark begins by noting that for the Aristo
telians all of the problems raised here were brought together by them as
instances of the activity of ends in Nature. The fate of those problems in
modern science varies greatly. The coordination of powers in generation is
now explained by theories of the regulation of gene expression and the
gradients that control tissue differentiation. Evolutionary theory has taken
over the explanation of useful behavioral traits in animals. The stability of
the atmosphere is referred to equilibrium principles and to the various

Authenticated
[186] Vicaria Dei
geological and biological processes by which carbon dioxide, for example, is

replenished or removed. The utility of animals and plants to humans falls to
political economy (in its anthropological guise) and the history of
domestication; its moral aspects fall largely outside science. The point is not
just that the problems are treated in distinct disciplines. With due allowance
for the relative absence of specialization, much the same could have been
said in the sixteenth century. It is rather that the modes of reasoning about
the problems have turned out to be heterogeneous, and the unification
achieved by treating them as instances of final causation has turned out to
be spurious. That, and not the almost equally spurious unification achieved
by treating them mechanistically (see Gabbey 1990), is the lasting lagacy of
Descartes and his coconspirators.
My last remark concerns the fading away of the question An natura agat
propterfinem [Whether nature acts on account of an end]. Toletus, Suarez,
and the Coimbrans all devote several pages or more to that question and to
others issuing from it. Later cursus, including those of Eustachius ( 16og),
Abrade Raconis (1617), Hurtado de Mendoza (1624), Arriaga (1632),
Poncius (1642), De Quiros (1666), and Dupasquier (1705), do not have
such a question. In part this may be because the structure of the cursus,
unlike that of the commentary, had no obvious systematic place for it. But I
suspect also that the consideration of the ends of Nature was beginning to
pass over into theology. There it flourished. In Butler's Analogy, the Boyle
Lectures, Derham's Physico-theology, Bernardin de St. Pierre's Harmonies,
and the Bridgewater Treatises written by, among others, Whewell and Bab
bage, the discoveries of the new science were assiduously catalogued into
evidences of the hand of the Creator. Cartesian strictures, as Tocanne notes,
had little effect on that industry. Unlike Descartes, most natural philoso
phers found no repugnance between efficient and final causal
explanations. 20
6.3. Character of the Final Cause

Writing on final causes in Scholasticism, Anneliese Maier concludes that in
Buridan, at least, the efficient cause and the necessity associated with it had
attained a primacy that presaged the eventual eclipse of the final cause in
the seventeenth century.2 1 Ockham had already argued, following Avi
20. Philosophers of the later seventeenth century, "with rare exceptions, were not con
scious of any opposition or conflict between the search for the efficient cause, the necessity of
which was not doubted, and the search for the final cause" (Tocanne 1978:76).
21. Maier 1955, c.5, especiallypp.325rr, and Lang 1989. Lang takes Buridan's "procedure,"
the literary form of his work, to reflect his substantive views: "the procedure of Buridan's

Authenticated
cenna, that the final cause acts only by virtue of existing in the intellect of an
agent; to which Buridan added that when it acts thus, it acts as an efficient
cause, and that where the agent is not such as to conceive the ends by which
it acts, there is no final cause at all, only efficient causes. To the argument
that if there were no ends in nature, then one thing would follow from
another haphazardly, Buridan replies (as we would) that efficient causes
suffice. "The regula," Maier explains, "from which we can read off whether a
natural action turns out talis qualis debet esse [as it should be], is no longer the
goal toward which the process is directed, but the law according to which it
is accomplished and according to which the same cause yields, ceteris paribus,
always the same effect" (Maier 1955:334). In inanimate nature and even in
animals, the necessity that earlier philosophers had referred to the opera
tion of ends is referred by Buridan, if not quite to laws of nature, at least to
the "univocality" of efficient causes.
The central texts present both a retrenchment in favor of more tradi
tional views and an accommodation to the conclusions reached by Buridan.
The case of rational agents, first of all, becomes the only case to which their
account fully applies. Only in rational agents do ends operate straightfor
wardly. Animals are held to be acted upon by ends as rational agents are, but
"imperfectly," because their ability to judge the goodness of things, and to
deliberate about means, is limited. Inanimate things are not acted upon by
ends except insofar as they are the instruments of God.
In the last of these propositions there is a significant step away from
Aristotle and toward Plato. ''The novelty," writes David Balme, "in Aristotle's
theory was his insistence that finality is within nature: it is part of the natural
process, not imposed upon it by an independent agent like Plato's world
soul or Demiourgos" (Balme 1987:275). That innovation is rescinded in
Christian Aristotelianism, no doubt because the Christian God provides a
counterpart to the Demiurge, and because it was fitting that the nonhuman
world should depend on God as an instrument on its maker (§7.2). If they
disagree with Plato, it is in denying that universals exist except in rem, or that
genuine knowledge can only be had of eternal Ideas. But when they assimi
late the inanimate world to a divine artifact, and final causation to inten-
Questiones is sequential: the questions follow one after another without any clear or significant
relation to a larger purpose. In this sense, his procedure is not teleological" (588). I doubt that
any such relation can be borne out. The "disputed questions" and the expositiowere standard
literary forms, used by philosophers of all persuasions; they were not "procedures," if by that
one means actual or ideal methods of inquiry. In fact Thomas was perfectly capable ofwriting a
set of questions with no more stmcture than Buridan's (e.g., the Sentencia de anima), and
Buridan of writing a mnning commentary like Thomas's on the Physics (see Buridan bXp. de
an.). Lang's argument reflects, I think, a rather naive understanding of the relation between
literary convention, individual authorial choice (especially in the Middle Ages), and content.

Authenticated
[188] Vicaria Dei
tional action, they hark back to him even as they unwittingly prepare the way
to the world-machine of Descartes.
Doubts about the existence of ends in nature came largely from outside
Aristotelianism. With few exceptions, Aristotelians agreed that the actions
of rational and irrational agents had ends. But it is one thing to say that an
action has an end, and another to say that it might be caused by its end.22
Here various suspicions had been voiced by Aristotelians from Avicenna
onward. For any genuine cause, it was thought, one must be able to answer a
standard list of questions about its nature, including the causalitas or
"causality" of the cause, its mode of existence, and the objects upon which it
could act. Suspicions surrounding the final cause arose largely from the
difficulty of answering the standard questionnaire, especially when the
agent is "irrational"-inanimate, or animate but inferior to humans. Where
for the other three causes-material, formal, and efficient-there might be
controversy over details, there was broad agreement that the questionnaire
could be sensibly answered. For the final cause there was no agreement.
The difficulties arose in trying to reconcile the character of ends with
three conditions thought to hold of all causes. Every cause is, first of all,
temporally prior to its effects (see, e.g., Eustachius Summa pt2, Physica
1tr2d1q4, 2:138). But in generation, say, the form does not exist until the
action it is supposed to be the final cause of has ceased. More generally, to
be a cause a thing must exist. But the ends to which people act often do not
yet, and sometimes cannot ever, actually exist. A person may strive, Fonseca
writes, "to make the diagonal [of a square] commensurate with its side"; yet
to such a thing real existence is "repugnant" (Fonseca In meta. 5c2q 11 § 1,
2:163bF). The third condition is that a cause be nobler or more perfect
than its effects (Eustachius ib.). But often the ends a thing acts toward are
less noble than the thing itself. Whatever end God acts toward, except
himself, must be less perfect than he is. In particular the world is less perfect
than God; yet it must be the end toward which he acts in creation.
In the answers to the standard questionnaire we find attempts to respond
to these anomalies. The Aristotelians denied, for example, when specifying
the causality of ends, that ends literally act, thus avoiding, or evading, the
first. The second they answered with what at first sight appears to be an ad
hoc broadening of the notion of 'real existence'. The third, and least
difficult, anomaly they handled using the distinction I have already men
tioned between the ends of things and the ends of their operations.
22. "Ut enim notat Gabriel [Biel] [ ...], finis et causa finalis non omnino sunt idem; nam
finis ut sic solum dicit terminum ad quem tendit operatio, vel ad quem motus ordinantur;
causa autem finalis est, qu;e movet agens ad operandum" [As Gabriel Biel observes ( ...) , the
end and the final cause are not entirely the same; 'end' as such means only the terminus toward
which an operation tends, or toward which motus are ordered, while the final cause is what
moves an agent to operate]" (Suarez Disp. 23§9'18, opera 25:884).

Authenticated
Finality and Final Causes [189]
1. Causality of the final cause. To each kind of cause, the Aristotelians

believed, there corresponds a peculiar causalitas, a "formal reason" by which
the mode of causation of each kind differs from the rest. Since every cause is
a "principle that imparts being" (principium quod influit esse: Suarez Disp.
23§9'110, 386), causality in general is "nothing other than the influx, or
concurrence, by which each cause in its kind actually imparts being to its
effect [ actu influit esse in effectum]. "23 The material cause imparts esse, as does
the formal cause, simply by being a component of substance: "matter is like
a beginning, or foundation of being itself, while form consummates and
completes it." The efficient cause is "like a source and principle which per se
imparts esse to its effect." As for the final cause, that's a problem, Suarez
admits, promising to discuss it later. 24 In the disputation on final causes, we
get a precise definition. The causality of the final cause consists in a "meta
phorical motion" of the will; that motion is nothing other than the real
motion of which the will is the efficient cause, and which consists in being
attracted to, or desiring, or loving, the end.2 5
The notion of metaphorical motion was common property. The
Coimbrans write, "there are two kinds of motion, one proper, and which
occurs by a true and genuine action, as when fire heats water; the other
improper and figurative [translatitiam], and of that sort is the motion by
which something is said to move which, by inspiring a love for itself, attracts
and draws to itself [the soul]."26 Why "metaphorical"? Fonseca notes that
23. Suarez Disp. 12§2'113, opera 25:387. I translate injluere by the rather unsatisfactory
'impart'. 'Inflow' (meaning 'cause to flow in') is obsolete (the only quotation in the OED is
Hobbes asking what it means). Eileen O'Neill has, however, revived the word in her useful
suiVey of influx as a model of causation (O'Neill 1993).
24. One problem that is easily solved is how the end can be a "principle." Principles
(Aristotle's apxa[) are supposed to be first, yet the end is last. Suarez's answer, the standard
one, is to affirm that although it is the last "in execution," it is the first "in intention"; it "moves
the agent to act," and is therefore a true principle (389).
25. Suarez Disp. 23§4'l[8-9, Opera 25:861. "Est ergo tertia sententia, qua: constituit etiam
hanc finis causalitatem in motione metaphorica. Addit vero, hujusmodi motionem non poni in
actu secundo, nisi quando voluntas in actu secundo movetur, et quando sic ponitur in re non
esse aliquid distinctum ab ipsomet actu voluntatis": 'There is therefore a third position, which
also takes the causality of the final cause to consist in metaphorical motion. But it adds that
motion of this sort is not posited in a second act, except when the will is moved in a second act,
and when it is posited in the thing it is nothing other than the act of will itself." Mter citing
numerous authorities in support of this sententia, Suarez adds that "null us tamen ita clare et
expresse pra:dictam declaravit sententiam, sicut Ocham, in 2, q. 3, a. 2, ubi ait causationem
finis esse movere efficiens ad agendum [a commonplace]; illud autem movere, non esse aliud
nisi ipsum finem amari ab agente, vel aliquid propter ipsum": "No-one has so clearly and
expressly put fmward the position above as Ockham has in 2, q. 3, a. 2, where he says that the
causation of the end is to move the efficient cause to action; but the term 'move' means
nothing other than that the end is loved by the agent, or something on account of the end."
26. "Ut qua:nam finis caussalitas sit scrutemur, animadvertendum est ex Divo Thoma [ST
1pt2q9a1] & Scoto [In sent. 1d1q4] duplicem esse motionem, unam propriam, qua: fit per
veram, & germanam actionem, ut cum ignis aquam calefacit; aliam impropriam, & trans
latitiam, cuiusmodi est ea, qua movere dicitur id, quod sui amorem iniiciendo allit, trahitque

Authenticated
[I go] Vicaria Dei
some ends might well also act on an agent as efficient causes. Food will
attract an animal by its smell and look. But many ends have no efficacy and
yet are capable of moving an efficient cause to act: "in such a way a lower
place draws a stone out of a higher, though it impresses nothing in it, and
most fictitious things, which have per se no effective power, move many
agents to obtain them" (Fonseca In meta. 5c2q 10§4, 2: 16obC). Yet there is,
at least in rational agents, a genuine motuswhose reason is given by the end.
The will desires in potentia whatever the intellect is capable of presenting to
it as an end. To desire an end in actu is therefore the actualization of potentia,
or in other words a motus. The will itself is the efficient cause of that motus.
But since the will is the efficient cause of all its motus, particular determina
tions of the will must have an additional reason-the end they are directed
toward, or that ceteris parilnts they would effect. In that sense, acts of the will
depend not only on the will but on ends; metaphorical motio is that
dependence.27
Speaking of the action of ends as metaphorical opens one avenue to
solving the problem of the nonexistence of ends. If one could make sense of
metaphorical action, perhaps one could make for them a reasoned excep
tion to the rule that causes must exist to act. It also allows one to speak of
items otherwise devoid of action, like places, as having effects. What it does
not do, despite Fonseca's unfortunate example, is allow one to explain the
action of ends on inanimate things. Metaphorical motion, whatever it may
be, moves only the will. Suarez explicitly restricts his discussion to the action
of ends on the wills of created agents, thereby excluding God (who cannot
be acted upon even metaphorically) and inanimates. The Coimbrans, con
fronting the objection that "even things that lack cognition operate on
account of ends [...] and yet they are not attracted to an end" (In Phys.
2c7q21a1, 1:308), reply that "natural agents [...] are not attracted per seto
the end, nor do they strive toward it per se, but only insofar as they are
directed by the first cause, in which their ends, and the knowledge and love
[of those ends] pre-exist" (a3, 1:311). A natural agent, a stone say, is "at
tracted" to the ground not per se, not considered in itself, but only in
relation to God, who has caused it to be endowed with a quality that will, as
an efficient cause, impel it downward. It "strives" toward the ground only per
accidens, as a key is said per accidens to turn a lock.
We may find the talk of metaphorical motion obfuscating. The real story,
we think, is given in the example of an animal's attraction to food: an end
ad sese" (Coimbra In Phys. 2qq21, 1:307). Eustachius plagiarizes this passage (Summa pt2,
Physica 1tr2d2q6, 2:143).
27. Conversely, when there is no actualization of a potentia, there is no motio, even meta
phorical: such is the case for divine acts. Suarez writes that in such instances "the whole
manner of speaking is metaphorical" (Disp. 23§9'16, opera 25:883).

Authenticated
attracts by efficiently causing a desire that in tum efficiently causes actions

that will, if the world cooperates, result in the end's being attained. Since
the Aristotelians themselves agree that the cognition of an end is a conditio
sine qua non for its being a cause, the anti-Aristotelian need only argue that
the cognition is notjust a condition of, but the cause of the volition that was
said to result from the metaphorical movement of the will by ends, and that
the cognition of an end will have been efficiently caused by the end.
Though there is still reason to talk of ends, final causality is eliminated in
favor of efficient causality.
2. Mode ofexistence ofthefinal cause. The accountjust proposed was not only
known to the Aristotelians, it was rejected by them. Avicenna and Soncinas
had held that an end acts only "according to the existence it has in the soul,
not outside it" (Soncinas Q meta. 5q3, 55a). A standard way of stating the
view was to distinguish two modes of existence: esse intentionate or objectivum,
the mode of existence things have when they are objects of thought, and esse
reate, or existence independent of thought or of the soul. 28 The claim, then,
is that the end is a cause with respect to esse intentionate, not with respect to
esse reate. At least one earlier philosopher, Guido Terreni, had gone on to say
that the end, existing only in intention, acts by efficiently causing a desire or
appetite.29 But Soncinas did not, and the central texts do not.
Instead they insist that the end causes according to esse reate. Only esse
reate, and not esse intentionate, satisfies three criteria. It alone fulfills the
condition of being that on account of which an agent acts. The esse reate of
the end, or its absence, determines whether the act is successful or frus
trated. The means are chosen not to accomplish the esse intentionate of the
end, but its esse reate. A doctor who by hypnosis succeeded only in making his
patients think they were cured would be a quack. so
Esse intentionale, then, is not the esse according to which an end acts. But
being thought of is a conditio sine qua non for final causality. Even God by his
absolute power, Suarez argues, could not bring it about that the will should
be drawn toward an unthought end (Disp. 23§7!6, opera 25:876). The
28. "Scholastici Ens Intentionale appellant Ens, quod sola intellectus conceptione & con
sideratione inest; seu Ens quod est intra animam per notiones [...) , cui oppositur Reale, quod
reperitur extra anim<e notiones" (Goclenius Lexicon s.v. 'Intentionale', p256; cf. also 157). Ens
(esse) intentionale and objectivum are used interchangeably.
29. In a Q}todlibetum of 1315, Terreni argued that the "the causality of the end with respect
to the movement ofthe agent by an appetite of the soul is efficient causality" (Maier 1955:286,
citing Terreni's Q}todlibeta 3q2). Efficient causality, he says, is that "by whose presence and
motion what was in potentia is now formally in actu." But the end is "that by whose presence [in
apprehension] a certain appetite that was in potentia is now in actu." It is thus the efficient cause
of that appetite. I should note that none of the central texts cites Terreni or explicitly states his
view.
30. "A sick person does not desire the esse intentionale of health, which is long since awaited
in his soul, nor does he take walks and abstain from food for the sake of it, but for the sake of
the esse reate of health" (Coimbra In Phys. 2c7q23a2, 1:316).

Authenticated
Vicaria Dei
metaphorical motion of an end on the will is founded, he writes, "on a

natural agreement and sympathy of the intellect and the will," in which it is
necessary that the cognition of the end should precede the inclining of the
will toward it ('l[2, 875). Suarez and the Coimbrans both compare the
"proximity" of the intellect and will to the spatial proximity necessary in
efficient causation: 'just as in the efficient cause local proximity is a require
ment, so in the final cause proximity in the soul, vital proximity [approx
imatio quasi animalis, seu vitalis] is a requirement" (Suarez Disp. 23§8'l[ 10,
opera 25:881; cf. Coimbra In Phys. 2c7q23a2, 1:315). Vital proximity is not
merely the joint presence of two powers in one individual. It is, rather, that
each is necessary to the other's operation and part of its reason for being.
Intellect without will would be gratuitous, but so too will without intellect. 31
The analogy brings out what is most plausible in the idea of metaphorical
motion, and helps make clear why final causality could not be reduced to
efficient causality. Suarez, noting that the mere words 'metaphorical mo
tion' do not clarify the nature of final causality, explains it thus: "by the very
fact that the goodness of the end is sufficiently understood and proposed to
the will, it excites [the will], and according to its force, draws the will to
desire it [...] [This relation between cognitions of goodness and acts of
will] seems to be established in experience, and its foundation seems to rest
on a certain natural sympathy between the intellect and the will, insofar as
they are rooted in the same essence of the soul. "32 The understanding that a
thing is good just does, as a matter of fact, incline the will toward it. But it is
not the thought itself that inclines the will. It is, rather, the goodness of the
thing understood. Suarez, as I said, identifies the metaphorical motion of
the end with the real motion by which the will produces particular desires.
They are the same instance of "causing," viewed with respect first to final
and then to efficient causation. Their joint effect-the particular desire,
31. In arguing that goodness is the formal reason of the final cause, and thus that what is
bad cannot be willed as bad, Suarez writes that for the will to be moved by the bad as bad would
be "repugnant to [the will] itself': "The end or institution of the will is that through it man
seeks what is agreeable to himself, and flees what is disagreeable; if [the will] were to accept an
inclination tending to the disagreeable insofar as it is disagreeable, it would formally and
directly contradict itself and its end" (Suarez Disp. 23§5i4, opera 25:865). The will can intend
an evil under the false impression that it is good, but not an evil as such, no more than the
intellect can assent to falsehood as such. Suarez's argument, I should note, is teleological. A
similar argument could be made to show why the intellect and will are in "vital" proximity.
Earlier Suarez speaks of the judgment of the intellect as "bringing the end close enough that it
can cause" the will to act ( 25:86o). Such ways of talking, I take it, are intended to relieve the
discomfort one might feel if one thought that ends would sometimes have to act at a distance.
The recognition of an end, whether through sensation, imagination, or pure understanding,
gives it an existence in the soul; and then sympathy brings it the rest of the way.
32. Suarez Disp. 23§41!4, opera 25:86o; cf. also §s1 14, 867. Suarez eventually rejects the
particular explanation of metaphorical motion that this passage is in aid of. But he does not
reject the description itself.

Authenticated
directed toward the end-is thus immediately caused by the end, under the
formal reason of goodness.
All this only makes more urgent the problem of nonexistent ends. If an
end acts according to real existence, and under the formal reason of good
ness, then how could anyone want jam tomorrow or try to square the circle
or wish (as those in hell might) that they didn't exist? The answer is at first
sight disappointing: "Esse reale is taken in two ways: in one way for that which
is a true and positive being, and not dependent on the notions of the
intellect; in the other, so broadly that it embraces any being that in some way
is possible to obtain or at least is presented to appetite as possible or not
impossible" (Coimbra In Phys. 2c7q23a2, 1:315). Fonseca, too, writes that
the term should be understood to mean what is possible to be obtained, or
seems so, "even if it is not truly real, but a being of reason," like beingjudged
great, "or even negative," as when someone wretched wishes to die, or
impossible (though not seeming so). 33 Though there are ways to treat some
of these cases without stretching esse reale, 34 the Aristotelians acknowledge
that esse reale must include future existence, potential or possible exis
tence, and-most striking-apparently possible (but in fact impossible)
existence.
The best way to understand their thinking is to remember that being real
does not entail being actual. 'Real', in the present context, denotes what
exists in re, as contrasted with what exists only in anima. Jam tomorrow, or
Quine's possible fat man in the doorway, are real insofar as their being what
they are does not depend on whether, or how, they are thought about.
'Mind-independent' is the closest current term. In this respect the Aristo
telians were thoroughgoing realists: future things, potential things, possible
things, all have real though not actual existence.
But not so thoroughgoing as to admit impossibilia. Here they insist only
that what we pursue we pursue as if it had real existence and goodness. A
33· "Adverte tamen, nomine esse realis veri aut apparentis, intelligendum esse in hac materia,
esse possibile obtineri, aut quod obtineri posse videatur; etiam si illud non sit vere reale, sed
rationis, quale est concipi ab aliis, magni <estimari, et similia: aut etiam negativum, quale est
miseriarum, ut ita dicam carentia, propter quam a plerisque mors eligitur. [...] Illud etiam
adverte, possibile tam late quoque intellignedum hie esse, ut etiam pro eo accipi debeat, quod
non videtur impossibile. Multa enim intendimus, de quibus dubitamus, num re vera obtineri
queant, ut si quis dubiter, num possibile sit circulum quadrare: & tamen conetur eius quadra
tionem invenire" (Fonseca In meta. 5c2q11 §4, 2: 16gD-F). Suarez writes that "neither genuine
essence nor possible existence [ neque esse essentim verum, aut esse existentim possibile] is necessary
to final causation; apprehended being [esse apprehensum] suffices" (Disp. 23§8'17. opera
25:88o).
34· The Coimbrans handle negative ends by arguing that every privation or negation (like
wanting to be hungry or to cease to be) is desired only per accidens, as a consequence ofdesiring
something positive. The damned wish not to exist in order to avoid eternal torment (In Phys.
1:316). An end, therefore, can incline the will either toward itself (if it is apprehended as good)
or away (if it is apprehended as bad).

Authenticated
Vicaria Dei
person who attempts, as Hobbes did, to square the circle is moved not by the
concept itself, or by the existence of a squared circle in thought (which he
already has achieved), but by what we would call the content of that thought,
the squared circle conceived, though falsely, as possible.35 Although the
squared circle may not have esse reate, it acts as a final cause secundum esse
reate.
Since the ends to which irrational agents act are at least possible, I will not
develop the point further. What I want to retain is the thought that ends,
even in the actions of rational agents, act immediately as ends, under the
formal reason of goodness. Cognition of the end is a necessary condition
and indeed absolutely, not just naturally, necessary. But just as the local
proximity necessary to efficient causation does not constitute the imme
diacy with which an efficient cause operates, so too the "vital proximity"
introduced by cognition does not constitute immediacy in the operation of
a final cause.
3· Objects offinal causation. The object of a cause is the thing on which it
acts. To include created rational agents-human, demons, and angels
among the objects of final causation was clearly not subject to debate. They
were the paradigm. Yet we have also seen that the actions, powers, and forms
of irrational agents are continually explained in terms of their ends or the
ends of Nature. In reading questions on the final cause, one is struck by the
slippage between the practice of teleological explanation and the theory of
that practice. The one occurs everywhere, the other seems to allow only for
final causation in finite creatures with intellect and will. The impression is
inescapable that while the ascription of ends may be going strong, their
place among causes has become, as already in Buridan, tenuous everywhere
except in human action.
Setting aside the case of God, there are essentially two kinds of objects to
consider: inanimate things and bruta, or animals as distinguished from
plants. The treatment in each case is what one would expect if ends could
only act on rational agents. Inanimate things and their actions have ends
only at second hand, by virtue of being God's instruments. The actions of
bruta are the objects of final causation only to the extent that they possess
something akin to reason. The Coimbrans grant that animals give the ap
pearance of judging ends and deliberating on means, but they do so by
sheer instinct. Even those who are inclined to a generous estimate of their
capacities, like Suarez, deny that animals recognize the good as good; they
are moved without knowing why.
The Coimbrans distinguish three "grades" of tendency to ends. The first is
35· Fonseca writes that if someone believes it possible that the side of a square should be
commensurate with its diagonal, and strives to demonstrate it, what moves him is not "ipsa
conceptio talis commensurationis, ut possibilis" but rather ~ipsa commensuratio concepta, ut
possibilis." Not the concept, but what is conceived, is the final cause (In meta. 2: 17oA).

Authenticated
that of rational agents, "who not only apprehend an end under the reason
of the good and the agreeable, but also recognize the aptness [habitudinem]
and proportion of the means for obtaining that end." Such agents alone
order themselves toward ends. The second grade is "of those who at most
perceive the end materially, that is, under the reason of the good and the
agreeable," but who neither discern the means nor order themselves to
their ends. Such are animals. The third and lowest grade is of those things
"that in no way recognize ends," because they lack both intellect and sense
(Coimbra In Phys. 2cgq2a2, 1:329). 36 Such things were sometimes called
natural agents.
For natural agents, the only way of being acted upon by ends is to be
"directed toward an end by some superior and more excellent cause, that is,
by the artisan of Nature herself, who comprehends the ends of all things,
and gives [to them] the propensity and powers to obtain those ends." The
work of nature is the work of an intelligence, "moved by divine art and
directed by ingenious reason." As the archer impresses upon an arrow the
impetus by which it attains the target, Toletus writes, using a figure bor
rowed from Thomas, "so glorious God has given to all things a nature by
which even without foreseeing their ends they are led unerringly to
them. "3 7 Suarez, citing the same illustration, makes the point more pre
cisely. In the actions of natural agents, "there is no final causality, properly
speaking, but only a tendency [habitudo] toward a certain end [...] The
adequate principle of these actions is notjust the proximate natural agent,
unless perhaps secundum quid [i.e., with respect to the order of efficient
causes] [...], but absolutely the chief [principle] is the first cause, and so
an adequate principle of such actions includes an intellectual cause intend
ing their ends" (Disp. 23§10~6, Opera 25:887). The analogy is familiar, its
advantages numerous. An instrument clearly need not know the ends to
which it is designed or used in order to be said in some sense to have those
ends. God clearly has the power and the knowledge by which to order things
to whatever ends he chooses. Since he is the rational agent par excellence,
there would seem to be no doubt that ends could act on him by metaphori
cal motion. More generally, an artifactual world is the natural counterpart to
a creator God, and evidently dependent on him not only in existence but in
essence.
Yet the analogy raises serious problems. One concerns the action of ends
on God. Although metaphorical motion is clearly not, say, alteration or
change of place, it is, all the same, a change in the thing moved: an "inclin
ing" or "drawing" of the will toward an end (cf. n.27 above). Yet God, being
36. See also Suarez Disp. 23§101 131r, Opera 25:589; Toletus In Phys. 2cgq12, opera 4:74v.
The three grades are found already in Thomas ST 1pt2q1a2 and q6a2, Parma 2:2, 30.
37· In Phys. 2q12, opera 4:75ra; cf. Thomas ST 1puqw3a1, Parma 1:395.

Authenticated
[1g6] Vicaria Dei
necessarily entirely in actu, can receive no impression from without. The

answer, in brief, amounts to accepting the objection. Divine acts of will are
not caused by ends, or indeed by anything, because they are nothing other
than God's will itself, which is entirely in actu. Metaphorical motion, which is
indeed the actualization of a potential to desire, applies only to created
rational agents.38
The second problem is more intractable. When Aristotle defines nature,
he mentions artifacts only to deny that they have natures (see §7.2). That
argument is in part an argument against Plato on behalf of the independent
reality and knowability of the natural world. The possibility of an episteme of
Nature qua Nature rests on finding the principles of natural change within
Nature itself, including the natures of individual things. Yet Suarez holds
that an adequate principle can be found only in God.
One easy reply is that Aristotle was wrong. There are actions that cannot
be sufficiently accounted for by reference "to the private properties or
inclinations of particular things." The horror vacui or the confinement of the
sea, Suarez argues, cannot be accounted for by the peculiar nature of water,
but only by "the end, which rests in the perfection of the whole universe."
That end "must be intended by a superior agency" (Disp. 23§ 10110, opera
25:888). If Aristotle did not acknowledge the existence of cosmic ends that
could not be accomplished incidentally in the pursuit of private ends, then
his physics was incomplete.
But the reply, even ifjust, only accentuates the difficulty. A more cogent
reply will show that some instruments, at least, inherit the ends of their
users. Part of the reply will turn on the differences between natural and
artificial instruments, or equivalently on the differences between human
industry and divine creation. That part, I defer to §7.2. Here I consider only
the relation of means to ends, and to final causes, in general.
The question, the,n, is whether means can really be said to share the ends
to which their users put them, or for which they make them. 39 Suarez,
arguing that all actions and effects of the will are effects of the final cause,
writes that "when by these actions there is produced a terminus that has an
enduring existence [terminus permanens in facto esse], this also is held to be an
effect of the end conceived beforehand, whether in becoming as it is actu
ally made or in existence in fact while it endures afterward. For that reason
g8. Suarez Disp. 23§g'l[g 1r, opera 25:882 1; cf. Coimbra In Phys. 2q q21a2, 1:gog.
39· Philosophers now ask an analogous question about intentionality. Some objects, like
thoughts, have "original" intentionality, while others, like written or spoken words, have only
"derived" intentionality. John Searle, for example, insists that no computer yet made has
original intentionality. The physical states of the central processor or of the video display have
intentional content only by virtue of being our instruments. In Aristotelian terms, they have
intentio only denominatively, as when we call a caricature cruel.

Authenticated
Aristotle said that instruments exist on account of an end, and likewise

houses and other artifacts are effects of some end conceived beforehand"
(Suarez Disp. 23§3~19, opera 25:857). That seems clear enough. Though
the immediate effect of an end, as we have seen, on created agents is the
metaphorical motion of the will, subsequent effects too are obviously on
account of that same end. Yet he then argues that the causality of the end
with respect to external actions and things is extrinsic and denominative:
"But acts merely commanded by the will (and all the more so their effects)
in no way are elicited immediately by the end itself, nor are they said to have
an intrinsic aptitude toward the end, but are said to be ordered to the end
only by an extrinsic denomination by way of interior acts [ofwill], as walking
is merely extrinsically ordered toward health" (~23, 858). Mter all, Suarez
notes, if walking or some other exterior act (or, presumably, an artifact
produced by such an act) occurred "on account of some other end, or by
chance," "in itself and in its being [entitate] it would not be transmuted or
have anything taken from it." Walking is walking, whether you walk for
health or to buy cigarettes. The exterior act is therefore not caused per se by
the end, but only per accidens, as a falling stone pushes air aside on its way
to the ground.
Arriaga makes the same point. Considered in relation to a given end, a
means may be said to be useful intrinsically, since its usefulness consists in its
being genuinely effective in bringing about that end. Cutting a vein is useful
to health because of its nature. But ifwe do not consider any end in particu
lar, then cutting a vein is neither useful nor useless. Means, in short, are
called useful or not only by "extrinsic denomination" (Arriaga Cursus 353).
Applying these conclusions to natural things, regarded as divine instru
ments, we would have to say that they too have their ends only extrinsically,
and are caused by them only per accidens. Suarez and the rest agree, as we
have seen, that even if God were as impassive as the gods of Lucretius,
particular efficient causes would not cease to have the effects they now have.
Suarez argues, moreover, that "if from the mutation of the cause, neither
the effect nor the action is changed, that is a sign that the cause does not
influence the effect [injluere in effectum] either immediately or per se" (858).
Natural agents, insofar as they may be characterized in terms of efficient
causes and effects alone, would not differ in the Lucretian world; nor are
they caused by divine ends immediately or per se.
A disturbing outcome, if you're an Aristotelian. The means of palliating it
will be found, I think, only by examining the differences between nature
and art (§7.3). Here I will just mention briefly one other way out. The
argument proves only that if there were a Lucretian world containing indi
viduals of the same sort that now exist, it would (setting aside collective and

Authenticated
Vicaria Dei
cosmic ends) precisely resemble the actual world. The Aristotelian may still
argue, then, that although the actions of individual natural agents do not
have intrinsic ends, the natures of those individuals do. Only if-and this
would have seemed quite unlikely, as well as unorthodox, to an
Aristotelian-natural kinds could be shown to emerge through the opera
tion of natural efficient causes alone, would there be reason to doubt that
natural kinds and their instances had intrinsic ends. We have seen that in
fact the Aristotelians believed that sublunary agents were incapable even of
reproducing their forms; still less would they be capable of creating new
ones. The case for appealing to ends in natural philosophy, therefore, rests
not on their being needed to explain particular events, but on their be
ing needed to explain the existence of kinds. Cosmic and collective ends,
not particular ends, will be the primary instances of finality in natural
agents.
With animals, the stakes are quite different. That their activities were
guided by ends, and that they to some degree recognize those ends, was
amply confirmed in experience. Though there were attempts, which I will
examine shortly, to assimilate the causes of their actions to those operating
in inanimate things, the dominant view was that efficient causes alone
would not suffice. The point of dissension was instead whether their actions
were guided, like ours, by a rational intellect in proximity to a free will, or by
innate instinct.
Instinct is not itself a final cause but a means, analogous to metaphorical
motion in rational agents, by which ends could incite, via sense and imagina
tion or phantasia, the motive powers of the soul into action. Even in humans,
all actions that we accomplish "without the command or movement of the
will" occur either in the manner of inanimate agents or by instinct (Suarez
Disp. 23§34][18 and §10'1)5, opera 25:856, 88g-8go). What interests me
here is not the psychology of instinct, but rather the reasons given for
introducing a third way for final causes to operate, distinct both from meta
phorical motion and from the instrumentality of inanimate agents.
Suarez notes that certain unnamed philosophers, in their eagerness to
avoid attributing rational souls to animals, have gone so far as to deny that
animals recognize their ends at all. Instead these philosophers hold that
their actions are brought about either by internal powers analogous to
weight, or by external powers analogous to magnetism. Suarez immediately
dispatches the view with the comment that it is "absurd, contrary to evident
experience, and indeed to divine Scripture" (88g).
Nevertheless Buridan came close to proposing just that view. The actions
of animals depend entirely on the efficient causal powers of their own
natures, the heavens, and God:

Authenticated
Finality and Final Causes (Igg]
Concerning natural things I believe that the swallow, when it mates, nests,
and lays eggs, no more thinks of the young which are to be generated than
a tree, when it leafs and flowers, thinks of its fruit. Nor do the mating,
nesting, and egg-laying of the swallow depend in their existence and order
on the young. On the contrary, the young do not determine the swallow to
operate thus; rather the form and nature of the swallow and celestial
bodies at the appointed times and God by his infinite wisdom determine
the swallow to mate, and from that follows the generation of eggs, and
then, when the swallow is so disposed by its nature together with celestial
bodies and God, all of them determine it to nest-building and then to egg
laying [...] All these [effects] issue from divine art and celestial bodies
and particular agents, both extrinsic and intrinsic (like the substantial
forms of natural things themselves). (Buridan In Phys. 2q13, p4orb; cf.
Maier 1955:326)
Drawing on the familiar maxim that what is first and better known to nature
is later and less well known to us, Buridan explains that although we know
and intend the ulterior consequences of our immediate actions, in nature
"there is no intention or appetite [...] aside from a potentia and determina
tion to produce an effect"-namely the first effect, which is naturally bound
to follow, and from which all later effects will come.
The Coimbrans' story differs only a little. Plants, which no one credits
with knowledge of their ends, generate leaves and fruit as if they foresaw the
needs of their offspring. The swallow, a stock example in these discussions
(see Phys. 2c8, 1gga26), "looks at [intuetur] mud; an image of the mud
enters its imagination [phantasiam]; then judging it useful the swallow picks
up the mud and carries it to this or that place," and so forth. In the progres
sion of actions, each ofwhich-we might say-provides the stimulus for the
next, ·~udging" amounts only to a determinate action of the image on the
motive power of the swallow, "a certain imitation of .[human] judgment,
which animals exercise in a crude and simple act,"without the operations of
composition and division characteristic of rational thought (Coimbra In
Phys. 2cgq4a2, 337-338).
Suarez, on the other hand, grants that animals recognize the ends to
which they act. But only "materially," not "formally": they do not recognize
the "formal reason of agreeableness or utility, and they are not so moved
that they could order one thing to another, or desire a thing formally as
lovable in itself."40 The swallow sees mud, and is inclined, willy-nilly, to pick
40. SuarezDisp. 23§101 15, opera 25:88g, citing Thomas ST1pt2q1a2 andq6a2 (Parma 2:2,
go); see also tpuqsgat. Earlier the Coimbrans make the same threefold distinction (Coimbra
InPhys. 2cgq2a2, 1:329), but in their question on instinct they do not refer to it. Their primary
interest is in refuting the claim, which they attribute to Jean Gerson, that instinct is a "special
motion, by which God concurs with them in their works" (2cgq4a1 and g, t:gg6, 338).

Authenticated
[200] Vicaria Dei
it up. The mud is in fact useful for nest-building; but even though the
swallow does use it to that end, it doesn't form the thought that the mud is
useful, or choose it rather than some other building material.
Two conclusions may be drawn from the discussion of lnuta and natural
agents. The first is that everything except rational agents, and in a quite
limited way, animals, is acted on by ends only as an instrument of God.
Finality occurs everywhere, final causes only in nature's highest forms. The
second is that final causality is, for the Aristotelians, always dependent on
rational cognition, human, angelic, or divine. It follows that the Aristo
telians are not open to the charge of animism: that charge rests on a misun
derstanding of the role of analogy in their natural philosophy. 41 It follows
also, however, that knowledge of the ends of nature is not, as it would have
been for Aristotle, a knowledge of one sort of intrinsically natural property
among others; it is instead knowledge, however deferred and uncertain, of
the mind of God.
6-4- Teleological Reasoning

In this subsection I will detail three instances of teleological argument: the
slogan 'Matter desires form'; natural limits on size; and the production of
monsters. The first two are straightfmward enough; the last requires a bit of
preamble. Monsters themselves, or rather monstrous forms, are not them
selves ends, unless perhaps of Nature or of God. Like other chance events,
they are the incidental outcomes of natural changes whose ends are quite
different. Part of the point in treating them is to delineate the limits of
finality, as a prelude to discussing natural and violent change in §7; the
other part is to illustrate in some detail the conjoint roles of material,
efficient, and final causes.
Together the examples illustrate three points. First, in natural agents, the
Aristotelians were inclined to treat tendencies to form as nothing other than
the dispositiones or proportiones of proximate matter which were discussed in
§5·3· That is not to deny that those agents have ends. The term dispositio is
itself charged with finality. Rather it is to say that 'having such and such an
end' and 'having such and such a material constitution' are alternative
41. Mter quoting Thomas's example of the arrow tending toward its target (n.37 above),
Weisheipl writes that "the scholastic terminology was commonly attacked in the seventeenth
century [... ] as the expression of animism and anthropomorphism; this was due to a miscon
ception of analogical usage-a human necessity" (Weisheipl1g85:2 2, n.gs). Descartes did not
deny that analogies were useful and even necessary to physics (see, e.g., Principles 4§201 and
203, AT 8 1 :324, 326, 9:319, 321). That was not the issue; the issue was which analogies. But I
agree that Descartes's serious misreading of his predecessors' views indicates that, among
other things, he mistook the role of intentional idioms in their physical reasoning.

Authenticated
descriptions of the same accidents, one with respect to form as their normal
effect, the other with respect to the combination of elemental and derived
qualities which has such effects.
Second, cosmic ends like that implied in the slogan 'Nature does nothing
in vain' are in certain respects functionally equivalent to laws in Cartesian
science. Not only do they supply the warrant for inferences from what has
been observed to occur to what could or could not occur; they also find a
similar ground in the distinction between the absolute and ordained powers
of God.
Finally, in examining responses to the question ofwhether nature intends
monsters, I will uncover an interesting disparity between Toletus and the
Coimbrans. For the Coimbrans, the role of monsters is the traditional one of
adding variety and thus beauty to the world; for Toletus, they emerge inev
itably from the proportioning of God's concurrence with second causes
according to their strength. The difference, I think, is significant: Toletus's
account not only avoids appeal to an aesthetic motive, but also permits an
explanation of monsters to be given, if we can measure the "strength" of
causes, in entirely physical terms.
1. Matter desiresform. In this slogan from Physics !we see flagrant misuse of
intentional idiom, and gross confusion between soul and body. The passage
in Aristotle reads: "[Matter] desires [form] just as the female [desires] the
male, or ugliness beauty. But [matter] is neither ugly nor female per se, but
rather per accidens" (Phys. 192a22ff). 42 The immediate context is an argu
ment that matter and privation are distinct, or, equivalently, that matter is
notjust nonbeing. Form, Aristotle argues, does not desire itself, since it itself
is the desirable; privation could not desire form, since that would amount to
desiring its own destruction. That, the Coimbrans say, is "most alien to the
laws of nature" [quod anaturce legibus quam maxime est alienum]-one of the
infrequent occurrences of that phrase (In Phys. 1: 149). Hence what desires
form is neither form nor privation but a third principle, namely, matter.
The Coimbrans' explanation provides ample pretext for the accusation of
animism. Beauty, they say, is "nothing other than the splendor ofa thing, the
sight of which attracts the soul": so Plato has defined it. But matter alone,
being nuda potentia, has no power to produce accidents by which to be seen.
Only when it has received form and with it color and other visible qualities,
can matter be beautiful. On that account it desires form. Matter, moreover,
can assimilate itself to the divine, from which it is otherwise "most distant,"
42. "Sed hoc, est materia, perinde appetens illud ac si marem foemina, & quod turpe est,
appetat pulchrum. At nee turpe, nee foemina per se, sed ex accidenti" (Coimbra In Phys.
1:148). Other translations differ, but not materially (cf., e.g., Albert In phys. 1tr3CI6, Opera
4.1:71). Interestingly enough, all the commentators ignore the analogy with female desire,
preferring to talk only about beauty.

Authenticated
[202] Vicaria Dei
only by receiving form, and so too the actuality and activity of that form. On
that account also matter desires form.
Matter, it would seem, can envision the benefits of form, and though it
cannot act on its desire, it can (according to Scotus) be called 'satisfied'
when that desire is fulfilled. But the Coimbrans know as well as Descartes
that matter-and a fortiori prime matter-cannot literally be said to desire
anything. The appetite of matter for form, they remark, following Thomas,
is "nothing other than the inclination of matter to receive form," an inclina
tion "deeply rooted in matter," and in fact not distinct from it (In Phys.
1cgqsa1, 1:16o). Toletus, more precise on this point, distinguishes ap
petites into intellectual, sensitive, and natural. Natural appetites are the
inclinations of things that can recognize their ends neither formally, as we
do, nor materially, as animals do. They are nothing more than the propen
sities toward the good which God, who knows the ends of all things, has
given to them. Matter in particular is "proportioned" to receive all forms;
and by virtue of that can be said, even when it already has one, to desire
those it does not have (Toletus In Phys. IC7q17, opera 4:38ra).
Whatever Aristotle may have meant by his dictum, it is clear that the
Aristotelians do not literally attribute desire to matter. "When we say that
earth desires the lowest place, or matter desires form [... ] , this is true only
according to a transferred usage [per translationem]." Of such things it is true
only that it is as if they had appetites in the proper sense (Fonseca In meta.
Iciq1§3, 1:62aE). The intentional language picks out a class of
properties-which we still call propensities or inclinations or tendencies
by virtue of their analogy with a part of the class intimately known to us
our own, intellectually guided tendencies. The phenomena thus picked out
in natural agents can also be described in the more neutral vocabulary of
proportio and dispositio. But the Aristotelians, unlike some recent philoso
phers, feel no obligation to give up the analogical, "transferred," way of
speaking. The analogy is stronger for the Aristotelians than for us because
they do not hesitate to call that which a thing by nature inclines toward its
good, whether that thing is human or not. To every appetite, moreover, there
corresponds a state that may be called, properly or by translatio, the "satisfac
tion" of that appetite, and whose outward sign is the spontaneous cessation
of motus. When that state is not reached, words like 'frustration' or 'miscar
riage' may be used, whether the agent is rational or not (Phys. 1ggb; cf.
Coimbra In Phys. 1:320, 339). In those respects, then, the use of intentional
language to refer to features of the inanimate world highlights a genuinely
common pattern of behavior. Its use would be dangerous only if one some
how forgot the all-important differences between· rational and natural
agents: that we know the good as good, that we have the power to refrain
from following our inclinations.

Authenticated
2. Limits on size. A question on the upper and lower limits to size in

material substance seems to have become a set piece in Physics commen
taries of the sixteenth century, only to disappear in the seventeenth. I will
examine only part of the question, that having to do with limits on the size
of living things.
The Coimbrans' argument, once one sorts out the logical distinctions
they make between upper (or lower) bounds and least upper (or greatest
lower) bounds, comes to this. Every animal has a definite essence and
perfection; its nutritive faculty, which controls its growth, does too; hence
there is an upper bound. As for lower bounds, every soul requires organs to
carry out its functions, and "the consequence is that these are determinate,
and so the soul also needs a definite minimum quantity." The arguments, in
other words, point out that among the perfections of the soul is that of
having a quantity within certain bounds. 4 3 But inanimate things are
thought not to have upper bounds, and yet they too have their perfections.
So the question at hand seems begged.
Toletus does better (In Phys. 1qqg, opera 4:24vb). Mter citing au
thorities, he argues first that the manner by which living things grow im
poses limits on their size. The "instrumental cause" of growth is "natural
heat." Natural heat is necessary to the parts of the body; as it is lost, it must
be restored by the vital heat that, generated in digestion, is distributed to
the rest of the body by the liver. The larger the parts, the weaker and less
intense the vital heat will be in each, and so at some point more heat will be
lost than gained. Hence there is, for each species, a definite upper bound
on size.
The argument has a nice feel to it, perhaps because it compares quan
tities and appeals to efficient causes only. In the background, however, there
is a teleological principle. It is logically possible that a soul should, so to
speak, be programmed to grow so large that it will inevitably die. But
nothing-to cite a maxim mentioned a few pages ago-can naturally tend
to its own destruction. Such a soul cannot naturally exist.
More explicit is the finality appealed to in a later argument: "that nature
has never done a certain thing up to now is a sign that it cannot; but no one
has ever come across men larger than mountains, or smaller than ants;
nature therefore cannot do this, since if it could a potentia would exist in
vain, unless it sometime achieved actus" (Toletus In Phys. 1qqg, opera
4:24vb). We have first a rule of thumb: what has never been known to occur
43· Faced with the objection that the crocodile never stops growing, they answer, correctly
but irrelevantly, that each crocodile will have an upper limit to its size-namely, however big it
is when it dies! They also argue that in fact it must stop growing before it dies, on the grounds
that the nutritive faculty must fail before the moment of death. That argument too is plausible
but irrelevant (Coimbra In Phys. 1qq 1, 1:110).

Authenticated
Vicaria Dei
probably can't. Then an application: Brobdingnagian and Lilliputian peo

ple have never been known to occur, so they can't. To that Toletus appends
a justification for the rule. Nature, that is God, would not provide a form
with a power and then never allow that power to be exercised, because then
that power would have been created for no reason.
The argument is not a mere induction. It is not of the form 'No Brob
dingnagian human has been obsetved, ergo no human (past, present, or
future) is Brobdingnagian'. If it were, the last step would be superfluous.
The conclusion is in fact not that no human is Brobdingnagian, but that no
human naturally can be, or that no human is in potentia Brobdingnagian.
For that the last step is needed. We have seen many times a potentia
inferred from an actus. Water spontaneously cools: therefore it has the
potentia to be cold. Such inferences presuppose that the obsetved state is the
outcome of a natural change, but they are otherwise straightforward if the
natural can be discerned from the unnatural without knowing what powers
a thing has. To infer the absence of a potentia requires more machinery.
There is no inconsistency in supposing that a potentia has never manifested
itself. To infer from the failure of a supposed potentia to manifest itself that
there is no such potentia, one must suppose that every one has, or has had, its
moment; and that inference will in turn rest on the economy with which
God pursues his ends.
One must suppose also that the obsetvations made so far are an unbiased
sample of the occasions on which the supposed potentia might have man
ifested itself. That raises an interesting question. The created world yields
evidence of the magnitude of divine power. But no inference of the form
'God has never been known to do X; therefore God has not the power to do
X can be sound. In particular the evidence, otherwise overwhelming,
against the unique events recorded in the New Testament, shows only that
those events lie outside the scope of God's ordained power. Induction re
tains its validity only on the assumption that God continues to adhere to his
self-imposed order; the inference from actual to potential likewise relies on
that order. If that order is altered, those inferences may no longer be sound.
No human body had the power to promote salvation until the first Commu
nion, no bones the curative effect of saintly relics. Every statement that
proposes to circumscribe the powers of things, or list their ends, must be
implicitly relativized to an order we know to be contingent, and whose
motives may outstrip our comprehension. What we can be sure of is that
God's ends are fixed and that the means he chooses to accomplish them are
not gratuitously varied. That suffices to ensure that the powers he has given
to each thing to pursue its good will not change, and it warrants the in
ference from the actual to the potential.
3· Monsters. Chance events are an acknowledged exception to the reg

Authenticated
ularity with which natural agents pursue their ends. 44 Yet monsters, at least,
are not an exception to the natural order. Not only are they caused by
natural agents, but their ends, although distinct from the ends to which
normal births are directed, can be seen to lie within nature.
A monster, Toletus writes, is "a mistake [peccatum] of nature acting on
account of some end, from which it is frustrated by some corrupt principle."
Or, more precisely:
A monster is a natural effect which degenerates from the correct and usual
disposition consonant with its species [effectus naturalia recta & solita secun
dum speciem dispositione degenerans] . It is called a "natural" effect, because in
art there are no monsters, properly speaking, but only by analogy and
resemblance to natural things. [...] It is said to be that which "degene
rates from the correct disposition" because a natural effect is not called a
monster unless there is some defect, obliquity, or deviation from what it
was going to be according to the reason of that effect, and on this account
monsters are called "mistakes of nature" [peccata naturr.e: cf. Phys. 199a33
b4], because nature falls short and wanders in bringing about the effect.
The disposition is said to be the "usual" one because what is monstrous
occurs rarely and does not resemble the other effects that proceed fre
quently [from the same causes]. I add "consonant with its species," since
what is monstrous does not preseiVe a specific resemblance to its proxi
mate principal cause, so that if a dog were born of a human female, it
would be a monster. (Toletus In Phys. 2cgq 13, opera 4:75rb)
The outward signs of monstrosity are rarity and lack of resemblance with
other, more frequent, effects of the same causes. A monster, I should note,
differs from its normal kin not by its form (it is still a member of the same
species) but by its dispositio-the arrangement of its parts, or perhaps the
temperament of its elemental qualities. The inward ratio of monstrosity is a
departure of the motus of generation from the actus implied in the powers
44· What follows is intended only to address questions about finality. A suiVey of textbook
questions on monsters would be a sizable undertaking, although they all draw on Aristotle,
Plutarch, Pliny, Galen, and Albert. The standard work on philosophical theories of monsters in
the Renaissance is Ceard 1977; cf. also his edition of Ambroise Pare's Monstres. Daston and
Park 1981 discuss learned and popular belief; Darmon 1977 details a number of seventeenth
and eighteenth-century cases. Pare, Daston and Park, and Darmon include contemporary
illustrations. Reisch's chapter on monsters includes a well-known depiction of five more or less
credible instances (Marg. phil. 8e19, p345; also at 405, in the chapter [gqo] on human
deformities). Goclenius's entry, which treats the term in its broad sense of 'prodigy' or 'por
tent', presents an exhaustive classification of monstra in every department of nature (s.v. 'Man
strum', 708 1r). The locus classicus in Aristotle (Phys. 199b4ff) is examined briefly by Lerner
(Ig6g:I6gf). In his introduction to Nicole Oresme's De causis mirabilium, Bert Hansen gives a
lucid presentation of the relation in Aristotelian thought between maiVels and natural order;
on monsters in particular cf. p62 and notes 37, 38. Oresme himself devotes a long section of
the De causis to monsters and their causes (Mira. 3§6, 229-248).

Authenticated
[206] Vicaria Dei
that produce it, whether that is to be ascribed to the principal cause itself
the seed-or to the concurring causes and conditions sine qua non.
The Coimbrans list five causes or conditions which must concur in gener
ation (Coimbra In Phys. 2cgqsa2):
(i) the formative virtue of the seed, which "delineates and builds the
whole body limb by limb";
(ii) the matter "from which the fetus coalesces";
(iii) the womb or receptaculum;
(iv) the primary qualities "on whose temperament the health and integ
rity of animals depend";
(v) extrinsic factors like the "salubrity or inclemency of the region, the
influx of the stars," and so forth.
Of these five, (i) and (v) are· efficient causes, the rest are aspects of the
material cause; I will refer to all but (i) as "impeding" causes. Deficiencies or
abnormalities in these conditions will produce characteristic defects. If the
seed is crude or sluggish, for example, the offspring may have confused,
useless parts, or revert to a lower form, as when a man is born with a ram's
head. In such cases "nature, since she cannot attain a perfect finish, con
tents herself with an inferior grade" (Coimbra In Phys. 1:341).
Sometimes the matter of several seeds from different species is promiscu
ously confused, and half-men or half-beasts are born. This happens in Afri
can rivers when the water is low (Toletus says it happens in India). Some
times there is too much matter, or too little, and dwatves or giants result. If
there is too much heat in the womb, the offspring may be born with a beard
(which has, the Coimbrans add, "occurred in our own Portugal"); too little,
and it will age prematurely. The heavens, whose crucial role in generation
was studied in §5-4, can affect the parts of the offspring in many ways.
Sometimes, the Coimbrans write, "when those stars dominate, that most
favor the generation of men, it happens (if we believe Albert in his De
mineralibus, and Uohn of] Jandun in question 24 of this book [i.e., in his De
mineralibus]) that the animals generated then, and even the stones that
congeal, partly imitate the likeness of man" ( 1:342). 45 One last cause is the
parents' power of imagination, "which occasionally makes the formative
faculty wander from its target, and imprints upon the fetus absurd or alien
45· Pare cites from a French translation of Aristotle's Problemata a story of Albert the Great.
In a certain village a cow gave birth to a calf that was half human. "The villagers, dubious about
their pastor, brought him to judgment, and meant to bum him and the cow together." Albert
persuaded them that the monster had come about "by a special constellation," and so the
pastor was "delivered and purged from the imposition of so execrable a crime." Pare for his
part doubts that Albert was right (Monstres 68; for the source, see 1720118). Whatever its
detractors may have said, it is clear that Aristotelian science was not entirely useless.

Authenticated
figures." Such are the marques d'envie that Mersenne asked Descartes to
explain. 4 6
Thus far the story contains nothing but efficient and material causes.
Finality enters when one asks why there should be monsters, or, as the
Aristotelians put it, whether Nature "intends" monsters. In the Coimbrans'
response, it becomes clear that the sense they give to the question is 'what
part of Nature in particular intends monsters?' There are, as it turns out,
four possibilities: the principal cause, the impeding cause or causes, the
principal and impeding causes together, and God. The principal cause does
not intend the monster: "it strives to produce without error or vice an effect
similar to itself' (In Phys. 2cgq6a1, 1:344). But a monster, as we have seen, is
by definition dissimilar to its principal cause. The impeding cause, too, does
not intend the monster, for similar reasons; by itself, it would produce
nothing of the sort. Only the two taken together-call this the total cause
can be said to intend the monster, or rather what the Coimbrans call its
fundamentum, the respect in which a thing is a monster, like two-headedness
or six-fingeredness. That Nature should intend some effect "is nothing
other than that it should incline toward it"; the total cause does incline
toward the monstrous effect, and in that respect so does Nature.
That, however, leaves the initial question unanswered. It is true that the
total cause, here as elsewhere, tends toward its effect. How could it not? But
the question was rather: why should such total causes, having the charac
teristic effects they do, exist at all? The answer, as one might expect, consists
in showing that God intends that they should. With respect to his ends,
nothing, the Coimbrans affirm, happens "fortuitously or by chance." The
frustration of the action ofa particular cause is not the frustration ofa divine
end, because God "does not absolutely will the ends of all particular causes."
Some he promotes; others, "by reason of their weakness," he allows to fail.
His ends, which are universal, are never frustrated. 47
And what are those ends? The only one mentioned is the beauty of the
whole. The theater of the visible world is ornamented by "the variety of
images not only of elegant things, but also of the deformed," just as the
charm of pictures is augmented by adding dark colors to light (a2, 1:346).
46. To Mersenne 29]an. 164o,AT 3:2o; 3o]ul. 164o,AT 3:120; cf. Diop. 5,AT 6:129. For a
grotesque example, see Pare Monstres 36, fig. 28 (= Darmon 1977:176, fig. 41). This is a child
born in 1577 (Pare says 1517, but Ceard corrects it to 1577) with the face of a frog, because its
mother conceived it while holding a live frog to cure her fever. Darmon lists a number of
similar instances.
47· A passage from Thomas's Summa cited by the Coimbrans here reads: "Since God is the
first and universal cause not merely of one kind of thing, but universally of all beings, it is
impossible that something should occur outside the order of divine governance; but if some
thing seems to escape, when considered according to one cause, from the order of divine
providence, it is necessary that it should have been sliding back into that order according to
another cause" (Thomas ST 1puqw3a7, Parma 1:398).

Authenticated
[208] Vicaria Dei
Toletus answers the two questions rather differently, the second in rather
less Panglossian fashion. What intends the monster is the impeding cause.
While it is true that it is not the principal cause of the monster, it is that part
of the total cause from which the monstrous effect frequently follows. The
conjunction of the two may be rare, but given that the first is operative, if the
second acts, the same deformity will necessarily occur (Toletus In Phys.
2cgq 13, opera 4:76rb). That, for Toletus, suffices to say that the impeding
cause "intends" its effect. Unlike the Coimbrans, he does not insist that what
intends the effect should resemble it, or that it should issue in that effect for
the most part unconditionally. Although a closer study would be needed to
establish the point, it does seem that Toletus, in ignoring resemblance, and
in requiring the 'for the most part' condition to hold only conditionally, has
a more modern view of efficient causation.4S
His answer to the second question strikes me in a similar way. Toletus asks
how God could "concur with both the impeding cause and the generating
[i.e., principal] cause." It would seem that the one concurrence impedes
the other, and that God is frustrating his own action. We have seen that the
Coimbrans answer, in effect, that God subordinates particular to universal
ends. Toletus's answer is that although God concurs with every cause, he
concurs with each according to its nature, so that "with stronger causes he
concurs more strongly, and with weaker causes more weakly." Hence the
one concurrence impedes the other "not from the side of the concurrence,
but from the side of the causes." If there were, he adds, "a supreme Em
peror, who had beneath him various Kings, who would seek arms and
troops, and receive them according to their condition and manner; and
when two Kings contended, would overthrow the one who was weaker," that
Emperor would be like God, who, though he assists all things, does do
according to their nature and power and, when two contend, allows the
stronger to win. Though the Coimbrans note in passing that God allows the
weaker cause to fail, their emphasis is on the universal and the particular,
and on the beauty of the whole. Since in doctrinally similar texts like these,
emphasis yields most of the differences from one to the next, the shift is
worth remarking.
No one who has read Deleuze's Nietzsche could fail to be struck by this
48. Two recent discussions ofAristotle's phrase 'always or for the most part' and its contrary
'rarely or never' differ on a similar point (Mignucci 1981,judson 1991). Mignucci holds that
'always or for the most part' should be taken to indicate a high absolute probability: normal
births happen most of the time, monsters only a few times.Judson argues that "always or for the
most part" should instead be taken to denote a conditional probability: given that an event is
an instance of bovine generation, the likelihood of its being an instance of the generation of a
normal calf is high. The Coimbrans, then, side with Mignucci: what is "intended" by a cause
follows from that cause most of the time, whatever other causes are operating. Toletus, on the
other hand, holds that an effect is intended provided only that it occur most of the time given
that the other causes .are also operating.

Authenticated
passage. Though Toletus speaks of providence, the justice of his world

perilously resembles that ofThrasymachus. The beauty that the Coimbrans,
following Augustine, take to be God's leading motive is nowhere to be seen.
Divine concurrence, moreover, is nothing other than a ratification of rela
tions intrinsic to nature: as Toletus puts it, nature's outcomes are deter
mined ex pane causarum, not ex pane concursus.
Perhaps God has arranged it so that the more noble, the more lovely, is
also the stronger. The struggle of kings might then issue in beauty. But only
per accidens, at least from our standpoint. In his question on fate, Toletus
defines fate, following Boethius, as the "inherent disposition of mobilia [sc.,
corpora, natural things] according to which divine providence binds them to
its order" (In Phys. 2c6q 11, Opera 4:6gva). That disposition, considered with
respect to the divine mind that arranges it, is called "providence." With
respect to the things themselves, it is called "fate." All things under the
heavens are ruled by fate-"all movements, and mutations, and other pas
sions." Human bodies are, as we have seen, subordinate to the heavens,
especially in generation, which is of all the soul's powers the most distant
from the intellect and will. Are we too, then, ruled by fate? No: the will is
only inclined by the body, never compelled; only God rules it. Yet if the
heavens can impede every bodily movement, or at least every movement not
immediately caused by the will, then the sovereignty of the will is restricted,
as Descartes would later hold, to the soul itself. We escape fate, and are
stronger than the stars, but only in what we will, not in what we happen to
accomplish.
Whatever their differences, Toletus, Coimbra, and the other central texts
agree that every event that looks like a counterexample to finality can be
analyzed in one of two ways.
The event may be the outcome of two causes acting simultaneously on the
same subject, or in the same place, each ofwhich, were it to act alone, would
fit the scheme of natural change. Here the model is the encounter of two
people at a market. Their meeting is not the end ofeither one's actions; but it
can be analyzed into two actions each of which does reach, or would have
reached, a natural terminus. The generation ofa monster that results from the
mixing of seed, or from the intervention of celestial causes, can likewise be
analyzed in to the frustrated, but still directed, actions ofeach kind ofseed.
Or the event may be the outcome of a single cause acting upon matter
that is not fit to receive that action. The quantity of seed is not an active
power, yet an overabundance of seed can frustrate the action of the vis
formatrix embodied in it. What opposes the active power is not another
active power, but the absence of a corresponding passive power. The action
of that one active power is, clearly, directed to the same end it would have
attained had the matter been fit.

Authenticated
[210] Vicaria Dei
In neither case is there an action that lacks an end. There are only actions
that fail to attain their ends. The treatment of monsters here is of a piece
with the treatment of chance events generally. Chance, Suarez writes, is not
"a peculiar cause instituted per se to such an effect, but can be any created
efficient cause, insofar as by accident and without intention [prmter inten
tionem] there is co~oined with its per se effect another rare and fortuitous
effect" (Suarez Disp. tg§t2t5, opera 25:743). No effect is a chance effect
with respect to God, who foresees and intends all that happens; only with
respect to particular things can there be chance events. In monstrous births,
Toletus holds, the generator fails to achieve its end; the celestial influence, if
it is present, does achieve its end, which is to dispose terrestrial matter as it
sees fit. That it makes the seed deviate from its usual path is an outcome that
happens to be conjoined, because of the simultaneous action of the genera
tor on the same matter, with its intended effect (Toletus In Phys. 2c6qg,
opera 4:67rb).
I said at the outset of this subsection that ends in Aristotelian physics
supply a functional equivalent to laws in later physics. I will conclude by
making that claim more precise. Let us suppose that in both the earlier and
the later science, experience, whether it is the common experience ap
pealed to in many Aristotelian arguments or the experimental results used
in modern science, warrants not merely singular claims but counterfactuals
of the form 'Whether or not S occurs, if S were to occur, T would occur'.
I take it that in many instances, Descartes and his predecessors would
have agreed that such counterfactuals-call them phenomenological
regularities-are true. But thereafter they diverge.
The Aristotelian looks for an agent-sometimes a specific nature, some
times Nature as a whole-and a description of Sand Tsuch that Twould be
an intended effect of a certain operation characteristic of that kind of agent,
given that S has occurred. The generality with which a phenomenological
regularity is credited derives from the fact that agents of that kind will all act
according to the same ends, whether these are the purely formal ends of
§6.2 or ends peculiar to that kind. The necessity implicated in the counter
factual form of the regularity derives from the fact that the effect, given that
the agent has triggered an appropriate efficient cause, that no impeding
cause overcomes its action, and that the matter it operates on is adequate,
will be bound to follow. For example: the birth of a calf, given that its parent
has the end of reproducing its form and has set the seed in motion, that the
heavens do not intervene in an unusual way, and that the matter operated
on is not excessive, and so forth, is bound to occur. That is the kind of
warrant the Aristotelian would offer for the commonsense belief that if
Bessie were to give birth, she would have a calf.
The Cartesian looks for a description of S and T which will permit a

Authenticated
derivation of the conditional 'If S occurs then T occurs' from the laws of
nature together with assumptions about the natures of the things involved.
(I leave further details for Part II.) Where in Aristotelianism the nature of a
thing entails that it shall have certain ends, according to which it will act
of necessity if it is an irrational agent-, in Cartesianism the nature of a
thing entails that certain characteristic effects must follow if the thing or
certain of its parts is moved in certain ways. Fire consists of small parti
cles moving with great speed; placed next to wood, those particles must
break off particles of similiar size and speed from the larger particles of
wood they collide with. The necessity of that effect is inherited from that
of the laws they instantiate. Thus the Cartesian justifies the common
sense belief that if a flame were put next to something wooden, that
thing would burn.
The claims of the opponents ofAristotelianism that its appeal to ends was
superfluous, or that it must have arisen from a confusion of matter with
soul, are, it seems to me, warranted only after one has adopted the Cartesian
way of explaining phenomenological regularities (or something similar).
Considered from within Aristotelianism, there seems to be neither super
fluity nor confusion in the appeal to ends. One has indeed the option, if one
is explaining the actions of inanimate agents, of construing talk of ends in
terms of material dispositions and what follows from them. So we saw in
Toletus's treatment of the dictum 'Matter desires form'. That option seems
to have been increasingly favored by the Aristotelians, although one does
not find them pursuing it as far as some of their Medieval predecessors had.
But to insist, as later philosophers did, on choosing that option to the
exclusion of the appeal to ends which is also available would be to give up
the unification ofinanimate, animate, and rational agency which the appeal
makes possible, and to abjure the Aristotelian notion of nature altogether.

Authenticated
Nature and Counternature
N ature, as everyone knows, comes in two sizes: one size fits all, and
extra large. The first, standardly glossed as 'essence' or 'quid
dity', is that which defines each individual substance. In Aristo
telian physics, it is the principle of motion and rest-provided
that we understand by motion natural motion, and that the individual is
correctly described. The second, glossed as 'world', 'universe', or 'cosmos',
is the system of things which have natures in the first sense. It is a system, a
unity rather than a mere aggregate, by virtue of efficient causal relations,
like the relations between celestial powers and terrestrial souls which were
studied in §5.4, but more significantly by virtue of the teleological relations
studied in §6.
In the first part of this section I will examine the various senses of the
words 'nature' and 'natural' acknowledged in the central texts, and certain
of the contrasts in which the natural participates. Their complexity was a
topos for philosophical lexicographers almost from the start. Out of that
motley, two themes, pervasive in Aristotelianism, emerge: the theme, which
we have already seen in several guises, of order; the theme of intrinsicness,
which I interpret in terms of explanatory autonomy.
I then turn to the contrasts of 'natural' and 'supernatural', and of 'natu
ral' and 'preternatural', 'contranatural,' 'violent'. I show that the first con
trast is ambiguous: there are events that, though their causes lie outside
nature, are "owed to nature." Nature enjoys a certain autonomy with respect
to them. There are, on the other hand, truly miraculous events whose
explanation lies, but for the fact that they occur to natural things, entirely
outside nature.
Preternatural, contranatural, and violent changes all have causes within
nature. The contrasts between them rest, as I will show, on the ways in which
[212]
Authenticated
a thing can fail to be autonomous with respect to the changes that occur to
it: in the contranatural, with respect to final causes; in the violent, with
respect to efficient causes. The preternatural, that to which a thing's nature
is indifferent, is violent but not contranatural. Although preternatural
changes neutralize, as I will put it, the difference between "according to
nature" and "against nature," and are therefore not subject to evaluation in
terms of ends, the exception they yield to the directedness of natural
change is always subsumed under a higher end.
In §7.2 I turn to the definition of nature, and its relations with essence
and finality. Nature, as Aristotle defines it, comprises matter and form, a
passive and an active principle. But in keeping with the importance of
agency in the Aristotelian conception of natural change, the active principle
predominates. One immediate consequence is that essence and definition
depend primarily upon what a thing does, and not on what can be done to
it. Ifwe look forward to Cartesian physics, we see that because agency has no
role in the nonhuman world, Descartes must find other ways to account for
the obvious fact that the world includes distinct natural kinds. I then ask to
what extent finality does and must enter into the definition of natural kinds.
In defining nature Aristotle contrasts it with, and yet thinks it in some
ways analogous to, artifactual form. Turning to that analogy in §7.3, I show
that, by virtue of their mode of production, which consists chiefly in the
spatial rearrangement of their material, artifacts have, in the proper sense,
no form-and, therefore, no nature. The absence of nature in artifacts,
which for Descartes was the key to their use in the understanding of natural
things, renders problematic the time-honored analogy between human art
and divine creation, and the conception, which in §6 we saw was essential to
giving ends a role in physical explanation, of natural agents as instruments
of God. A reconciliation of that conception with the fundamental belief in
genuine natural agency was indeed achieved. But for anyone who insisted,
as Descartes did, on the absolute freedom of God's will, the underlying
tension, and the alternative of employing the analogy to erase natural
agency, would have been apparent.
7.1. The Uses of Nature

Virtually every philosopher who attempts to define 'nature' or its sister
words in other languages remarks on the variety of its uses. Some attempt to
regiment that variety by picking out one use as central. Some, like Boyle,
urge its elimination, suggesting other terms thought to be less equivocal. 1
Yet the word has remained, its complexity undiminished. A century after
1. Boyle, "A Free Enquiry into the vulgarly received notion of nature" §2, Papers 17gr.

Authenticated
[21{] Vicaria Dei
Boyle's "Free Enquiry," the Encyclopedie retains his eight senses and adds
several more for good measure. Since then, things have only gotten worse
for the fastidious. Raymond Williams calls 'nature' "perhaps the most com
plex word in the language," adding that a complete history of its uses "would
be a history of a large part of human thought."2
Complexity, and no doubt complaints about complexity, have accom
panied 'nature', natura, ¢ums, and their kin all along. One need not search
far for the reason. Nature was, as G. E. R. Lloyd has argued, not waiting to be
discovered by Xenophanes or the authors of the Hippocratic corpus. It had
to be invented, and once invented, it was "repeatedly contested" (Lloyd
1991:418,432).3 Though the medical authors and their rivals are long
gone, notions ofthe natural, and the numerous contrasts in which 'nature'
participates, have proved to be of permanent utility. At the same time they
have proved to be permanently contentious, a point nicely underlined in
Lloyd's essay by his references to the environmental movement.
Aristotle's definition of ¢uaLs in the second book of the Physics is, there
fore, more the fresh gambit of a new hand at an old game than a deliverance
of empirical inquiry or conceptual analysis. So too, though they have the air
of well-rehearsed, and fixed, matches, are the qut£stiones devoted to that
definition in the commentaries. But before I turn to them, I want to survey
the territory of the natural of which these texts are but the most learned
province.
The diligent Goclenius, at the outset of his entry on natura, lists ten
significations, virtually all of which can be found at the heart of one or
another controversy. 4 Slightly systematized, they are:
(i) "the essence of each being, whether it be substance or accident,"
2. Williams 1985:219, 221. On the roots of the natural-supernatural distinction, see Lubac
1946, Auer 1964.
3· One mark of an invention is that not everyone has it. Rolf Schonberger remarks that
"anyone who looks at the texts of Judaism must observe that a Semitic equivalent for the
concept 'nature' is altogether absent" (1991:216).
4· Goclenius Lexicon s.v. 'Natura', P739· The loci classici for the various senses of 'nature'
were Aristode Meta. 5C4 ( 1014b17IT) and BoethiusDua. nat. p1341rr. With these precedents to
justifY it, listing the senses of 'nature' becomes a standard part of arguments on nature. The
Coimbrans, saying there are many more, list just six (In Phys. 2c1q1a1, 1:203), collapsing
Goclenius's senses (i) through (v) into their second, his (ix) and (x) into their third, omitting
his (viii) (no doubt because it is not immediately relevant to physics), and adding two new
ones: in the first place "the divine mind, maker and founder [panms] of all things," and in the
last place "the generation of living things, which is called nativitas," which is included to
motivate the etymology from nascendo. Abra de Raconis lists six, distinguishing natura in the
sense defined in the Physics, which corresponds to Goclenius's (ii) and (iv), from natura in the
sense of essentia, which is applicable not only to corporeal substances but to God, mathematical
objects, and accidents (Abrade Raconis Phys. 67). I am not sure what the significance of these
differences is, or even whether they are significant-in some instances we may have simply
variation for its own sake.

Authenticated
This is the site of an old contest indeed. The Coimbrans link the use of
natura to denote essence with the theological dictum that "in the three
divine persons there is one nature," which was made dogma by the Council
ofChalcedon in 451. 5 In such contexts natura is allied with the more techni
cal term subsistentia (which is found almost exclusively in metaphysics). 6
(ii) "substance," as when Pliny says "Nature is the maker of things";

(iii) "both substance and its qualities, affections, operations of nature";
(iv) "the matter or form of a natural body separately";
(v) "the form of inanimate natural bodies only, as when nature is op
posed to the soul, and distinguished from it."
These are the senses with which the definition given by Aristotle in the
Physics is most directly concerned. The last of them, which is often the
subject of a separate question, impinges immediately on the immortality of
the soul, since all that is complex and entirely natural is subject to
corruption.
(vi) "the quality and affection of a natural body, as when a magnet is said
to attract iron by nature."
Among those qualities is the "vital heat" of animals. This was the sense of
'nature' appealed to in the controversies among Hippocratic, temple and
folk medicine detailed by Lloyd.
(vii)
"natural efficient causes"-Goclenius's example is the slogan Nat
ura nihil facit frustra;1
(viii) "the state of a man not reborn [through baptism], as opposed to the
state of grace."R
This last sense is again the site of battles both old and new, grace being an
instance of the nonnatural with urgent and fundamental interest for the
5· In Phys. 1:203; cf. Denzinger 1976, nos. 300-303 (Cone. Chalcedonensis, Actio 5). The
fourth Lateran Council ( 1215), in a Definitio contra Albigenses et Catharos, uses the form cited by
the Coimbrans: "tres quidem person<e, sed una essentia, substantia seu natura simplex om
nino" (ib. no. Boo, Cone. Lateranense IV, Cl; cf. also no. 1330, Cone. Florentinum, Bulla
"Cantate Domino" [1442]).
6. See Suarez Disp. 34§2-4, Opera 26:353-379, where the three terms natura, suppositum,
and subsistentia are defined and distinguished. In seventeenth-century cun·us one chapter in
the otherwise much-reduced metaphysics segment is typically devoted to subsistentia and
natura.
7. Similarly, the Coimbrans record the sense "natural causes, insofar as they act according
to an ingrained propensity," and list a number of dicta of the same sort.
8. On this sense, see Auer 1964:333. There is an ambiguity in this use of nature as between
the "pure" nature with which we-in the person of Adam-were created, and the debilitated
"nature" that we, through Adam, acquired in Original Sin.

Authenticated
[216] Vicaria Dei
descendants of Adam and Eve. 9 I will return to the question of nature and
grace below.
In each of these first eight significations, natura is said of an individual,
whether as totality, essence, form, or accident. I have a nature (or rather
several), you have a nature, the rocks and the earth they are made of have
natures. On the other hand, natura applies to all the individuals of the class
otherwise designated by terms like mundus, universum, and cosmos:
(ix) "the world onrav, that is, the totality [universitas] of things that the
world consists in," for which usage Goclenius's example is "In na
ture nothing is idle and ihaKTOV [sc., disorderly]" (cf. Phys. 8CI,
252a11);
(x) "the order of natural things ordained by God. "10
Nature in these inclusive senses I will denote by the capitalized word 'Na
ture'; for the other, individual senses, I will use 'nature'.
Although it may be futile to look for a unique core sense of natura, one
key is to be found not in what the senses share but in their collocation.
Although not all of them explicitly assert the existence of a cause or princi
ple of a thing's actions, to speak of a thing's nature rather than its essence or
substance is to bring the entity referred to into the field of cause or princi
ple, as opposed, say, to that of existence or definition. More important, it is
to imply that the cause is intrinsic.
This is clear in the physical uses of the word. The presupposition of the
use of 'nature' is that among the actual or possible changes associated with
each thing, some are recognized to belong to it, some not. Those changes,
unlike extraneous changes, find their explanation in the thing itself. More
precisely: though they may be elicited or triggered by another, the way they
proceed is determined from within. In a box I confine an object; at some
moment I open all sides of the box. The object, if predominantly earthen or
watery, will fall; if predominantly airy or fiery, it will rise. The fall or rising are
set off, allowed to manifest themselves, by something outside the object. But
thereafter the object itself determines what will happen.
For each thing, then, there is a set of at least possible happenings for
which a sufficient explanation of the course of those happenings (though
not perhaps of their inception) can be given by referring only to its features.
I will say that the thing has explanatory autonomy, or that it is autonomous, with
g. "By the term 'nature' we designate the natural faculty offreewill to choose the good, and
by the name 'grace' the assistance [auxilium] and gift infused in us by God for those tasks to
whose undertaking and completion nature does not suffice" (Soto De nat. & gratia 1c2, 4v).
10. Both the Aristotelians and their successors take pains to assure their reader that by
'Natura' they do not mean anything like a person or an anima mundi. On Alain de Lille's
personification of Natura in the sense of "cosmic principle," see Speer 1991:111 IT and Chenu
1 957• C. I.

Authenticated
Nature and Counternature [217]
respect to that set. The features that do the explaining, or some more
fundamental feature that gives rise to them, are the thing's nature; and the
autonomously explicable happenings are the natural happenings.
Typically those explanations will at some point advert to the kind that the
thing belongs to. This animal stores food because, being an ant, it is pro
vident; that animal flees because, being a hare, it is timid (see Coimbra In
Phys. 2cgq3a2, 1:334). Goclenius's senses (i), (iv), and (vi) all concern
nature in the sense of kind: a lodestone attracts iron by nature, which is to
say, by its nature, by virtue of being the kind of thing it is. As Montgomery
Furth notes, it is a datum for Aristotle that each individual, especially biolog
ical, is "permanently endowed with a highly definite specific nature [... ]
which it shares with other individuals" (Furth 1988:72; though Furth here
refers only to animates, and among the Aristotelians the datum holds uni
versally). What lodestone does by nature and what gold does are different;
attraction is natural to lodestone, not to gold. Natural kinds and natural acts
presuppose one another.
Explanatory autonomy, applied to the world, is a restricted version of
what is sometimes called "naturalism. "11 It holds that there is a delineable
subset of worldly goings-on whose causes, efficient or constitutive, are to be
found in the world itself. There may be other happenings whose causes are
only to be found in God. Aristotle argues that the origin of motus must be
traced to a prime mover; but in all other respects he supposes the world to
have explanatory autonomy. In §5.4 we saw that many Aristotelians were
convinced that the generation of humans, and perhaps even of all animate
beings, is not explicable by natural causes, while the generation of inani
mates is explicable.
The world is self-sufficient only to a rather limited degree; but to some
degree it must be if physics is to be more than descriptive. It is not trivial to
hold that the world of sensible things enjoys a degree of explanatory auton
omy. Anyone who denies that second causes are genuinely efficacious is
thereby denying that the fundamental level of explanation is to be found
within nature. One might admit that at another, less profound-call it
phenomenological-level, some sort of explanation is to be had. But if one
holds, as Malebranche did, that all genuine causation is mediated by God,
11. 'Naturalism' has, unfortunately, as many senses as 'nature'. Keith Hutchison defines it
as the view that "virtually all events in the material world, and many of man's moral attitudes as
well, can be accommodated by human reason without appeal to the supernatural." Newton,
believing as he did that the solar system needed a nudge from God every so often to remain
stable, was not a naturalist; Laplace, who did away with nudges, was (Hutchison 1983:297.
298). Naturalism, I think, is better stated as a claim about the causes of events and features in
the sensible world than as a claim about the scope of reason. If it is true that physics can do
without the hypothesis of a divine governor, that is because the causes of natural change are
themselves all to be found in nature, or, more circumspectly, that the differentiating causes are
found there.

Authenticated
[218] Vicaria Dei
then one is denying radically the sufficiency of any merely physical account
of natural change. The Aristotelians, though they believed that all second
causes require the concurrence of God, did not deny genuine efficacy to
them. They were to that degree more naturalistic than some of their
successors.I2
The applicability of 'nature', whether to individuals or to the world, rests
on there being changes that belong to a thing, on its explanatory autonomy
with respect to those changes. Nature is, therefore, a relative concept, rela
tive to the class of changes picked out as natural. To the objection that
natura denotes something absolute but is defined as a relative term, Buridan
responds that it is not, in fact, absolute: "the term 'nature' signifies matter
and form and supposes them [supponat pro eis] but it does not signify them
absolutely; rather [it signifies them] with respect to natural motion and rest"
(Buridan In Phys. 2q4, 21fva). Even though 'nature' denotes ("supposes
for") something that can be defined absolutely, it is itself a relative term, like
'element'. 13 It is, moreover, relative to natural motion and rest (and not,
e.g., violent or miraculous motion). Contrary to what one might think from
the order of definitions in the Physics, the concept of nature presupposes,
and does not give rise to, a demarcation of natural change.l4
That demarcation is pretheoretical, or at least prephysical. Its signifi
cance derives in part from what is not included: for the world as a whole, the
supernatural; for individual things, the pretematura[.l5
The supernatural was, in Aristotelianism, no simpler a notion than the
12. Hutchison emphasizes the naturalism of "radical" Aristotelianism (cf. Steenberghen

1991:325-335), as opposed to the "mitigated" versions proposed by Thomas and most other
Aristotelians. But because Hutchison underestimates the reinforcement that Thomism re
ceived in the latter part of the sixteenth century, at least in Catholic schools, and the well
honed ability of the central texts to achieve a prima facie consistent position on the role of the
supernatural or to push problems offstage, he exaggerates the extent to which seventeenth
century natural philosophers found it needful or useful to oppose the bad Aristotle ofAverroes
and Pomponazzi.
13. Buridan In Phys. 2q4, 21va; cf. In Meta. 2q, quoted in Schonberger 1991:217, n.I.
Buridan's argument was not new. Toletus (In Phys. 20 q 1, opera 4:46va) cites Albert (In phys.
2tr1c2) and Thomas in support of the claim that natura is a relativum.
14. The point can be reinforced by two observations about the Aristotelians' defense of
Aristotle's definition, which will be discussed later. The first is that several authors argue that
when Aristotle calls nature the principle of rest and motion, only physical motion is meant, and
not that of which angels are capable, for example, the motion of the will in desire or of the
intellect in contemplation (cf., e.g., Abrade Raconis Phys. 69). The second is that as Toletus,
following Albert, points out, the quies referred to in the definition is natural rest, not the
enforced rest of something prevented, say, from descending to its natural place.
15. As Schonberger puts it: "As a rule concepts are determined, insofar as they are demar
cated vis a vis others, and thereby defined [i.e., in the etymological sense of having their
boundaries fixed]. What is designated by natural being can only be determined if one sets it
against others" (1991:217). Whether or not this structuralist homily suits all occasions, for a
concept like nature, part ofwhose bite comes from its being denied ofsome things, understand
ing the various modes of not being natural is indeed essential to understanding the natural
itself.

Authenticated
natural. Woven into it were at least three strands: the immediacy of divine
action in certain events, the extraordinariness of the effects of such action,
and its portentousness, its significance as a revelation of divine intentions.
The interplay among them is complicated. To understand it, I turn for a
moment to the supernatural event par excellence, the infusion of grace by
God into his elect.I6
Soto, explaining why the attainment of grace requires supernatural as
sistance, begins by delimiting those acts that are "owed to nature": "The
general influx and concurrence of God, although it is a spontaneous and
voluntary [act] of divine majesty [...] is enumerated among natural causes,
because God has agreed to concur with the nature of things, so that [...]
they may perform their natural operations, as if this were owed to nature
[naturce debitus]. [...] On that account, when the philosophers and theolo
gians say that the sun by nature shines, the heavens naturally turn, fire burns
by its own nature, and the like, they include under the name of 'nature' the
general efficacy of God (Soto De nat. & gratia 1c2, 4v). If one considers
God's mercy alone, one will regard all creation, this "machine of things"
(5r), as a work of grace. But once God has created a certain form, he would
frustrate his own intention were he not to concur in that thing's operations.
In that sense one might say that God, though he remains of course free, has
bound himself to Nature.I7
In purely natural things, which in humans includes all that can be at
tained by reason, God concurs with whatever operations are needed to
accomplish them (Soto c3, 7ff). The effects of concurrence, one should
note, are determined entirely by the natures of the agents with which he
concurs. It is not at all surprising that God should concur, for example, in
the growth of our bodies, which is needed for our smvival and reproduc
16. Henri Lubac argues that the word supernaturalis achieves currency only with St. Thomas
and is inscribed into dogma only at the time of the Council of Trent (Lubac 1946:327).
Supernaturalis applies, on the one hand, to what exists in nature, but is beyond its power to
produce; and, on the other, to what exists, as God does, outside of nature: the "miraculous"
and the "transcendent." Grace is both, because it is neither implied in human nature nor
obtainable through natural action, and because it "contains a reality that already [i.e., in this
life] renders man a participant in divine nature" (401 ). My concern here is chiefly with the first
sense; the second will tum up in §7.2. What binds the two together is that what lies beyond
Nature's power can be effected only by something above Nature.
17. Suarez too argues that when God voluntarily concurs with second causes, his volition is
"not entirely absolute [ ...], but is accommodated to the natures of things, and as if in
accordance with a debt ofjust distribution [ ...] As also by God's will they have the power of
acting, but not by a volition entirely superadded and as if by grace adjoined to the volition by
which he willed their existence; instead the two volitions are connected according to a natural
debt and connection of things themselves. So God's having willed that fire should heat, and
water cool, is no bare volition but in its way owed [to them], on the supposition that he wanted
to create such things" (Suarez Disp. 18§11 10, opera 25:396). I tis not hard to see how someone
who insisted on the absolute freedom of divine acts would be inclined to deny that one act of
God could induce any sort of obligation to another. Cf. §7.3 and n.39 below.

Authenticated
[220] Vicaria Dei
tion. Since it is a factor common to all such events, concurrence can be

omitted from their explanation. One doesn't need to say, once the theologi
cal setting is understood, that the paper caught fire because it was heated
and because God concurred in its being heated. Nature enjoys, in short,
explanatory autonomy with respect to what is owed to it.
But humans, unique among creatures, are endowed with free will; and
Adam, our representative and type, chose, as the Bible says, to disobey God.
That sin, for reasons that Soto expounds at length, has tainted us all, and
cannot be remitted even if a person acts justly according to reason. Its
"disorder and deformity," whose most obvious outward manifestation is
death, has "remained as a habit, affixed to our nature" (cg, 30v), but not
since it resulted from a free act ofAdam, and not from God's creation of the
human form-genuinely natural.
Since our natural ends do not require that we be released from the
disorder of original sin, it cannot be said that God, merely by creating us as
we are, has bound himself to supply us the means to release ourselves.
Acting in accordance with reason or following the dictates of natural justice
can never suffice to achieve grace, although it may be a necessary condition.
If, when a person is baptized, the disorder is removed, that is not simply by
virtue of God's concurrence in whatever natural acts take place. God con
curs, for example, in the movements of the priest. But those movements
would not suffice to remove the disorder of Original Sin unless God added
to his usual concurrence a special "assisting" act. That assisting act is
supernatural.
Since second causes by themselves could never effect the remission of
Original Sin, God himself must not merely concur in remission but himself
effect it. One mark of the supernatural is the immediacy of divine action. But
the fact that an act is brought about by God without the aid of second causes
does not in itsel~ suffice to distinguish supernatural acts. Suarez holds that
God must "supplement" the power of celestial agents in the generation of
animals (Disp. 18§2137, opera 25:612; cf. §5-4 above). Even if God's act can
still be described as concurrence, it is, nevertheless, of a different order than
his concurrence in the generation of the elements. There natural causes
suffice entirely, and concurrence factors out; but in the generation of ani
mals, natural causes do not suffice. Nature is not autonomous with respect
to those actions.
But generation is not usually thought of as supernatural. The reason, I
suppose, is that even though in generation divine concurrence not only
allows but supplements the action of natural causes, still that concurrence,
just because it is necessary, is "owed" to them. It will arrive whenever the
natural causes are set to act, just as weaker kinds of concurrence do. Even
human generation, where the soul, because it is not educed from matter,

Authenticated
must be immediately created, is not supernatural. We can explain even that

creation in terms of the appropriate configuration of second causes, since
God's action follows as a matter of course: it would be a miracle if a human
soul were not created at the right moment. In that sense nature is autono
mous even with respect to those actions.
Thus the supernatural, when it does not amount merely to God's action
in those instances where he alone can produce a certain effect, is set apart
also by extraordinariness. Thomas argues that on the one hand "every work
that God alone can perform can be called miraculous." But in a stronger
sense, the miraculous requires that the "form induced [by God's action] is
beyond the natural power of the matter [in which it is induced]," and that
there should be "something outside the usual and customary order of
causes and effects" (Thomas ST lpt2qii3alO, opera [Parma] 2:454). God
concurs with, and even supplements, the powers he has implanted in cre
ated forms; but those are just the powers ofwhich Aristotle said that they act
always or for the most part. Even human generation, though it requires the
creation of a soul, is part of the natural order. Since concurrence is the
exercise of God's ordained power, we have what by the seventeenth century
was becoming a standard identification: that which occurs according to
potentia ordinata is the ordinary. It is not surprising that some philosophers
should have begun speaking of potentia ordinaria. Is
But that inclination, though understandable, is misleading. It is true that
when God does what is owed to nature, the acts will not be surprising; their
occurrence is, though not caused by natural things, explicable in terms of
their powers and deficiencies. But events supernatural in the stronger sense
may also be perfectly ordinary and are entirely to be expected. Priests and
parents did not doubt that, if the ceremony were performed correctly, the
taint of Original Sin would be washed away by baptism.
Immediacy and extraordinariness, then, are interdependent but distinct
strands in the notion of the supernatural. On the third, portentousness, I
can be brief: though it was much discussed in theology and in popular
literature, it does not figure greatly in the central texts. The supernatural, in
its guise of the extraordinary, was an invitation to exegesis. Notjust monsters
but prodigies of all sorts-comets, meteors, earthquakes, floods, frogs pour
ing from the sky-were examined for what they might reveal both about the
future and about the intentions and sentiments ofGod. 19 The central texts,
18. See Oakley 1984:57ff on the shift from ordinata to ordinary.

19. On the interpretation of prodigies, see, e.g., the views of Cornelius Gemma (De naturOi
divinis Characterismis . .. , 1575) analyzed by Ceard. Gemma proposed a new science, an ars
cosmocritica, which would be to the disorders of nature what symptomatology or semiology was
to the disorders of humans (Ceard 1977:37o1r). It is to be noted that events with perfectly
respectable natural causes, like eclipses, were also held to be portents: for some, but not all,
authors on matVels, the availability of a natural explanation did not close off the possibility, or

Authenticated
[222] Vicaria Dei
though they do not explicitly discourage the exegesis of prodigies, do not

encourage it either. Their steadily naturalistic standpoint may well have
been intended to inoculate students against too credulous a reception of
popular practices of divination and against ascribing strange events to the
actions of demons (whose existence they do not doubt).20
The contrast of the natural and the supernatural cannot be reduced to a
straightforward opposition between what is caused by natural things and
what is caused by God alone. Nor can it be reduced to the opposition of the
usual and the rare. Still less was it what it became in the seventeenth century,
and perhaps is even now: a simple contrast between what is according to the
laws of nature, and therefore explicable naturally, and what violates those
laws, and is therefore, from the standpoint of physics, essentially surd. A
long road must yet be traveled to reach Hume.
Unlike the supernatural, the contranatural is, in its definition, straightfor
wardly derivative from the natural. Natural change is according to the na
ture of a thing, contranatural change "against its nature and inclination"
(Toletus In Phys. 4c8text57, opera 4: 126rb). What is not so clear isjust which
changes are against a thing's nature or inclination. Since the immediate
context of Aristotle's longest discussion of contranatural change in the
Physics is a classification of the ways in which changes can be contrary to one
another, one might expect that contranatural changejust is change contrary
to one or another of the changes that follow from a thing's nature.
But two problems, one raised by Aristotle himself, the other by his com
mentators, reveal an unexpected complexity.
1. Since nothing aims at its own nonexistence, but instead tends to pre
setve itself, it would seem that corruption is always contranatural. In particu
lar, death is against nature. But Aristotle himself holds that although genera
tion and corruption are contraries, neither is natural or contranatural. In
fact corruption, since it is the inevitable outcome of the natural process of
aging, is natural. Then he adds: "But if what is violent [quod vi est, i.e., what
exists by force] is prceternatural [translating Aristotle's napa ¢uow], a cor
ruption will also be contrary to a corruption, namely [a corruption] which is
natural [will be contrary] to one which is violent" (Phys. sc6, 230a2gff;
Coimbra In Phys. sc6text57, 2: 157). The violent, quod vi est, was introduced
earlier in his arguments against the void: "First of all, every motus is effected
the need, for interpretation (see 488; for the contrary view, see 340, 443· 451, 473fT).
20. Daston and Park argue that "for the educated layman [...] and even more for the
professional scientist of 1700, the religious associations of monsters were merely another
manifestation of popular ignorance and superstition" (1981:24). The central texts, which
reflect of course the view of the well educated in the late sixteenth and early seventeenth
centuries, exhibit an earlier stage of that development: their view is that monsters have entirely
natural causes, though they can, perhaps, also be signs.

Authenticated
either by force or by nature. But it is necessary, if there should be violent

motus, that there is also natural motus, since in fact the violent is prceter
natural, and what is prceternatural is posterior to the natural" (Phys. 4c8,
215a1 f; Coimbra 4c8text67, 2:52). Stated only conditionally in Phys. 5, but
asserted in Phys. 4, is that violence is preternatural, "alongside" nature. (I set
aside for later the claim that the natural is prior to the violent.) Now the
sense of 'violent' is, as Thomas puts it, that a violent act is "from a extrinsic
principle when the patient does not contribute any force [a principia extrin
seco vim passo non confcrente]" (Thomas In Phys. 8lect7, opera [Leonina]
2:388). The patient, in other words, "does not have an inclination" to the
act (John of St. Thomas Nat. phil. 1qg~, Cursus 2: 188a).
The characterization of violence is ambiguous, as John of St. Thomas
notes, between a "negative" interpretation of' does not have an inclination',
according to which the patient neither helps nor opposes the extrinsic
principle, and a "positive" interpretation, according to which it does oppose
the extrinsic principle. If the patient neither helps nor opposes, then the
violence is preternatural, properly speaking. If it opposes, then the violence
is contranatural. Some Aristotelians doubted that there were any preter
natural actions: that is the second problem, which I will come to shortly.
Now back to corruption. It seems clear that corruption is not just preter
natural but contranatural. How could a thing not oppose its own destruc
tion? And yet Aristotle holds that some corruptions are natural, others
violent and therefore contranatural.
Recall that a class of changes was earlier said to be natural for an agent if it
enjoyed explanatory autonomy with respect to them. Violence is evidently
nonnatural on that definition, since the force behind it is extrinsic. So
violent corruption will be nonnatural, and corruption that starts from within
will be natural by contrast.
Yet it is difficult to understand, in the face of the general principle that
things tend to preserve themselves, how corruption could be natural. I think
we can make explanatory autonomy do the work here too: what is con
tranatural about corruption, whether it comes from within or not, is that it
goes against the good of the individual. An individual can enjoy explanatory
autonomy with respect not only to a certain class of changes but with respect
to a certain kind of cause. Violent acts are those for which a thing fails to be
autonomous with respect to efficient causation (I take 'vis' to denote the
efficient cause alone). "Contrafinal" acts are those for which a thing fails to
be autonomous with respect to final causation. Corruption from within
which in aging humans is brought about by the gradual loss of vital heat
has no final cause within the organism. One could tell a story about how
individuals must die to make way for their descendants, or some larger story
about the fallenness of the material world; but whatever story one tells it will

Authenticated
Vicaria Dei
not, and cannot, according to the principle that nothing intends its own
demise, include the ends of the individuals themselves.
Earlier I said that Nature was autonomous not only with respect to those
events whose efficient causes lie within it, but also with respect to those
events whose causes lie outside it, but which are "owed" to it. Those acts are
owed to natural things that, given that they exist and have the natures they
have, and thus the characteristic ends they have, must be performed if they
are to fulfill those ends. (Descartes's use of the thesis that God cannot be a
deceiver to vindicate our clear and distinct perceptual ideas of corporeal
substance is one instance of an act owed to Nature.) Such acts are, with
respect not to their efficient causes, but with respect to their final causes,
entirely natural. Corruption, on the other hand, even if it serves a collective
or cosmic end, is always contranatural.
2. The remaining problem has already been stated: are there, in addition
to natural and contranatural changes, preternatural changes as well? The
problem comes up in questions on the motions of the heavens. Aristotle
argues that because no violent motion can be perpetual, the motion of the
heavens, which is perpetual, must be natural.2 1 It also seems to have come
up in discussions of divine action, and especially of the punishments of
demons in hell (John of St. Thomas Nat. phil. 1qga4, Cursus 2:18gb). Be
yond those specialized questions is another: what scope was there already in
Aristotelianism for a concept of change, and thus of ens mobile, the object of
physics, that would neutralize the otherwise all-embracing framework of
activity, nature, and finality? The preternatural-that to which natures are
indifferent-is one part of this new physics, and its corresponding potentia,
the potentia neutra or obedientialis of matter to the manipulations of art is
another. Though there were doubtless a number of avenues to the new
physics of the seventeenth century, including the mechanics based in the
pseudo-Aristotelian Problemata, and the middle sciences, whose practice was
indfferent to many of the distinctions elsewhere essential to Aristotelianism,
it is significant that without departing from that tradition one could already
find, in its interstices, some elements of what was to come.
That the preternatural is a kind of neutralizer of key contrasts is apparent
when one considers the chief objection raised against it. Zabarella reports
that his Paduan predecessor, Marcantonio Zimara, in his Theoremata "en
tirely denies" that there is any middle ground between natural and violent
motion, where 'violent' is understood to mean 'contranatural' (Zabarella De
21. Zabarella De motu ignis 2, De relnts nat. 292A-D; Aristotle De cr£lo 2text17, 26gb6. Since
the question of the existence of preternatural movements does not obviously arise in the Physics
(though an able commentator would have no trouble finding pretext~ for raising it), and since
it is therefore absent from most Physics commentaries and cursus, a detailed discussion is
outside the scope of this work. My purpose here is to lay out the problem raised by the
preternatural and to lay the ground for referring to it in the Part II of this work.

Authenticated
motu ignis 4, De rebus nat. 293Cff). Aside from citing texts where Aristotle
treats the division as exhaustive, Zimara points out that the argument of De
ccelo mentioned above is invalid if there is a third alternative. One can infer
from the fact that the motions of the heavens are not violent that they are
natural only if natural and violent cover all cases.
John of St. Thomas reports that one of his colleagues at Alcala, a certain
Ioannes Ramirez, argues that every motus is either natural or violent in the
sense of not being from an intrinsic principle. That in itself does not, as we
have already seen, preclude the possibility of change which is neither natu
ral nor contranatural. But Ramirez goes on to argue that all bodies have an
internal propensity to move toward or away from the center of the universe;
if they "recede" from that place, they are being moved violently. The sense
in which "recession" is to be taken is not clear: but if it includes any motion
that is not toward the natural place of a thing, then every motion will be
either natural or contranatural.
Both Zabarella and John of St. Thomas, who would, I think, have agreed
on little else, argue that there can be preternatural change. Zabarella offers
a variety of arguments, of which the most interesting here is this: "[The
motion of fire around the heavens] is not natural if one takes the natural to
be that which is brought about by an internal principle; but it is not so apart
from nature [prceter naturam] that it opposes or injures nature; instead it
nourishes nature and preserves it. It is rather, similar to both [natural and
contranatural motion]" (ib. 8, 3ooE). Zabarella is contrasting motus whose
efficient cause is not contained in their subject with motuswhose final cause
is not contained in the subject, which is the contrast I drew a moment ago in
explaining how dying of old age could be at once natural and contranatural.
John of St. Thomas, quoting his namesake (Thomas In de Ccelo 1lec4,
opera [Leonina] 3:16) makes a similar point (ib. 195b). The circulation of
fire and air in the heavens is not caused by an intrinsic principle, but neither
is it contranatural; it is, instead, "in a certain way above nature, because such
a motus inheres in them through the impression of superior bodies, whose
motus fire and air follow according to perfect circulation, while water [fol
lows it] according to imperfect circulation"-whence the tides. John of St.
Thomas adds that it is not against nature because moving in a circle whose
center is the center of the universe is neither moving toward nor away from
one's natural place, whatever that may be.
Circular motion, then, is preternatural because it neutralizes the con
trariety of natural places. John of St. Thomas then notes that on this ac
count, a revolving wheel might be said to move preternaturally, and that
since only contranatural motions are resisted by a thing's nature, and thus
subject to quick decay, the revolution of the wheel, "once impressed [... ]
would never stop, unless it were held back from outside" ( tg6a). He answers

Authenticated
[226] Vicaria Dei
that the wheel would (like any projectile) be moved by an impressed "im
pulse" that because it is extraneous to its subject will eventually vanish. The
forced choice between natural and contranatural is rejected, only to be
replaced by an equally forced choice-so it would seem-between extra
neous and native or connate qualities.
The neutralization of contrasts is only temporary, aufgehoben by a new
contrast. A similar pattern is exhibited in John of St. Thomas's arguments
on behalf of the claim that when God acts as the "principle and root of every
motus" he cannot be acting violently (the punishment of demons is another
story). It is evident in experience, first of all, that some motus are not violent
though they result from an extrinsic principle. Such are the tides (whose
motion we have already seen described as preternatural) and the illumina
tion of the air by the sun. Contrary to what Aristotle thought, not every such
motion is violent (188b).
John of St. Thomas adds to that a posteriori argument an a priori reason.
The coordination, he says, of superior and inferior causes is itself natural.
Indeed an inferior agent inclines yet more to a motus coming to it from a
superior agent than to its own motus, "because it inclines more to the conser
vation of the whole than to its own." If, as everyone would agree, its inclina
tion to its own conservation is natural, "all the more so will its inclination to
the conservation of the whole be natural." But the conservation of the whole
depends on the connection of the parts, so that if an inferior part disobeyed
its betters it would threaten the whole.
We saw a particular case of what one might call the communal spirit of
natural things, and their submission to higher ends than their own, in the
Coimbrans' explanation of the horror vacui (§6.2; cf. John of St. Thomas's
own discussion in the present question, 1gob, and his vacuum question,
1q 17a1, 362b). Bits of matter will move, even against their natural inclina
tions, and even at risk to themselves, to ward off any threat to their spatial
continuity.
It is tempting to read off a politics from such arguments. I will content
myself, however, with pointing out the coincidence of two orders. The first is
the order of power: God's power infinite and unparalleled, that of the
angels greater than any material thing, ours greater than the animals', and
so forth. The second is the order of perfections: the part is subordinated to
the whole because the perfection of the whole is superior to that of its parts.
The two orders are linked: the more powerful an agent is, the more fully it
takes the perfection of the whole, its own contribution to which is propor
tionate to its power, as its end.
Here as before the preternatural, as well as the locally contranatural, are
aufgehoben through the intervention of a new contrast, this time between
part and whole. There is no place in Aristotelianism for a change that is

Authenticated
simply, and without further explanation, "outside" nature. Even if, more
over, a change is preternatural with respect to an individual, it will not be
preternatural if viewed with respect to the ends of something else, whether
they be those of the substantial form of a thing-the ground upon which
the extraneous and the innate are distinguished-or those of a whole or a
higher agent to whose ends the individual's own ends are subordinated. It
will therefore, in a broader sense, have an end, and will therefore terminate
when that end is reached. The preternatural, like the contranatural, is
therefore always transient.
7.2. Individual Natures

Aristotle, having in the first book of the Physics settled on matter, form, and
privation as the principles of natural things, proceeds in the second to
define 'nature' itself. The nature so defined is the nature of individual
things. Though the Physics acknowledges the use of 'nature' to denote the
system of natural things, it does not explicitly define that sense of the
word. 22 I will first examine what the commentaries have to say about the
definition itself, and certain issues that arise immediately from it. These
include alternative concepts of nature rejected in favor of Aristotle's, the
differences among animates, inanimates, and artifacts, and the applicability
of the notion to the soul. Following that I will enlarge the scope of the
discussion to consider the relation between nature and essence, and the
finality inherent in the classification of natural kinds.
1. The definition ofnature. Expositions and questions on Aristotle's defini
tion of nature can be seen to have two purposes. One is to make the concept
precise through successive parings-down. The tenninus a quo is the genus
'principle', the tenninus ad quem matter and form, with primacy accorded to
form. The other is to establish among all things, natural or not, a hierarchy,
and to ascertain the position of animates and of the human in that hier
archy. Fulfilling those two purposes yields the object of physics-ens natu
rale, according to the dominant view-and gives it a position in the order of
knowable things, below divinity and above art.
In the usual translation, Aristotle's definition reads: "Natura motionis, &
quietis, eius rei, in qua est, primum, & per se, & non ex accidenti, prin
cipium quoddam sit, & caussa [Nature is a certain principle and cause of
movement and rest in which it exists first, and per se, not accidentally]"
22. See, e.g., Phys. SCI, 252a13 (Coimbra In Phys. 8c2, 2:271 ): "For nature is the reason of
the order ofall things." It is actually not easy to find an occurrence ofcj>{ms in the Physics which
must be interpreted in this sense; nor is it very common elsewhere, although in many instances
it is not easy, perhaps not possible, to decide (Bonitz 1870, s.v. cj>ums).

Authenticated
Vicaria Dei
(Coimbra In Phys. 2Cl explanatio, 1:1gg). 23 Aristotle says 'principle and

cause', the Coimbrans note, so as to exclude privation, which is a principle
but not a cause. "Nothing else, except matter and form, is acknowledged by
Aristotle to be (a) nature" (2oq2a1, 1:205). Nature is called the principle
of motion and rest so as to exclude mathematical forms, which are not
causes of motion, and nonphysical causes of motion-separated substances
like God and the angels.
Such causes are excluded by the phrase in qua est, which signifies that
nature is an intrinsic principle. That there are intrinsic principles in Nature
is, as we have seen, no idle assertion. One alternative, briefly noted by the
Coimbrans, is the view that nature is a universal principle, manifesting itself,
to be sure, in individuals, but properly belonging to none. Albert writes that
certain philosophers hold that "absolute" nature is a "form and power issu
ing from the first cause through the motion of the [outermost] sphere,
which, after it comes forth, is diffused into all natural things and becomes in
them the principle of movement and stasis."24 Such theories, which were
not without their analogues among Renaissance philosophers, did not
strike the Aristotelians as serious competitors, because they effectively
denied efficacy to second causes.25 But they are worth mentioning, if only
because the combination of a denial of individual natures, especially active,
and an affirmation of a universal force proceeding from God, finds an echo
in Descartes, where the quantity of motion imparted to matter by God at the
outset is the only source of activity in the world.
The next part of the definition, per se & non ex accidenti, setves to exclude
both art, in the sense of skill, and its products. Since I discuss artifacts in
particular in §7.3 below, I will not dwell on them here. Medicine is a skill
and a cause of the actions of those who practice it. But medicine in the
2 3· As usual, the cited text varies slightly from one commentary to another, or even within a
commentary. The version that the Coimbrans analyze in their qucestioon the correctness of the
definition is "Natura est principium & caussa, ut id moveatur & quiescat, in quo prima, per se,
& non ex accidente inest" (2CJq2, 1:205), a close paraphrase of the translation into better
Latin. Their Greek text coincides with the Loeb edition.
24. In phys. 2tnc5, opera (lnst.) 4.1:83, see also 49· As authorities for the view, which he
refutes, Albert cites the Pythagoreans, Plato, and Hermes Trismegistus. His immediate source,
however, which is also mentioned by the Coimbrans, is Avicenna (Suff. 1c7, 17vb). Toletus
writes that for Plato, "nature is life, or a certain force suffused through bodies, their governor
and director," a "substance proceeding from God, and divided amongst lower things," yet in
itself one substance (In Phys. 2CJq 1, opera 4:46ra).
25. Among others were Francesco Patrizi da Cherso (Ingegno 1988:257) and Robert
Fludd, with whom Mersenne engaged in an extended controversy. Mersenne reports that
according to Fludd, "God, while he is a certain light diffused through the whole world, does
not enter anything unless he has first assumed like a cloak an ethereal spirit [...] God forms a
composite with this ethereal spirit. He resides with it especially in the Sun, from which
he radiates outward to generate and vivify all things. In that way God [is] the form of all things
and does all things, so that second causes do nothing per se" (A de Baugy 26 avr. 1630, Corr.
2:441).

Authenticated
doctor who treats herself is not her nature or part of her nature. She only
happens to be both the subject in which the skill resides and that which is
changed through the exercise of the skill. But when the soul acts on the
body, or when active powers emanate from substantial form, the object of
such actions cannot but be the body or the substance itself in which the soul
or the form resides.
The relation between an artifact qua artifact and the natural changes that
happen to it is similar. Artifacts do undergo change. Sculptures erode,
paintings fade, saws get rusty. But it is not, or so Aristotle argues, by virtue of
being a saw that the saw rusts. It is by virtue of being made from iron. A saw
only happens to be that which rusts; iron rusts by nature. The principle of
that change does not reside in the saw, but in the stuff it is made from, its
materia ex quo.
So far we have it that natures are intrinsic per se causes of natural change.
The goal, as I said, is to show that matter and form alone are nature in the
truest sense. Hence the word prima in the definition. The Coimbrans, like
other commentators, construe it as an adverb modifying in qua est. A nature
must exist in its subject prima, which the Coimbrans gloss as "not by another,
or so that nothing is in [the thing] beforehand." We have seen that use of
primo in the definition of substantial form, the first form joined with matter,
or at least the first that yields a complete substance. Here the standard
doctrine that substantial form and prime matter are alone "first" in their
respective constitutive roles yields the conclusion that they alone are
natures.
But which is the nature of a thing? One way to raise the issue is to ask
whether 'nature' applies to active principles, passive principles, or both. In
the commentaries, that query was prompted by an oddity in Aristotle's
definition. Nature is the principle not of moving but of being moved. In the
Greek, it is the apx~ TOU KLVE'iaem, with KlVEL6aL a middle/passive infinitive;
in the usual Latin translation, KLVE'ia6m becomes motio, glossed in the
Coimbrans' paraphrase by ut id moveatur, a passive form.2 6 The definition
looks incomplete, since form, which is active, is surely nature if anything is.
Aristotle himself says it is nature more than matter (Phys. 193b7). Why,
then, the passive verb?
This is no mere quibble. As it stands, the definition suits matter rather
than form. But the Coimbrans note that prime matter, being indifferent to
26. On the forms ofKLVEW, see Waterlow 1982:I61 1T. She argues that KLVEa9m "is passive as
to its grammatical form, but not necessarily passive as to its meaning." It does not, in other
words, presuppose an agent. Nature could be a principle of intransitive change, an internal
principle of the procession of phenomena in a thing-not unlike the essence of a Leibnizian
individual substance. But since (see §2.2) for Aristotle every intransitive change is at the same
time a transitive change, KLVE1a9m should here be taken-as all the Aristotelians take it-to be
passive.

Authenticated
Vicaria Dei
all forms, is nature with respect not only to natural motion but to violent
motion, "and so there is no motus which is not natural with respect to
matter" (Coimbra In Phys. 2c1q2a3, 1:2o8-2og; cf. Abrade Raconis Phys.
75). If nature were matter alone, all change would be "natural"-a good
outcome for Descartes, but not so good for an Aristotelian, since then
directedness and naturalness would be radically distinct.
The Coimbrans defend the definition by making a threefold distinction.
Some things, they say, are above nature: God, the angels, the intelligentia.
Though angels are capable of local motion, they can only be moved from
outside; they do not move themselves. Some things are below nature: ar
tifacts, "which are so ignoble that they have neither a principle of being
moved themselves or of moving others" (1:208). The middle place is oc
cupied by "physical things, which are neither so abject as not to lay claim to a
principle of any motion, nor so perfect that they cannot be moved." They
can move others, for example, in generation, and themselves, locally or by
growth. In that they resemble their betters. But they also have a principle of
being moved, and in that they resemble neither their betters nor their
inferiors. That is why Aristotle mentioned only a principle of being moved
in his definition, and not of principles of moving. He did not intend to deny
that nature includes active principles. He only intended, as the immediate
context of the definition shows, to distinguish natural things from artifacts.
There is, however, a complication. If all natural things, animate and
inanimate, have an active principle as well as a passive, then how do ani
mates differ from inanimates? Aristotle, after all, says that the differentia of
the animate is that it moves itself; but if heaviness, say, is construed as an
active power, then heavy things, animate or not, move themselves. One
answer, that of Averroes, was to suppose that animate things alone have the
power not only to act, but to initiate their actions (Toletus In Phys. 2c1q2,
opera 4:4 7rb). Against that are not only a number of Aristotelian texts, but
the basic principle that nothing acts on itself. Animates can move them
selves, as Aristotle says elsewhere, only because they are inhomogeneous.
Under scrutiny their apparent self-movement turns out to be the movement
of one part by another (4 7va). Inanimate things, being extended, have
potential parts. But the "bits" of one individual are not different enough to
satisfy the condition ofthe basic principle. So they cannot move themselves
even in the somewhat Pickwickian sense in which an animal can.
Toletus concludes that Averroes's stratagem is superfluous: "Neither the
form nor the matter ofinanimates is an active principle of their movement,
but only a passive [principle], and in that respect such forms differ from the
forms ofliving things, which are active principles of their own movement: so
that when a stone is borne downward, or fire upward, they are not actively
moved either by their form nor by their matter" (47vb). That solution does

Authenticated
not contradict the claims made earlier (cf., e.g., §4.3, and Toletus In Phys.
2c6q 1o, Opera 4:61 ra) that heat, cold, and various other qualities are active
powers. They are, but only as extrinsic causes (cf. also Coimbra In Phys.
6Ciq 1a2, 2:224ff). The heat offire does not act on the fire itself (which is, in
any case, as hot as it can be). It acts only on other things. Weight is a more
difficult case, since it follows from the substantial form of a thing, and
appears only to act on the thing itself. Without entering into detail, I will
note that in the standard treatment of falling bodies, the generator of a
heavy body is the principal efficient cause of falling, and the heaviness of a
body only an instrumental cause (Toletus 4:48ra). The basic principle is
preserved, and with it the denial of active powers to inanimate things.
We arrive at a fourfold classification of substances according to the pres
ence or absence of active and passive principles, as in Figure 7. The classifi
cation is at the same time a ranking. Artifacts, at the lowest, the most
"ignoble" level, have no nature at all: "no artifact, as such, has in itself a
determinate and certain principle of its own motion, either active or pas
sive." Since a statue, say, is clearly not animate, it has no active principle. It
has no passive principle, either, qua statue. It tarnishes or rusts because of its
material; and since the same kind of statue can be made out of different
kinds of matter, some of which may rust and some not, proneness to rust
belongs to the material it happens to contain, not to the statue itself
(Coimbra In Phys. 2c1q2a2, 1:211). At the next rank is the inanimate,
which, having only a passive principle, is least among natural things. The
animate occupies a middle station. It is superior to the inanimate by virtue
of having an active principle of motus, but it is inferior to separated sub
stances by virtue of having also a passive principle. Passivity, in short, is
ignoble, but lacking power altogether yet more ignoble.
What is left is to ascertain the place of the human soul in the hierarchy.
That is no surprise. The soul is in part natural, in part "transnatural," as the
Coimbrans put it (2cgq4a2, 1:213). It is indeed, like other animals' souls,
the principle of vital functions, like growth, nutrition, and sensation. But
other actions, like grasping intelligible things by divine illumination, which
"befit [the soul] when it is separated from the body and outside the dregs of
matter," are tacitly excluded from the definition of nature. The soul, to the
extent that it is the principle of those actions, is not a nature. Some, notably
Avicenna, had argued that it is not a nature at all. But true philosophy and
the Church agree that the soul-all of it, and notjust its lower parts-is the
substantial form ofthe body, and therefore its nature (2cgq4a1, 1:212). So
long, at least, as it remains joined with the body, it is, therefore, part of the
natural world.
2. Nature and essence. Virtually all lists of the senses of natura include
essentia and quidditas among them. The relation of nature to the other two is

Authenticated
Vicaria Dei
Principles
Substance
Active Passive
Supra God, angels -
naturam +
Habens Animate + +
naturam Inanimate - +
Infra Artifacts - -
naturam
Fig. 7· Classification of substances
succinctly stated by the Coimbrans: "essence implicates [importat] an order

ing toward the existence of a thing, quiddity toward its definition, and
nature toward its operations. "27 But since nature includes both form and
matter, and since it was a disputed question whether form alone, or both
form and matter, were comprised in the essence of material things, it is
worth examining the question more carefully. Mter doing so, I will argue
that natura, which corresponds reasonably well with what philosophers now
call the 'natural kind' of a thing, must be defined in relation to ends. That is
not surprising, but it is worth bringing out because it helps set the terms for
the Cartesian project of a physics without finality: such a physics will have to
persuade its audience either that natural kinds are not needed in explana
tion (as some recent philosophers have argued) or that they can be simu
lated by other means.
The Coimbrans note that some of Aristotle's predecessors, who were not
fortunate enough to have discovered form, held that essence consisted in
matter alone (In Phys. 1cgq5a1, 1:162). To my knowledge no one within the
Aristotelian tradition agreed with them. The entire question, therefore,
turns on whether matter should be included in essence. If it is, then essence
and nature coincide; if not, then they will diverge, and the role of nature, as
opposed to form alone, in the classification of substances will be correspon
dingly diminished.
The question was an old one. It had already drawn the attention of the
early Greek commentators Themistius and Alexander of Aphrodisia. Be
cause it pertains to doctrine on the Incarnation, it also shows up, in various
forms, in patristic sources. By the time the central texts were being written

Authenticated
the arguments had, as usually happens when a question is both well aged
and theologically sensitive, become quite complex.28
Nevertheless they fall into one of two patterns. The first is this: matter can
change while essence remains fixed; if matter were included in essence, that
would be impossible, so it is not. The most obvious instance is the growth
and nutrition ofanimals: "if matter was contained in the essence of Socrates,
for example, that singular matter would be of his essence, which is proved to
be false. First, because the matter of Socrates is perpetually dispersed by the
action of heat as it consumes moisture: but the essence of a thing must be
fixed and stable" (Coimbra In Phys. 1cgq5a1, 1:162). We don't consider
Socrates to have become a different kind of individual just because he loses
a pound or two. Yet losing and gaining weight are changes in matter, or so it
would seem. Another instance, less obvious to us, is resurrection: "if that
singular matter were of the essence of this man [Socrates], the same man
could not be called back to life when bodies are resurrected, unless he
assumed the same body; but that seems false, since the same matter, at least
partially, can belong to many [people], as is clear among those who feast on
human flesh." The resurrected Socrates will have the same essence as the
Socrates ofAthens only if their bodies are identical. But that will be true also
of Caesar, say, and Charlemagne, whose matter, we may suppose, incorpo
rates some bits of Socrates' body. If the essences of those three individuals
include their matter, then their resurrected bodies will have to contain the
same parts they did on earth, including the bits that have successively been
incorporated into all of them. The picture is unsettling, to say the least.
A third, extreme case that would likely not occur to us is this: "A worm is
sometimes generated out of the species of the Eucharist. Such a worm is
certainly a natural being. But it is not constituted from matter, since the
matter of the bread vanishes after consecration, and thus no matter is the
substrate of [ subsit] the generation of the worm. Therefore not every natural
being is constituted from matter, from which it follows that matter does not
pertain to the essence of a natural being" (ib.). If a thing can have worm
essence without having any matter, then clearly worm-essence does not
include even the property of having some matter or other, let alone a
particular matter.
The second pattern of argument against including matter in nature
proves the converse: essence can change while matter remains fixed. So
crates the man and Socrates the cadaver share the same matter. But they
differ in essence: one is rational, the other not. More generally, matter is all
28. Defenders of the exclusion of matter from essence include Averroes (In Meta.
7comm21, 34, arguing against Avicenna),John ofJandun, Nifo (Di!.p. 7disp13); Soncinas,
rather surprisingly, does not take a stand, offering argument~ and refutations for both sides ( Q.
meta. 7q26, p157r-v).

Authenticated
Vicaria Dei
of one species (§4.1). People and horses, since they are different species,
must differ in form alone.29
The Coimbrans stage their response to these arguments in three conclu
siones: that matter is a part of the essence of natural things; that "singular"
matter is included in the essence of singular things; that the included
singular matter is not "indeterminate and vague," but determinate and
identical throughout the existence of the thing.
What is theologically at stake in the question is that the human being,
including the incarnated Christ, should be in the fullest sense an ens natu
rale, a natural being. Suppose that matter were not included in the essence
of human beings. Then "the soul would be an unum per se, and complete in
itself," and the union of soul and body, contrary to received doctrine, would
not be (see §5.2). It would follow that since the soul is immortal, and the
union only an ens per accidens, no substance is destroyed in death, and
human beings are in every way immortal. If, furthermore, the essence of the
human does not include specific matter, the essence of an individual will
not include its singular matter, where 'singular matter' denotes the individ
ual bit of stuff informed by individual form. Were this so, then 'human',
which designates an essence consisting of both matter and form, could not
truly be affirmed of 'Socrates' (In Phys. 1cgqsa2, 1:163).
To reach their final conclusion, the Coimbrans start by defining intrinsic
and adventitious unity. A thing has intrinsic unity simply by virtue of exist
ing: as the commonplace puts it, 'being' and 'one' are mutually convertible.
Adventitious unity is conferred on a plurality by something else, roughly in
the way that a certain unity is conferred on a plurality of causes by an end
toward which they all tend. The first is always the same, the second is
"various and fleeting" ( 164). Some philosophers, notably Peter of Lom
bard, overlooking that difference, found it necessary to hold that the matter
that a person receives from his parents persists in him all his life. But the
Coimbrans believe that a unity of "continuation" suffices, by which they
mean that although parts are continuously being destroyed and replaced, at
no one time is Socrates' matter replaced in toto.30
29. Since it is not clear why Socrates the man and Socrates the cadaver cannot share part of
the.ir essence, namely, their matter, the argument is incomplete as it stands. What needs to be
filled in is that individual form corresponds to individual or specific difference and that
'essence' is to be taken as that by which a thing is first constituted as an actual individual (see
Suarez Disp. 2§{16, opera 25:8g). Socrates' corporeality and his animality do not distinguish
him from other things; nor does the species of his matter; only his individual form does so.
That individual form, moreover, is the first form (here the working assumption seems to be the
Scotist theory of a plurality of forms; cf. §4.1) by which the individual Socrates is an actual
existing thing.. It alone, therefore, is his essence.
30. These arguments do not suffice to handle the Eucharistic worm. For that the
Coimbrans, following Thomas (ST 3q77a5), have recourse to a special act of God by which at
the very moment the worm is generated either new matter is created, or existing matter
translated from elsewhere, or the matter of the host re-produced (ib. a3, 166).

Authenticated
3· Nature and ends. Nature, then, includes both form and matter. But they
are not of equal weight. Aristotle argues that "that by which a natural being
is in actu is nature more" than matter (Phys. 20, 193b6; Coimbra In Phys.
2c1 explanatio, 1:202). The underlying reason is that actus is the specifica
tion of potentia (§3.1), and thus closer to defining a thing than potentia. As
Aristotle says, a thing is said when it is in actu: we normally say 'the dog' or
'the statue', not 'the body' or 'the bronze': 'said' in such usages means
'designated according to its definition'. Definitions are perspicuously ex
hibited essences. But in individual substances, form is the actus, and there
fore the specification, of matter; and so form is nature-in its guise of
essence-more than matter.
The relation of nature to ends has, therefore, two aspects: nature as form,
nature as matter, the first having greater weight. The question I will be
asking here is whether the Aristotelians in fact appeal to ends in definition,
and to what extent they must. How deeply, in other words, does teleology
penetrate into the workings of Nature?
Though every material form has among its ends its own sustenance and
the generation of its like in new matter, those are purely formal teleological
ends that do not serve to distinguish one material form from another. By
'purely formal' I mean that for every form, one can enter its name in the
blanks of'-- acts so to sustain itself' and '--strives to reproduce itself',
without knowing anything at all about the form, except perhaps that it is a
material form. If such ends were all that obtained in the world, teleology
would serve only to establish in broadest outline the character of material
substance; specific natures would be defined in other ways.
What we are looking for is ends whose definition will not lead us back to
the form itself. There are two ways that could happen. One is if there are
ends independent of the formal ends just mentioned. The other is if there
are ends that, though instrumental to the formal ends of self-sustenance
and reproduction, can be defined independently. I will present samples of
each.
In Metaphysics commentaries one often finds, at the beginning, a question
elicited by the opening sentence of that work: "All men by nature desire
knowledge." In showing that Aristotle was right, the commentators typically
conclude that not only do people desire knowledge, but that "metaphysics is
most desirable to man insofar as he is man, both by a natural appetite and by
a rational appetite ordered in the best way" (Suarez Disp. 1§61J[ 21, opera
25:63-64). A thing can be described as having an appetite for something
either properly, if it desires that thing through the recognition of its good
ness or apparent goodness, or improperly, if it merely has a propensity to
the thing (Fonseca In meta. 1c1q 1§3, 1:62; cf. §6.3). Among appetites prop
erly speaking, some are natural, in one of several ways. Art appetite may

Authenticated
Vicaria Dei
simply be given in a thing's nature, without, as Suarez puts it, being effected
by any action of that thing (!4, 54). It may exist potentially in a thing by
nature, but actually only under certain conditions, like hunger-the con
trasting term in such cases is adventitious or extraneous. It may simply agree
with a thing's nature, like a desire to listen to music, in which case the
contrasting term is contranatura[.3I
Clearly only the first two kinds of natural appetite pertain to the defini
tion of a thing. But people do not always actually desire knowledge, nor are
they incapable of turning it down when it is offered-the Aristotelians do
not have so rosy a view of human nature as to overlook our occasional desire
to be secure in our ignorance. On the other hand, the desire for knowledge
cannot but be elicited by the appropriate objects in a person who is neither
distracted nor physically impaired nor depraved; it will never knowingly be
hated, although one may freely refrain from acquiring it (Fonseca §6,
1:66B-E; Suarez t8, 55).
The reason that a desire for knowledge will be elicited is that knowledge
will necessarily be seen as good. Indeed, for reasons that do not need to be
spelled out here, the contemplation of truth is held to be the highest good,
and its discovery the most excellent operation of the soul. Knowledge is
useful, of course, but Aristotle, with the wholehearted concurrence of the
Aristotelians, holds that it is desired most of all for its own sake. Our natural
felicity, he says in the Ethics, is the contemplation of God and separated
substances (cf. Suarez 1:34· 63; Fonseca §6, 6gC).
The soul is that which has the power to live, sense, and think, says De
anima. But since thinking is the most excellent of its operations, and since
each thing exists on account of its operations, the soul exists on account of
thinking. That thinking, moreover, itself has an end: knowledge, of which
the highest form is knowledge of God. Could one then define the human as
that form whose end is knowledge of God? Not quite. Other species have
that end too-the angels, perhaps God himself. The human is that whose
end is knowledge of God, but to define us specifically one must add some
thing about the means. That, as we will see, requires reference to the mate
rial part of our nature. What matters here is that the contemplation of God,
though insufficient to define human nature, is not a purely formal end.
There are also ends that, though instrumental to purely formal ends, can
be defined independent of them. The end of sensation is to bring species of
31. Four technical terms are used here. An appetite is innate if it exists actually or poten
tially without any previous cognition of the goodness of its object, and elicited otherwise. An
appetite is said to be necessary quoad exerdtium if it will always arise under the appropriate
circumstances and, if once it is elicited, the will cannot but actually be moved by it; necessary
quoad speciem (or quoad specifimtionem) if it will always arise under the appropriate circum
stances, and the will is free only to be actually moved by it, or to withhold its consent, but not to
be averse to it. See Fonseca In meta. 1e1q1§4-5, 1:62-64; Suarez Disp. 1§615, opera 25:54.

Authenticated
sensible things to the judgment, or the vis restimativa that in animals corre
sponds to the human power of judgment, so that things sensed may elicit
desire or aversion in the soul. Although it is true that ultimately sensation
aids self-sustenance, there seems to be no need to refer to that end in
defining sensation.
Having sensation is in fact part of the definition of 'animal', as contrasted
with 'plant'. Among animals, the possession of all five senses, whose ends
can be defined in terms of their proper sensibilia, distinguishes higher from
lower. To the extent that the senses are defined teleologically, so too the soul
will be. Yet having sensation does not distinguish one species, or even one
genus, from another. For that we must advert to matter. And if we ask why
sensation, rather than some other operation or even none at all, should be a
means to self-sustenance, again we must advert to matter.
All roads, then, lead to matter, and prime matter first of all. Suarez, in his
question on the relation of matter to essence, argues that matter "in some
way contributes to the operations of substance," since if it didn't, the soul
would notjoin with itto make an unum per se (Disp. 36§21 11, opera 26:485).
Its contribution, however, is quite limited: matter "is of the essence of mate
rial substance as a certain beginning [inchoatio] and foundation of that
essence, which is completed by form" (112, 485). It seems, therefore, to
contribute to the essence of material substance no more than what follows
from being a material substance-corporeality, which consists in having
dimensions, and occupying space.
Now one could say that the corporeality of material substances is for the
sake of their forms. Certain of the operations that belong to those forms by
nature require that the form be "incorporated": the singular matter to
which an individual material form is joined would be for the sake of those
operations. In that rather lax sense the material part of the thing's nature
would be defined in terms of the thing's ends. The human body exists so
that the soul may have sensations, so that it may thereby know the world and
eventually God. But there is a twist in that account, which will be best
brought out by turning to proximate matter.
Proximate matter is the disposed, quantified matter to which a given
specific form, if it is to exist naturally, must be joined. We have seen that the
relation of dispositions to form was controversial. But everyone agrees that
matter, even proximate matter, cannot exist naturally without a unifying,
end-supplying, form. (The "organized body" of which Aristotle says the soul
is the form is not a single thing until it has form; experience confirms that
after the soul departs, the body, lacking its principle of unity, rather quickly
falls apart.) Now it is true that the existence in this body of such and such a
disposition is explained by the operations of its soul. The roots of a plant are
beneath so that it may receive nourishment from the ground. But the

Authenticated
Vicaria Dei
disposition itself, so it seems, can be defined apart from any consideration of

those operations.
Yet the question is not quite so simple as that. Consider a giraffe, which is
twelve feet tall. 'Being twelve feet tall' makes no reference to an end. But for
that very reason, an Aristotelian might argue, it is not the most scientifically
informative designation. If we call that quantity "tree-grazing" height we
understand better why giraffes have that height and not some other: gi
raffes, being the sort of animals that nourish themselves by eating the leaves
off trees, will achieve that end only if they are of a certain height.
So we have got from height to eating. The end of eating is rather more
obvious: animals eat in order to preserve their vital heat and to grow. And
what is vital heat? As the name implies, it is the active quality that powers, so
to speak, the various operations of the soul. Even memory and imagination
require the circulation of animal spirits in the brain, and animal spirits
cannot be produced but by heating.
There is a certain pattern to these answers. They all propose that the
operation or feature we want to explain is for the sake of some operation:
but they also seem inevitably to refer to particular facts about the matter of
the animal. John Cooper has characterized Aristotelian teleological expla
nations as appealing to what he calls 'hypothetical necessity' (Cooper
1987). The "hypothesis" is that some operation is to be performed, and the
necessity consists in its being the case that given such and such material
means, the operation can only be carried out thus. Given what light is and
how colors work, if you want to see colors you had better have a transparent
sense organ, since anything itself colored would not be capable of sensing
all colors. The transparency of the crystalline humor of the eye is explained
jointly by the end, vision, and by the nonteleologically defined properties of
its materials.
It would seem, then, that some features of the proximate matter of a
substance can and should be defined teleologically. But if Cooper is right
the examples seem to bear him out-then there will always be a non
teleological component in the definition of any material natural kind. Aris
totle himself would not be troubled by that conclusion. But a Christian
Aristotelian, convinced that the world included purely spiritual creatures
and dedicated to making teleological sense of our existence, would have a
further issue to worry about. Our highest aim, which we share with the
angels, is the knowledge of God. Having a body, being a partly material
form, is clearly not needed to achieve that aim, since the angels enjoy a
vision of God to which only the most saintly among humans can aspire. Our
bodies do provide instruments by which to know God. The senses provide
evidence of his power, intellect, and benevolence; yet they are prone to

Authenticated
error and, because of their link with the passions, to leading us astray. So
even if God has given us what is due to us by nature, a person might still,
however unjustifiably, envy the angels and question the justice of the God
who gave us so inadequate a nature. Original sin goes some way to explain
ing our condition. But Adam was, after all, human: he knew God through
his senses just as we do.
The only justification I can see is that God in the plenitude of his good
ness and power will have chosen to create natures in every degree of perfec
tion. That I, this individual human substance, should quarrel with the na
ture he has bestowed on me would be to demand of God that he should
omit the human altogether from the ranks of created things, since every
person may make the same claims. I have no more cause to quarrel with my
status than do angels to revel in theirs.
The question at the outset was to what extent natures are defined in terms
of their ends. Purely formal ends will not do, because they take this form:
whatever this thing is, it will act in order to preserve itself. Such ends do not
distinguish one thing from another: what distinguishes things is the means
by which purely formal ends are accomplished, and defining the means
seems to presuppose a nonteleologically defined material nature. Among
human ends that are not purely formal, the knowledge of God is the high
est. Yet that end, although it distinguishes us from every other material
nature, does not distinguish us from the angels. Again the means must be
referred to: the senses and imagination by which, barring divine illumina
tion, we come to know God. Sense can hardly be defined except in refer
ence to our material nature. The question then becomes this: to what end
should there be things that can indeed recognize their Maker, but only by
way of matter? The answer seems to be: to no other end than that God
should have manifested his power by creating all possible forms.
7·3· Artifacts, Human and Divine

Art, in the broad sense of human industry or the products thereof, is at once
a model for the understanding of nature and a paradigm of the non
natural.32 In Aristotelianism the comparison serves to glorify and justify our
32. On the analogy and disanalogy of art and nature in Aristotle, see, among many others,
Fiedler 1978, c.s; on Renaissance versions, Close 1969 and Schmitt 1983, c.s. A wide-ranging
examination of the shifting balance between imitation and creation in the understanding of
art is found in Blumenberg 1957. Goclenius, I should note, lists three primary senses for the
word ars, of which only the first is relevant here: "Habitus TTOLT]TLKOS' (factivus) seu >LTJXUVLKOS',
rro[ TJ>LU seu >LTJXclVTJ>LU aliquod post se relinquens nobile vel ignobile, ut JEdificatoria." A forma
artificialis is "qme fit ab arte in materia" (Goclenius, s.v. 'ars' and 'artificiale', p125-126).

Authenticated
Vicaria Dei
labors, and at the same time to emphatically call to mind the infinite
difference not just between our powers and those of the Creator, but be
tween our powers and the powers in Nature to whom God has, so to speak,
delegated responsibility for the natural order. That order we can emulate,
and even bring to fruition where it falls short, but we cannot alter it in any
essential way. In what follows I will first examine the products of art, their
causes and kinds, showing why artifacts, though they resemble in certain
respects the works of God and of our own natural powers of reproduction,
fail even to be substances. That conclusion was foreshadowed in the hier
archy laid out earlier: artifacts, having neither active nor passive powers,
have neither form nor matter to call their own.
I then turn to the analogy and disanalogy of art and nature, first with
respect to the artifacts and then with respect to the act of production. It will
become clear that the Platonic denigration of art is not entirely absent from
the Aristotelians' manner of thinking. Human industry is, in their view also,
secondary and superficial. There is, on the other hand, in the conception of
art as a supplement to nature, the germ of a quite different view, one that
came to predominate in Descartes's thinking.
The twofold analogy between God's labors and ours, and between God's
works and ours, though treated as benign, leads to a certain difficulty that
was fully exploited by Descartes. The trouble can be brought out by con
sidering the effects of reversing the order of the two members of the com
parison. One can say-and this is what Aristotle meant-that art is like
nature, whether by resembling it in form or in its production. That seems
harmless enough. But if art is like nature, then so too nature is like art. Yet
artifacts are bereft of any intrinsic principle, active or passive, of movement.
If nature were indeed like art, nature too would lack such a principle: it
would have, in short, no nature. That is what Descartes meant in his version
of the well-worn adage (PP2§23, AT 8/1:53). At the end of this subsection I
will examine the Aristotelian response to that argument.
1. The causes of art. Artifacts in the strict sense result only from human
industry. Humans are the only natural agents fully capable of recognizing
their ends (which requires, as we have seen, a formal recognition of their
goodness). That restriction has two important consequences for the classifi
cation of artifacts and the relation of artifactual to natural kinds.
The first applies to those artifacts that "imitate" nature by virtue of an
intended resemblance to natural things. In natural production, the effect
resembles the cause by virtue of being specifically the same form as some
form that exists really in the cause. In art, the resemblance must be medi
ated by thought. But resemblance mediated by thought can only be a di
stant, attenuated version of resemblance in form. Toletus, echoing the Meta
physics, notes that of sensible accidents, many are common to several kinds

Authenticated
of things-colors, tastes, and so forth. Only figure is "diverse in diverse

species," and even in diverse individuals. He continues:
Man, the king of this world, and for whose sake [natural things] are made,
depends in his understanding of them on the senses, and with the senses
he perceives substances only through their accidents; truly he would al
ways be deceived and would not be able to distinguish forms and sub
stances, if there were not some accidents by which substantial forms are
designated: such are the figures and exterior forms of things [...]
Figures, therefore, follow substantial forms [ ...] They are so conjoined
with them, and attendant upon them, that many judge them to be the
substances themselves [ ...] But in fact they are not substantial forms,
they are only indices [of form]. Hence a perfect similitude of two things is
most noticeable in their figures, so that those things seem to be of the
same species that participate in the same figure, and are named by the
same name. (Toletus In Phys. 2c2q6, opera 4:54va)
It is for that reason that arts like painting and sculpture, which aim at
likeness, aim at likeness in figure above all. There is a natural relation of
designation already in the relation of figures to forms; that designation can
be imitated in the artificial designation of that which has both the figure and
the form of a lion by something that has only the figure (54vb). The latter,
the artificial designation, is anticipated in the common sense or the imag
ination, which like art, can designate forms only by figures. When the re
semblance of the final cause to its effect is mediated by sense or imagina
tion, the most secure ground of resemblance, if not the only ground, is
figure.
That restriction, though telling, would not be decisive were it not for the
second consequence of the human origin of art. Human activity is quite
limited in its effects. Cutting, shaping, and joining change only the figure of
things, painting or gilding only their surfaces, moving only their place. Even
though we can, however imperfectly, come to conceive the forms of things,
we have not the power to introduce those forms into matter. "Art," the
Coimbrans write, "never forms true things, like a true tree, but imitations of
the true, and, as someone rightly said, truly false [ vere Jalsas]" (Coimbra In
Phys. 2c1q5a2, 1:216). Resemblance mediated by thought, even if it were
not limited in thought itself, would be limited by the means of production
available to us.
These two limitations in the cause of art will serve to demarcate artificial
forms from natural. But before I develop the Aristotelian notion of artificial
form, I should mention three kinds of exception. Only the last raises serious
difficulties.
Some arts have no product. They aim, as Toletus says, only at actions:

Authenticated
Vicaria Dei
singing, dancing, navigating. Such arts cannot be said to imitate nature by

representing it. Their practitioners are said, as we will see, to imitate nature,
or the Creator, in their mode of acting. 33 In any case, such arts will not be
relevant to an account of artificial form.
Certain other arts "accomplish nothing except to apply natural agents, so
that upon their being applied, nature produces some effect, as from the
mixture and distillation of bodies, or in the production of certain animals"
(Arriaga Phys. 2d6§2, Cursus 319a). In all such instances, the action is en
tirely natural. The human contribution consists only in bringing about the
necessary proximity of the agent to the matter it acts on, and the form
produced is not different from the form that would have been made had the
agent and the matter come together without our help. 34 Again nothing new
is added to the stock of forms.
Some arts, finally, rather than imitate preexisting natural effects, imitate
those "which ought to have pre-existed" (Toletus In Phys. 2c2q6, opera
4:55ra). They "supply those things that nature lacks, as in the art of making
houses, clothes, and similar necessities of life." That might seem to be
stretching the notion of imitation (cf. the quotations from Cusa below,
n.41). In what sense ought houses to have existed, and how could the
houses we build resemble houses that merely ought to exist? The answer, it
seems to me, can be gleaned from an argument that turns up in questions
about the ends of nature. If nature acted according to ends, she would treat
us, her most noble members, the most beneficently. But she does not; "she
treats animals like a mother, us like a stepmother. [...] nature clothes
beasts, but casts man naked upon the shore as if from a shipwreck. To strong
and pugnacious beasts she gives arms, to timid ones cleverness, quickness,
and cover: but man she sends to fight in the arena weaponless, slower than
the swift, weaker than the well-armed" (Coimbra In Phys. 2cgq1a2, 1:325).
The reply, of course, is that Nature has given us the best weapon of all,
intelligence, and the organon of instruments, the human hand. But here the
point is not how Nature made up for casting us naked upon the shore. It is
33· If such arts have a product, it consists in a certain skill or habitus acquired by the artist.
We saw that in the definition of nature such skills, because they are only principles per acddens
of being moved in the thing that has them, are excluded. Skills are not active powers. They only
modifY the actions of active powers: "the movement of dancing, which is performed by the
motive faculty inhering in the limbs, owes to art the fact that it is performed elegantly, sym
metrically, and rhythmically" (Coimbra, 1:218). Arts of skill merely introduce adverbial modi
fications of operations we were already equipped to perform. They do not add new powers to
those we possess by nature. If one were to consider the dancer herself to be a kind of artifact,
human material formed by teaching as a bed by the carpenter, the artificial form dancerwould
be as inefficacious in its own right as a bed is.
34· The most spectacular example of applying natural agents, mentioned in several texts, is
the invocation of demons through amulets or astrological figures (see Coimbra In Phys. 2c1q7,
1:217°').

Authenticated
that some such things were owed to us, both by Nature-that is, God the
Creator-and by nature-according to what is naturally required for us to
flourish. Nature ought to have given us coats. Instead she gave us the ability
to imagine them, and hands to make them.
The arts that supplement Nature, then, imitate the instruments she has
given other forms, but not to us. That account helps to remove a certain
difficulty that would otherwise mar comparisons between artifacts and natu
ral things. Clement Rosset argues that for Aristotle "the artificial object
possesses the paradoxical privilege of having more to tell us about nature
than any natural object" and that Aristotle's conception of nature is there
fore anthropomorphic (Rosset 1973:241-242). Aristotle does say that "ac
cording as a thing is done, so it comes about naturally" (the Latin reads: ut
quidque agitur, sic natura aptum est, the last translating Aristotle's TIE<j>UKE). But
he also immediately says, in a passage Rosset omits, "according as a thing
comes about naturally, so it is done [sc. by humans: ut natura aptum est [...] ,
agitur]" (Phys. 1gga8ff). There is a symmetry here that Rosset, eager to press
the charge of anthropomorphism, overlooks.
Nor is art Aristotle's only model for the understanding of nature. In the
biological works, comparisons of nature to art do occur, but Aristotle is far
more interested in comparing animals to each other. More significantly,
although it is true that, in keeping with the precept to start with what is
better known to us, one may explain the workings of the lungs by reference
to a bellows, the workings of bellows will have been, on the argument I am
presenting, originally modeled on those of some natural pump, perhaps
even the lungs themselves. What we know better about bellows is how air
moves through them, or what ends might be served by them, and that we
know because we have more experience with them. Aristotelianism is no
more anthropomorphic in its understanding of nature than physiomorphic
in its understanding of art. If art truly does imitate nature, even where she
has left us wanting and forced us to use our intellect, then to use artifacts as
an analogy is simply to use a familiar bit of Nature to understand an un
familiar bit.
The products of art, then, imitate what Nature has done or what Nature
ought to have done. Imitation, as we have seen, consists in the first instance
in likeness of figure; in the second, in likeness of function. A fur coat is
"made" not so as to resemble the coat of an animal in size or shape, but to
keep us warm as the animal's coat keeps it warm. Here it is significant that
the analogies of nature to art which do occur in Aristotle are not from, say,
animals to pictures ofanimals (which would be gratuitous), but from organs
to artifacts of the supplemental sort. Those are, so to speak, the extrinsic
organs we make for ourselves. If we could not conceive them by looking at
their natural prototypes, and fashion them by cutting and joining and so

Authenticated
Vicaria Dei
forth, they would, because owed to us by nature, have been made for us by
nature, just as the heart and lungs are.
2. Artificial forms. All art, whether imitative by figure or by function, is
secondary, superadded. The Aristotelians explicate that secondariness, and
justify it, in several ways. I will start with the one cause I have not yet
examined, the material cause, move to the efficient cause, and then to the
formal cause.
The central texts agree that the material of artistic production, unlike the
matter of natural generation, is already a complete substance. (I will use the
word 'material' to describe the matter of artificial forms, which, unlike the
matter of substantial forms, is a complete substance.) The Coimbrans in
stitute a threefold comparison between divine creation, natural generation,
and artistic production: "As art supposes nature, so nature supposes God. In
other words, just as art effects nothing unless a Physical composite [sub
stance] is available to it, in which it arranges an artificial form; so nature
generates nothing without an underlying matter created by God, in which it
induces a natural form. And so through these grades [of existence] things
come to light: God produces from nothing, nature from potential being, art
from perfected being; God by creating, nature by generating, art by bring
ing together or arranging" (Coimbra In Phys. 2oq5ai, I :2I4). Art is so far
like nature. But just because the "forms of artifacts supervene on actual
being, on an already perfected and absolute thing," namely, composite
substance, they bring no alteration to that thing's nature. They cannot add
to or remove its powers. They can only modulate them.
Two examples will illustrate the point. The Coimbrans consider whether
art can undertake the works of nature. The instances adduced on behalf of
the affirmative include automata, of which they give a half-page list of stock
examples from antiquity, and astrological images that are said to have magi
cal powers not accounted for by their material. In their response, they argue
that in both a natYral power is simply being modified or elicited. Automata
"are not moved by their forms, nor by art, but by nature," as in clocks driven
by weights. Art does not effect those movements, it only "modifies and
tempers" them (Coimbra In Phys. 20q7a3, I :2I9). Astrological images and
other such magical instruments, to the extent that the marvels attributed to
them are real, arise "from the industry of demons," who for nefarious rea
sons of their own wish to perform them (a2, 2Ig). 35
The reasons for this limitation will become clear if we consider the move
ments by which artifacts are made and the forms that result. Arriaga, affirm
ing that artificial form is not distinct from the parts and locations [ ubica
35· Reisch (Marg. phil. 7tr2c2o, p316) likewise argues that the names inscribed on such
images "have no power [virlus]"; they are "signs of an occult pact with demons." For some
detailed descriptions of figures used in divination, see Agrippa De oa1.tlta soo 11 ·, and lib. 4·

Authenticated
tiones] of the artifact, argues first that to understand which picture we see in a
painting, it suffices to understand its colors "with such and such a location
on the canvas" (Arriaga Phys. 2d6§2, Cursus 31gb). Similarly, to understand
which word has been written on a page, "nothing else need be understood
but that the ink is located this way or that."
What is significant in those arguments, and in rest of the question, is the
terms ubicatio, ubicatum, and so forth, which I have translated by 'location',
'located', and so on. These are learned formations, originally coined in
discussions of the category Ubi or 'Where'. Ubicatio, 'where-ing', is, like many
nouns in -tio, used to denote both the action of the corresponding verb
ubicare and the result of that action. To 'ubicate' something is to change its
Ubi, that accident of a thing which consists in its being here or there (Suarez
Disp. 51§1!14, opera 26:g76; cf. §4.2).
Artificial forms are produced through ubicationes of existing substances.
In the making of a wooden or marble statue, Arriaga writes, "through the
cutting away ofsuperfluous or impeding parts [...] , the others that remain,
without any new form preserve the distance and expressive proportion of a
man, and effect a statue" (31gb; cf. Suarez Disp. 16§2118 and 17§2111,
opera 25:580, 588). Hence there is no artificial form "distinct from the parts
ofthe wood thus located, and the absence ofthe redundant parts." Cutting
away the redundant parts can produce no new substantial form in those that
remain; after all, if other parts had been cut away at random, the action
would have been the same, and no one would say that a new substantial form
had been made.
More precisely, the artificial form is not distinct from the relative loca
tions of the parts. If the whole statue is moved, in the absolute sense, the Ubi
of every part is changed, but the distances between the parts are not. So
"morally speaking," the figure is not changed (Arriaga 320). The Elgin
Marbles are not different sculptures than the ones Lord Elgin packed away
in Athens. (Though one could imagine a conceptual artist creating a work
called 'Five Thousand Two Hundred Eighty Busts of Homer' by moving the
same bust one mile, a foot at a time.)
The action of the artist consists in nothing more than moving existing
parts here and there. For an Aristotelian that is not the sort of change that
could ever yield a new form. While it is true that, as Aristotle says, every
natural change presupposes a concomitant change of place (Phys. 8q,
26oa28ff), including generation and corruption, generation and corrup
tion cannot be reduced to the concomitant changes of place. The action of
the artist yields at most a new figure, a new dispositio of the material. Artifac
tual kinds, therefore, or specific artificial forms, are nothing other than
types of figure. Though we may denominate a work according to its subject,
whose figure it resembles, that denomination is no more than a convenient

Authenticated
Vicaria Dei
way of designating a certain figure-type, of which both the work and the
model are tokens.
Figure, being a mode of quantity, is entirely inert (§4.3). "Just as quantity
by its own character [suopte ingenio] is idle and inert, and given to nature, as
if it were a second matter, to sustain accidents, so the form induced by art
receives no efficacy" (Coimbra In Phys. 2oa6a2, 1:217). Since, as we have
seen (§7.2), the nature of a thing depends primarily on its substantial form,
and on the active powers attendant on that form, figure is perhaps as far
from nature as an accidental form could be. A classification, moreover, of
individual substances according to their figures would only accidentally be a
classification having any use in physics; and even though each species is
marked, as it were, for our benefit by a characteristic figure, a taxonomy of
figures would be articulated not on the nature of things, but on one of its
last and least concomitants. One could as usefully classify things by the first
letter of their names.
The tenuous place of figure in the natural order, and the corresponding
exiguity of our ability to affect that order, is underlined by the relation of
artificial form to its matter. The material of art, as we have seen, consists in
substances perfected already by nature; the matter of an artificial form is the
matter of the substance or substances it is composed out of. Suarez asks
whether artificial forms can, like other forms, be said to be educed from the
potentia of matter (see §5.3). His answer has two parts. The first is that when
a person produces an artifact, he does so "by a natural motive force," and
thus the induced artificial form does not exceed the natural potentia of the
material. Even if the movement needed to produce the artifact is violent, as
lifting a stone would be, still the form intended in that movement, the Ubi of
the stone, is educed from an inherent potentia obedientialis of the material
(§2 .1). The artificial form, on the other hand, because it is intended only by
the maker, and not in the end of any of the movements of the material,
"comes about merely as a result," and so is not, properly speaking, educed
from matter-a distinction it shares with the rational soul, but for quite
different reasons. So there is, Suarez concludes, no natural potentia of matter
or material to artificial form per se, "for although the intention of the maker
tends toward them per se, and thus in such manner directs through art his
actions [...] still that action, by which he executes his design, does not
terminate per se and immediately in such a form, but in some other mode,
from which such a form results" (Disp. 16§2'![ 18, opera 25:580). The artifi
cial form is intended only in thought. What is intended in nature, or in
other words, the ends per se of the movements that go into making an
artifact, is at most the various ubicationes of the parts. 36 The form "results"
36. "In the instruments of art it sometimes seems that by the form [i.e., the shape] of the
instrument the form of the effect is proximately achieved, as when the seal imprints wax or
money with a figure similar to its own, although then also in fact figure is not brought about per

Authenticated
from them, as the shape of a constellation may be said to result from the
motions of the stars it contains, but it is not the actualization of any natural
potentia in its material parts.
Indeed it is doubtful that any figure, artifact or not, is intended per se in
the inanimate world. Even the spherical shape of the earth, though evi
dently the result of natural movements, was not intended by any of the
mobilia. It came about per accidens from the movements of individual bits of
earth, each of which strives to reach the center of the universe. So too the
globular shape of a drop of water might result per accidens from the "love of
conjunction" of its particles (Coimbra InPhys. 4cgq1a3, 2:62; cf. §6.1 ), each
of whose movements aims only at achieving the highest degree of unity with
its companions.
Artificial form, then, is nonnatural several times over. It is superadded to
existing complete substances by God so that their forms may be designated
to us in sensation. It is neither an active nor a passive principle, serving at
best to modulate the action ofsuch principles, and is therefore never part of
a thing's nature. It is never intended per se in natural action, but only in the
thought of rational agents; toward it, matter has no natural but at most an
"obedient" potentia. Artifactual form, which is to say figure, is the perfect
substitute for natural form if, like Descartes, you don't believe there are
substantial forms, natures, natural ends, or potentice.
3· Divine art. The parallel between human art and divine creation, and
between artifacts and creatures, is not without its dangers. If human artifacts
are devoid of activity, and fail to constitute genuine natural kinds, then to
conceive of God as an artificer puts injeopardy the activity and substantiality
of his artifacts.
There are indeed a number of similarities between human and divine
modes of operation. Human art, as we have seen, is accomplished according
to a cognized form. The substrate material has only a potentia neutra toward
that form. The form itself is no more than figure imposed upon an already
complete substance. The resulting artifact lacks all spontaneity: what it does
it does only denominatively, as a key is said to turn a lock. 3 7
In divine production, too, an idea of the thing precedes its production.
Though in creation matter and form are produced simultaneously, the idea
is of the form, and prime matter, like the material of art, has only an
undifferentiated potentia, natural no doubt but "neutral" in the sense of not
tending to one form rather than another. If we restrict our attention to the
se, but rather place [Ubi] [...] Therefore more generally it is said that while it is true that these
instruments do not achieve per se figure or artificial form, still that is not because they concur
instrumentally, but because the form [i.e., shape] is not per se producible in any case" (Suarez
Disp. 17§21ll, Opera 25:588).
37· "Forma: naturales sunt actuosa:, & quasi viva:; forma: vero artefactorum tanquam
stolida:, & emortua:, nullam effectricem vim habentes" (Coimbra In Phys. 2c1q5a2, 1:217).

Authenticated
Vicaria Dei
production of inanimate things, or even to irrational (material but not

human) things, divine artifacts, though they do not lack all spontaneity, lack
the highest degree, free will. There is, finally, some reason to consider
natural things to be instruments by which God accomplishes his ends (see
§6.2). All created things "strive toward the end which the first agent has
proposed for himself," and because he operates all things on account of
himself (Corinthians 1c12), they, too, all regard that end.38
In what respect, then, do creatures, especially inanimate creatures, differ
from artifacts? In particular, since artifacts have no active powers per se, in
what respect can inanimate creatures be said to have such powers?
The last question impinges on the broader question of the efficacy of
second causes. 39 From that complicated question I take only the conclusion
regarding inanimate things: though their active powers lack the untram
meled spontaneity of will, they too can perform those acts that are propor
tionate to them, just as our senses tell us they can. For although they are
inferior to spiritual things, and act always of necessity, they are not "max
imally distant from God [...] ; according to their form bodies have a greater
perfection and similitude to God [than prime matter, which does lack all
efficacy], and so according to [their form] they can have effective power"
(Suarez Disp. 18§ 1t 14, opera 2 5:597). Since not every action requires infi
nite power, and since it is not repugnant to material forms to have finite
power, it is possible that God should have "communicated an active perfec
tion" to them in addition to the perfection of existence (18,, 10, 595-596).
Indeed if he had not, not only would the variety of forms and qualities have
been superfluous, it would have been only apparent: "for if fire does not
heat, but rather God does in its presence, he could equally naturally heat in
the presence of water; and therefore from that action [of heating] we can
no more conclude that fire is hot than that water is" (16, 595). And if heat
were not in the nature of fire, nothing would be: fire would have, in effect,
no nature. Since it is obvious to sense, and not at all repugnant to reason,
that the world is populated by things having genuinely different natures, it
must be that those things are genuinely capable of acting as well as of being
acted on.
38. "Sicuti agentia secundaria nihil absque influxu prim.e caus.e efficiunt, ita ad eum
nituntur finem, quem primum agens sibi propositum habet; sicque ut agens primum semetip
sum tanquam ultimum suorum operum finem intendit; quia universa propter se operatur; ita
eundem finem secundaria agentia spectant" (Coimbra In Phys. 2cgqn1, 1 :328).
39· Suarez's arguments, along with many others, on the efficacy of second causes are
treated in Freddoso 1g88. On the argument in 1[6, see esp. p.wg: "To Aquinas, Molina,
Suarez, or any other robust Aristotelian, denying active causal power to an entity amounts to
nothing less than denying that entity that status of being a substance." To deny that, Freddoso
obseiVes a bit later, is "tantamount to denying that there are any material entities at all and a
fortiori to denying that scientific knowledge is possible."

Authenticated
That, then, is bedrock: inanimate things do have the active powers they
seem to have. What remains to be explained is how they can, nevertheless,
be called instruments of God, and how God's action in producing them
differs from the action of a human artificer.
Suarez lists five ways in which principal causes might be contrasted
with instrumental causes (Disp. 17§2~7-17, opera 25:587-591):
(i) the principal cause is "that to which the action is properly and simply
attributed";
(ii) the principal cause is that which "by its own virtue flows into the
effect"; 40
(iii) the principal cause is that which "proximately and by its own influx
flows into the effect," or, in other words, that from which the effect
receives its being immediately and per se;
(iv) an instrumental cause is that which "only acts insofar as it is moved by
another," and the principal cause that which "per se and without the
motion of another has the force to operate";
(v) an instrumental cause is that which "concurs in effecting or is raised
up [elevatur] to effect something nobler than itself, or beyond the
measure of its own perfection and action," as when heat concurs in
the making of flesh.
According to the first four of these ways, instrumental causes do not have
any efficacy of their own. The influx of esse characteristic of true causes, or in
the fourth, the motus, comes from the principal cause alone. Anyone who
held that irrational agents were instruments of God in any of those ways
would be denying genuine efficacy to them.
Suarez, not surprisingly, favors the fifth way of contrasting instrumental
and principal causes. Doing so allows him to deny that creatures are divine
instruments in any sense that would deprive them of efficacy. They are
indeed subordinate to God, serving his aims even at the expense of their
own, but they are principal causes with respect to all that is within their own
powers to accomplish. Since we have seen that it is not repugnant or con
trary to experience to hold that some actions are within their power, it
follows that irrational agents do have genuine efficacy.
What remains is to show how God's action in producing them differs from
those of human artificers. Art, we have seen, either imitates Nature in the
sense of depicting or designating natural things, or supplements Nature,
bringing forth through the deliberate collocation of natural causes what
Nature ought to have made. It is clear that supplementing Nature to supply
40. 'Influx' or 'flowing in' is Suarez's most general term to designate the causality of causes,
or the proper reason according to which they are causes (Disp. 12§2'1 10 & 13, opera 25:386,
387).

Authenticated
[250] Vicaria Dei
needs that our own natures do not suffice for, like clothing or shelter, is
irrelevant to divine action. God has no needs. His art is instead like that
which issues from exuberance rather than exigency, the unwritten song that
springs spontaneously to the lips of someone fortunate. Like that song it has
no prototype. The exemplar of a created thing that existed eternally in
God's intellect before creation is not "expressive or indicative of the thing";
on the contrary, like the spoon imagined by Cusa's idiota, it is regarded in
intuition and imitated by the thing.41 "But the expression of things them
selves is attributed to divine ideas in a peculiar way: because the divine
essence which grasps the definition of the idea, insofar as it is known by God
to be imitable by created things, contains the man, for example, eminently,
of whom there is an idea, insofar as it is imitable by him, and similarly other
things, and perfectly represents them" (Coimbra InPhys. 2c7q3q2, 1:247).
God's ideas "express" created things because God knows that such ideas can
be brought forth in material forms ("imitable" by Nature). The imitative
relation is exactly converse to that which obtains in human art: nature
imitates the divine intellect that eminently contains it. Since God is perfectly
capable of realizing those forms that are suited to matter, his ideas represent
them "perfectly."
Human art, as Plato said, is doubly imitative: first of nature, of which it
can realize only the figures (color, which is also used in art, is imitated only
in its arrangement: the red of a painting is not an artificial quality, but the
thing itself applied to a different object), and then of the divine mind. What
is more important, however, is that God alone has the infinite power re
quired to realize, in the full sense of that word, the substantial forms of
which he has the original ideas. Created things, except for the most inferior,
lack, as we have seen (§5.4), sufficient power even to reproduce their own
forms. Animals require the assistance of celestial powers, humans that of
God. But even if humans or animals had the power to reproduce their
forms, they do not have the power to produce them afresh. Only by God,
one of whose names is causa sui, are the forms of things not only repro
duced but originated. Divine art differs from human art in having the
capacity to produce autonomous instruments, in whose actions it will in
deed concur, but which it does not need continually to operate. 42 The
41. "This art [i.e., handicraft] is indeed an imitation, but not an imitation of nature; it is an
amotion of the ars infinitaofGod himself, and like that art originary, primitive, creative, though
not insofar as it has created the world. Coclear extra mentis nostrre ideam [aliud] non habet exemplar
[The spoon has outside our mind no other exemplar: Cusanus ldiota de mente 2, Werke 5:240].
The spoon, no high product of art, is still something absolutely new, an eidos not hidden in
nature. The mere 'layman' [idiota] is the one who brings it forth: non enim in hoc imitor figuram
cuiuscunque rei naturalis [for in this I do not imitate the figure of any natural thing]" (Blumen
berg 195T268). Not the figure, indeed: but it can imitate the function.
42. "Formre naturales sunt actuosre, & quasi vivre; formre vero artefactorum tanquam
stolidre, & emorture, nullam effectricem vim habentes" (Coimbra In Phys. 2Ciq5a2, 1:217).

Authenticated
closest we can come to that capacity, apart from generation, is when, by

putting a projectile into violent motion, we temporarily impress upon it a
motive quality not unlike the heaviness that God has permanently im
pressed on earthy and watery things. But that quality, or impetus, is evanes
cent, bound to be overcome by the natural powers that oppose it. In all
other instances the most we succeed in doing is to modulate the activity of
the natural powers that only God or Nature can impart to matter.
Aristotelianism reached an accommodation of sorts between two strongly
held fundamenta whose harmony was less than certain: the omnipotence of
God and the experienced activity of natural things. A few earlier philoso
phers had found them incompatible, believing that "to whatever degree
efficacy is granted to creatures, to that same degree it is taken away from the
divine power of the Creator" (Suarez Disp. I8§1t2, opera 25:593). Any
efficacy in things would indicate an insufficiency in God's power. Against
that Suarez argues that it is through no insufficiency that God allows second
causes to act in his stead, but "through a voluntary and most prudent appli
cation of his power": it would be wasteful were infinite power to be em
ployed where finite power will do ('l[ 10, 596). The omnipotence of God is
preserved, since he not only could have chosen to apply his power directly,
but can at any time prevent second causes from operating by withdrawing
his concurrence, as he did when the companions of Daniel walked un
scathed through the furnace of Nebuchadnezzar (Daniel 3, a frequent
example in questions on second causes). At the same time the manifest
activity of Nature is acknowledged and given a rationale.
It should be clear that the analogy of nature and art, as a tool for the
scientific understanding of nature, requires delicate handling. It will not do
even to make instrumentality the ground of the analogy, since artifacts,
having no proper efficacity, cannot even be instrumental causes per se
according to Suarez's definition. 43 Artifacts cause nothing per se, and are
caused by nothing per se except the intentions of their makers. The com
parison of natural things and artifacts can yield no information about their
natures, but only about the manner in which those natures express
themselves.
43· Suarez says that "in all orders of things there can be instruments of this sort"-that is,
according to the fifth way of contrasting instruments and principals-and lists natural, super
natural, and artificial forms. But what follows makes it clear that what he means is that artificial
forms can have instrumental causes, not that they can be instrumental causes themselves
(SuarezDisp. 17§2'117, Opera25:5go).

Authenticated
Authenticated
[8]
Motion and Its Causes
D escartes's physics was largely discredited by the beginning of the

eighteenth century. Except among the sect of Cartesians, which
included, ironically enough, professors of the sort that
Descartes had once numbered among his opponents, the pre
dominant view was that of Huygens, who in 1693 told Bayle he could find
"almost nothing I can approve of as true in all the physics, metaphysics, and
meteores" of Descartes (To Bayle 26 Feb. 1693, Appendix, CEuv. 10:403).
Huygens treats his early "preoccupation" with Cartesian philosophy in terms
not unlike those that Descartes had used of his preoccupation with Aristo
telianism: a childhood disease, in which verisimilitude was taken for truth.
Even at the time of the Principles, some of Descartes's correspondents
were doubtful, especially about the fourth collision rule, which seems man
ifestly contrary to experience. According to that rule, a smaller body in
motion can never so much as nudge a larger body at rest, however quickly it
moves. The Latin Principles already includes an elaborate explanation in
tended to show why the rule does not hold of bodies immersed in fluid. In
the French Principles the remarks following the rule itself are amplified in an
apparent attempt to lay to rest the misgivings of Clerselier and others.
The falsity of the rules has preoccupied historians; their response has
often been to cast about for some obvious physical truth that Descartes
somehow overlooked: the relativity of motion, or the failure, already noted
by Leibniz, to satisfY the requirement that to continuous variation in the
cause there should correspond a continuous variation in the effect. Al
though it is true that a person who did not recognize those truths could have
committed the mistakes Descartes did, the failure to recognize them hardly
suffices to explain why he made just those mistakes and not others.
Descartes's successes-notably the first and second laws of motion-are of

Authenticated
Bodies in Motion
a piece with his failures. The demonstrations scarcely differ; nor, for that
matter, does the kind of empirical evidence adduced in illustration of their
conclusions.
I start with Descartes's definition of motion (§8.1). Descartes, explicitly
marking his distance from the Aristotelians, acknowledges only motus localis.
But even that does not quite capture the magnitude of his departure. Local
motion, for the Aristotelian, when it is natural has, like any natural change, a
definite terminus. A body out of its natural place is in potentia toward being in
that place; when it reaches its natural place, it stops. Violent change too
comes to an end, because it is always opposed by a tendency to some con
trary natural change, as when a heavy body is thrown upward. Only a few
preternatural motions are exempt from the requirement, and it is to them,
and to the incessant motion of the heavens, that one must look to find the
persistence that Descartes ascribes to every motion.
For Descartes, motion is always entirely actual, the instantaneous rupture
of a body from its neighbors. It is, moreover, from a definite place, but never
by nature to a place: it has, as I will show in the discussion of the second law,
a direction but no terminus. Nevertheless Descartes, like the central texts,
holds that motion is a genuine mode, though a peculiar one, of the mobile. It
is peculiar because it is a joint mode of the two bodies that are separated by
it, or, in Descartes's own words, reciprocal. Many commentators have misin
terpreted that claim, and taken Descartes to hold that motion is relative
while at the same time insisting that rest and motion are genuine contraries.
Like Daniel Garber, I think that Descartes's view is consistent; passages
where he seems to be acknowledging the relativity of motion turn out to be
passages in which he rejects the vulgar notion of "place" on which the
supposed relativity is founded.
The discussion of the laws of motion and the rules of collision I divide
into two parts. The first (§8.2) concerns the mathematical description of
motion, or what Kant called phoronomy, and the second concerns the causes
of motion, in particular the role of God. In a discussion of the first two laws,
I show that what is conserved according to them is not motus itself, which
Descartes defines as the instantaneous mode of "rupture" between a body
and its surroundings, but the modes of that mode-its direction and speed.
I then examine in some detail the third law, the rules of collision, and
Descartes's conception of quantity of motion. I emphasize three points:
first, that the archetypal phomomena to be saved in the rules were reflection
and the loss of motion by a body in a resisting medium, of which the leading
special case was refraction. The notorious Rule 4 was arrived at in part
because Descartes realized that pure reflection-in which a reflected body
changes direction without transferring motion to the reflector-is impossi
ble under the rules of collision he had formulated before he began work on

Authenticated
Motion and Its Causes [257]
the Principles. The second point is that the classification of physical situa
tions in the rules of collision works quite differently than in classical physics;
the consetvation of motion setves only to determine how motion is trans
ferred, and not what I call the outcome of the collision. Unlike his onetime
collaborator Beeckman, Descartes never treats speed as a signed algebraic
quantity, or rest as a limiting case of motion. Contrariness of direction and
contrariness of rest and motion, determine outcomes independently of
speed and volume. Finally, I argue that the geometric representation of
speed, which Descartes continues to conceive in the Aristotelian manner as
an intensive quantity, made natural a "hydrostatic" model of the transfer of
motion in collisions.
In §8.3 I turn to the causes of motion and the proofs of the three laws of
motion. Like Hatfield and other recent commentators, I emphasize the
metaphysical underpinnings of Cartesian physics, turning to Aristotelian
accounts of the three kinds of divine action Descartes appeals to in his
proofs: creation, consetvation, and concurrence. The principal issue among
commentators has been whether force pertains to God, to bodies, or-in
the most recent interpretation by Garber-to nothing at all. There is, I
argue, a way to reconcile the alternatives. What has made the issue difficult
is that in consetvation and concurrence, as in the actions that Soto describes
as being "owed to nature" by God (§7 .1), there is a determination of change in
one body by the features of another, but no causation except by God.
8.1. The Definition and Mode of Existence of Motion

Descartes's conception of motus has perplexed commentators, not least of
all because in the view of many it is internally inconsistent. The difficulties
in understanding it are several. There is, first, a significant change in
Descartes's conception from the early works to the Principles. Mter vig
orously defending against Morin the rather unorthodox claim in the
Dioptrics that motus and actio are one and the same, Descartes effectively
conceded the point in the Principles, criticizing his own former view.
The second difficulty lies in the discussion of place in the Principles.
Descartes, in good disputational style, argues his own definition in opposi
tion to two others: explicitly he opposes a relational notion of place, while
implicitly he opposes the Aristotelian notion of Ubi. Supposing that motus
localis is to be defined as change of place, neither of those notions of place
will do. Instead place is to be defined as the surface common to a body and
those immediately contiguous with it.
The third difficulty arises out of the definition of motus itself. Many com
mentators, mistaking the reciprocal translation of a body out of its neighbor

Authenticated
Bodies in Motion
hood for the relative motion of a body with respect to a reference frame,
have argued that Descartes cannot both define motus as he does and hold
that it is a mode of bodies. Although that criticism, as Garber has argued, is
misguided, still the logic of the mode in question, which seems to belong to
two bodies at once, and its temporality raise serious doubts.
Contrary to the Aristotelian conception, and indeed to Descartes's own
treatment of it elsewhere, motus, defined in the Principles as the rupture of
two bodies, is not successive. Reciprocal translation presents the strange
aspect of no sooner beginning than it must end. For all that the definition
tells us, two bodies that are already ruptured cannot be said to move with
respect to one another or to be at rest. I will show that Descartes handles the
successiveness essential to motion by an appeal to the notion of trajectory.
That notion, it should be noted, is not relative to a reference frame; speak
ing anachronistically, one might liken it to the much more recent notion of
a world-line.
The peculiarities of motion as it is defined in the Principles should put the
reader on guard against reading classical conceptions into his physics, even
if one takes them to be present only in ovo. In particular the notion of
reference frame is, I believe, simply not pertinent to his thinking. Far more
important is the task Descartes's definition was meant to perform, which was
both to provide a substitute for the Aristotelian definition of motion, with its
inadmissible machinery of actus and potentia, and to introduce an instanta
neous notion of change to seiVe as the subject of the laws of motion.
1. Vulgar conceptions. In the second part of the Principles, after showing
that "all variation in matter, and all the diversity of its forms, depends on
motion," Descartes turns to the nature of motus. The nature of motus, he
advises us, must be carefully distinguished from its causes, of which the first
is God.l The "vulgar" conception, according to which motion is the "action,
by which a body migrates from one place to another," must be replaced by a
conception that issues from the "truth of the thing" (§25, 8/1:53).
Somewhat unusually, the vulgar conception is not that of the Philoso
phers but of his own earlier self. In the Dioptrics he had called light "a certain
movement, or a very quick and lively action" (AT 6:84). Defending this use
of 'action' against Morin, Descartes responds that "the signification of the
word action is general, and comprises not only the power [puissance] or the
inclination to move, but also movement itself. "2 In his response, Morin
accuses Descartes of equivocating between motion and its cause, or between
potentia and actus. 3 Descartes stands his ground, and replies simply that
"movement is the action by which the parts [of luminous bodies] change
1. 2§23 and 36, AT 8/1:52, 61.

2. ToDescartes22 Feb. 1638,1:543, To Morin 13]ul. 1638,2:204.
3· To Descartes 12 Aug. 1638, 2:291, reading 'forme' for 'force', as Descartes did.

Authenticated
their place. "4 The exchange ends there, with Morin believing that Descartes
can't tell actus from potentia, and Descartes believing that 'action' is still the
proper term by which to designate what is transmitted from the sun to our
eyes. 5
The Descartes of the Principles, who has recently refreshed his knowledge
of Aristotelian terminology, writes that "I have always held that it is one and
the same thing that, when referred to the terminus a quo, is called 'action',
but when referred to the terminus ad quem or in quo recipitur [in which it is
received], is called 'passion'" (To*** [Hyperaspistes) Aug. 1641, 3:428). A
few months later he advises Regius in similar terms: actio is motus designated
with respect to the mover. 6 Motusis "always in the mobile, not the mover" (PP
2§25, 8/1:54); it must therefore be distinguished from its causes, which are
typically distinct from the mobile. None of those claims would dismay an
Aristotelian (cf. §2.3). They too were careful to distinguish between motus
and the causes of motus.
Though the sources on which Descartes drew in formulating his early
views are not obvious,7 the motivation seems clear enough. Descartes
wanted a single word to cover both actual motion and what he would later
call the tendency or conatus of a body not actually moving toward motion. In
the Dioptrics the particles through which the tendency he identifies with
light is transmitted are taken to be at rest; later he took them to be in actual
circular motion, which according to the second law of motion gives rise to a
conatus directed outward. The earlier view had forced him to assert that
motion and tendency to motion, or "action," were subject to the same laws
and, in the letter to Morin, to an assertion about the signification of the
word 'action' which to an Aristotelian could only seem an expression of
ignorance.
4· To Morin 12 Sep. 1638, 2:364.
5· The relation between motion, tendency to move, and action in Descartes's earlier work is
discussed in Prendergast 1975. Prendergast appears not to believe that the definition of
motion in the Principles amounts to a repudiation of the earlier view, in part because he takes
'force' (or "the power causing motion," as in the letter to More of April 1649, AT 5:404) and
'tendency' to denote the same entity.
6. ToRegiusDec. 1641, 3:454r. Later he adds that some might call the "force [vim] by which
[a hard body] admits the movement of other bodies" an action, but that to do so is incorrect,
since then the action would be in the patient and the passion in the agent (455). Though the
Thomist Soncinas believed that resistance to change is an actio of the patient contrary to that of
the agent (§2 .4), I know of no one who called the mere reception of motus an action, for the
very reason that Descartes gives.
7. See Garber 1g88:346n.6. Goclenius notes that actio can be used to denote motus, but calls
the usage "improper"-a solecism for actus (Goclenius Lexicon s.v. 'actio', P44l· But he also
records a classification of actiones under which local motion is an actio creaturarum naturalis
nonviolens. Toletus attributes to t:Je twelfth-century author Gilbertus Porretanus the view that
motus belongs to the category of action (Opera 4:83ra), but does not take the view seriously.
Henry More, on the other hand, writes that unlike Descartes he judges motus to be "that force
or action, by which the bodies you say are moving are pulled apart from each other" (More to
Descartes 23 Jul. 1649, AT 5:380)-so perhaps the "vulgar" conception had some currency.

Authenticated
[260] Bodies in Motion
Yet Descartes's concessions on motus and the identity of motus and actio
did not preclude his continuing a more fundamental disagreement with
Aristotelianism. The other vulgar conception Descartes rejected was the
conception enshrined in the definition of motus as the actus of a being in
potentia. Although Morin was not obtuse in taking the "action" of light
particles to be an Aristotelian potentia, the corresponding notion of conatus
in the Principles is clearly not a being in potentia (see §8.2).
Nor is motus itself the actus of a being in potentia. Descartes, not without a
certain slyness, suggests that his revision of the Aristotelian theory of natural
change retains local motion while rejecting the other kinds of motion. In
fact what he retains is at most an extensional equivalence with the Aristo
telian concept. More precisely: a body will be changing in place according
to Descartes at every moment in which it is changing in place according to
the Aristotelian definition; the converse will also hold provided that motion
in a vacuum is excluded. That equivalence, however, belies the underlying
difference in concept.
2. The definition of motion. Descartes begins his study of motion in the
Principles in good Aristotelian style: first a definition, and then an explana
tion of its parts. What one should understand by motus, "so that a determi
nate nature is attributed to it," is
a translation of one pan of matter, or one body, from the vicinity of those
bodies, that immediately touch it and are regarded as being at rest, to the
vicinity of others. Where by one body or one part of matter I understand all
that which is transferred at once, even if again that [body] itself should
consist of many parts that have other motions in them. And I say it is a
translation, and not the force or action that transfers [the body], to show
that it is always in the mobile, not in the mover, because these two are
usually not precisely enough distinguished; and it is merely a mode [of the
body], not some subsistent thing,just as figure is a mode ofa figured thing
or rest [a mode] of the thing resting. (PP 2§25, AT 8/1 :53-54)
The "translation" of a body is not the action that produces or stops move
ment: "motus and quies are nothing other [...] than two diverse modes of
the mobile" (2§27, 8/l:ss). Yet it is said to be a translation with respect not
just to any other body, but with respect to contiguous bodies "regarded as
resting [quiescentia]." With respect to those bodies, it is "reciprocal": "one
cannot understand a body AB to be transferred from the vicinity of a body
CD unless at the same time CD is understood to be transferred from the
vicinity of the body AB," and the same force or action is required for both
translations. Hence, "if we wish to attribute an entirely proper nature to
motus, and not [a nature] related to something else [ad aliud relatam], when

Authenticated
two contiguous bodies are transferred, one to one part, the other to an
other, so that they are mutually separated, we should say that motus is just as
much in one body as in the other" (2§2g, 8/l:ss-s6). But that would
offend against usage. We don't say when we walk that we and the earth both
move, but only that we move. Mter explaining why usage often treats motus
as proper to just one of the two things it really belongs to, Descartes con
cludes that "everything that is real and positive in bodies that move, and on
account of which they are said to move, is found also in the other bodies
which are contiguous with them, but which we merely regard as resting"
(§3o, S/I:s7).
Descartes then argues that although a body can "participate" in innu
merably many motus by virtue of being part of other bodies, there is a
"unique motus of each body that is proper to it." Though it may be useful to
divide a motus into parts, as Descartes himself does in his proof of the rule of
refraction, still "absolutely speaking, one should count but one [motion] in
each body" (§32, 8/I:sS).
Descartes's definition and his explanation of it thus raise two questions.
The first is more logical than physical: can motus, if what is real and positive
in it must be attributed equally to two things, be a mode of either? Leibniz
didn't think so: "if nothing else inheres in motion, except that respective
change, it follows that no reason is given in nature why motion should be
ascribed to one thing rather than to others. Consequently motion is nothing
real. And so in order to say that something moves, we require not only that it
should change position with respect to others, but also that the cause of
change-the force or action-should be in it" ("Animadversiones," ad
2§25, Ph. Schr. 4:369). 'Nothing real', as the context makes clear, means that
the attribution of motion depends on an arbitrary choice of the body re
garded as being at rest. The distinction between motion and rest is a distinc
tion in reason only.
The key step in Leibniz's argument is that since translation is reciprocal,
there is no reason to say that it belongs to one of the two bodies that are
separated rather than the other. Instead it belongs to neither. The ground
of that inference must be that nothing real can belong to two really distinct
things at once. Leibniz therefore reverts to the view that the basis in reality
of ascriptions of proper motion is the vis or action that causes it.
Descartes seems to hold that it is a mode of both. That position, if
Descartes had chosen to make it explicit, would have required explaining; I
will later attempt to provide one. But even if we grant it, a second question
presents itself, this time to the claim that a body has one motion at a time.
Many commentators have found it inconsistent to hold both that motion is a
mode of res extensa and to attribute to it the kind of relativity exemplified by
passages like this:

Authenticated
Bodies in Motion
And indeed, in order to determine the position [situs] of a body, we must

do so in respect of certain other bodies, which we regard as immobile: and
insofar as we do so in respect of various [bodies] , we can say that the same
thing at the same time both moves and does not move. So that when a ship
travels on the sea, a person sitting in the stern remains always in one place
[locus], if one defines that place by the parts of the ship among which he
keeps the same position, and the same person continually changes place
[locus], if one defines it by the shorelines, since he continually recedes
from one and approaches the other. (PP 2§13, 8/1:47, cf. §24, P53)
Change of place defined in this manner will not be unique, nor will it be
genuinely contrary to rest: a single body can "move" in contrary directions,
or be both at rest and in motion, depending on which bodies are taken to be
at rest. Yet Descartes, forgetting, as it were, what he admits in 2§ 13 and
2§24, defines in 2§25 a nonrelative conception of motion. His misunder
standing of the relativity of motion not only vitiates that definition but also,
according to the critics, leads him to formulate empirically mistaken rules of
collision. Indeed it is not hard to show that the laws yield inconsistent results
if equivalence of situations is defined as it is in classical mechanics. But
although Descartes's third law of motion and the collision rules derived
from it do not accurately describe what happens when bodies collide,
Descartes's recognition of a certain relativity in some conceptions of motion
is not inconsistent with the definition of 2§25. Cartesian motion, as Garber
has argued, is reciprocal but not relative.
I will argue further that the notion of frame ofreference, which is central to
the classical conception, has no application in Cartesian physics. I will offer
a reading ofthe crucial phrase 'regarded as being at rest' (tanquam quiescen
tia spectantur) to show that Descartes is not in fact taking the bodies so
designated to b~ a local frame of reference in relation to which motion is
defined.
3· Place and vicinity. One reason later philosophers have failed to under
stand what Descartes intended in defining motus is that they take him to
conceive place and space in classical terms. He does not. Since the defini
tion of motus rests on that of place, any error in understanding Cartesian
place will redound on the understanding of Cartesian motion. Descartes's
view is indeed a revision of the Aristotelian conception; but it is not, for all
that, the conception ofHuygens and Newton. I will start by briefly reviewing
the Aristotelian position, and then develop Descartes's.
The Aristotelians allotted to three terms of art the senses in which a thing
might be said to be at or in a place: Ubi, locus, and situs. Ubi, as we have seen
(§4.2), is that by which a thing is said to be here or there. It alone is a
genuine mode of bodies. Locus is defined in the Physics as the surface of the

Authenticated
bodies surrounding a thing. The third term, situs, defined in the Categories as
the orientation or posture of a thing, is of little importance in physics.
The Coimbrans, to establish that the terminus of local motion is Ubi,
consider three alternatives. Some philosophers hold that it is "a relation of
distance to the poles, [or] of the thing contained to the surface that con
tains it." The Coimbrans reject that out of hand. There is no motus in the
category of relation: relational properties are always acquired by way of
nonrelational fundamenta. Other philosophers take the terminus to be an
"entity resulting in the mobile from the surface of surrounding bodies," or
locus regarded as a mode of the mobile. The Coimbrans reject that on the
grounds that genuine local motion in a vacuum is possible. What remains is
Ubi, or "existence in space," which is just "the quantity of the mobile, accord
ing as it exists in this or that part of space, or a certain mode which quantity
takes up according as it coincides now with this, now with that part of space,
genuine or fictitious."
Space, in their view, includes not only what actually contains bodies, but
the "imaginary" space beyond the heavens, in which no body, but only God,
is present (In Phys. 8ooq2a4, 2:36gf). In imaginary space, there is Ubi but
no locus. Similarly, angels can undergo change of Ubi, but since they do not
occupy space, they cannot undergo change of locus. Although a new locus
will be acquired in the local motion of bodies (except in a vacuum) along
with a new Ubi, it is acquired only per accidens (3c3q2a2, 1:352). Local
motion, therefore, is change of Ubi, not change of locus.
Descartes, on the other hand, holds that corporeal substance and exten
sion or quantity coincide. Substance and space differ only as the singular or
specific from the generic: ''when a stone is removed from the space or locus
in which it exists, we suppose its extension also to be removed, insofar as we
regard it as singular and inseparable. But meanwhile we judge that the
extension of the place, in which the stone existed, remains and is the same,
although now that place is occupied by wood, water, air, or some other
body." What we call the same "place," generically, is nothing but a volume,
situated among other volumes, which can be filled with different individual
bodies (2§11-12, 8/1:46). It is not a mode of anything, except insofar as a
body happens actually to occupy a generic place. Singular place, or locus, to
which I will return in a moment, is a mode of the "singular" extension of a
thing-its extension, as opposed to extension in general-, and thus a
mode of the thing.
Simplifying a bit, I take Descartes to introduce two terms of art, situs and
locus, to replace the Aristotelian terms. Only locus, as it turns out, is relevant
to the definition of motus.
Situs is the space, singular or generic, that a body occupies, designated
with respect to "certain other bodies, that we regard as immobile." We can

Authenticated
Bodies in Motion
designate a situs without defining its boundaries precisely; in that loose

sense, the same "place" may be occupied by bodies having different figures
and magnitudes, and the same body may occupy differently designated
"places" at the same time.
Locus, on the other hand, is "the surface that immediately surrounds the
thing placed [superficie quce proxime ambit locatum]" (2§15, 8:48). That sur
face is "nothing other than a mode" of the surrounded body, if by 'surface'
we mean the entire surface common to the surrounded body and those that
surround it, and if we consider it to be the same surface so long as it keeps
the same magnitude and figure. If a ship in a stream is held up by the wind,
one may "easily believe that it remains in the same place, although all the
surrounding surface [i.e., the combined surfaces of the surrounding
bodies] changes" (49). That point will be crucial to establishing that the
earth is at rest.
So defined, locus, like Ubi and unlike the receptaculum of certain Platonists,
is not a substance but a mode of substance. But unlike Ubi it is implied in the
very idea of determinate extension (which Descartes always understands to
be actual, not potential). Since a vacuum-including the supposed imagin
ary space outside the heavens-is not just naturally but absolutely impossi
ble, Aristotelian arguments for distinguishing place from Ubi carry no
weight. Wherever there is motion, there is change of place, and conversely;
Ubi is superfluous.
Situs, too, is rejected, not because it is superfluous but because change of
situs is for several reasons not a suitable idea of motion. The first is that situs
is not a mode of bodies, but only a way of designating their extensions. If
motion is to be a mode, it cannot be change of situs. The second reason is
that situs, unlike locus, is not unique (that is the point of 2§13). If motuswere
defined as change of situs, each body that moves would not have a unique
motion, and motion in one direction would not be genuinely contrary to
motion in the opposite direction. The third is that if place, properly speak
ing, could be designated by relative locutions like 'outside of', then place
and extension would be distinct. As the Coimbrans note, those who deny
the existence of imaginary space do so on the grounds that "a distance or
interval capable of receiving body, but without body, cannot be understood,"
and for the same reason they deny that a vacuum can be produced, even by
God's absolute power (In Phys. 4cgq3a1, 2:68). That is precisely Descartes's
position: no vacuum even by God's absolute power, no imaginary space, no
place without body. s
8. Denying the existence of imaginary space allows Descartes to dismiss the problems raised
by Buridan for Ockham's account of motion (see §2.2). Ockham had denied any real distinc
tion between motus localis and the mobile; Buridan argues that the outermost heavens can move
through imaginary space, or rotate, and that motus is therefore not just a mode of the mobile,

Authenticated
The definition of motion refers not to locus itself, but to the closely related
notion of the "vicinity" ( vicinia) ofa body. A body is in the vicinity of another
if the place of the one coincides in part with the place of the other-if, in
other words, they are contiguous. Descartes explains his use of that term,
rather than locus, in the definition of motion: "I have added, moreover, that
translation occurs from the vicinity of the bodies contiguous to it to the vicinity of
others, and not from one place to another: because, as I explained above,
'place' is taken in various ways, and depends on our thought: but when by
'motion' we understand the translation that occurs from the vicinity of
contiguous bodies, then since only one set of bodies can be contiguous [to
the mobile] in the same moment of time, we cannot attribute several motus to
the mobile at the same time, but only one" (2§28, 8/I:ss). It is essential that
what counts as one body be "all that which is translated at once." If it were
not, then in Figure 8 below the putative body BC, as it slid along the hollow
cylinder KL, would not be translated from the vicinity of KL, AB, and CD to
that of another set. But since AB and CD are translated with it, the mobile is
actually AD or a body that contains AD. Once that is noted, motus appears to
be unique, since the set of bodies touching the mobile is at each moment
unique, and is different just in case the mobile moves.
That would settle the question of the relativity of motus, but for two
problems. The first is that Descartes insists (2§29) that the translation oc
curs out of the vicinity of notjust any contiguous bodies but "those only, which
are regarded as resting." The 'regarded as' would seem to introduce the rela
tivity to thought that Descartes intended to exclude from the strict sense of
locus. 9 The second is that what Descartes calls the determination of motion
cannot be uniquely fixed merely by reference to the vicinity out of which a
body is translated, and the others into whose vicinity it is translated can be
uniquely defined only after the motion is done. That problem, which is
crucial to understanding what becomes of the directedness of motion in
Cartesian physics, will be treated in the discussion of the second law in §8.2.
Here I concentrate on the relativity of motion.
4· Translation. The absolute fact by which mutual motion and rest are
distinguished Descartes calls translatio. Two bodies A and B have been trans
lated from each other's vicinity during a certain interval of time if at the
beginning of that interval they are contiguous and at the end they are not.
'Being contiguous' and its negation are symmetric relations; so too is trans
lation (2§2g, 8/I:ss), and there is no reason to regard it as inhering in one
but really distinct from it. In Descartes's view the world can neither be translated nor rotate,
since it does not, properly speaking, have a place at all.
9· In addition to the commentators cited in Garber, see Prendergast 1972 in Moyal1991,
4:104- Prendergast holds that "ifwe are to take this text seriously the reality of rest and motion
is destroyed." That, as we have seen, was Leibniz's argument.

Authenticated
K L
Fig. 8. Vicinity
body rather than the other. It inheres, independently of our manner of

conception, in both if it holds in either.l 0 Descartes calls it "reciprocal."
Both the logic and the temporality of translation are peculiar. Let the
mode in A which consists in being contiguous with B be denoted c, and the
mode in B which consists in being contiguous with A be denoted c'. Then
presumably even God cannot bring it about that one should exist while the
other does not. So c and c' are not even modally distinct. The same holds of
that mode sofA which consists in being apart from Band the correspond
ing mode of B. The same too must hold of translation, which for A consists
in having c and then s (which is contrary to c). Yet Descartes holds that in
general modes, among them motus, which belong to really distinct sub
stances are themselves really distinct (1§61, 8/1:30). It is no wonder Leib
niz decided that motus a la mode Descartes is nothing real.
The logical peculiarity shows up in a somewhat different way in an exam
ple that More took to be a kind of reductio of Descartes's view. 11 Mter
arguing that motus is reciprocal, Descartes considers why we don't or
dinarily say the earth moves when we lift our feet: "The principal reason for
this is that motus is understood to be of the whole body that moves; nor could
it be thus understood to be of the whole earth on account of the translation
of certain of its parts out of the vicinity of smaller bodies to which they are
contiguous: since often several such translations, mutually contrary, may
occur in it" (PP2§3o, AT 8/1:56). It's not hard to come up with examples.
In Figure g, the central body Bmust, with respect to the left-hand body A, be
said to move from left to right, although with respect to C, it must be said to
move from right to left. On the face of it those are contrary motions that a
single body cannot undergo simultaneously. The resolution of the paradox
10. "Motion and rest differ truly and modally, if by 'motion' is understood the separation of
two bodies from each other, and by 'rest' the negation of this separation. But when only one of
two bodies that are separating from each other is said to move, and the other to be at rest, in
this sense motion and rest do not differ except in reason" (AT 11:657; "Cartesius," a collection
of notes gathered together by Leibniz, was probably written in 1642, cf. 11:647). Garber has
emphasized the importance of this passage (Garber 1992:167).
11. "This article [i.e., 2§3o] seems to contain a most evident demonstration that translation
or local motion (unless it were merely a respect of bodies) is not in anyway reciprocal" (More to
Descartes AT 5:385). Unfortunately Descartes never responded to this part of More's letter.

Authenticated
I II
Fig. g. Contrary motions
lies in considering which motions those directions are the directions of.
Direction, .as we will see later, is a mode of motus. The motus that B has by
virtue of its mutual translation with respect to A and the motus it has by virtue
of its mutual translation with respect to C are distinct modes of B. The
directions are thus directions of distinct motus, and are therefore not
contrary.
It may seem that the proposed resolution flies in the face of Descartes's
insistence that each thing has but one motus proper to it ( 2§31). One
answer suggested by the passage quoted above is that the two motus might be
regarded as belonging not to the entire body Bbut to its parts-the top and
bottom halves, say. But that is unsatisfactory: B is solid, and its parts are at
rest with respect to one another. A second answer is suggested by a remark at
the end of §31. Descartes considers the motions of the wheels of a watch
carried by a man walking on a ship at sea. Each wheel has, he says, one
motion proper to it-that which corresponds to the change in its singular
place, or, in other words, to its translation as a whole out of the vicinity of
certain bodies touching it. It also has other motions by virtue of being
"adjoined to the man walking," and to the ship, and to the sea, and finally to
the earth, "if indeed the whole Earth be moving." "All these motions," he
concludes, "truly are in these wheels; but because so many are not easily
understood at once, nor can even all of them be known, it would suffice to
consider in each body that unique motion, which is proper to it" (2§31,
8/ 1:57). Since the translation by which motion is defined need only be with
respect to some contiguous bodies, not all, "regarded as resting," perhaps
neither A nor Cis suitable for defining the proper motion of B, even though
the reciprocal motions in B defined by them are truly in B. I suspect that
ultimately Descartes would, if presented with situations like that of Figure g,
or others yet worse, simply deny that B has a proper motion, which is to say:
B is at rest. 12
12. That solution is not satisfactory either. The situation of Fig. 9 occurs all the time in the
circularly moving concentric vortices that surround each star and planet. Take A, B, and Cto lie

Authenticated
The temporality of translation is just as problematic. Translation, though

it would seem that it can take place only at an instant, can never occur in an
instant. There is no medium between touching and not touching, and in
that sense the change from one mode to the other can only occur at an
instant. Nevertheless, after it has occurred, the two bodies will be a finite
distance apart, with other bodies intervening. Since every body that moves
through a finite interval of space does so in a finite interval of time, the
translation could not have taken place in an instant. Descartes is cognizant
of that peculiarity, I think, when in the proofof the second law of motion he
writes that although "no motion occurs in an instant," still we can "desig
nate" the instants at which a body is moving (2§39, 8/1:64).
The odd thing is that, if we follow Descartes's way of thinking, we can in
fact only pick out interoals of time in which we know that at some instant or
other two bodies now separated had the reciprocal mode of translation.
That mode no sooner appears than it disappears; its nature, it would seem,
is to be "borne toward its own destruction," to quote Descartes's erroneous
criticism of the Aristotelian concept of motus (see the discussion of the first
law in §8.2). Although what one might call a completed translation can never
occur in an instant, the translation referred to in 2§25, which is in a certain
sense never a completed motion, can only occur in an instant.
5· Succession. Although the motus defined in 2§25 is a temporally point
like translatio, Descartes elsewhere clearly takes motus to be successive. To
understand how he thought he could generate successive change from a
definiendum in which succession is lacking, it is worth considering first how
the Aristotelians treated succession. We have seen that in Aristotle's defini
tion in Physics 3c1 motus is understood to be a process that-if change of
substance be set aside-occurs through time. More precisely, it is a mode of
the forms taken on successively during a change, consisting in the tendency
of ~ach to lead to the next by way of being on the way to the terminus ad quem.
At least one Aristotelian, Fonseca, characterizes that tendency in terms of
the instantaneous introduction offorms one after the other. The intermedi
ate form reached during a change, considered in itself, ''will not obviously
satisfy the definition of motus or of acquisition, but only that of a form of a
certain perfection just acquired." But ''when [the form] is taken, not just
insofar as it has been introduced, but also insofar as it is immediately to be
introduced, now it will satisfy in every way the true and relevant definition of
motus" (Fonseca In meta. 5e13q9§2, 2:718-719). Since in local motion the
in adjacent layers of the vortex, A in the innermost, Bin the next, Cin the outermost. The layer
containing A moves faster than that containing B, and that one in tum faster than the layer
containing C. If at a certain moment A, B, and Call lie on a line through the center of the
vortex, as in Fig. 9 (I), a short time later they will be as in Fig. 9 (II). Yet B cannot be said to be at
rest, since if it were, it would have no conatus to move outward from the center.

Authenticated
"form" is the place attained at each instant in motion, we have it that the
mode of the mobile with respect to which it will satisfy the definition of motus
is that of being just about to take a certain place.
But for the Aristotelian motus is essentially successive. As the Coimbrans
put it, we cannot conceive the way and tendency to a form without succes
sion, since it is "the mobile gradually attaining the perfection of the form,"
where 'gradually' means "part after part" (Coimbra In Phys. 3c2q2a2,
1:336). What unites the successive acquisitions of partial forms into a single
motus is clearly the perfected form, the terminus ad quem common to all those
acquisitions. A partial form that is, in Fonseca's words, "immediately to be
introduced," is so because it is the next step on the way to the terminus.
Setting aside for the moment the conatus or inclination to move which
Descartes introduces with his laws of motion, and the conseiVing power of
God, Descartes has no such means by which to unite one instantaneous
change with another. There are no incomplete actus, no partial forms, in his
physics. There are only the actual joinings and disjoinings of bodies at
various times. Or indeed, since in each instant two bodies that touch are as
one, there are only actual bodies at various times. Change consists in the
rearranging of boundaries within the one big body, the indeterminate res
extensa we call the world.
It is not surprising that the dtifiniendum in Descartes's definition of motion
should be a punctual event and not an essentially successive entity like
Aristotelian motus. What is surprising is the untoward consequence I have
mentioned: motion with respect to something is defined only in that instant
when two bodies are rupturing. Yet he clearly thinks of some motions as
continuing through time. One of the examples he uses to illustrate the
claim that each body has but one motion is the following: "If the line AB is
carried toward CD, and at the same time the point A is carried toward B, the
straight line AD, which the point A describes, depends on the two rectilinear
motions-from A to Band from ABto CD" (pp2§32, 8/1:58, see Figure 10).
The motion of the point A toward B or toward D, and the motion of the line
AB, cannot be brought within the scope of Descartes's definition except by
extending it.
Now I think Descartes's definition can be extended consistently with the
claim that each thing has but one motion. The examples in 2§32-the one
given above and the example of a point on a rolling wagon wheel-show
that the successive loci taken up by a point continually moving define a
geometric figure such that to each instant of motion there corresponds
exactly one point of the figure. That holds independently of how the mo
tion is conceived.
To make things simple, consider two spheres E and F, which are at first
contiguous and sometime later not contiguous. That situation must have

Authenticated
Bodies in Motion
Fig. 10. Combined motions (PP 2§32, AT 8/1 :58)
arisen as a result of successive ruptures of E or For both with the bodies

surrounding them. We consider the bodies defining the place of Eat each
instant to be at rest, not only with respect to E, but with respect to each
other. (If Eis immersed in a fluid, for example, the bodies it is in the vicinity
of must be moving with respect not only toE, but to each other.) Abstracting
from whatever other motions they happen to be involved in at that moment,
we can consider them to be one body, call it G. There is no reason not to
include Fin that body. E, as it is translated out of the vicinity of G, is thus
translated out of the vicinity of a body ofwhich Fis part. It will leave behind a
series of places, in the generic sense, which joined together constitute a
tubelike virtual body, which I will call the physical trajectory of £.13
The physical trajectory of E is in fact independent ofF, which serves only
to determine the first vicinity included in the trajectory. Because it was
defined in abstraction from all movements but those of E, the trajectory is
nothing other than a potential part of the one big res extensa. In Figure 10,
the diagonal line AD is the physical trajectory of the point A through the
interval of time represented in the figure, and the whole rectangle ABCD is
the trajectory of the line through that interval. Similarly the "very intricate"
line traced by a point on the circumference of a rolling wheel is the trajec
tory of that point, if the point be regarded as distinct from the wheel. If we
imagine the trajectory to be generated by a succession of translations, each
of them adding its increment, then the trajectory, like its parts, will be
defined absolutely.14
13. Trajectories, as readers acquainted with Special Relativity will note, are like worldlines.
In both Galilean and Einsteinian kinematics, the fact that two bodies have separated is invar
iant under the admissible transformations. The distance between two points (or space-time
points) is likewise an invariant. For periodic motions, like the revolution of a point on a wheel,
the statement in the text would have to be modified, since in some reference frames the
trajectory will be a finite closed curve, and in others it will be an indefinitely long open path.
14- It is tempting to think that in the example of the wheel, Descartes is implicitly appealing
to a frame of reference in which the earth is at rest. But there is no taxis in the diagram; there is
only the line AD, which represents as simultaneous parts of one body the spaces successively
occupied by the point-body A as it moves. The diagram is not intended to represent A's

Authenticated
'Moving with respect to' can now be defined more generally. Suppose that
Fis contiguous with the trajectory of E. Then Ewill have left the vicinity ofF
at some time, and traversed a certain amount of space. Not only will it have
moved with respect to F in the narrow sense described earlier, but in a
broader sense it will have moved with respect to F even after it ceased to
touch F, by virtue of traversing the spaces intermediate between the space
contiguous with Fand its present locus. The locution 'moving with respect
to' continues to denote a joint mode of E and F, since a similar argument
shows that F has been moving through the same period with respect to E.
The phrase 'regard as resting' is a red herring for those acquainted with
later developments in kinematics. It does not mean 'regard as a rest frame',
not even in the eccentric sense in which, according to Garber, a moving
thing's rest frame is constantly moving with it (1992:171). It is true that
Descartes, perhaps to accommodate the Aristotelian conception of local
motion as an exitus from one locus to another, allows his reader momen
tarily to regard the motus as inhering in the mobile alone, and not in its
vicinity also. But the nonrelational facts upon which judgment about mo
tion are based are reciprocal facts of touching and not touching. Descartes
wants to define motion in the Aristotelian manner as the successive acquisi
tion of place-but without adverting to potentia and actus.
Descartes, as he was elaborating his succedaneum to the cursus of Eu
stachius, aimed at a particular sort of rapprochement with the Aristotelian
definition, to which the phrase 'in the singular instants which can be desig
nated while it moves' is the key. He takes the Aristotelian definition to be a
way of designating a thing that moves at just those times it is moving. That
Aristotle does so by invoking the machinery of actus and potentia is irrelevant.
What matters is to find something which can plausibly be called a mode of
extension and which will exist in bodies when and only when they are
moving. That something is translation. Though he calls it motus, it resembles
the object Aristotle was trying to define only in that respect.
There were, of course, other motives. Many critics, seeing what they re
gard as a sophistical application of the definition of motus to demonstrate
that the earth has no proper motion, have held that the definition was
devised ad hoc to yield that result. Like Garber, I am inclined to think that
Descartes was not led to his definition merely by the desire to reconcile his
physics with the Church's rejection of Copernicanism.l5 But neither was he
motion with respect to a reference frame in which A, B, and Care at rest, but rather A's motion
simpliciter, the "world-line" of the point-body A as it moves from the generic place A to the
generic place D.
15. Though Descartes was aware that the decision of the Court of the Inquisition against
Galileo was not "immediately thereby an article of faith," he nevertheless treated it as if it were
(To Mersenne Apr. 1634. AT 1:285). The light of faith he treated as superior to the light of

Authenticated
Bodies in Motion
simply working out problems internal to the physics he had developed up to

the point of embarking on the Principles. What he would have discovered or
rediscovered in his reading of Eustachius, Abra de Raconis, and perhaps
other central texts in the interval between the Meditations and the Principles
was that a host of problems, of which the controversy with Morin had
already given him a taste, attended the concept of motion. Those problems
he had ignored when he was writing Le Monde, either because he had
forgotten them or because he thought that he could appeal to an immedi
ate intuition of the nature oflocal motion (Monde7,AT 11:39-40). Some of
them he continued to ignore, like the debate about the interpretation of
potentia. Others, like that of defining the instant of motion or of determin
ing whether succession is essential to motion, had to be confronted,
whether because his own physics demanded an answer or because an Aristo
telian audience would expect one.
Descartes should not be treated as doing ineptly what Huygens and New
ton later did well. He is trying to do better what in his view the Aristotelians
had done poorly: to explain what is true of a thing at each moment of its
absolute motion. Excising potentia from the body of physics, however, leaves
him with nothing but Fonseca's 'form about to be introduced', the instanta
neous, punctual translatio from the vicinity of one set of bodies to another,
and the thought that by gluing together the spaces successively occupied by
a body one can construct a physical trajectory to which the relative concep
tion of place is irrelevant.
8.2. Persistence, Conatus and Quantity of Motion

The causes of motion are two: "the general cause of all the motions that
exist in the world," and "the particular cause, by which it happens that single
parts of matter acquire motions that they did not have before" (pp2§36,
8/1:61). The general cause is God himself. The "particular and secondary"
causes of diverse motions in single bodies are "certain rules or laws of
reason, and since the pronouncements of the Church on matters offaith were authoritative, its
expressed view on the movement of the earth functioned in Descartes's physics as an un
doubted truth, similar in that respect to the identification of matter with extended substance.
A comparison with the theory of Le Monde shows that the materials for formulating the
theory in the Principles are already there: the main vortex of the sun, which carries the planets
with it, the subsidiary vortices of the planets, the containment of each body always within its
own vortex (Descartes Monde 10, AT 11:6g-7o). The only missing ingredient is the definition
of motus that Descartes did not find it necessary to formulate (Monde 7, 11 :39). It is undoubted
ly no coincidence that the definition he does formulate in the Principles should allow him to
assert that he has "withdrawn all motion from the Earth more truly than Tycho and more
carefully than Copernicus" (PP 2§25, AT 8/1:53-54). But the definition also serves ends
unrelated to that of managing a prudent measure of agreement with the Church.

Authenticated
nature" ( 2§3 7, 8/1 :62). In this subsection I treatthose laws and the concep
tions of conatus and quantity of motion which accompany their proof and
application. In §8.3 I will examine the general cause, and the relation
between divine action and force.
1. Persistence: the first law. The first law, in both its early and later versions,
applies notjust to motion but to any "state," including figure and size. I give
the versions in Le Monde and the Principles:
Each part of matter, in particular, continues always to be in the same state,

as long as an encounter with others does not constrain it to change. 16
Each thing, insofar as it is simple and undivided, remains, quantum in se est,

in the same state always, nor is it ever changed [ mutari] unless by external
causes. 17
The later version differs from the earlier in two respects. Its scope is not
restricted to parts of matter, and it replaces the phrase en particulierwith the
phrase quantum in se est, which I have left untranslated.
Although the substitution of res for partie de la matiere seems not to have
had any particular significance in the parts of the Principles Descartes man
aged to write, the phrase quantum in se est is of some significance not only for
Descartes but for Newton's response to Descartes. Alan Gabbey, drawing on
I. B. Cohen's interpretation, argues that the phrase has two connotations.l 8
The first is "the limitation of the body's power to remain in this or that
state"-the fact that it must interact with others. The second is equivalent to
sua sponte, ex natura sua, and sua vi, where the stress is rather on what is
natural or innate to a thing.
In Descartes's thinking, as in that of his predecessors, persistence can be
regarded under a negative or a positive aspect. Negative persistence is the
permanence of a state in the absence of external causes. I will call this simply
persistence. Since in Aristotelian terms change by an external cause is violent
change, persistence will have, if it has a natural cause at all, a cause arising
from a thing's own nature. Persistence will in that sense be the natural
condition of the thing. If, on the other hand, a thing were to cease to exist
even though no external cause acted on it, its perishing, on the same
reasoning, would likewise be natural. Descartes, in an argument I will
16. "La premiere [regie] est: Que chaque partie de Ia matiere, en particulier, continue
toujours d'etre en un meme etat, pendant que Ia rencontre des autres ne Ia contraint point de
Ia changer" (Monde 7, AT 11 :38).
17. "Unamquamque rem, quatenus est simplex & indivisa, manere, quantum in se est, in
eodem semper statu nee unquam mutari nisi ii causis externis" (PP2§37, AT 8/1:62).
18. Gabbey 1971:32n.g1; Cohen 1964:148.

Authenticated
Bodies in Motion
discuss at the end of this subsection, holds-erroneously-that the Aristo

telians take motus to perish by its own nature.l9
Positive persistence is the permanence, or the tendency to permanence,
of a state in the face of external causes. This I will call resistance. Although
Descartes's first law asserts only that a thing will persist, in his second and
third laws it is clear that a thing will also resist certain kinds of external
cause. Resistance, as we have seen, was construed by some authors to be
passive, by others as active, or, more specifically, reactive. That ambiguity
remains in Descartes's physics, and, as I will argue at the end of §8.3, its
underground smvival helps to explain why he formulated the rules of colli
sion so as to create a marked asymmetry between motion and rest.
In §2.4 I discussed potentia resistendi, the power each thing has to remain
as it is and to resist being changed. Zabarella takes it to be the consequence
of an active nisus or striving of each thing to preserve itself as it is: "Re
sistance is, properly speaking, the repulsion [propulsatio] of an action apart
from any consideration of reaction; such repulsion has no other being than
privative, since it is the non-admission or the non-undergoing of action, and
follows upon the nature of the form striving to conserve itself' (De relnts nat.
440F). Striving to preserve oneself is primary; the nonadmission of the
action of external causes is, as Zabarella later makes clear, only one means of
self-preservation. Galileo, on the other hand, construes resistance passively.
In his treatise on the elements he writes that resistance is "permanence in a
proper state against a contrary action," which he does not distinguish "from
the thing's very existence whereby it endures" (Galileo Early Notebooks 244).
That is in certain respects very close to Descartes's view. The persistence
asserted in the first law is, once God's role in conserving created things is
understood, nothing other than their existence with respect to the continu
ing action of God according to his immutable will.
That things persist in their natural states, and resist being deprived of
them, was not a new idea. The Coimbrans, in their question on divine
conservation, agree that "to have an inborn propensity not to exist is con
trary to the nature of created things, since all incline to the opposite, and
have [... ] an ingrained desire to perpetuate themselves, insofar as that can
happen" (Coimbra In Phys. 2c7q10a3, 1:26g). Elsewhere they speak of
desiring one's own destruction as "most alien to the laws of nature" (In Phys.
1: 149). We have seen that in the forms of corruptible things the primary
expression of the tendency to self-perpetuation is generation, the reproduc
tion of forms in new matter, by which they achieve a simulacrum of the
immutability of God.
Motus, too, at least under certain circumstances, was thought to persist, or
19. See §8.3 on the role of self-preseiVation in arguments about divine conseiVation.

Authenticated
at least not to cease ofitself.20 Considering whether the circular motion of

the heavens could have been going on eternally, the Coimbrans note that "it
is not a contradiction, either from the side of the mobile, or from the side of
the motor, to suppose that motion in a circle should exist from eternity." The
reason that the motion of the heavens could not have been going on eter
nally is not that it must have ceased of itself in a finite time, but only that
there is a "repugnance" in supposing, for example, that there have been
infinitely many days and nights (8c2q7a3-4, 2:284ff). In a few quite special
cases, even terrestrial bodies might be thought to persist indefinitely in
motion. The Coimbrans argue that the continuing motion of a projectile is
owed to a vis impressed upon it by the projector, and not, as Aristotle
thought, to the impulsion of the medium (6c2q 1a8, 2:246f). They hold, as
we have seen, that nothing tends by nature to its own destruction, or equiv
alently that a thing will persist if not acted upon by others. They argue that a
body can move in a vacuum, that such a movement can be successive rather
than instantaneous (4cgq4-5), and that a vacuum is absolutely possible (In
Phys. 4cgq2, 2:67f); in fact they use such a motion to argue for the vis
impressa.21 It would seem to follow, though the Coimbrans do not say this,
that if a projectile moves in a vacuum, then the vis impressa, since it is
opposed by nothing and since it will not tend to its own destruction, should
not decrease. The projectile should therefore continue to move indefi
nitely.22
Abra de Raconis gives the argument a thorough airing, only to reject it:
This quality [i.e., the vis impressa] cannot be corrupted. It would either be
corrupted by its contrary, but it does not appear to have a contrary, since it
is well conjoined with gravity and levity, and these alone, it seems, could be
contrary to it. Or else by the absence of the projector, on which it would
depend for its conservation; but that is not so, since if the projector is
20. For late Aristotelian views of the persistence of motus, see Maier 1951, esp. pp. 295
304, Clavelin 1968:112"·. and Wolff 1978:212"'; for Galileo's view that circular motion persists,
see Maier 1967=488"', Clavelin 1968:239, and Wallace 1981:323. Wallace's chapter 15, a
discussion of Maier's essay on the impetus theory in Galileo, includes a detailed examination of
questions on projectile motion by several of Galileo's predecessors at the Collegia Romano.
21. The argument is also found in Abrade Raconis: "It can happen, as in a vacuum given by
divine power, that an Angel or someone else should hurl a missile, and this motion would not
be by something urging it on from behind, since there is no such thing; it therefore comes
from elsewhere, namely from an impressed impetus" (Phys. 248).
22. Mutius Vitelleschi, one of the professors at the Collegia Romano studied by Wallace,
argues on similar grounds that a preternatural motion, namely the circular motion of fire
around the heavens (see §7.1), could go on forever (Wallace 1981:335, 337). I am not
suggesting, ofcourse, that Descartes knew ofVitelleschi's unpublished notes, but that a version
of Descartes's first law would not have been alien to one strand of Aristotelian thought.
Descartes would more likely have come across such reasoning in Galileo's Dialogo sopra i due
massimi sistemi del mondo (1616; cf. Maier 1967:488); but it is with Beeckman that a fully
attested connection can be made.

Authenticated
Bodies in Motion
absent or even immediately destroyed, the impulse impressed on the

projectile can remain and bring about its motion. Or else on account of
the resistance of the medium, but that too is false, because otheiwise,
since in a vacuum there is no resisting medium, and since in it there could
occur the movement of projectiles and impressed impulses, an impulse of
that sort would never cease; therefore if that impulse were produced in a
projectile, it would persevere in it perpetually. (Phys. 249)
Referring to Suarez, he argues that nevertheless the "impulse" would cease,

"namely from the repugnance of the body in which it is impressed" (250) .23
The Aristotelian account thus postulates a force or impetus which is given
to a projectile by the projector. Some philosophers held that the force is
under certain circumstances opposed by nothing, and therefore will con
tinue to cause motion; others that it would, since it is extraneous, be op
posed under all circumstances and would always cease to act.
It is by way of opposing the very idea of such a force that Descartes's
sometime mentor Beeckman first formulates his version of the first law.
Commenting on an argument of Scaliger, he writes:
[Postil: Coelum semel motum semper movetur.] On Scal[iger], de Sub. exerct.

68, r [Scaliger Exer. 68'l[1, 105r-w6r].-It seems that one should hold
that the heavens are moved not by the Intelligences, nor by a continuous
volition [ nutu] of God, but rather by their own nature and that of their
place, they can never by themselves come to rest, once moved. (Beeckman
joumal18]ul. 1612, 1:10)
The argument is that since the heavens-by which he means the heavenly
spheres-are homogeneous, and everywhere equidistant from the center of
the universe, "there is no reason why they should be said to be able to come
to rest per se." They have no reason to rise or fall, and they are, so long as
they merely revolve, always in the same place. It follows that, if somehow
they are moved, they will continue to move forever. Sometime after June
1613, five years before his encounter with Descartes, Beeckman generalized
the conclusion. "No thing," he now writes, "once moved, ever comes to rest
unless on account of an external impediment" ( 1: 16). So the sun will con
tinue to move, even if it is not "bound" to a heavenly sphere. More signifi
23. Abrade Raconis's reference is inaccurate, at least in relation to the modern editions of
the Disputationes. The passage he probably meant is Disp. 21§3~27 (Opera 25:801 ). But Suarez
is there attempting to explain why the impetus of a body thrown upward disappears, a somewhat
different question. Nevertheless his conclusion suits Abrade Raconis's purposes: "because this
quality is in a contrary subject that resists its activity, on whose account alone this quality is
impressed by way of an instrument, it does not require always to be conserved there, and
because otheiWise the subject always resists it and its action, the nature of such a thing requires
that gradually it should cease from being conserved."

Authenticated
cantly, a stone, "propelled in a vacuum, would move perpetually; but the air,
which always strikes it anew, hinders it, and thus brings it about that its
motion is diminished" ( 1 :24). Beeckman now included not only all bodies,
celestial or not, but both varieties of motion, circular and rectilinear.
The Aristotelians never took the absence of external causes to be a suffi
cient account of persistence. Beeckman does. Against Scaliger, who had
argued that because the heavens are always in their natural place, their
movement has no natural cause and is therefore voluntary,2 4 Beeckman
replies that if we set aside the cause of their beginning to move, the cause of
their continuing to move is simply that there is no reason for them not to. 25
One need not suppose that God or the celestial intelligences constantly
intervene to keep them going. Beeckman cites the Ockhamist maxim, by
now common property: quod ergo fieri potestperpauca, male diciturfieri perplura
(]oumalt:w). Although the Aristotelians suppose that continuing motion
requires the continuing presence of a vis impressa, Beeckman holds that it
requires no cause at all.
Descartes, for his part, was careful to deny that conatus denotes anything
like a vis or power. Such language is to be interpreted as a kind of shorthand
for certain counterfactual claims: "When I say that globes of the second
element strive to recede from the center around which they turn, one
should not suppose that I impute any thought to them, from which that
striving would proceed; but only that they are so located, and incited to
motion, that they would really go in that direction, if no other cause im
peded them" (P.Jls§s6, 8/1:108). In the next paragraph he adds that "fre
quently many diverse causes act at once on the same body, so that some
impede the effects of others." Indeed, since there cannot be a vacuum, any
body that moves is notjust frequently but of necessity acted upon by others.
But even if it is not clear how the antecedent in the counterfactual should
be interpreted, the intention is clear: to construe conatus counterfactually
serves to eliminate active powers, especially voluntary, from corporeal
nature.
24. "What moves, moves so as to come to rest. It cannot come to rest except in its place. It
will therefore not move from its place, if motion has already occurred in order to attain that
place. So if a part of the Heavens moves by a purely natural motion, it is not in its place. But all
parts of the Heavens are in their place, and so they do not move so as to occupy their proper
place. Hence it is necessary that they should be moved by a motion other than a purely natural
motion" (Scaliger Exer. 68, p105).
25. In Le Monde Descartes denies, in language not unlike Beeckman's, that one need "give a
reason for the fact that a stone continues to move for a certain time after it leaves the hand of
the person who throws it" (Monde 7, AT 11:41). Nevertheless he also believes that every
existence, continuing or not, needs a cause. To that extent it is Beeckman who has the more
modern view. But in other respects, notably in their view of the naturalness of circular motion,
Descartes looks to be the more modern, in part just because he ascribes the conservation of
motion to God, whose action, he believes, is simple. The contrast illustrates well the difficulty,
or rather the imposition, in the notion of scientific advance or "modernity."

Authenticated
Bodies in Motion
Yet Descartes does not quite believe that the persistence of a body in its
present state needs no cause save the absence of reasons to change. Every
existence, whether new or merely a continuation, needs a cause at every
moment of its existence-the sustaining power of God, a voluntary cause
par excellence. The proof of the first law rests on that necessity. But since
the cause in question is supernatural, he agrees with Beeckman in denying
that continued motion requires anything like a vis impressa.
One last point. Descartes in Le Monde 7 says the Aristotelians believe that
motus, "contrary to all the laws of Nature, strives of itself to destroy itself'
( 11 :40). He liked this argument well enough to repeat it in the PP ( 2§3 7,
8/1:63). The Aristotelians agree that a self-destructive tendency would be
"most alien to the laws of nature" (Coimbra In Phys. 1:149; cf. §6.4). They
say, moreover, that motus tends toward quies, which is to say that it tends
toward a terminus whose attainment coincides with the cessation of motus
("motus of itself tends toward the quies of the terminus ad quem, at which it
remains") .26
But it would be surprising if such an outright contradiction had escaped
the scrutiny of the Aristotelians or their Medieval predecessors. In fact it did
not. Quies is twofold: there is the not-yet-changing quies of the terminus a quo;
there is the already-changed quies of the terminus ad quem. Only the first is
genuinely contrary to motus. 27 A seed that has not yet germinated has a quies
contrary to the motus of growth, or, in other words, the privation of the motus
to which it is naturally inclined. A full-grown plant, on the other hand,
though it is no longer growing, does not have the privation of growth, but
only the negation of growth. In §3.1 we saw that not just any negation of a
property is a contrary of that property: this is a case in point. The quies
toward which motus tends is not its contrary, but only its negation.
The point can be put another way: motus tends per se to its terminus, and to
quies only per accidens. What is intended in change is a certain end, whose
attainment is typically accompanied by the cessation of change. But in one
instance at least, the end is attained precisely by the perpetual continuation
26. Toletus In Phys. sc6text54 ( 229b2grr), Opera 4: 163rb; cf. Goclenius Lexicon s.v. 'quies',
P944· Some modem critics of Aristotle would agree that motion tends to its own destruction.
Kosman writes that motion is "auto-subversive, for its whole purpose and project is one of self
annihilation" (Kosman 1g6g:57; Waterlow tg82:1o6, 123). The Aristotelians would regard
that as a confusion between the per se tendency of motus to its terminus and the per accidens
tendency to rest.
27. "Motus can be called a kind of quieting [quietatio], that is, a way toward quies, which
partly coexists with the motus itself. \Vhat moves is partly coming to rest with respect to part of
the terminus ad quem and partly moving toward the remaining parts: it is therefore not contrary
to quies itself: for a contrary does not tend to its contrary" (Toletus In Phys. sc6text54, Opera
4: 163rb). John of St. Thomas distinguishes between a quies that consists in the conseiVation of
an acquired state, and to which motus is ordered per se, and a quies that consists in the privation
of motus (Nat. phil.tq tat, Cursus 2:16, 172; cf. Coimbra In Phys. 1:33, where the distinction is
credited to Scotus and Durandus).

Authenticated
of change: the motion of the heavens. Their end, which is the imitation of
God and the promotion of the well-being of sublunary substances, is accom
plished in the motion itself (Coimbra In Phys. 2qq22a2 and cgq2a1, 1:314,
3 27). In general motus does not, simply by virtue of being motus, tend toward
rest. If a certain kind of motus does naturally issue in rest, that is only by
virtue of intending a specific end, not rest itself.
Descartes's first law was no novelty in the physics of the period, not even
in Aristotelian physics. Though there certainly were Aristotelians, notably
Suarez, who would have disagreed with it, there was, especially in the treat
ment of preternatural motion, ample precedent. Nor was it a novelty to treat
motus as a state. Many Aristotelians took motus to be a mode either of the
form acquired in passing or, in local motion, of the mobile itself (see §2.2). 28
What is new is Descartes' denial that motion is in any sense an incomplete
entity; his application of the persistence of motion to every physical situation,
so that a body in motion, whether its motion is constant or not, always has a
tendency, determinate in direction and speed, to continue its motion; and
finally his insistence that the cause of any change in motion must be exter
nal to the mobile, so that in Aristotelian terms no change in motion is natural.
2. Direction, tendency, and the second law. It is in the second law that
Descartes begins to depart both from his predecessors and from his contem
poraries Galileo and Beeckman. That law takes the following forms:
When a body moves, even though its movement occurs most often in a
cuiVed line and though it cannot even make any [motion] that is not in
some way circular [...] , still each of its parts in particular tends always to
continue its own [movement] in a straight line. And so their action, that
is, their inclination to move, is different from their movement.29
Each part of matter, considered separately, never tends in such a way that
it would continue to move according to any cuiVed lines [lineas obliquas],
28. John Wild writes that for Descartes motion "does not have the structure offrom-to, but
is a fixed mode or quality, like figure which is either present or not present." He regards that
conception as a mistake that "may be traced back to [Descartes's] youthful inability to under
stand the Aristotelian descriptions of potency, as mediated by the late scholastics with whom he
was familiar" (Wild 1941 in Moyal1991, 4:32). But if to call motusa mode is a mistake, then not
only Descartes but virtually all the central texts make it-and Aristotle, who sometimes calls
motus a passion, a quality, or a quantity, does too. Descartes, though he jokes about the
incomprehensibility ofAristotle's definition, did not misunderstand the notion of potency. He
understood it quite well, and rejected it. Garber, too, seems to think that Descartes's treatment
of motus as a mode distinguishes him from the Aristotelians (Garber 1992:196). But although
Descartes certainly rejects the conception of that mode as a via or fluxus, by nature incomplete
and tending to a determinate terminus, the ontological status assigned to motus is precisely the
same.
29. "Lorsqu'un corps se meut, encore que son mouvement se fasse le plus souvent en Iigne
courbe et qu'il ne s'en puisse jamais faire aucun, qui ne soit en quelque fa<:on circulaire [...] ,
toutefois chacune de ses parties en particulier tend toujours ii continuer le sien en ligne
droite" (Monde 7, AT 11:44).

Authenticated
but only according to straight lines, even though many [parts] are often
made to tum aside through their encounter with others, and though
[...] in every motion a certain circle is made by all the matter which is
moved at the same time.30
Although the first law applies to every state of a thing, the postil to 2§39
makes it clear that the second law is concerned to spell out just what God
conserves when he conserves motion. "Every motion," it says, "of itself is
straight" (63), to which one should add that since every straight line has a
direction, every motion of itself has a direction. Determining that direction
is no problem: if a particle actually moves in a straight line, the direction of
its motion is at every instant the direction of the line. The direction in which
it "tends" to move is presumably also the direction ofthe line. The purpose
of the second rule is to show, by way of defining what God conserves at each
"instant," that there is such a tendency even when it is not manifested
without interference.
We have seen that the rupture or separation of two bodies, which is the
absolute fact underlying ascriptions of motion, cannot but occur in an
instant. I understand 'instant' here in the standard Aristotelian sense: an
instant is the boundary of a determinate duration or interval of time.31 The
instant of rupture is the boundary between the interval during which two
bodies are joined (or the duration of the one body compounded of the two)
and the interval during which they are apart. It is in that sense that rupture
occurs "in" an instant-not because an instant is akin to an infinitesimal
interoal of time but because rupture, like an instant, is a boundary between
durations-that of the compound body, and those of its separated parts.32
Descartes, as we have seen, treats the boundary as a mode, which in that
respect is analogous to the surface shared by contiguous bodies. That mode
30. "Unamquamque partem materi;e, seorsim spectatam, non tendere unquam ut secun
dum ullas lineas obliquas pergat moveri, sed tantummodo secundum rectas; etsi mult;e s<epe
cogantur deflectere propter occursum aliarum, atque [...] in quolibet motu fiat quodam
modo circulus, ex omni materia simul mota" (PP 2§39, AT 8/1:63; cf. g/2:85).
31. "Instans physicis est idem, quod non tempus, seu extremum temporis, quod improprie
dicitur tempus momentaneum, & indivisibile, seu individuum" (Goclenius Lexicon s.v. 'in
stans', p245; cf. Fonseca In meta. 5C13q10§2, 2:730A). Time was typically treated as a con
tinuous quantity. Just as a line segment was said (though not always by the same philosophers)
to be both bounded by and composed of indivisible points, an inteiVal of time is both bounded
by and composed of indivisible instants (needless to say, there were serious difficulties in
devising a coherent doctrine of indivisibilia). See Suarez Disp. 45§5, 1, Opera 26:551, 559rr (on
points) and §g1t, ,g, 26:586,589 (on time as a quantity).
32. All the difficulties that attended contemporary attempts to understand how a line could
be composed of indivisible points would, of course, also attend any attempt to understand how
a motus enduring through time could be composed ofinstantaneous ruptures. It is not clear to
me that Descartes ever confronted the problem (which is understandable, given his hostility to
the paradoxes of the infinite and to infinitesimals in mathematics). I do think, however, that
Descartes was not committed by his view to "discontinuous" time, and still less to the continual
re-creation of the world at each instant (cf. n.88 below).

Authenticated
is, in Descartes's way of thinking, itself capable of having modes. 33 One of

them is direction, denoted by expressions like ad illam partem; the other,
which figures in the third law, is speed.
Descartes calls that mode a conatus or inclinatio. The reason, I think, is
this. Suppose a body ruptures first from the vicinity of a certain collection of
bodies N, and later from the vicinity of another collection N. The mode of
being joined-with-and-then-separated-from N cannot be numerically identi
cal to the corresponding mode with respect to N. Descartes explains to
More that "there is one mode in the first point of a body A which is sepa
rated from the first point of a body B; and another that is separated from the
second point; and another that [is separated from the third], and so forth"
(To More 30 Aug. 1649, AT 5:405). It is not clear what sort of physical
situation Descartes has in mind-either a sliding motion or else the simulta
neous rupture of one surface from another conceived as infinitely many
simultaneous point-ruptures. In any case, it would surely follow that if A
ruptures first from B and later from another body C, a fortiori the two
modes will be different. The first law tells us, then, not that numerically the
same motus (defined as in 2§25) will persist, but that the same kind of motus
will persist. Presumably that would be a motus having the same direction and
speed. To have, at a certain instant, a conatus in a certain direction, then, is
to have in that instant the mode of rupture in that direction; the word
conatus connotes the persistence of that mode according to the first law.
What the law asserts of conatus is quite strong: at every instant the conatus
of a moving body has a unique direction in the direction of the tangent to
the trajectory of the thing at the point corresponding to that instant. It
asserts, moreover, that the manifestation of conatus in a body not acted on by
others would be rectilinear motion in that direction. What is proved, how
ever, in his presentations of the second law is rather less.
There is, first of all, no argument to show that the direction assigned to
the inclination should be that of the tangent. A look at two well-known
Cartesian textbooks shows that the gap was noticed. Regis and Rohault start
by obse:rving that if a body A is determined at a certain instant to move
toward a point B, and if its determination does not change until it reaches B,
it will describe a straight line from its starting point to B. If a body has, on the
other hand, traveled in a square, "one must conclude that at the four angles
where it changed its determination, it was deflected by encounters with
33· In a slightly different context (a response to the objection of Hobbes that determina
tion cannot be an accident of motus, since motus is itself an accident), Descartes writes that
"there is nothing disagreeable or absurd in saying that an accident is the subject of another
accident, as when one says that quantity is the subject of the other accidents [of corporeal
substance]" (To Mersenne 12 Apr. 1641, AT 3:355; on quantity as the subject of accidents, see
§5.3 above). Gabbey argues on the basis of that passage that speed and determination are
modes of the mode motus (1g8o:257).

Authenticated
Bodies in Motion
certain other bodies that resisted its movement and determination" (Regis
Cours 1pt204, 1:337; cf. Rohault System 1c13t5, 1:7g). A circle can be
regarded as an infinite-sided polygon; a body that moves in a circle there
fore "undergoes a continual violence through encounters with other
bodies," since otheiWise it would not change its determination at every
instant (Regis 1:338; Rohault 1:8o).
According to Rohault, "the only Determination that is natural to a Body in .
Motion" is a straight line ( 1:79). The problem then is to determine which
line the body will follow if released; that line will also represent its inclina
tion even when it is not released. Rohault claims that the line will be the
tangent. Unlike Descartes, he gives a reason: the tangent is that line that
makes the "least Angle" with the curved path that the body was traveling on
until now. 34
Regis's argument is, to my mind, more Cartesian in spirit. If the body A in
Figure 11 moves around the circle to F, it must be deflected at all the points
between. But suppose that at Fit encounters no obstacle. Though A will
have taken on many different determinations as it moves, "the later [ones]
destroy the earlier," so that at Fit will have only the "last" determination.
"But this carries it toward G, because it must take the inclination that the
curved line has at the point F, which is measured by the tangent FG"
(1pt2c14, 1:338).
Descartes, though he knew that mathematicians sometimes treated the
circle as an infinite-sided regular polygon, rejected the use of infinitesimals
in mathematics. On the other hand, he does treat the motion of falling
bodies as if it consisted in a series of segments of uniform motion, each at
slightly greater speed than the last, and each arising from the impact of a
particle on the falling body. Circular motion, too, arises either because a
body "is retained by something which obliges it to keep always at the same
Distance" (Rohault System 1:8o), or because it is at each moment resisted by
others along its path. Although it would be incorrect to suppose that
Descartes believed that there are infinitesimal motions, it would not be
entirely misleading to say that conatus, or the "first preparation for motion"
as he once called it (PP 3§63, AT 8/1: 115), is treated as if it were an
infinitesimal motion.
The arguments I have just examined presume that inclination, or "deter
mination," as Rohault calls it, is rectilinear; Regis's argument in particular
presumes that motion in the small must be in a straight line. That may seem
34· The argument is circular if the angle between a line and a curve meeting in a point Pis
defined as the angle between the line and the tangent to the curve at P. But one can show at
least that for any other line through Pbesides the tangent at P, the tangent will be between that
line and the curve. On the measurement of angles, see Hobbes Critique 23§2 1T, p27orr, and the
references cited there, as well as De corpore 2c14§7 11·, Opera 1:159rr.

Authenticated
obvious now. It is built into the differential triangle used by later authors to
calculate derivatives, and thus instantaneous speed and direction. But some
of Descartes's contemporaries did not find it so. During his first collabora
tion with Descartes, Beeckman wrote that
what is once moved, in a vacuum always moves, either according to a straight

line or a circular line, either on its own axis, like the Earth in its diurnal
motion, or around a center, like a ring. For since any minimal part of the
circumference is curved, and curved in the same way as the whole periph
ery, there is no reason why the circular motion of the Earth should aban
don this curved line and proceed on a straight line. For a straight line has
no more natural and regular a nature and extension than does a circular
line, because the parts of the circumference stand to the whole as the
parts of a straight line to the whole. (Beeckman]oumal23 Nov.-26 Dec.
I6I8, I :253; Cf. 256)
Oddly enough, Beeckman had earlier argued that a stone set on a revolving
wheel and suddenly released will not continue to move in a circle but "in a
straight line toward the place toward which it was directed in the moment it
was released" (]oumal16 Mar. 1618, 1:167). Though Descartes used such
examples to argue that motion in the small is rectilinear, Beeckman evi
dently either did not consider them in that light or denied their relevance.
Like Galileo, he continued to believe that circular motion is as "natural" as
rectilinear motion, and that motion in the small can be either curved or
rectilinear.
Descartes, then, could not take it to be obvious that motion, as it is
"comprised in an instant," must be rectilinear. In Le Monde he offers this
proof:
God conserves each thing by an uninterrupted action; consequently he

does not conserve it such as it may have been some time before, but
precisely such as it is in the same instant he conserves it. But of all move
ments the straight line is the only one which is entirely simple and whose
nature is comprised in an instant. To conceive it, it suffices to think that a
body is in action to move toward a certain direction, and that is the case in
each of the instants that can be determined during the time it moves. To
conceive circular motion, on the other hand, or any other that might
exist, one must consider at least two of its instants, or rather two of its
parts, and the relation between them. (Monde 7, AT II :44f)
The nature of the straight line "comprised in an instant" is simply to be

toward a certain direction. By 'direction' I mean the relation that holds
between a point A and any other point Bon a straight line through that point.

Authenticated
Bodies in Motion
F
Fig. 11. Movement along a tangent (Regis Cours 1pt2cq, 1:338)
There are many paths from A to B. Of those paths only the line itself is given
merely by giving the point A and the direction. Any other path-the circu
lar minimum imagined by Beeckman, say-requires more information, an
additional point Cat least (see Figure 12). The directions from A to C and
from C to B will be different. It is in that sense that only a straight line is
"entirely simple."
In the Principles the proof is this: 'The cause of this rule is the same as that
of the preceding, namely the immutability and simplicity of the operation
by which God consetves motion in matter. For he does not conseiVe it,
unless precisely such as it is in the very moment of time in which he con
setves it, without regard for that which may have occurred just before. And
although no motion occurs in an instant, still it is manifest that whatever
moves is determined, in the particular instants that can be designated while
it moves, to continue its motion toward some part, according to a straight
line, and not according to any cutved line" (Descartes PP2§3g, AT 8/1:64).
The simplicity here is not that of the motion but of God's "operation," and
Descartes merely claims that it is "manifest" that what is determined to move
toward some part will be determined according to a straight line. In the
example that follows, Descartes adds only that none of the "cutvedness" of
the stone's motion prior to the instant at which it is released can be under
stood to remain in it. But the point is not whether the determination of
motion in one instant includes its determination at earlier instants; it is
whether the instantaneous mode of rupture could include, among its
modes, a mode according to which it would move on a cutve. The whole
weight of the argument therefore rests on identifying the consetvation of
the mode of rupture at each instant with a simple operation of God; and the
simplicity of that operation must rest not merely on the fact that motion in
an instant has a direction, but that its natural path along that direction is a
straight line. We are thus led back to Le Monde and its criterion of simplicity.
The Aristotelians employed different criteria. In De ccelo Aristotle argues

Authenticated
c
~---·---~
A~ ~B
Fig. 12. Simplicity of rectilinear motion
that circular motion is simpler than rectilinear because its terminus a quo and
terminus ad quem are one: "Circular motion is perlect, while rectilinear mo
tions are imperfect, because they do not return to their starting point, like
circular motion, but rather they have a terminus which is most distant and
contrary to their starting point: and for that reason, as at the starting point
they begin to move, so, when they are at their terminus, they begin to come to
rest" (Aristotle in Thomas In de C(J!lo 2lect1, opera [Parma] 19:78). When
the naturalness of place is understood as its relation to the center of the
universe, the heavens-or any body moving in a circle around the center
though they are moved locally, do not move from one natural place to
another. Their motion, once begun, can be understood without reference
to any other place but the one they occupy. In that sense it is simpler.
A second criterion of simplicity, this time in the sense of unity, is found in
the Physics. Aristotle there defines the unity of motus in terms of the unity of
the mobile, the unity of the time through which the motus occurs, and the
unity of the form. In local motion, the form acquired in motion is place, and
a local motion is therefore one if it is to one place (Coimbra In Phys.
7qqsa2, 2: 151; see also q4, 2: 147ff). The motion of a falling body, there
fore, is one motion, as is the motion of an animal toward its food. But the
motions of the animal's parts, considered in relation to their elements, are
typically not one, since the animal moves its parts both upward and down
ward. More generally, when the motion of a body is neither upward nor
downward, its unity does not depend on the place it tends to but on the end
with which the agent sets it in motion. For inanimate things, the only
natural local motions unified with respect to those things themselves are the
fall of heavy bodies and the rise of light bodies.
Descartes denies that any local motion in itself has a terminus. Even the
tendency to move does not have a terminus. It only has a direction. Neverthe
less, if we consider the direction to be operating in the role of a terminus,
then Descartes's criterion of simplicity-a simple motion is that which has
one direction-appears to be a transposition into his physics of the Aristo
telian criterion of unity. Neither criterion, I should note, tells us anything
about the path of motion. But if a thing moves in one direction only, and if
direction is understood, as I suggested earlier, in terms of the relation of
points on a straight line, then clearly motion in one direction only will be

Authenticated
rectilinear. The tendency to simple motion will be a tendency to rectilinear

motion.
The interpretation I have given of Descartes's proofs of the second law
shows why Cartesian conatus is not potentia in disguise, but rather a substitute
for it. The difference, despite Descartes's insinuations in 3§56 and else
where, is not that to attribute potentia to a thing is confusedly to attribute
thought to it. I have said enough in Part I to show that the Aristotelians were
not committing any such error, and that Descartes's descriptions of them
are tendentious. The difference is rather that Cartesian conatus, though
determinate in direction (and in quantity), is directed toward no definite
place. In this it resembles not potentia but impetus, which is, as we have seen, a
kind of bare striving forward with no natural terminus.
3· The third law: contrariness and quantity of motion. The third law and the
rules of collision that derived from it have received far more attention than
the first two laws. On the face of it there is an obvious reason, succinctly put
by Paul Tannery in his notes to the French translation of the Principles:
"While the two preceding laws are today considered to be truths scien
tifically acquired, the third has been rejected [ruinee] from the seventeenth
century on through the work of Huygens on the impact of bodies. It is on
this point that the principal error of Descartes' physics bears" (AT g/ 2:86,
note c). The falsity of the third law infects the rules of collision as well, and
all subsequent explanations that depend on them.
But the clear sense of many historians that Descartes's adoption of the
third law calls for explanation in a way that his adoption of the first two does
not arises, I think, from a dubious method of interpretation. The method is
to start from the true laws, or Newton's laws, and to understand Descartes's
laws as deviations from them, as a movement of thought that fell short of its
terminus ad quem. It is exemplified in the appendix by Tannery added to the
French translation of the Principles. Tannery begins with the classical laws of
the conservation of the movement of the center of gravity and the conserva
tion of kinetic e11ergy. He then shows that the rules of collision lead to
deviations from those laws (g/2:328ff). Numerous subsequent treatments
have followed the same path, explaining Descartes's views in terms of what
one might call the aberrational pattern. The pattern reflects an almost Carte
sian sense that error results from the perturbation of a faculty that would
otherwise track the truth.
Not only does Desmond Clarke, for example, following an almost univer
sal practice, rewrite the rules in modern notation (instead of using propor
tions, as Descartes does-the point of insisting on the difference will be
come clear later). Clarke makes it his primary task to uncover the "deep
conceptual confusions" that led Descartes astray, and to show that, however
painfully, he was "genuinely groping his way towards a satisfactory dynamical

Authenticated
account of familiar natural phenomena" (Clarke 1982:213, 226). He ex

plains the "unsatisfactory character" of Descartes's rules by taking them to
be "tentative efforts to formulate a dynamics." Efforts they were, but tenta
tive they were not, at least in presentation. 35 There were, in any case, many
other ways Descartes could have gone wrong, so the tentativeness of the rules
tells us little about the way he did go wrong.
The aim of the analysis that follows is to restore some of the context in
which Descartes worked. It will, I think, illuminate Descartes's thinking in a
way that no analysis for which Newtonian physics is a terminus ad quem has
done. I will analyze the law and its corollaries in terms of a hierarchy of
contraries: state, direction, volume, velocity (or total speed), and quantity of mo
tion. The reader familiar with the third law will note that quantity of motion
is at the end, not the beginning of this list. I will then examine the con
trariety of rest and motion and the Cartesian treatment of quantity of mo
tion, showing that the rules of collision are to be understood to a large
degree in the nonquantitative setting of the Aristotelian notion of con
trariety, and that even the quantitative part of the rules is best understood in
terms of the Aristotelian treatment of quantity of motion as an intensive
quantity.
It is here that Descartes's use of proportions becomes significant: his term
"degree of speed" (gradus celeritatis, degre de mouvement) typically denotes the
aliquot parts of a total speed that is never specified. Hence the use of propor
tions rather than algebraic formulas like Huygens's ax+ by in his determina
tions of the exchange of quantity of motion. 36 The conclusion will be that
the neglect of the relativity of motion, or of the continuity of outcomes that
Leibniz expected, is not a conceptual failing on Descartes's part, as if he had
overlooked an aspect of motion that would have been obvious.
The third law (in Le Monde, the second law) is stated in the following
terms:
When a body impels another, it cannot give it any motion without losing at
the same time the same amount of its own motion; nor take from it any,
35· Descartes consistently took the difficulty that others found in understanding the rules
to be a sign not that they were incorrect, or only approximately true, but that they were not
sufficiently clear (see To Clerselier 17 Feb. 1645, AT 4: 183; Entretien avec Burman ad 2§46, AT
5: 168). Gab bey, I should note, has argued that the rules of collision were added very late in the
writing of the Principles (Gabbey 1980:262~"). Costabel concludes that the rules are "only a
sketch," the principal part of an "attempt at translation[ ... ] that aims to put simple and true
instruments at the disposal of the average person" (196T249). But they were not, of course,
presented as mere attempts-a point Costabel himself underlines.
36. See the "Troisieme Partie" of the 1652 draft ofHuygens's De motu corporum ex percussione
(OEuv. 16:gg). The published De motu adheres to the proportional mode of presentation. I
know of no passage where Descartes uses a formula like ax+ ify to denote the total quantity of
motion of two colliding bodies.

Authenticated
without augmenting its own by the same amount. (Descartes Monde 7, AT

11:41)
The third law ofnature is this: where a body that moves meets another, if it
has less force to continue according to a straight line than the other [has]
to resist it, then it is deflected in some other way, and while retaining its
motion gives up only its determination; but if it has more [force], then it
moves the other body with it, and however much it gives [the other] of its
own motion, it loses the same amount. (Descartes PP2§4o, AT 8/1 :65; the
French does not differ significantly)
The explanations after the rules bring out two points. The first is that the
primary experiences upon which Descartes bases the rules comprise two essen
tially different outcomes of encounters among bodies. The first is reflection;
the second I will call absmption, of which refraction is a special case. Since
the rules tend to be treated in isolation from their application, and since
preserving the phenomenon of reflection was crucial to the formulation of
the problematic Rule 4, it is worth dwelling on these two archetypal
outcomes.
Reflection consists of the reversal of that part of a thing's motion which is
normal to the reflecting surface.The motion AB of a particle toward a
surface CD is divided into a component AFparallel to CD and a component
AE normal to CD (see Figure 13). Only AE is opposed by the surface CD.
That surface, one should note, is part of a body which is supposed to be
immovable (inebranlable, says Regis); otherwise the law of reflection will not
hold precisely. The vertical component, successfully resisted by the immov
able CD, is turned into its contrary BF. The horizontal component AF, which
is unopposed, persists, so that in the interval of time tsucceeding and equal
to the interval in which the particle moved from A to B the horizontal
component of the particle's motion will be BH. Since the entire motion of
the particle must in the interval t bring it to the circumference of the circle
CAGD, the vertical component of its motion can be constructed by drawing
a vertical line upward from H to G. That segment is equal to BF (Regis Cours
1pt2c181), 1:352f; cf. Descartes Diop. 2, AT 6:g7, 102-104).
Reflection, then, is the archetype of the first part of the third law. 3 7 The
point of dividing the motion AB into the components AE and AF is to
remove from oblique reflection that which distinguishes it from perpen
dicular reflection, thus preparing the way to a straightfmward application of
37· Garber credits John Schuster with remarking on the importance of reflection in
Descartes's understanding of collision (Garber 1992:36on.4o). Descartes, as I will argue later,
discarded an earlier rule of collision in which the situations of Rules 4• 5· and 6 received
uniform treatment because, according to that rule, a body colliding with another body at rest
would never be reflected.

Authenticated
c D
E B H
Fig. 13. Reflection (after Regis Cours 1:352)
the first part of the law. The reflecting body, assumed to be immovable, has
all the force it needs to resist the reflected particle, and so the particle
changes direction without changing speed. Since it gives up none of its
motion to the larger. body, and does not change in volume, quantity of
motion is trivially conserved in reflection, just as the third law would have it.
The archetype of the second part is the movement of a solid body
through a resisting fluid. In Le Monde, Descartes turns the tables on those
who worry about an impressed force to carry projectiles along: "Having
supposed the preceding [rule], we are exempt from the difficulty that the
Doctors [i.e., the Aristotelians] find themselves in when they want to give a
reason for the fact that a stone continues to move itself some time after
leaving the hand of the one who threw it: for one should rather ask us why it
does not continue always to move. But the reason [for that] is easy to give.
For who could deny that the air in which it travels puts up some resistance to
it?" (Monde 7, AT I I :42). Certainly not the Aristotelians; for some of them,
in fact, the absence of resistance would entail that the movement was instan
taneous. Descartes adds:
But if one fails to explain the effect of [the air's] resistance according to
our second Rule, and if one thinks that the more a body can resist the
more it would be capable of arresting the movement of others [...] : one
will right away have a great deal of difficulty in giving a reason why the
movement of the stone dies away rather when it encounters a soft body,
whose resistance is mediocre, than when it encounters a harder body,
which resists it more. And also why, as soon as it exerts a little effort against
[the harder body], it turns back immediately as if to retrace its steps,
rather than stopping or interrupting its movement on that account. (ib.)
The second part of the third law, then, explains why motion in a fluid
continues but is slowed down. That, in Aristotelian and Cartesian physics, is the
only condition under which local motion in terrestrial bodies can naturally
occur. Refraction, the object of Descartes's first successful work in mathe
matical physics, is a special case of motion in a fluid medium: it results from

Authenticated
[2go] Bodies in Motion
the momentary slowing down of a particle moving across the interface be
tween media of differing hardness.
In accordance with the two archetypes, the rules of collision have effec
tively two kinds of outcome:
(i) reflection: reversal of direction, without loss of motion (Rules 1, 4, and

the second part of 7);
(ii) absorption: continued motion in the same direction with transfer of
quantity (Rules 2, 3, 5, and the first part of 7).
Rule 6 is exceptional in combining both reflection and transfer of motion,

but that, as we will see, is because its conditions are a perfect compromise
between those that give rise to (i) and those that give rise to (ii). In what
follows I will first quickly lay out the hierarchy of contraries that I will use in
my analysis, and then give a brief account of each law, after which I will take
up the contraries rest and motion.38
The hierarchy is summarized in Figure 14. The first two entries, state and
direction, are two-term polarities. Even though Descartes takes any direction
to be contrary to any other, in the rules the only contrariety that matters is as
between one direction and the opposite direction. In keeping with his
terminology I call the opposites 'right' and 'left'. The last three rows are
continuous quantities. What matters in the rules is whether one body, desig
nated B, has a greater, smaller, or identical quantity. It is to those relations
that the contest model delineated by Gabbey applies.3 9
I have arranged the contrarieties according to two principles. The first is
that the rules of collision can be perspicuously presented in terms of a series
of "tiebreakers" that successively determine the outcome of contests. Ac
cording to that model the "stronger" of the two contestants wins-which is
to say that it manages both to persist in its condition as far as possible and
that it forces the other to change in such a way as to make its own persistence
possible. This accounts for the order of the last three rows. The second is
that Descartes treats changes of state, direction, and velocity (volume is
fixed, and quantity of motion is therefore proportional to speed) as qualita
tively different kinds of change.
To the list above I add one more pair of contraries, union and disunion.
When the outcome of a collision is the continued motion of the two bodies
38. The longer version ofGabbey's essay lays out a "two-fold contrariety model" ofthe rules;
a close reading of the account shows that there are three contrarieties: rest and motion,
direction, and quickness or slowness of speed (Gabbey 1g8o:26of). The latter, though it is
mentioned in 2§44 as a kind of contrariety-slowness being to quickness as rest to motion-is,
in Rule 7, absorbed into a more complicated comparison of quantities of motion.
39· See Gabbey 1971:16ffand Gabbey tg8o:243rr, citing Heriveltg65:4gr.

Authenticated
Contrariety Terms Symbol
State Rest, motion s,s*

Direction Right, left d,d*
Volume <,>,= M
Velocity (total speed) <,>,= v,v*
Quantity of motion <,>,= Q
Fig. 14. Contrarieties in the rules of collision (symbols with a* denote the condition
after collision)
in the same direction, they will in fact be moving with the same speed and
will therefore be at rest with respect to one another according to the defini
tion in 2§25. But that amounts to saying that they become one body or, more
pregnant, that the loser is forced to give up its integrity and to become part
of the winner. Descartes, in a paragraph I will examine more closely later,
writes that each body-one should say "notional" body-has, ifjoined with
another, a force [vis] to resist attempts to separate it from the other, and
that if it is disjoint from another, it has a vis to remain disjoint. But being
joined with another is nothing other than being at rest (in the strict sense of
2§25) with respect to it, and being disjoined or ruptured from another is
being moved with respect to the other. The overcoming, therefore, of a
body's state of rest, and the dissolution of its integrity, are the same occur
rence viewed under different aspects. But because union and disunion are
rather the effects than the causes of change, I have left them out of Figure
14.
For the sake of convenience I now give the rules in full, as they appear in
the Latin Principles (2§46-52, AT 8/1:68-70). 40
[Rule I] If two bodies, say B & C, were entirely equal, and moved equally
speedily, Bfrom right to left, and Ctoward Bfrom left to right, then when
they met each other they would be reflected, and aftetward strive to move,
B toward the right and C toward the left, giving up none of their
quickness.
40. For the French, which incorporates significant additions by Descartes, see 9/2:89-93
and Alq. 3:196-204. I translate velox and its derivatives by 'speedy' and so forth, celer and its
derivatives by 'quick', and so forth. Picot uses vitesse, vile, and so on, for both.

Authenticated
Bodies in Motion
[Rule 2] If Bwere even a little greater than C, and the rest as before, then
only C would be reflected, and both bodies would move toward the left
with the same quickness.
[Rule 3] If the volume [mole] were equal, but B moved even a little more
quickly than C, not only would both strive to move toward the left, but also
there would be transferred from B to C half of the quickness by which [B]
exceeded [ C]: that is, if there were at first six degrees of quickness in B,
and only four in C, after their collision each would tend toward the left,
with five degrees of quickness.
[Rule 4] If the body Cwere entirely at rest, and were slightly larger than B,
then however quickly B moved toward Cit would never move C; but would
be repelled by it in the contrary direction: because a body at rest resists a
greater quickness more than a lesser quickness, and this in the ratio of the
excess of the one over the other; and so there would always be a greater
force in C to resist, than in B to impel [ C].
[Rule 5] If the body Cat rest were smaller than B, then, however slowly B
moved toward C, it would move [ C] with it, namely by transferring to [ C]
part of its motion, so that both [bodies] would afterward move equally
quickly: namely, if Bwere twice as large as C, it would transfer to [ C] a third
part of its motion, because that one-third part would move the body Cas
quickly as the two remaining parts of the body B which is twice as large.
And thus after Bmet with C, it would move one third part more slowly, that
is, it would require as much time to move through two feet of space as
before to move through three. [...]
[Rule 6] If the body Cat rest were most precisely equal to the body Bwhich
is moving toward it, it would be partly impelled by it, and partly repel it in
the contrary direction: namely if B came toward C with four degrees of
quickness, it would communicate to Cone degree, and be reflected with
the remaining three toward the opposite direction.
[Rule 7] If B & C were moving in the same direction, C more slowly, B
following it more quickly, so that it caught up with [ C], and if Cwere larger
than B, but the excess of quickness in B were greater than the excess of
size inC: then Bwould transfer only as much of its motion to C, as [would
be required] for both afterward to move equally quickly and in the same
directions. If, however, on the contrary the excess of quickness in B were
less than the excess of magnitude of size in C, then Bwould be reflected in
the contrary direction, and retain all its motion. And these excesses are
computed thus: if Cwere twice as large as B, and B did not move twice as
quickly as C, it would not impel [ C], but would be reflected in the contrary
direction; but if it moved more than twice as quickly, it would impel [ C].
Namely, if C had only two degrees of quickness, and B had five, two
degrees would be taken from B, which, being transferred to C, would
effect only one degree [of quickness], because Cis twice as large as B: and
by that it would result that the two bodies Band Cwould afterward move
with three degrees of quickness.

Authenticated
To keep matters simple, I concentrate on the Latin Principles, leaving aside

the French additions and revisions, and the reformulated proof of Rule 4 in
a letter to Clerselier.41
The rules fall first of all into two groups according to the contraries rest
and motion. Rules 4, 5, and 6 treat the cases where C is at rest and B in
motion; 1, 2, 3, and 7 the cases where both are in motion. 42 In the first
group, with C at rest and B in motion, contrariety of direction is inapplica
ble, and we have three cases: M8 < M 0 M 8 > M 0 M 8 = M 0 corresponding to
Rules 4, 5, and 6, respectively. I use the word outcome to denote the states
and directions of the two particles after the collision, whatever the quan
tities of motion may be. The outcomes, according to the contest model, are
as follows:
(i) [Rule 4] If M8 < Me then B loses. The direction of B after the

collision, d* 8 , will be contrary to its direction ds before, since that
ends the conflict between B's state and Cs state, which will not
change since C wins. C, it should be noted, maintains its integrity.
(ii) [Rule 5] If M8 >Me then Bwins. Cmust change its state from motion
to rest. The question then is how much motion B gives to C.
Descartes holds that Cnot only changes its state, but loses its integrity
as well. It therefore must move equally quickly as B.
(iii) [Rule 6] If M8 = Me, then since there is no comparison of motion
with rest, the remaining tiebreakers-speed and quantity of
motion-are irrelevant; with no tiebreakers left, the only possibility
is that both bodies lose and both win. C is forced to take on the
contrary state, that is, to move, while B is forced to reverse its direc
tion. Both retain their integrity.
It is essential to notice that in these three rules-the "rest" rules-the

velocity of B is irrelevant to the outcome, where by 'outcome' I mean change
or lack of change ins, d, and integrity. Descartes subsumes Rule 4 under the
third law by supposing that the resistance of C toward being moved is always
greater than the force of B toward moving it, on the (insufficient) grounds
that Cs resistance is proportional to B's speed. 4 3 I will return to that claim
later.
41. See To Clerselier 17 Feb. 1645, AT 4:183-187. An analysis and translation of the French
rules and the letter to Clerselier are given in Garber 1992:248-262. See also Clarke 1982,
appendix 2 (reprinted in Moyal1991, 4:110-122).
42. The following analysis is closest in spirit to those of Costabel and Gabbey, although I
have taken more liberties with the order of the rules. Costabel subordinates the contrariness of
rest and motion (which he calls a "contrariete de vitesse" on the basis of 2§44, see Costabel
196T243) to the contrariness of direction, which allows him to take up Rules 1-3 before Rules
4-6.
43· The grounds are insufficient because they apply equally well to the situations of Rules 5
and 6. Garber suggests that the force of resistance is proportional also to the volume of the
resisting body, so that Cs force of resistance will be equal to Mev (Garber 1992:240, 358).

Authenticated
Bodies in Motion
In the remaining four rules, where both bodies are in a state of motion,
contrariety of direction is applicable, and so they in turn fall into two
groups. In Rules 1, 2, and 3 the directions of Band Care contrary. The
symmetry of right and left allows the cases MB < Me and MB > Me to be
regarded as one. We therefore have two cases: MB >Me and MB =Me. When
the volumes are equal, the next tiebreaker is speed, and again symmetry
allows the cases % < vc and % > ve to be combined.
(iv) [Rule 2] If Ms >Me then Closes. It therefore changes direction and

is joined with B. Descartes considers only the case where tis = vc. In
that case the velocity of the resulting body CBwill be lis· My suspicion
is that the other two cases would have been treated analogously to
Rule 5·
(v) [Rule J] If Ms = Me then speed is the tiebreaker. If tis > Vc then C
loses. It changes direction and is joined with B.
(vi) [Rule 1] If Ms = Me and tis= Vc, then there is no tiebreaker (the
quantity of motion will be the same for Band C). At least one of the
two bodies must change direction, since they cannot interpenetrate.
But B does not resist the action of C more than C resists the action of
B. So they both change direction, and neither yields any motion to
the other. Nor, of course, does either lose its integrity.
Quantity of motion, it will be seen, has played no role in determining

winners and losers. Its role is limited to determining, in those instances
where a transfer of motion is mandated, the amount to be transferred. Any
arbitrariness is removed either by supposing that the two bodies move as
one, or that (in Rule 6) the outcome of a tie is a mixture of the two ways in
which the tie could have been broken.
Only in the last rule is quantity of motion explicitly a tiebreaker. In Rule 7,
both the state and the direction of the two bodies are the same. Both are
moving, and moving to the left. Clearly if Bis to collide with C, it must have a
total speed greater than that of C; but in the only cases considered under
the rule C has a greater volume than B. Quantity of motion is therefore the
only tiebreaker left. 44 There are three cases: Qs < Qc, Qs > Qc, Qs = Qc. The
third case is taken up only in the French Principles.
Since Es force of motion is equal to M8 v, and Me > M8 , Cs force of resistance will always
overcome Es force of motion (Gabbey has a similar account but with a different estimate of Cs
force of resistance; see Gabbey 1971 :27f). That is clearly the intent of the French version of the
proof of Rule 4; the Latin is not so clear.
44· Descartes does not in fact state the condition explicitly in terms of quantity of motion in
Rule 7. He writes: "if the excess of quickness in B were greater than the excess of size in C."
'Excess' denotes a ratio greater than one, so we have lis> Vc, Me> M8 , and VsiVc >Mel M8 .
That is equivalent to Qs = M8 v8 > Mcvc = Qc.

Authenticated
(vii) [Rule 7a] If (1 > Qc; then C loses its integrity, while B gives up
sufficient motion so that the two bodies can travel at the same
speed.
(viii) [Rule 7b] If (1 < Qc;, then C keeps its integrity, while B is forced to
change direction; the speed of each body remains what it was. 45
The analysis so far is summarized in Figure 15. What is striking is that,

contrary to what one might expect from the third law, quantity of motion
has only a minor role in determining the outcomes of collisions. It is the
determinant per se of the outcome only when both bodies are in motion
and in the same direction. In Rules 4, 5, and 6 quantity of motion has no
role in determining the winner of a contest; it serves only to determine how
motion will be allocated when there is a transfer.46
The rationale of the rules has two heterogeneous components. One,
qualitative, serves to classify physical situations. Unlike a classical physicist,
Descartes takes the situation ofRules 1, 2, and 3, in which the directions of
motion are contrary, to be physically distinct from that of Rule 7, in which
they are not. More egregiously from the classical point of view, he
distinguishes the rest cases in Rules 4, 5, and 6 from the others.
The other component, which is quantitative, serves to determine a precise
amount of motion to be transferred when there is transfer. The third law
merely tells us that the amount by which the motion of one body is
decreased (or remitted, to use the proper Aristotelian term) and the
amount by which the motion of the other is increased (or intensified) will
be equal.
I will now examine first the contrariness of state, and then the concept
and use of quantity of motion.
State. A physicist now will write vc = o to denote the state of the body C in
the rest cases (Rules 4-6).47 There are innocent anachronisms, no doubt,
but this is not one of them. Instead it conceals the fact that Descartes did not
incorporate the relatively new conception of rest as a limiting case of motion
into his physics.
In Aristotelian physics, as I have mentioned, there are two quite different
ways of being without motion. When a thing is in potentia such and such its
quies is a genuine contrary to motion. But when a thing that was in potentia
such and such and is now entirely in actu such and such, its quies is only the
45· The three cases of Rule 7 are clearly meant to parallel the three rest rules (4, 5, 6), with
7a corresponding to 5 and 7h to 4· If we work backward from 7a to 5, and from 7b to 4· then
the counterpart to Qc: in 4 and 5 should be a quantity greater than Qn if~:> M8 and less than
Qn if Me > M 8 . The French Principles make it clear that the resistive force of Cis equal to the
quantity of motion that would have been transferred had C and B both moved off to the left at
equal speed. I return to this question below.
46. See Clarke 1982:22otr.
47· So, for example, Clarke 1990:212 andJammer 1991:316.

Authenticated
State Direction Volume Speed Q Winner Outcome Union Rule
MB<Me c d*B contra dB 0 4

sB contra sB,
sB =motion, B s*e contra se
se =rest MB>Me • 5
MB=Me - d*B contra dB, s*e contra se 0 6
SB = SB, dB contra de, MB>Me VB= ve (QB > B d* e contra de 2

both in dB from right Qc)
•
motion to left
MB=Me VB> ve (~> B d* e contra de 3
Qc) •
VB= ve (QB = - d* B contra dB, d* e contra de 0 1
Qc)
dB= de, (Mc,~MB) (VB> Ve) QB> Qe B - 7a I

right to left
•
~<Qe c d* B contra dB 0 7b 1
Fig. 15. Outcomes of collision (asterisk * indicates the value of a state after collision; symbol • denotes union or the loss of integrity in the
loser, while 0 denotes the absence of union and the maintenance of integrity; conditions in parentheses are not explicitly mentioned)

Authenticated
negation, not the contrary, of the motus that led up to it. Though both kinds
of quies are qualitatively, not quantitatively, distinct from their correspond
ing motus, the first is a privatio, the second a natural cessatio.
For Descartes, since motus has in general no terminus ad quem, the distinc
tion between the two kinds of quies has no pertinence. At each instant a body
is either being moved or not, either undergoing rupture from the vicinity of
neighboring bodies or not. If it is not, then it is in fact united with them. It is
only a notional part of a larger body. The difference, for a given piece of res
extensa, between being at rest and being in motion is as between potential and
actual distinctness. Potential here is not to be understood in the Aristotelian
physical sense of in potentia. That would imply that a potentially distinct
body had actual distinctness as its terminus. What is potentially distinct is
merely that which, by God's absolute power at least, can be made to exist
separately. This use of in potentia one finds already in Aristotelian discussions
of the metaphysics of indivisibilia like points and lines. 4 8
Cartesian quies is therefore not a limiting case of Cartesian motus. There is,
as Leibniz noted later with disapproval, a saltus between the two. That saltus
one can trace back to Descartes's attempt to define motus as a respectivum, as
a joint mode of two bodies. He is committed thereby to constructing, from
properties admissible in his physics, a scheme that will yield a kind of abso
lute change that can plausibly be called change of place. Since he admits
neither Aristotelian natural place nor anything like Newtonian absolute
space, the alternatives are few. If one really conceives of body as res extensa,
the only temporally varying relations among bodies are distance and adja
cency. Relations of distance, however, require a choice of reference points
from which to measure distance, and there is, as Descartes himself argues,
no reason to choose one set of reference points rather than another. Only
adjacency remains.
An immediate consequence of Descartes's definition of motion is that the
inception of motion is always accompanied by the joining and disjoining of
parts. When in Rule 5 the smaller body at rest is set into motion by the
larger, andjoined to it, the smaller body must be at rest with respect to some
collection of neighboring bodies, and in fact part of them. In the letter to
Clerselier, Descartes writes that "by a body which is without movement, I
understand a body that is not in action to separate its surface from those of
the other bodies that surround it, and, consequently, that forms part of
another hard body which is larger" (To Clerselier 17 Feb. 1645, AT 4:187).
The remark, whatever its success in vindicating the fourth rule, shows that
48. Suarez distinguishes potential existence from potential isolation (namely, by division). All
of the infinitely many points in a line exist in actu, but no point exists in isolation, divided from
all other points, except perhaps by God's absolute power. See Suarez Disp. 40§5129, 33rr, 44•
Opera 26:559· 561, 563.

Authenticated
[2g8] Bodies in Motion
setting a body in motion and separating one body from another are coextensive. It
should not be surprising that in the Principles, resistance to rupture and to
being moved are mentioned together: "But here one must carefully note
what the force of each body to act on another or to resist its action consists
in: namely in this alone, that each thing tends, quantum in se est, to remain in
the same state it is in [... ] For this reason what is conjoined to another has
some force to impede its being disjoined; that which is disjoined, to remain
disjoined; that which is at rest, to persevere in its rest, and consequently to
resist all those things that can change it; and that which moves to persevere
in its motion, namely, in motion with the same quickness and toward the
same direction" (PP2§43, AT 8/1:66). Later Descartes argues that there is
no "glue" stronger than rest by which to hold the parts of a hard body
together: "No other [mode] can be more opposed to the motion by which
these particles might be separated than their rest. And aside from sub
stances and their modes, we acknowledge no other kind of thing" (2§55,
8/1:71). The "force of rest," to which I will return in §8.3, and the "glue"
that binds together the potential parts of one body are one and the same.
That point is often overlooked because in the rules of collision Descartes
takes the bodies Band C to be "so divided from all others that their motion
would be neither impeded nor helped by the others around them" (2§45,
8 I 1 :67). Yet in Rule 4, at least, as the letter to Clerselier shows, he cannot
entirely forget that Cis united with a larger body.
Rest is not motion in potentia. Aristotelian potentia, which in Descartes's
view is fatally infected with finality and obscurity, is supplanted by the color
less potentiality of the parts of quantity. Every part of a body is itself a poten
tial body; all that is needed to make it so is the "action to separate its surface"
from the whole. The contrast of union and disunion, and so of rest and
motion, is of a different order than the contrast between one degree of
motion and another. It resembles what is now called a phase transition-from
solid to liquid, or liquid to gas-rather than the continuous increase of
velocity suggested by the physicist's v = o. Like a phase transition, the transi
tion from rest to motion requires a quantum of force greater than that
needed merely to bring about change within a phase. But unlike any phase
transition, it sometimes cannot be accomplished by any quantum of force,
however large.
Yet the suggestion that a body at rest resists the attempt of another to
move it by virtue of being part of some larger body cannot be the whole
story. The outcome, as we have seen, is continued rest only if the body Cat
rest is larger than the body B moving toward it. Then it will remain at rest no
matter how quickly Bis moving. If, on the other hand, Cis not larger than B,
then Cwill be moved no matter how slowly Bis moving. But Cwill still be part
of some larger body.

Authenticated
Motion and Its Causes [2gg]
The problem is not just that the reason offered cannot account for the
difference in outcomes. It is that when Cis at rest, there is no obvious reason
to pick it out as the body with which B collides rather than some larger or
smaller part of the whole that C belongs to. In the simplest case, where B
and C are brick-shaped objects meeting face on, one might suppose that
only those points of Cwhich lie along the line of motion of some point of B
are the part with which B actually collides. Call that part G*. What Rules 4
through 6 would tell us, if this is right, is whether B will break off G* and
absorb it, or else be reflected by G*, so that Cremains whole.
I said that Descartes cannot entirely forget that Band Chave other bodies
around them. The point can be put more sharply: Descartes cannot regard
the bodies surrounding C, when it is at rest, as neither impeding nor assist
ing its motion. Not, at least, if he wants potential bodies mutually at rest to
cohere. Later authors have often chastised Descartes for "overlooking" the
relativity of uniform motion. That is incorrect: he recognized that defining
motion in relative terms was possible and rejected it (2§15, 28). The
difficulty is rather in the handling of the conditions under which the laws
would hold.
A look at Descartes's response to an experience offered by Mersenne will
clarify the issue. Mersenne, sometime in late 1641, has written to Descartes
that a ball A, colliding with two balls B and C, with B as large as A, and C
smaller, will "push the small one C by means of the large one B, almost
without making B move [sansfaire quasi mouvoir B]" (To Mersenne 17 Nov.
1641, AT 3:452). The experience is a counterexample not to the collision
rules in the Principles, but to an earlier rule: when a body B collides with a
body Cat rest, it always communicates to Cwhatever motion is needed for
the two bodies to proceed with equal speed in the direction of B's motion. 49
I will call Descartes's earlier rule, which is, as an early annotator of the
manuscript letter noted, "contrary to his principles," the I639 Rule, al
though Descartes may well have arrived at it before 1639.
Mersenne takes the 1639 Rule to entail that in Figure 16, Band C should
move off to the right with equal speed. But in fact B stays almost at rest and C
alone moves. Descartes's answer is this: "Although at the first moment when
the two balls Band Care touched [by A], they no doubt move with equal
speed, still, because B is heavier than C, it is much more hindered by the
inequalities of the plane on which they roll, and it is those inequalities that
stop the ball Band are not capable of stopping the ball C; even if the 2 balls
were of the same size, C could go faster than B, since all the inequalities of
the plane that resist it also resist B following it, and they use their forces
49· 1o Mersenne 25 Dec. 1639, 28 Oct. 1640, AT 2:627; cf. 1o Debeaune 30 Apr. 1639, 2:543
and To Mersenne 28 Oct. 1640, 3:211.

Authenticated
om
A B
Fig. 16. Collision (AT 3:452)
conjointly to overcome them; but what resists B does not hinder C, which for
that reason can immediately move away from [B]" The conditions of Mer
senne's experience, in other words, are not ideal. Had the plane been per
fectly smooth, Band Cwould have rolled off to the right togetherjust as they
were supposed to.50
Perfect smoothness, or perfect inelasticity, is a condition that could be
fulfilled, at least by divine power. But the condition of Rules 4, 5, and 6,
which is both that a body be at rest and that its movement be unimpeded by
others, cannot be, not if mutual rest is also supposed to explain the cohe
sion of the parts of a simple body. 51 If Descartes gives up the assumption
that the body at rest is unimpeded, then he has no obvious way to determine
its quantity. But if he gives up the equivalence of mutual rest and cohesion,
he cannot explain why the outcomes in Rules 4, 5, and 6 should differ.
The dilemma is striking not least of all because it is absent from his earlier
thinking. The 1639 Rule, the one to which Mersenne objects, yields the
same outcome whatever the relative sizes of the body at rest and the body in
motion. The two will always move off with equal speed in the original direc
tion of motion. The rule is still incorrect, but it is at least less counterintui
tive than Rule 4, if one is to judge from the objections to that rule.
There are, it seems to me, two reasons for what from our point of view is a
step backward.52 The first is that only Rule 4 (and the related Rule 6), and
not the 1639 Rule, yields reflection rather than absorption when one of the two
bodies is at rest. From the time he was writing the Dioptrique onward, Des
cartes always takes the reflecting body to be at rest; but the 1639 Rule would
yield absorption. 53 The second is that, perhaps at the same time he realized
50. In his response to another counterexample, this time to Rule 5 (1o Mersenne 23 Feb.
1643, AT 3:634f), Descartes appeals both to the imperfection of the plane on which the balls
roll and to their elasticity. In a later letter he actually attempts to estimate the contribution of
elasticity to the outcome (To Mersenne 26 Apr. 1643, 3:652f).
51. Gabbey notes the dilemma, adding that "coherence seems not always to have been
Descanes' strong suit" (Gabbey 1g8o:314n.I62). I am suggesting that although the position
may have been physically incoherent, it had serious motivations.
52. The interpretation given here differs from that of Clarke, who takes Descartes to have
used already a version of the "least modal change principle" in formulating the collision rules
in the Latin Principles (Clarke 1982:223rr). I am inclined to take the letter to represent a later
stage in Descartes's thinking.
53· In a letter to Mersenne from 1640 Descartes notes that the reflecting surface must be
"hard and immobile," and that in nature "the reflection of an ordinary ball never occurs
exactly in equal angles, nor perhaps that oflight" (To Mersenne 28 Oct. 1640, AT 3:208; cf. also

Authenticated
that the 1639 Rule would not work, he was reading Aristotelian treatments of
motion. Although he rejected the finality implicit in the Aristotelian scheme
(§3.1), he incorporated into the contest model, already present in Le Monde,
the contrariety of rest and motion and of direction. He was able, therefore,
to adapt to his own purposes a language familiar to the Philosophers, whom
he hoped would adopt the Principles in teaching. The result is a hybrid, in
which the contests that determine change of state and change of direction
are prior to the comparison of quantities of motion; the conseiVation of the
quantity of motion does not determine outcomes but seiVes only to deter
mine the allocation of motion when motion is transferred.
But it is not evident that the world could not be as Descartes describes it.
His laws are not, indeed, invariant under Galilean transformation from one
rest frame to another moving uniformly with respect to it. 5 4 They are invar
iant only under the mirroring of right and left, which he uses to reduce the
number of cases to nine (Rules 1 to 6 together with the three cases of Rule
7). 55 Nor do they satisfy Leibniz's continuity condition: to the continuous
passage from MB/ Me = 1 + e to MB/ Me = 1 there corresponds a passage
from absorption (Rule 2) to reflection (Rule 1), a gross disparity in out
come (Leibniz Ph. Schr. 4:378; cf. also jammer 1990:317). That, however, is
what the contest model would lead one to expect, when the stakes-the
direction of Cs motion-are discontinuous. I am not sure why God could
not have created a world in which the laws of physics were not invariant
under Galilean transformations (in fact Maxwell's laws are invariant only
under Lorentz transformations), or in which continuous change of physical
quantities yielded discontinuous outcomes (as in the photoelectric effect).
Perhaps he didn't; but not because he couldn't.
Speed and quantity ofmotion. The third law tells us that whenever motion is
transferred, the total quantity of motion does not change. But it does not
tell us directly how much will be transferred. Nor does it tell us how quantity
of motion is to be measured. All that the Principles tell us in general is that the
To Mersenne for Hobbes, 21 Jan. 1641, 3:289, and Diop. 2, AT 6:94-95). Both Mersenne and
Bourdin ask Descartes why in oblique reflection the vertical part of the body's movement is
changed while the horizontal part is not (cf. To Mersenne 29Jul 1640, 3:107r and 111, and To
Mersenne 30 Jul. 1640, 3:129), to which his answer is that the reflecting surface opposes only
the vertical determination of the movement of the reflected body.
54· A Galilean transformation has the form x:" = x - vt, y* = y, z* = z, t* = t, where the *
denotes coordinates in a frame F* moving at velocity v along the x-axis of the original frame F
(Rindler 1979:3). A law is invariant under Galilean transformations if it remains true when x:",
etc., are substituted for x, etc. A Lorentz transformation is the counterpart of the Galilean
transformation in special relativity (ib. 31).
55· He leaves out the case where both bodies are in motion, in opposite directions, MB >
Me, and vB"' vc. The case vB > Vc could presumably be regarded as analogous to Rule 3· The
case % < Vc cannot be decided except perhaps in the manner of Rule 7, by comparing Qn and
~;· I find it significant that he did not do so: it support~ the claim that in Rules 1 to 3, where
directions are contrary, the contest is between volumes and speeds rather than directly between
quantities of motion.

Authenticated
Bodies in Motion
force with which a body perseveres in rest or motion "should be judged

sometimes [tum] by the magnitude of the body in which it exists, and the
surface according to which that body is disjoined from another; sometimes
[tum] by the quickness of the motion, and the nature and contrariety of
modes by which diverse bodies meet one another. "56 I will set aside the
problem of measuring the force of resistance of a body at rest and consider
only the force of a body in motion. That force, Descartes tells us, is directly
proportional both to speed and to volume or moles: ''When one part of matter
moves twice as quickly as another, and the other is twice as big as the first,
then the same motusis in the smaller as in the larger; and by whatever amount
the motus of one part becomes slower, the motus of some other [part]
becomes quicker" (PP2§36, AT 8/1:61). Provided we know how to individu
ate bodies, the nature and measurement of volume present no serious
difficulty. Speed, on the other hand, is, like motion, not obviously a mode of
extension, nor is its measurement simple. In what follows I will, by compar
ing the Aristotelian and Cartesian treatment of speed, show that Descartes
remains closer to his predecessors than is usually recognized. I will show also
that quantity of motion was conceived in geometric rather than in algebraic
terms, and that the rationale of its allocation in the collision rules becomes
much clearer once the geometric conception is worked out.
Descartes treats the terms 'quality' and 'mode' as having the same denota
tion. They differ only in connotation: when we consider the substance
affected or varied by a mode or quality, we call it a mode; when we denomi
nate it by way of the variation, we call it a quality (PP 1§46, AT 8/1:26). I
know of no passage in the Principles where Descartes calls motus a quality
rather than a mode. In Le Monde, however, he writes that rest and motion are
both qualities, distinguishing them not from qualities simpliciter but from
"real" qualities, a term used to denote sensible qualities as they inhere in
things (Monde 7, AT 11:41; cf. Goclenius Lexicon 912). This is in keeping
with the definition in 2§25 of motusin the strict sense of rupture: rupture or
the absence of rupture is not itself a quantity but a mode of the surfaces of
the things ruptured.
Earlier I left it unresolved whether Descartes could define speed and
direction for motus in the strict sense. I suggested only that by appealing to
the notion of physical trajectory one might be able to define them without
shifting to motion in the vulgar sense-motion relative to an arbitrary frame
of reference. The first thing to note is that Descartes inherits from the
Aristotelians a twofold notion of speed. The first is average speed, measured
56. "Visque ilia debet a:stimari tum a magnitudine corporis in quo est, & superficiei secun
dum quam istud corpus ab alio disjungitur; tum a celeritate motiis, ac natura & contrarietate
modi, quo diversa corpora sibi mutuo occurrunt" (PP 2§43, AT 8/1:67). The force of tum ...
tum . . . is ambiguous: it can indicate either alternative or coordinate members of an
enumeration.

Authenticated
by comparing how far two bodies will travel in a given time. In Rule 5, for
example, Descartes writes that "after B collided with C, it would move one
third slower than before, that is, it would require the same time to move
through two feet of space as before it required to move through three"
(2§50, 8/I:6g).57 The second notion, alone referred to in the collision
rules, is instantaneous speed. Instantaneous speed, whether alone or com
bined with volume into quantity of motion, is what determines the outcome
of collisions in the rules where both bodies move; in Rules 5 and 6, volume
determines the outcome, but the allocation of motion after a collision
depends on the instantaneous speed of the moving body.
In the Aristotelian tradition, instantaneous speed was held to be an inten
sive quantity-the intensity of impetus, which was typically taken to be a
quality analogous to heat. 5 8 Extensive quantity is quantity of bulk [quantitas
molis]. Intensive quantity of virtue [ quantitas virtutis] is that, "according to
which one [virtue] is more perfect than another[...] Similarly, one quality
is more intense than another, or weaker: and according to that there is also
proportion. For we say that one heat is twice as intense as another" (In Phys.
7c5q5, opera 4:2o5ra). The chief difference between intensive and exten
sive quantity is that the parts of an intensive quantity, unlike those of an
extensive quantity, cannot be separated to yield smaller quantities of the
same kind. But one intensive quantity can-so the Aristotelians believed
be said to stand in proportion to another, whether of the same quality and
subject, or of others. In particular, an intensity can be divided into propor
tionate parts or degrees, and smaller or larger intensities measured according
to those degrees. 59
The same quality can have both an extensive and an intensive quantity.
The heat of a warm body, for example, has an extensive quantity, the volume
of the body, and an intensive quantity that may vary from part to part. For a
body having but one spatial dimension-a line, in other words-Nicole
Oresme represented the extensive quantity by a horizontal line segment
and the intensive quantity at each point by another line perpendicular to
the first line at that point. The area of the resulting figure Oresme treats as a
quantity that may be compared with other quantities (De config. 3c5,
p404 ff). One could call it the total quantity of a certain quality in a subject.
57. The measurement ofspeed so defined does not require measurements ofelapsed time,
which could rarely be made with sufficient accuracy. It requires only that certain events-the
inception of motion, the passing of a mark-be observed to be simultaneous. The measure
ment of speed in rectilinear motions by distance traveled in equal times comes from Aristotle
(Toletus In Phys. 7C4q4, opera 4:2o3rb).
58. See Maier 1958, c.3 and Murdoch & Sylla 1978:231-241 for surveys of Medieval
treatments of motion as an intensive quantity.
59· The third book of Mersenne 's Nouvelle pensees de Galilee illustrates the geometric, rather
than algebraic, manner in which relations between time, speed, and distance traveled were
treated (Mersenne Nouv. pens. [167'T]).

Authenticated
Bodies in Motion
One application of that method of representation is to motion. In a simple

body, motion will have a uniform intensity over all parts of the body. The only
extensive dimension along which its intensity varies is time. Oresme repre
sented motion by diagrams in which the horizontal axis is the duration of the
motion and the length of the perpendiculars is its intensity, which is to say, its
speed. Descartes had some knowledge of the technique. Diagrams similar to
Oresme's appear sporadically in his works, notably in the 1629 proof of an
incorrect law of falling bodies, and in a correct law included in a letter to
Mersenne in 1634.60 Descartes also used such diagrams to represent what we
now call work as une force adeux dimensions.6I
The instantaneous quantity of motion in a body, which is the measure of
its impulsive or resistive force, can be represented by a rectangle whose base
is a line proportional to volume, and whose height is proportional to the
speed of the whole body. The third law tells us in effect that the sum of the
areas of the two rectangles representing the quantity of motion of the two
interacting bodies before the collision will be equal to their sum after the
collision. It does not tell us what they will be.
We have seen that the outcome of a collision is either a reflection or an
absorption. The last six rules can be divided accordingly into three groups.
In the first group, Rules 4 and 7b, there is pure reflection, that is, change of
direction of the loser with no exchange of motion (Rule 1 yields a reflection
of both bodies). In the second group, Rules 2, 3, 5, and 7a, there is pure
absorption, that is, the two bodies move in the direction of the winner with
the same speed. It follows that there will be an exchange of motion between
6o. To Mersenne 13 Nov. 1629, AT 2:72, To Mersenne 14 Aug. 1634, 2:304; cf. also the early
Cogitationes privata! 10:219 (a correct proof of the law, in response to a problem set by Beeck
man). The diagram in Beeckman's version (fouma/1:262) differs from an Oresmian diagram
in making the vertical axis represent time and the horizontal ordinates speed. In the letters to
Mersenne, Descartes, rather confusedly, makes the vertical axis represent both the total dis
tance traversed (a quantity that in Beeckman's diagram is represented by the area under the
cmve representing velocity) and the speed of the motion at instants during the motion,
represented by the horizontal axis.
Even in 1634, when he states the law correctly, he still gets the diagram wrong (Mersenne,
on the other hand, drawing on Galileo, both states the law correctly and gets the diagram right
in his Harmonie universelle [2pr2, Mersenne Harm. univ. 1:89], which was completed in 1634).
Only in 1639, in a letter to Mersenne on the motion of strings, do we find a diagram in which
intensity (the "force" applied to the string) is plotted against time, not distance, and distance
represented by areas (To Mersenne 30 Apr. 1639, 2:535). Later Descartes uses a similar diagram
to illustrate the speed ofdescentoffalling bodies (ToHuygens 18 or 19 Feb 1643, 3:620; on this
letter see Nardi 1986). In other less carefully drawn figures it is impossible to tell what the
vertical axis is supposed to represent (e.g., To Mersenne 11 Mar. 1640, 3:36).
61. In letters to Mersenne on the relation between force a une dimension and force a deux
dimensions, the "one-dimensional" force required at each point of a body's motion to lift it is
plotted against the path traversed; the area of the resulting rectangle represents the "two
dimensional" force or the work required to lift the body through the whole distance (To
Mersenne 12 Sep 1638, 2:357, 359; cf. Carteron 1922:245, 252). Beeckman had earlier used
the Flemish word cracht 'force' to denote the product ofweight and distance moved (foumal1
8 Oct. 1628, 3:93; cf. DeWaard's note on Beeckman in Mersenne Corr. 2:123).

Authenticated
them. In the third group, which includes only Rule 6, there is reflection and
transfer of motion; Descartes characterizes the outcome as a compromise
between the outcomes of Rules 4 and 5·
Given that the outcome includes a transfer of motion, the conservation of
total quantity determines the allocation of motion after collision only in
combination with the supposition that the outcome should be absorption. I
repeat the relevant portions of Rules 5 and 7:
if B were twice as large as C, it would transfer to [C) a third part of its

motion, because that one-third part would move the body Cas quickly as
the two remaining parts of the body B which is twice as large. And thus
after B met with C, it would move one-third part more slowly, that is, it
would require as much time to move through two feet of space as before to
move through three.
[...] if Chad only two degrees of quickness, and B had five, two degrees
would be taken from B, which, being transferred to C, would effect only
one degree [of quickness], because Cis tWice as large as B: and by that it
would result that the two bodies Band Cwould afterward move with three
degrees of quickness.
There are, clearly, allocations of motion that satisfy the third law and leave B
moving less quickly than C. Only the third law together with the require
ment that after the collision v 8 * = Vc *determines how much motion should
be allocated to C.
One could immediately represent the outcome in Rule 5 by
vB *-- Vc *-- M MBM %·

B+ C
But it seems to me to be more in keeping with Descartes's thinking to

proceed in the following way. In Rule 5, he writes that one-third of B's
motion will move C, which is half the size of B, as quickly as the remaining
two-thirds of B's motion move B. Represent B's original quantity of motion
Qs by a rectangle R of arbitrary base and height, and represent Cs original
quantity of motion, which is nil, by a line half as long as the base of R, as in
Figure 17. Descartes tells us in effect that "degrees" of the motion of B can
be converted into degrees of the motion of C in proportion to their vol
umes, so that in Rule 5 one degree of B's motion yields two degrees of Cs
motion. In Rule 7a, on the other hand, Cis twice the size of B, and so two
degrees of B's motion yield one degree of Cs. The rationale of the conver
sion is easy to see if one imagines converting areas on the base ab in Figure
17 into areas on the base cd, as if one were pouring fluid from one gradu
ated cylinder to another.

Authenticated
s ''
'
R
''
'
'
'
' Vs
'
'
'
'
'
'
'
'd a b
Me Ms
Fig. 17. Calculating motion after collision (Rule 5)
The problem of determining how much motion should be transferred to

C then becomes the problem of transferring a certain degree x of the
original motion Vs of B to C according to the conversion factor Ms/ Me so
that the new motion Us - x of B, or the level offluid in R, will be equal to the
new motion (Ms/ Me) x of C, or the level of fluid in S. A little calculation
shows that
x= M McM Vs·
s+ c
The new motion of the two bodies will be (Ms/(Ms +Me) )us,justas before.
Similarly, in the example used by Descartes in Rule 7a, we have initially the
situation in Figure 18(a). The motion ofB is divided into five degrees, of
which two convert into one degree of motion in C. Since C has by assump
tion two degrees already, diminishing B's motion by two degrees and trans
ferring them to C will leave both B and C with three degrees of motion. We
end up with the situation in Figure 18(b).
If Descartes had adhered to the 1639 Rule mentioned earlier, in which
motion is always transferred when a body at rest meets a body in motion,
then Rule 7a could be reduced to Rule 5 by taking the initial velocity of Cto
be zero and that of B to be uc- Us. That rule, however, has the fatal defect of
making it impossible for reflections ever to occur. Instead we have the
roundabout comparison of us/ Uc with Mel Ms, and the division of cases
according to the obscure thesis that relative slowness "participates in the
nature of rest" (PP2§44).
Taking minor liberties with the term, I call this the hydrostatic model of the
transfer of motion. In it a simple body is treated as a graduated cylinder with
base proportional to its volume, into which or out of which degrees of
motion can be poured. The level of the liquid in the cylinder corresponding
to one body can be compared with that in others; in practice this means

Authenticated
Qc
Qc
I
(a) (b)
Fig. 18. Calculating motion after collision (Rule 7a)
comparing the "degrees" or fractional parts of one body's total speed with
equal levels of liquid in the cylinder corresponding to another. Since
degrees need never be converted into absolute units, one need not take the
variable v to be in units like meters per second, any more than in a geometri
cal diagram one need take the lengths of lines to be measured in meters or
yards.
The hydrostatic model yields some insight into Descartes's comments on
"natural inertia" in a letter to Debeaune in 1639. Mter having denied in an
earlier letter that there is any "natural inertia or slowness [ tardiuete1 in
bodies" (To Mersenne Dec. 1638, AT 2:466), Descartes now explains that
"since if each of two unequal bodies receives as much motion as the other,
this same quantity of movement does not give so much speed to the larger as
to the smaller, one may say in this sense that the more matter a body
contains, the more Natural Inertia it has; to which one may add that a body
which is large can better transfer its movement to other bodies than a small
[body], and that it can be moved less by them. So that there is a sort of
Inertia, which depends on the quantity of matter, and another [sort] that
depends on the extension of its surface" (ToDebeaune 30 Apr. 1639, 2:543-
544). The "other" sort, which is clearly related to the "force of rest"
described in the Principles as depending both on magnitude and on "the
surface according to which this body is disjoined from others," I will return
to shortly. The inertia that depends only on the quantity of matter is, in
effect, nothing other than the conversion factor MB/ Me, the relative size of
the cylinder into which or from which motion is poured. 'Large' means
'relatively large'; one can more easily fill a small cylinder from a large one
than a large cylinder from a small one. So although there is nothing in any
body that opposes motion, either in the sense that a body, considered in
itself, cannot be moved by even the smallest force, or in the sense that it has

Authenticated
of itself some quality contrary to motion, the transfer of a certain amount of

motion to a large body will result in fewer degrees of speed than the transfer
of the same amount to a small body.
That much is consistent with supposing that body is extension and noth
ing more. The force of rest, on the other hand, seems to imply that a body
can have a mode, namely rest, which is not only contrary to but actually
opposes motion. It still remains puzzling how the resistance of rest to mo
tion, which in Rule 4 must be taken to be indefinitely large, can be
coherently fitted into the system just outlined. On the face of it, that system
would be far more congenial to the 1639 Rule than to the rules Descartes
arrived at in the Principles.
Here I think one should not disregard, as commentators tend to do, the
reference to the surface area of the resting body. I suggested earlier that a
body genuinely at rest is in fact part of a larger body (which need not be at
rest), and that the change from rest to motion is at the same time a change
from union to disunion with the larger body. Even in the letter to Clerselier,
where Descartes is improvising an alternative explanation of Rule 4, he
notes that "a body that is not in action to separate its surface from those of
the other bodies that surround it" is therefore "part of another hard body
which is bigger," referring Clerselier to the definition of motion at 2§25.
How big is that bigger body? As big as it needs to be for Rule 4 to hold-in
other words, indefinitely large. In material added to the French Principles,
Descartes works out, in accordance with the hydrostatic model, the amount
of motion that would have to be transferred to the body C if it were to be
moved as quickly as B: "So, for example, if Cis twice as large as B, and B has
three degrees of movement, it cannot impel C, which is at rest, unless it
transfers [to q two of its degrees, namely one for each half [of q, and
retains for itself only the third, since it is no larger than each half of C, and
can go no faster afterward than they do" (2§49; 9/2:91). The third law does
not forbid the transfer of more than half of a thing's motion. Indeed it
would seem to enjoin the transfer in situations like that of Figure 19, if the
two bodies move afterward with the same motion. But a body at rest exhibits
a double resistance: the resistance described a moment ago, which belongs
to every body, whether at rest or not, and a second resistance specifically to
the change from rest to motion. The second resistance is proportional to
the quantity of motion the resting body would receive from the moving
body if there were a transfer of motion according to the hydrostatic model.
That quantity-two degrees in the example above-is effectively all that the
moving body can give (if it gave more, say the amount denoted by the
rectangle ab in Figure 19, and moved C no faster, there would be a net loss
of motion, which would violate the third law). It is, in other words, the
measure of the force that the moving body possesses to impel the body at

Authenticated
Motion and Its Causes [309]
Qc Qc
Before Mter
Fig. 19. Calculating motion after collision (Rule 4)
rest to move. Since it is, in fact, not greater than the force that the resting
body has to resist it, the moving body must instead turn back, retaining, as
the first part of the third law tells us, all of its motion.
The additions in the French Principles were not Descartes's first attempt to
clarify Rule 4· In the letter to Clerselier, who apparently recognized that the
outcome in Figure 19 is not ruled out by the third law alone, Descartes had
invented an entirely new principle to explain Rule 4: "when two bodies that
have incompatible modes meet, there must indeed occur some change in
those modes to make them compatible, but this change is always the least
that can be, that is, if, when a certain quantity of these modes is changed,
they can become compatible, no larger quantity will change" (To Clerselier 17
Feb. 1645, AT 4:185). Motion and rest are incompatible modes; they can be
made compatible either by pure reflection or by absorption. The body in
motion must either be reflected or give to the body at rest sufficient motion
so that the two will afterward move as one. But to do so it must give up more
than half its motion, and (Descartes adds) more than half its "determina
tion to move from the right to the left, insofar as this determination isjoined
to its speed." That change would be more than the change of all its deter
mination, and so it changes all its determination. The result is therefore
reflection.
Though some writers find this proposal an advance upon the doctrine of
the Latin Principles, I am inclined to think that it is, like the original, a kind
of hybrid, this time of the hydrostatic model and what Pierre Costabel calls a
"principle of economy" to which Descartes is otherwise unfriendly (Costabel
1967:248). Like many hybrids, this one proved to be sterile. Since the
quantity of determination in the direction of motion seems to be just the
speed of the motion, speed effectively enters into the calculation of the
"total change" twice. To say, as Costabel does, that "no other passage is as

Authenticated
Bodies in Motion
close to the vectorial conception of speed" is too charitable. 62 The very fact
that Descartes treats determination and speed as distinct quantities capable
of addition shows that he is far from a vectorial conception as ever. He was
closer to the truth-if that matters-in his answer to Bourdin's objections
against the proofs of the laws of reflection and refraction. There he writes
that "in speaking of the determination to the right I mean the whole part of
the movement that is determined toward the right" (To Mersenne 3 Dec.
1640, AT 3:251). The determination to the right and the determination
downward together make up the entire determination. Even so, in the proof
of the law of reflection itself, Descartes does not derive the motion after
reflection by adding together the unchanged horizontal component and
the new vertical component. He derives it by construction from the horizon
tal component and the circle at whose circumference the reflected body
must arrive if its total motion is unchanged. 63 If having a vectorial concep
tion requires the composition as well as the resolution of motions, then
Descartes does not have one.64
The hydrostatic model, then, was retained by Descartes through every
stage in his thinking about the laws of motion. But in both the Principles and
in the letter to Clerselier it was combined with heterogeneous conceptions.
The Principles subordinates the transfer of motion to the contrarieties of rest
and motion and of direction, while the letter attempts to unite the transfer
of speed and determination under a "principle of least modal change."
Garber takes the letter to be a step fmward, on the grounds that the princi
ple enables the outcome of collisions to be calculated "directly, without a
detour through force for proceeding and force of resisting and the balance
of forces in an impact contest" (1992:247). Indeed, because of the double
role of speed in determining the quantity of motion and the quantity of
62. Costabel, however, remarks on the "strange declaration that says that determination is
joined with speed arrd that permits Descartes to speak of 'half' in relation to a modality that
seemed not to be subject to quantification" (Costabel 1967:248). Gabbey, on the contrary,
argues persuasively that determinatiois a quantity (Gabbey 1980:258). The determination in the
direction eof a body moving in the direction dis, in modem terms, the projection ofits velocity
vector onto the direction e (ib. 259). In Fig. 3, for example, the horizontal determination is
measured by A.F, and in the vertical direction by AE. The quantity of its determination in the
direction d itself is just its speed.
63. See Gabbey 1980:256. The version ofthe prooffrom Regis cited earlier in this section is
essentially Descartes's.
64. In a response to Hobbes, Descartes writes that even though he could have spoken ofthe
composition of two motions, he did not, "for fear that it might perhaps be understood that the
quantity of these velocities thus composed, and the proportion of one to another, would
remain" (To Mersenne for Hobbes 21 Jan. 1641, AT 3:288). The ensuing exchange reveals little
about his understanding of the composition of motions. The only other examples I know of in
which motions are composed occur in letters to Huygens ( 18 or 19 Feb. 1643, AT 3:624), and
to Mersenne (23 Mar. 1643, 3:640~"), both concerned with the trajectories ofjets of water. In
the letter to Huygens, he writes: "In the water drop, two movements are composed, without the
quickness or slowness of the first changing anything in the second."

Authenticated
determination transferred to the body at rest, one can regard the principle
as an attempt to make the hydrostatic model determine not only the alloca
tion of motion but the outcome of collision. But as Garber himself carefully
notes, the principle does not cover all the situations envisaged in the rules of
collision. Nor does Descartes apply it to changes in the state of the resting
body, since he still has no way of quantifYing the modal change from rest to
motion.
The path to classical mechanics was strewn with difficulties, some of which
can be recognized only ifwe forget where it was headed. Descartes's first law,
often heralded as the discovery of inertia, was in fact something else: it was
the claim that the negative resistance of a thing to change-the persistence
of its states in the absence of external causes-needs no natural cause. That
was not entirely new, though the application to motion was relatively un
usual. What distinguishes Descartes from some of his predecessors, and in
particular from the Aristotelians, is the claim embodied in the second law:
instantaneous rectilinear motion, or conatus, alone persists.
The immediate consequence of that claim is that any motion that is not
rectilinear must have a cause. Since Descartes admitted only local motion,
and denied that any body moves itself, that cause could only be contact with
another body whose state is incompatible with the first. Here we find what is
perhaps the most deeply compromised part of Descartes's theory of motion.
The negative resistance, or persistence, argued for in the first and second
laws does not entail a positive resistance to change. Yet virtually everyone
would have agreed that efficient causal change requires power to overcome
the positive resistance of an otherwise persisting state. That inference
from negative to positive resistance, and then to the requirement of an
overcoming power in efficient causes-underlies the use of the contest
model.
Descartes did his best to eliminate active and passive powers in Nature.
But he was only partly successful, even in the heart of his physics. The
success lay in his understanding of "natural inertia," represented by the
conversion factor MB/Me in the hydrostatic model of the transfer of mo
tion. Such inertia offers no positive resistance to change and requires no
power merely to be overcome. As he says in the French version of Rule 5,
and had earlier maintained even for the situation in Rule 4, even the small
est force suffices to put a body in motion. The failure was in his understand
ing of rest. A body at rest offers, in addition to its natural inertia, a positive
resistance to change, quantifiable only ex post facto. In the situation of Rule
4, it turns out to have been however much was needed to resist the moving
body B; in the situation of Rule 5, it turns out to have been equal to natural
inertia. In either case the body at rest is still conceived to be acted on by the

Authenticated
Bodies in Motion
body in motion; the asymmetry of the agent-patient relation in Aristotelia

nism is carried over into the asymmetry of the force of rest and the force of
motion.
Consider again the paragraph of the Principles in which the vis cuiusque
corporis ad agendum vel resistendum (the force of each body for acting or
resisting) is more nearly specified, after having been introduced in the third
law. The two vires are said to arise "from this one [thing], that each thing
tends, quantum in se est, to remain in the same state in which it is, according
the law stated first" (PP2§43, AT 8/1:66). But that common origin is belied
by the difference between them. The vis ad agendum can only be the force of
persevering in a motion "of the same speed and in the same direction." All
the other forces mentioned, including that of persevering in rest, are re
sistive. Though each body in a collision can be neutrally described as per
severing in its present state as much as possible, given the incompatibility of
its state with that of the other, Descartes still thinks of the body at rest as
receiving or refusing the action of the body in motion. The body at rest is to
that extent the patient in the collision. In keeping with the Aristotelian
conception of transitive change, moreover, there is no reaction of the pa
tient on the agent, except per accidens. Reflection, the key instance, is not a
reaction of the reflecting body on the reflected body. It is the continuation
of the reflected body's state as far as possible given that its present direction
of motion is incompatible with the persistence of rest in the reflecting body.
The fact that, from a later point of view, Descartes's laws are not invariant
under transformation from one reference frame to another in uniform
motion with respect to the first, or that they do not satisfY Leibniz's condi
tion of continuity, is only symptomatic of his retention of the asymmetry of
agent and patient. Disregard of the obvious, even if Descartes were con
victed of it, would hardly suffice to explain the rules he did devise. Careful
attention to the Aristotelian physics he did not entirely leave behind, and to
the parallel efforts of contemporaries like Beeckman, yields a fuller and
more credible account.
8.3. Natural and Divine Agency: The Problem of Force

Perhaps the most vexing issue in Descartes's physics is the significance of
force. 65 The Latin vis and the French force preside at the center of an
extraordinarily complex network of concepts and denotations. They pre
sent difficulties beyond those offered by technical terms like velocitas and
65. The study afforce in Descartes's physics has been advanced by a number of historically
and philosophically distinguished contributions, among which Carte ron 1922, Gueroult 1934,
Gueroult 1970, Westfall1971, Hatfield 1979, Gabbey 1980, Grosholz 1991, and Garber 1992
have been the most important to my thinking on the questions discussed here.

Authenticated
determinatio. Aside from the problem of sorting out the variety of contexts in
which vis and its cognates occur, 66 there is the fundamental paradox em
phasized by Martial Gueroult. Descartes insists that in body there are no
modes except those that follow from extension, its principal attribute, or
from its being a substance. The modes of extension are figure, size, and
motion; those that follow from being a substance are duration, order, and
number. But he also attributes to some bodies a vis agendi, and to others a vis
resistendi. Since in collisions the contest between the vis ad pergendum of the
"active" body and the vis ad resistendum of the "passive" body determines the
outcome, the quantity of those forces has genuine effects. It would thus
seem that force must be yet another mode of body. But it is, evidently,
neither figure nor size nor duration. As for motion, Descartes has taken
some pains to ensure that motion as he defines it will not be confused with
the "force or action" by which motion is transferred. 67
It would appear that force, since it has genuine effects on bodies, must be
either a body itself or a substance apart from bodies, or a mode of bodies.
Descartes ridicules what he conceives to be the Philosophers' view, accord
ing to which force, like weight, is a quasi substance, a "little soul." But force,
although it has no obvious place among the modes of extension, cannot, it
would seem, be dismissed entirely in the way that he dismisses weight or
occult qualities. Call this the problem offorce.
The solution of the problem is intimately bound up with the role of God
in bringing about natural change. That role is characterized most precisely
in the sections of the Principles leading up to the laws of motion, and the
proofs of those laws. Descartes there adapts the view predominant in Aristo
telianism: every created thing depends at each moment on God for its
existence. He uses an analogy with the conservation of the quantity of
matter, which was more likely to gain assent, to argue that divine immu
tability entails the conservation of the total quantity of motion in the world
as well.
66. A useful vademecum is found in Westfall 1971, appendix B. Unlike Westfall, I don't think
that an "incompatible chaos of meanings" is attached to the word 'force' and its cognates. Like
most terms not owned by a dominant theory or assigned standard criteria of application (like
chemical names, when used to denote samples of reagents whose purity is established by
standard methods),forceand vis were context-dependent. Their ambiguities were resolved in
the way competent speakers usually do-by reference to surrounding words, the subject
matter, the speaker's identity, and so forth. No doubt vis had in some contexts a sense it cannot
have in others. To that degree its senses were incompatible. But I am not convinced that they
were chaotic. Merely to criticize, moreover, the manner in which the word was used may well
lead one to overlook an important development-the development in the seventeenth cen
tury of new methods in scientific discourse for stabilizing the senses of key terms.
67. PP2§25, AT 8/1:54· Gueroult concludes: 'There is indeed a paradox, since forces are
all entirely reduced to modes of extension, while, on the other hand, they are defined, insofar
as they are 'forces', in opposition to [modes of extension]"-in opposition because forces
cause modes (1970:88).

Authenticated
Bodies in Motion
Since the outcome of collision is said to depend on the relative strength

of the force of acting in one body and the force of resisting in the other,
commentators have looked to divine action as the key to solving the prob
lem of force. Some, like Gary Hatfield, have identified force with divine
action. Others, like Martial Gueroult and Alan Gabbey, have taken force to
have a double aspect: as a cause of change, it is a mode of divine action; as
the effect of change, it is a genuine mode of extended substances. Most
recently, Daniel Garber has argued for what one might call a "nominalist"
view: to say that a body has a force of acting or of resisting is nothing other
than to say that God will conserve that body in accordance with the first law;
to say that in collision there is a contest of forces is just to say that God
resolves incompatibilities in the manner prescribed by the third law, and
allocates motion so as to conserve its total quantity.
It seems to me that since a close reading of Descartes's own texts has not
been decisive, it will be worthwhile to approach the problem of force
obliquely, by considering the manner in which God was thought by
Descartes and his contemporaries to conserve created things, as well as the
relation between divine and natural action. Though Descartes does not
explicitly bring out the connection, the problem of force, insofar as it is a
problem of locating the causes of natural change, is one part of the vast
question of the role of second causes. The connection does become clear
later, when certain aspects of Descartes's thought are taken up into the
occasionalism of Malebranche and others.
1. The first law and immutability. Descartes prefaces the exposition of the
laws of motion in the Principles with a paragraph on the causes of motion:
The nature of motion having thus been examined, it is necessary to con

sider its cause, which is twofold: the first, universal and primary, which is
the general cause of all the motions that exist in the world; and then the
particular [cause], by which it happens that single parts of matter acquire
motions they did not have before. Concerning the general [cause], it
seems manifest to me that it is not other than God himself, who in the
beginning [the French adds: by his Omnipotence] created matter along
with motion and rest, and who now, by his ordinary concurrence alone,
conserves as much motion and rest in the whole as he then put [in it]. For
although that motion is nothing other in the matter moved than its mode,
still it has a certain and determinate quantity, which we easily understand
to be able to be the same in the whole universe of things, even if it changes
in particular parts. [ ...] We understand also that it is a perfection in God
that he should be not only immutable in himself but that he should
operate in the most constant and immutable way: so that with the excep
tion of those changes that evident experience or divine revelation render
certain and which we perceive or believe to occur without any change in
the Creator, we should suppose no other [changes] in his works, to avoid

Authenticated
Motion and Its Causes [3 1 5]
attributing some inconstancy to him. From which it follows that it is most

agreeable to reason that from the very fact that God moved the parts of
matter in diverse ways when he first created them, and now conseJVes that
matter in just the same·way and according to the same reason that he first
created it, we should suppose that he also conseiVes always the same
amount of motion. (PP 2§36, AT 8/1:61; g/2:83)
The argument of §36 starts with an uncontroversial thesis. God is the first
cause of all motion. 68 The Creation was not of a quiescent matter to which
motion was added, or ofa matter capable of moving itself, but of matter and
motion together. 69 Since Descartes, as we will see shortly, holds in agree
ment with the central texts that the action of conservation differs only in
reason from the action of creation, it follows that if God created motion
then he conserves it as well. Now "evident experience" tells us that some
individual parts of matter change their motion. So God's operation in con
serving motion must change to that degree. But the total motion shared by
the parts of matter can still be conceived to remain constant.
Descartes builds a parallel with a more widely accepted principle of con
servation. The Aristotelians believed that prime matter could be created
and annihilated only by God. In the absence of reasons (like the received
doctrine of the Eucharist) to believe the contrary, it is most in keeping with
God's immutability to suppose that God, once having created a certain
quantity of matter, would conserve that same quantity. 7o Although the quan
tity contained in a particular individual may vary, the total quantity of prime
matter shared among individuals will not. But (Descartes insists) matter and
motion are here on a par; and since motion, though it is only a mode, does
have quantity, its total quantity too is conserved. 7l
68. That God is the ultimate cause of all that exists was unquestioned. There was, however,
controversy about the identification of the Prime Mover, whose existence is demonstrated in
Physics 8 and Metaphysics 12, with God, since some characteristics attributed by Aristotle to the
Prime Mover-in particular that the Prime Mover does not have infinite power-are incom
patible with those of God (cf. Coimbra In Phys. 8cwq3, 2:37off).
6g. Since 'part of matter' and 'sharing the same motion' are correlatives in Descartes's
physics, one could regard the cocreation of matter and motion in Cartesian physics as a
transposition of the Aristotelian view that matter and form were created together (see §5.1).
Gueroult has emphasized that the doctrine of the divine tweak [chiquenaude], setting into
motion a previously immobile, and therefore featureless, matter is no part of Descartes's view
(Gueroult 1970: 12 5 ff).
70. Since the central texts do not accept the identification of matter with quantity, the
common view that matter can be neither created nor annihilated entails nothing about the
conservation of the quantity of matter. Toletus, however, argues that "there is only as much
matter as there was at the beginning of the world, and will be at the end"; the argument seems
to be that in generation or corruption there can be division into parts or joining of parts, but
not change of quantity (ln Phys. 1CN16, Opera 4:37va).
71. Toletus concludes, in his question on creation, that "the World, together with its mutus
and time, was made by God," and later that "to the production [of the world] there corre
sponds a unique instant, which was the last instant of the non-being of motus, because then
[i.e., at that instant, conceived of as the backward limit of time] there was no motus and

Authenticated
Bodies in Motion
The components of the argument are: (i) the immutability of God; (ii)
the thesis, implicit but necessary to the argument, that creation and conser
vation are only distinct in reason, so that in particular they have the same
object and-barring evidence to the contrary-are subject to the same laws;
(iii) the experience of change in motion; (iv) the characterization of mo
tion as a quantity, summable across individuals. It is (i) and especially (ii)
that interest me.
The immutability of God has in the derivation of the laws of motion a
double role. It ensures that they have the character of laws: they hold at all
times and places and for all bodies (or, in the first law, all things). Descartes
clearly feels no need to argue that the "ordinary concurrence" of God in
natural change will, given his immutability, indeed have a lawlike character.
The second role of immutability is more remarkable. In §36 Descartes
argues that since it is a perfection in God that he should operate always in
the same way, we should regard his action as changing only if revelation or
evident experience require us to do so. Now the first law says that a simple
undivided body "never changes unless by external causes [nee unquam mu
tari nisi a causis extemis]" (PP 2§37, AT 8/1:62). This will follow from the
proposition argued in §36 if "evident experience" does not require us to
postulate any internal principles of change in bodies. But experience re
quires us only to suppose that body is res extensa, and extension contains no
principles of change. We then have no reason to suppose that God's action
will change, and his immutability as a reason for supposing it will not.
Immutability returns in the proof of the second part of the third law.
Since I discuss the proof in more detail later, I note one detail. At the end of
the proof Descartes adds: "And so the continual mutation of creatures is
itself an argument for the immutability of God [Sicque hr£c ipsa creaturarum
continua mutatio immutabilitatis Dei est m;gumentum]" (PP 2§42, AT 8/1:66).
What leads to that fillip is an argument to the effect that changes of speed
will be necessary if the total quantity of motion is to be conseiVed; that such
changes do occur is, paradoxically, evidence for the absence of change in
their cause.
It is a nice trick to make the continual mutatio of things an argument for
divine immutatio. More than that, it is a trick the Aristotelians thought could
not be pulled off. Suarez, citing Scotus, concludes that even from the per
petual movement of the heavens one cannot prove God is immutable, since
the mover of the heavens could itself be movable in space, if not actually
moved (Disp. 30§81g, opera 26: 115). He begins his own proof of
immediately after there was" (In Phys. 8c2q2, 212va, 2I3fb). Since tempus is that which is
measured by the regular motions of the heavens, the creation of time is the creation of motus,
and Toletus assumes, not without reason, that time existed as soon as the world did.

Authenticated
the immutability of God with the statement, "I do not believe that [this
attribute] can be evidently proved from any immediate effect, but from the
other attributes it can be demonstrated quasi a priori" (1:2, 113). Descartes,
who was not averse to beating the Aristotelians at their own game, may have
had such claims in mind when he put fmward the paradoxical argument. 72
The question to which that argument addresses itself is ancient. Suppos
ing that a perfect, and therefore immutable, being is at every moment the
conserver or creator of all created things, how can there be change in those
things? Would that not imply change in their ultimate cause? The Aristo
telians typically take up the question in relation to the biggest change of all,
the creation of the world. As Fonseca puts it: "How could God, when before
the foundation of the world he was not a cause in actu of things, have begun
to be their cause in actu without any change?"73 Since it was part of revealed
truth that the world is not eternal, one obvious answer-that the world is
coeternal with God, who is therefore always its cause in actu-is ruled out
from the start.
The standard answer has two parts. First of all, since motus is in the mobile,
not the mover, action does not change the agent per se (see §2.3). This
principle applies to creation as well as to those actions in which the object of
the action exists beforehand. As Toletus puts it: "Even in inferior agents, an
agent may pass [pervenire] from a proximate active potentia to actus without
any change in itself[...] In the same way God from not producing becomes
a producer, since the production is not in him but in the product" (In Phys.
8c2q2, 4:212vb). Fonseca, who considers the question at some length, ar
gues first that the ground upon which a cause is said either to be causing or
not causing-the ratio causandi-is a "concurrence" of the cause with its
effect. The concurrence of fire with the heating ofwater "is really the same
as the form of fire, or heat, which are the principles ofheating"; those forms
are produced in the water, not the fire. Concurrences are neither sub
stances, nor relations, nor real accidents; they add "no entity to things which
are causes in actu" (In meta. 7c8q4§2, 3ogaB). But "in order that a thing
should genuinely change [mutetur vera mutatione], even if [the change] is
not physical [...] it is necessary that it should acquire or lose some entity"
(§6, 31 7bE). So the change from being a cause in potentia to being a cause in
72. Descartes undoubtedly believed that there were proofs of divine immutability akin to
Suarez's (who calls them "quasi" a priori because no attribute of God is, properly speaking, the
cause of another). Perhaps the quickest is from God's simplicity: in whatever changes there
must, by the arguments of §3.1 and §4.1, be something that remains and something that is new
(since otherwise we would have no reason to say there was change rather than annihilation and
creation), and the thing that changes will therefore be a composite of the two (Suarez Disp.
30§812, opera 26:113).
73· In meta. 7c8q4, 3:3o6a (with reference to Aristotle Phys. 5c2text3 for the distinction
between causes in potentia and causes in actu). Cf. Toletus In Phys. 8c2q 1, opera 4:21 ora, 213ra;
Coimbra In Phys. 8c2q4a2, 2:271.

Authenticated
Bodies in Motion
actu is not a genuine change. In particular, when God creates the world, or
even if he re-creates it at each moment (a view some commentators ascribe
to Descartes), no genuine change occurs in him.74
Even though the act of creation requires no change in God, still it could
be, and was, objected that the act required a new volition and would there
fore again be inconsistent with his immutability.75 The response is that
through all time God can will that an action should occur at a certain time:
"Nor is it surprising that the volition of God to produce the world should
always have existed, even though the world has not; for God has had from
eternity a volition to create the world such that he did not have the volition
to create it from eternity."76 The logical point is obvious: willing that some
thing should occur always (or in 4004 B.C.) is distinct from willing always
(or in 4004 B.C.) that it should occur. But beyond that is a view of divine
volition brought out in Fonseca's remark: "God, when he begins to be a
cause in actu, does not begin to be otherwise in himself [non aliter se habere
coepit in se ipso], but with respect to [in ordinem ad] creatures, formally by
various relations of reason that suppose nothing in him, and virtually by the
divine essence itself, which by its virtue and eminence contains all the
possible modes of being toward creatures" (Fonseca In meta. 7c8q4§7,
3:32oaC). A bit later Fonseca adds that even though it is logically possible,
for some propositions, that God could either have willed that the proposi
tion be true or that it not be true, it does not follow that he can at one time
will that it be true and at another time will that it not be true: "even if in God
there was eternally a potentia logica toward opposite free acts of will and
intellect, still it was not toward them successively but only as alternatives"
(7c9q5§6, 3:332aE).
7 4· Garber writes: "If God really is immutable, then, it would seem, the world should not
change at all, and one state should be exactly like the one that preceded it; a genuinely
immutable God should create an immutable world, it would seem" (1992:282). Since the
successive re-cre;nion of the world in different states-if that is indeed what Descartes meant
by 'conselVation' -raises no difficulty not already raised by the first act, it should be clear that
the Aristotelians anticipated such questions and went to some lengths to answer them. Their
answers should put us on guard against assuming that an agent must vary if its effects do.
75· "If[...] God began to produce the world [de navo produxit mundum]: therefore he
began to have some proportion [to the effect, i.e., the world], which he did not have before:
therefore he has changed: if not according to place, at least according to concupiscentia and will,
which he newly has in producing the world" (Toletus In Phys. 8c2q2, Opera 4:2 1ora; cf. Coimbra
In Phys. 8c2q4a2, 2:271). Objections to the creation of the world de navo that do not come from
Aristotle himself are ascribed to Proclus, Simplicius, Avicenna, and Averroes. A brief outline of
radical Aristotelian, Augustinian, and Thomist views is given in Grant 1978:269r.
76. "Nee mirum, quod voluntas divina producendi mundum semper extiterit, nee tamen
semper mundus fuerit; sic enim Deus ab a:terno voluntatem habuit mundum procreandi, ut
non habuerit voluntatem eum procreandi ab a:terno" (Coimbra In Phys. 8c2q4a{, 2:273; cf.
Toletus In Phys. 8c2q2, Opera 213va; Fonseca In meta. 7c8q4§7, 3:319bC gives a very clear
exposition of the point). Among the authors cited on the question are Justin Martyr, St.
Isidore, and Hugh of St. Victor: it was not at all new.

Authenticated
The language is complicated, but the thought is reasonably straightfor

ward. The varying relations of God to the world that result from mundane
change are nothing in God but modes of being-"pure" modes, Fonseca
calls them, to distinguish them from figure, intensity, and so forth, which
are also called modes. But insofar as they are not nothing, they are, like
corporeal things themselves, "eminently" or ''virtually" contained in the
divine essence. 77 A proposition describes a state of affairs, to which God
would stand as cause to effect. That relation corresponds to a pure mode in
God. For each proposition that is not logically true or false, God eternally
wills either that it be true or that it be false, since it is inconsistent with his
immutability that he should will one and then the other. Of those proposi
tions that are willed some are, from the point of view of a temporally or
dered world, "executed" at a particular time and place. But that does not
count against their having been eternally willed. 78
2. Concurrence and the ordained power ofGod. It is still a long way from this to
the argument of §36. The immutability of God, as I have just presented it,
has two components: the ontological claim that the varying dependencies of
things on God are pure modes of his essence, and the formal claim that
whatever God wills, whether it be an eternal truth or a change occurring in
time, he wills eternally. Neither component entails that God's action will
result in the conservation of any quantity. The missing link can be found in
the phrase 'ordinary concurrence' [concursum ordinarium]: Descartes says
that God "by his ordinary concurrence alone" conserves as much motion
and rest in the whole world as when he created it. I will set aside for the
moment the reasons why quantity of motion especially is said to be conserved.
The problem then is how the additional notion of the ordinary concurrence
of God, now conjoined with the immutability of God, is supposed to yield a
law according to which the total quantity, whether of matter or of motion,
remains constant through time.
The ordinary concurrence of God is an exercise of his potentia ordinata,
one result of which is the conservation of the amount of motion with which
the world was created. In what follows I will briefly attest the relation of
77· To use a traditional analogy (Dionysius Div. nom. 4§4, opera 6g7/6g8; Thomas ST
tq 104a1, opera [Parma]1 :4oo): that a plant should be sustained by the sun is true by virtue of
the plant's nature, not the sun's; and if the plant is at first sustained and then destroyed by the
sun's heat, that need not imply any change in the sun's action, or in the sun. Yet the "eternal"
possibility of being a sustainer of plants is contained in the sun's essence, since if it had a
different essence it might not be even virtually a sustainer of plants.
78. One common way to put the point was that God's action is "tenninated" in a temporally
located effect, but its principle in God remains eternal. "Divine causalities, eternal in them
selves, we apprehend as temporal, since we conceive them to be terminated in temporal
effects" (Fonseca In meta. 7c8q4§6, 3:316bA). A person's will can be determined to some
action long before the action is performed; in such a case the action will be "terminated" at a
different time than the determination (ib. §7, 3:31gaF).

Authenticated
Bodies in Motion
potentia ordinata to law and then trace a path from concurrence to the
conservation of motion. It is at that point that the identity of creation and
conservation will become relevant.
One needn't look far to find potentia ordinata (or potentia ordinaria) jux
taposed with law. According to Suarez, the Theologians call
the power of God 'absolute' [when it is considered] in itself without any

other supposition or determination of his will, and apart from any respect
toward the nature of things or toward other causes. The ordinary power
they explicate in two ways: the first is that the absolute power of God is
called 'ordinary' insofar as it is joined with the knowledge and will, by
which God has decreed that these effects and not others shall be done,
whatever order they belong to-whether natural or supernatural. The
other more usual way is to call 'ordinary' the power of God insofar as it
operates according to the common laws and causes which he has estab
lished universally. (Suarez Disp. 33§17'132, opera 26:216, citing Thomas
ST 1q25a5, opera [Parma] 1:114)
The intended contrast is not between two kinds of power, but between the
same infinite power, considered first only in itself, and then in relation to
what God, in accordance with his goodness, has chosen ordinarily or regularly
to do. In the Aristotelians, the notion of law, though sometimes used to
explain the "order" God has imposed on his own will, is not emphasized. 79
Words like 'usual' (usitatus) and 'customary' (consuetus) are as common.
In Aristotelianism the appeal to concurrence had two motives. One was to
achieve a balance between the efficacy that, according to the central texts,
must on empirical and moral grounds be attributed to second causes (see
§7.3) and the equally compelling thesis that the infinite power of God is
limited only by logical impossibility. The other motive, more familiar, was to
reconcile the liberty of rational agents, and the evils that sometimes result
from their free actions, with the perfect goodness of God. That problem
belongs rather to metaphysics than to physics, but it is worth remembering
that the Aristotelians used the same tool for both.
Concurrence stands to action as conservation to existence. Many of the
arguments for concurrence rest on that analogy. Supposing that the neces
sity of conservation has been proved, Suarez argues that "if the cause de
pends on God in existence, so too does the effect, because each is a being by
participation [in God's being]; as the cause depends [on God] in the in
stant in which it acts, so the effect does in the instant in which it is brought
79· The Coimbrans write that if God acts according to his ordinary power, then "if fire
should be properly applied to a suitable matter, God cannot but concur with it in burning,
since that is demanded by the common and ordinary law, which produces a concurrence of this
sort in created things." But they define the potentia ordinaria of God as that which "respects the
common and usual course and order ingrained in things" (Coimbra In Phys. 2c7q 16a1, 1:287).

Authenticated
about, because in that instant too each is a being by participation; every

effect, therefore of a second cause depends on God in becoming [in fieri],
and consequently a second cause does nothing without the concurrence of
God" (Suarez Disp. 22§ 1 ~ g, opera 25:804). Every coming-to-be is analogous
to creation. Second causes have the power to generate substance from
substance, or to bring about accidental change; nevertheless the new being
they effect depends on God for its existence no less than do things brought
about immediately by God.
Some philosophers, notably Thomas and his followers, held that concur
rence is an action of God distinct from and prior to the .action of a second
cause. According to the Coimbrans, they believed that "all second causes
before they operate receive from God a certain influx and motus [...], by
which they are incited to produce their actions," in the same way that an
artisan moves his tools: "though the ax be sharpened and ready to cut the
wood, it will certainly not cut unless it is set in motion by the impulse and
motus of the artisan."80 The central texts all dissent from this opinion, for
two main reasons. The "premotion" of the second cause by God is super
fluous and cannot be assigned an object. In rational agents, moreover, if
there were such a premotion, it would be irresistible, and no agent would
act freely. 81 Instead they hold that God's action and the action of the second
cause are identical: "God acts with a creature by an action numerically
identical to that performed by the creature, in the same way that a carpenter
not only moves the saw [...] but also acts with the saw by an action numer
ically identical [with that of the saw], namely the cutting itself. "82
God's contribution to that unique action, according to some authors, is
the ultimate determination of the action of the second cause:
The action of second causes is determined per se and first by God with
respect to singularity. [...] This fire, for example, of itself is not deter
mined toward a certain heating, numerically [distinct from others], both
because it can elicit many [beatings], as is clear; and because nature does
not definitely intend this singular [heating] [...] Therefore, in order
that this heat should be determinately produced, it must be determined
otheiWise [than by the fire]. But among second causes there is nothing by
which it could be determined, so one must revert to the First Cause [ ...]
So. "Namque esto dolabra pra:acuta sit, & ad expoliendum lignum idonea, haudquaquam
tamen id expoliet, nisi artificis impulsu, motiique cieatur" (Coimbra In Phys. 2c7q13a1,
1:279). Cf. Thomas ST 1q105, 2pt1q109a1, Opera (Parma) 1:405, 2:429.
81. See, for example,Abra de Raconis Phys. 151; CoimbrainPhys. 2qq13a2, 1:279r. Suarez
refutes the opinion while denying that Thomas himself adhered to it (Suarez Disp. 22§2'1 7 ,49,
52, Opera 25:811, 823, 824). Fonseca likewise refutes the view while attributing it only to the
Thomist Ferrarius (Fonseca In meta. 5c2q9§3, 2:142bF).
82. Fonseca In meta. 5c2q9§3, 2:143aD; cf. Suarez Disp. 22§3'12, Opera 25:826.

Authenticated
Bodies in Motion
for whom it is indeed an ordinary and established law to supplement this

defect, so to speak, of nature. (Coimbra In Phys. 2c7q15a2, 1:284) 8 3
What precisely requires determination is not entirely clear to me. But the
doctrine seems to be that when fire heats something, for example, the
specific effect-that it is heating, not cooling-is determined by the nature
of the fire, but the degree of heating and perhaps its duration are not fully
determined either by the nature of the fire, by the place, or by the matter
acted on. The concurrence of God, therefore, is required to put the final
touches on the actual effect. 84
One last point will be useful. Divine concurrence is not necessitated by
second causes, even when all the conditions needed for their operation are
present. God "by his will supplies that concurrence to second causes"; his
will, of course, is free, and he can withhold concurrence if he chooses.
When the sons of the Hebrews were made to walk through the furnace of
Nebuchadnezzar, God withheld his concurrence from the flames, and they
emerged unscathed. 85 Nevertheless concurrence is natural in the same way
in which divine action in generation is natural: on the supposition that God
acts according to his ordained power, and within the limits just mentioned,
it is entirely determinate what God will do in any particular situation. Suarez
writes that divine concurrence is "according to nature" [ad modum naturtP]
for two reasons: "first, because in concurring [God] accommodates himself
to the natures of things, and to each supplies a concurrence accommodated
to its power [ vinus]; second, because after he decreed that he would bring
about and conserve second causes, by an infallible law he concurs with them
in their operation; that law, if it is taken absolutely, without supposing any
particular and definite will of God, does not induce necessity but is only like
a sort of debt of connaturality [quasi debitum quoddam connaturalitatis]"
(Suarez Disp. 22§4'll3, Opera 25:829). To which Suarez adds that the neces
sity of an effect is "natural" with respect to the second cause, but "only a
necessity ex suppositione or of immutability" with respect to the first cause.
The relation of cause to effect is necessary, in other words, only on the
supposition that God has bound himself to concur with the agent so as to
produce such and such an effect at such and such a time; the volition to do
so is, as we have seen, present in him from eternity (25:830).
83. Both the Coimbrans and Suarez speak of this view as nominalist, although Thomas is
also cited (Suarez Disp. 22§2i 10 & 30, opera 25:812, 818; Thomas ST1q105a1, opera [Parma]
1:403).
84. The connection between the Cartesian use of determinatio and the passages from
Thomas cited in n.83 was noted by Gabbey (1g8o:249r). Gabbey uses it in discerning what
determinatio might have meant; the role of determinatio in discussions of concurrence is not
mentioned by him.
85. See Oakley 1984 for information on this standard example and its role in debates about
God's absolute power.

Authenticated
To summarize: variation in the world implies no mutability in God. The

dependence of things on his power, whether in creation, conservation, or
concurrence, is in God a pure mode. The volitions, moreover, with which he
acts in those various ways are in him eternally; whether we can know God's
particular volitions or not, we can be certain that they do not change. That
formal condition does not of itself entail that any property of the world as a
whole must remain constant, since if it were to change (as indeed it would,
for example, if God were to annihilate any part of matter), that too would
have been willed eternally. But God has chosen to act according to a certain
order or law. In particular he has chosen, eternally, to concur in the actions
of natural agents. On the supposition that he has done so, and that his will is
immutable, the actions of natural agents follow by natural necessity from
their natures. Divine concurrence is, to echo the words of Soto in his theory
of grace (§7.1), owed to things. It follows that if we are given the nature of
the cause and the conditions under which it acts (including the nature of
the patient it acts on), the effect will-at least in specie, if not in all its
singularity-be determined. That determination is characteristic of a cer
tain sort of supernatural intervention in nature, notably in the generation of
animals and humans (§7.1).
To return now to the argument of Principles 2§36. In the Aristotelian
account of concurrence, the word 'law' seems to designate two somewhat
different orders: the mundane order, also called the "ordinary" or "customary"
course of nature; and the divine order, or the limits God freely abides by in
the exercise of his power. The mundane order depends jointly upon the
divine order and on the immutability of the will that binds itself to it.
Descartes's intention, however, seems to be to derive from the formal condi
tion of immutability, which would seem to yield only the permanence of the
mundane order, but not substantive claims about that order itself.
In Aristotelianism, substantive claims about the mundane order concern
not natural laws per se, but the effects consequent upon a thing's nature,
which is defined, as we have seen, to be the principle of its rest and motion.
The ordinary course of nature-that humans give birth to humans, or that
fire heats-rests on the natures common to the things we call human or fire.
Divine concurrence is determined, in the sense specified above, by the
natures of agents and, to a lesser degree, of patients. Descartes, on the other
hand, denies that corporeal things have natures in the Aristotelian sense,
though in the Principles he contrives a nominal agreement with the Aristo
telian definition (PP2§23 AT 8/1:53). The argument of§36 is intended to
yield a basis for natural regularities consistent with the conception of body
as res extensa.
From that conception it does not follow that any body moves. Experience
alone shows us that bodies do move, from which, since there is no principle

Authenticated
Bodies in Motion
of motion in bodies themselves, we may infer that they are moved by some
thing other than body, that is, by rational agents. From the immutability of
God's volitions, it would seem that the first law, at least, can be inferred.
Since bodies have no internal principles of change, all change must come
from outside; if no external cause acts on a body, it will not change. Yet for
all we know, God could eternally will that a body not acted upon by others
should nevertheless grow, or alter in shape, or accelerate. The formal condi
tion of immutability yields no natural law except in conjunction with some
further condition.
That condition can be found, I think, if we attend to the phrase quantum
in se est studied by Cohen (see n.18 above). Even though Descartes denies
that bodies themselves contain principles of motion and rest, still he takes
over from Aristotelianism the thought that God has given to bodies a "na
ture" that is sufficient to determine how they will act, but now only on the
supposition that they were created with whatever motion they have. That
nature consists in extension and its modes. Considering it alone, we find in
it nothing that would entail change in its modes: "if some part of matter is
square, we easily persuade ourselves that it will perpetually remain square,
unless something arrives from elsewhere that changes its figure" (2§37, AT
8 I 1:62). God, in the exercise of his ordained power, will give to things
precisely what is owed to them by nature-neither more nor less. Not to
concur in the actions to which they are fitted would be to frustrate his own
end in creating them. To do more than what is owed would imply that
something was lacking in the original act of creation. 86
But even if the first law can be vindicated, the argument of §36 and the
third law, which is supposed to follow from it, remain troublesome. It is here
that the identity of conseiVation and creation comes into play. Since that
identity is essential not only to the argument of §36 and the proof of the
third law, but also to the ontology offorce, it is worth examining Aristotelian
and Cartesian views on the question.
3· Conseroation and the third law. ConseiVation can be brought about im
mediately by God, or by God and created things conjointly. The former
applies most obviously to substances, but also to accidents created and
sustained apart from substances by God. The latter applies to accidents
inhering in a substance, or to matter and form; God's part amounts to
86. That God should give to things motions not implied in their nature, or in the original
donation of movement to the world, would also deprive physics of what certainty it has.
Concerning bodies, all we can be certain ofis what is contained in our clear and distinct idea of
them, which is to say, our conception of them as determinate res extensm, together with what
follows from the supposition, warranted by the experience of moving our own bodies, that
bodies do move. If bodies not acted upon by others were to change in state, the cause of that
change, since it is alien to our conception of them, would be entirely inscrutable; like the
"premotion" of the soul effected by grace, it could be known to us only through revelation.

Authenticated
concurrence in the continued production of an effect, as when light is

sustained by the sun.
Concerning the first, the Aristotelians concluded that they differ "not by
an intrinsic, but only by an accidental difference. "87 The difference is simply
that 'creation' adds to the core notion of a thing's depending on something
for its existence the connotation that it did not exist before; while 'consetva
tion' adds the connotation that it did (Suarez Disp. 21§2!2, opera 25:791).
Among the reasons given for the view, the most interesting is this: cre
ation and consetvation are one continuous action. Suarez writes:
Who would believe that the Sun, or a lantern, illuminate the air by a
different action when it is first applied, and afterwards? What would there
be in the cause, such that the first action should be interrupted or cease,
and the other begin? Or again: the first action ought to last only for an
instant of our time, because any reason according to which it lasted longer
would suffice for it to last forever. The subsequent action therefore either
would likewise last only an instant, and so there would be two immediately
succeeding instants, or else would last through the whole time that the
thing was conseiVed, and so the same or even a stronger reason can be
given to show that the first action lasts, and the superfluous multiplication
of actions will be avoided. (Suarez Disp. 21§2'13, opera 25:791)
ConseiVation need no more be a continual creation than holding a weight

need be a continual lifting. If a rock were suddenly to appear on my out
stretched palm, there would be no reason to say that the action of beginning
to hold it and the subsequent action of continuing to hold it are distinct.
The cause is unchanged, and the terminus of the action is unchanged.
Similarly, in creation and conseiVation, the terminus is existence, and the
cause, which is God, is immutable. To hold, as Descartes too holds, that
creation and consetvation are one and the same does not entail, as some
authors have supposed, that conseiVation is the re-creation of the world at
each instant; indeed Suarez's argument gives one reason to doubt the
coherence of such a view.ss
87. Coimbra In Phys. 8c2q104, 2:259. Suarez concludes that conservation is neither really
nor modally distinct from creation. It is distinguished "only by a certain connotation or
negation" (Disp. 21§212, opera 25:791). There was significant dissent from that view, repre
sented by Henry of Ghent (Qp.odlib. 1q7), Gregory ofRimini (In Sent. 2d2q6), and, among the
central texts, Abrade Raconis (Phys. 141).
88. Since I think Descartes agrees with the Aristotelians on this point, and since many
recent interpreters of Descartes, notably Gueroult (1970:272ff), have argued that Descartes
believes that time is "discontinuous" and that conseiVation is the re-£reation of the world at
each instant (ib. 257, 277), I will briefly state my reasons. An instant, in the Aristotelian
conception, is the boundary of an inteiVal of time, just as a point is the boundary of a line
segment. It has no existence independent of such intervals (though here one finds disagree
ment and ambiguity, as in Descartes). The term 'moment', used by Descartes in a number of

Authenticated
Bodies in Motion
The other sort of consetvation is a joint action of God and a created thing
in sustaining the existence of accidents or incomplete substances. It is, as I
have said, a kind of concurrence.89 Matter depends naturally on form for its
existence and conversely (see §5.1); if God were, by his absolute power, to
conseiVe matter without form, that would require an additional action to
supplement the causal activity of form (see Suarez Disp. 21§2,8, opera
25:793). According to the Coimbrans, 'The conseiVation, by which form
conseiVes matter, and the creation of that matter cannot not be
distinguished really among themselves. [... ] Form is said to conseiVe mat
ter, insofar as it actualizes matter: but that actualization is nothing other that
a certain mode, by which form insinuates itself into matter [...] and is the
same thing as the form; consequently it is really distinct from matter"
(Coimbra In Phys. 8c2q1a4, 2:259f). The creation of the matter, on the
other hand, is not really distinct from the matter. So the consetvation of the
matter by its form, and the creation of that matter by God, are-as one
might expect-distinct, being modes of distinct things.
More generally, the consetvation of a thing by a second cause, and its
production, can in some cases be distinct, while in others they cannot.
Elemental qualities are produced in matter by the heavens; once produced,
however, they are consetved by the substantial forms of the elements, which,
like all things, strive to preseiVe themselves. The most interesting case is
motus. Suarez, trying to explain why some qualities persist after their causes
cease acting (as heat in water) and others do not (as light in air), cites the
passages that have been adduced in favor of an "atomistic" view of time, can usually be read as
denoting an interval, perhaps short but not infinitesimal (see Garber 1992:270; two clear cases
are To Mersenne 11 Mar. 1640 and 2 Feb. 1643, 3:36, 614; Mersenne defines 'moment' to
denote what we call a second, cf. Harm. univ. 1pr7, 1:12). Even the crucial passage in Meditation
3 (7:49; cf. 109) does not require that conservation be interpreted as creatio de novo: one should
not ignore the fact that Descartes writes quasi rnrsus creet, not rnrsus creet. The point is that
because intervals of time are really distinct, each moment in which I exist can be conceived of
as the first or last moment (i.e., as an interval before which I did not exist, or after which I will
not exist), and that the action of God in sustaining my existence is-just as the Aristotelians
argue-not distinct except in reason from the action of creation. Indeed Descartes is there
taking up a point made by Suarez also: even an eternally existing thing can be created (see
Disp. 20§5'l! 11 1r, Opera 25:782IT). Finally, although Gueroult believes that continuous re
creation is needed to explain the difference between the Cartesian and the Aristotelian con
cepts of motion, I think the definition of PP 2§25 suffices to show that he rejects the view that
motion is a jluxus or via. Since the Aristotelian sources uniformly take conservation to be an
action continuous with creation, and therefore not really distinct from it, and since they do not,
so far as I can see, even take notice of the doctrine of continuous re-creation, it seems to me
that in the absence ofstrong reasons to attribute that novelty to Descartes, one should take him
to agree with them, even apart from the psychological arguments advanced by Beyssade (see
Garber 1992:268rr, 361; see also Arthur 1988 for a vigorous refutation, largely on internal
grounds, of the traditional view).
89. Descartes, it is worth noting, in a letter of 1641, uses the term concursus to denote the
action by which substances too are conserved in being (Descartes To *** [Hyperaspistes] Aug.
1641,AT 3:429). The term would be appropriate if one thought of the striving of each thing to
persist as a kind of cause of its existence, and of God as concurring with that cause.

Authenticated
following: "In motus one can easily give a reason why it always depends on an
actual agent, namely because its existence is not so much existence as
becoming [ ejus esse non tam est esse quamfieri]; and because becoming cannot
be without acting, so motus [cannot be] without a mover [quia ergo fieri non
potest esse sine agere, ideo nee motus sine movente]" (Smirez Disp. 25§31 14, opera
25:797). Suarez himself, in contrast with Thomas, does not take that to be
sufficient to explain every case. But it is worth noting that here again conser
vation and production are identical.90
The doctrine I have outlined can be summarized as follows. The action of
creation, with respect to complete substances (and to separated accidents
or separated incomplete substances like the soul after death), is, like any
action, nothing other than the dependence of an effect-in this case the
existence ofcreated things-on its cause. That dependence is the participa
tion in being that God, as the only necessary being, has chosen to give to
them. Conservation differs from creation only in connotation; it is otherwise
identical with, and indeed continuous with, creation. With respect to acci
dents and incomplete substances, on the other hand, the two divine actions
can differ in re. The creaturely actions of production and conservation
sometimes differ, sometimes not. The conservation of heat in water warmed
by fire was held to be distinct from its production; the conservation oflight
in air illuminated by the sun was held not to be distinct; the conservation of
motus in projectiles seems to have been decided differently by different
philosophers.
Descartes, therefore, when he embarked on the project of proving his
laws of motion by way of the traditional three modes of divine efficient
causation-creation, conservation, and concurrence-was faced with ambi
guity atjust the point where he required an unequivocal connection. But he
had, it seems, already a firm conviction that the nature of body consists in its
being a res extensa. Since, as I have said, nothing in res extensa entails that it
should move, he was thereby already committed to supposing that the
motion we find in the world could have come, in its entirety, only by way of
the immediate creative act of God. There is, moreover, no second cause in
the world that could sustain motion once God imparted it to things. In
response to More, Descartes writes that "I consider matter leftfreely to itself and
receiving no impulse from elsewhere to be entirely at rest. It is impelled by God,
who conserves as much motion or translation in it as he put there in the
beginning" (To More 30 Aug. 1649, AT 5:404; the passage in italics is a
go. That resemblance may have inspired the analogy made by Bonaventura and Thomas
between conservation and illumination (Thomas ST 1q104a1, opera [Parma] 1:400). There
was no dearth of "light-cosmogonies" in which the creative power of God was taken, more or
less literally, to be an illumination-Grosseteste was one prominent example (Hedwig
1g8o:136f). Descartes was aware of the analogy (To*** [Hyperaspistes] Aug. 1641, AT 3:429).

Authenticated
Bodies in Motion
quotation from More to Descartes 23]ul. 1649, 5:381). 'At rest', as the word
translatio shows, has the properly Cartesian sense of 'not separating (at a
designated instant) from the vicinity of neighboring bodies' (see §8.1).
Motus, then, like light in air illuminated by the sun, is the sort of quality that
cannot exist except through the immediate continued action of the agent
that produced it, which is to say, God.
Motion is thus entirely on a par with res extensa itself. Both are immediately
created by God, and must, through the continuation of the creative act, be
immediately consetved by God. 91 In a world in which a single body-the
world itself-were given one motion, as in certain Scholastic thought
experiments (e.g., Buridan In Phys. 3q7, sova; cf. §2.2), that would be all
one needed to say. The first law would suffice. But that is not what God did:
"it is obvious that God, in the beginning, when he created the world, not
only moved its diverse parts in diverse ways, but at the same time also
brought it about that some should impel others and transfer their motions
to them: so that now, consetving [the world] by the same action, and with
the same laws with which he created it, he consetves motion not always
implanted in the same parts of matter but rather passing from some to
others accordingly as they collide with one another" (PP2§42, AT 8/1:66;
cf. Monde 7, 11 :43). In this vest-pocket version of Genesis, Descartes affirms
first that God created matter and motion together; then that he distributed
motion differently to different parts; and finally that in that first instant
those parts not only moved but impelled others, and transferred their mo
tions to them, so that the laws of motion are instantiated even as the world is
created.
There is nothing in matter itself that would compel one to suppose that
when a body meets a body it would notjust stop, like dough thrown against a
wall. But if the parts of res extensa cannot interpenetrate, it does follow that
certain motions of those parts cannot persist together. Descartes supposes
that God set things up so that from the very beginning parts of matter in fact
have incompatible motions. Motion is transferred, and the law that governs
transfer instantiated, at every moment of the world's existence.
But what should that law be? Consider the analogy drawn in the argument
of §36 between the quantity of matter in the world and the quantity of
motion: "from the very fact that God moved the parts of matter in diverse
ways when he first created them, and now consetves that matter in just the
91. See Carteron 1922:275. This begins to answer a question raised by Garber, who asks why
quantity of motion, rather than various other modes of extension, is conserved (Garber
1992:282). Unlike figure, or (for somewhat different reasons) number, motion alone must be
considered to have been created independently along with body; figure, size, and the number
of bodies will all be entirely determined once matter and motion are supposed. See n.95 below.

Authenticated
same way and according to the same reason that he first created it, we
should suppose that he also conserves always the same amount of motion"
(PP2§36, AT 8/1:61; g/2:83). I think one should take seriously the use of
part here and elsewhere (notably in the definition of motus at 2§25) to
denote what might otherwise be called individual material substances; one
should also take seriously the use of the singular action to denote creation:
Descartes in §42 writes of the "immutability of the operation of God, con
tinually conserving the world by the same action by which he once created
it," and a bit later of God "conserving it [i.e., the world] by the same action."
Although it is true that elsewhere, notably in the Meditations, creation is
clearly an action that terminates in individuals, and that he speaks of God as
"moving" or "impelling" bodies, still I think it is promising to consider the
creation of matter in general, and thus of a certain fixed quantity of matter,
as a single action.
The creation of matter and its conservation will also be a single con
tinuous action. Although God could withdraw his conserving action from a
part of matter,92 thereby annihilating it, we have no reason to believe that
such a thing occurs. The law governing his creative and conserving action
we may easily imagine to be invariant with respect to the temporal termina
tion of that action. God, when he created the world, willed that a certain
quantity of matter should exist, and, since conservation is continuous with
creation, he wills it at every subsequent moment.
Earlier I argued that since the existence of motion, as opposed to its
possibility, is not entailed by the existence of matter, the creation of motion
was independent of the creation of matter. Although God could not have
created motion without matter (motion, where it exists, is a mode of mat
ter), he could have created matter without motion. I now suppose that, by
analogy with matter, the creation of motion can be regarded as a single
action, namely, as the creation of "a certain & determinate quantity [...]
which we easily understand to be able to be always the same in the whole
universe of things" (PP 2§36, AT 8/1:61). 93 That quantity is a whole of
92. See To More 5 Feb. 1649, AT 5:272, and the general remarks on conseiVation in To***
[Hyperaspistes] Aug. 1641, AT 3:429.
93· It is pleasant to find this interpretation of Descartes's argument confirmed by Regis,
who writes: "One must not put in God as many volitions as there are particular movements; for
we know that this multitude of volitions is repugnant to the simplicity of the divine nature: or, if
we do put several volitions in God, one must conceive that they are distinguished neither really
nor formally, but only by a distinction of reason, founded on the fact that because our mind
cannot understand the infinite extension of the volition by which God resolved to move
bodies, it divides [that volition] into as many parts as there are particular bodies that God
wishes to move; which does not accord with the idea of a perfect being whose extreme
simplicity excludes every sort of real and formal composition, and admits only composition in
reason, which is consequently the only [composition] one may attribute to God. So, for

Authenticated
Bodies in Motion
which the quantities of motion in parts of matter are parts, just as the
quantities of the parts of matter themselves are parts of the quantity of the
whole.
God, then, when he created the world, willed not only that a certain
quantity of matter should exist, but also that a certain quantity of motion
should exist. Since creation is continuous with creation, he wills the same at
every subsequent moment. The whole quantity is conserved, even though its
division into parts changes when particles collide. Though it is not a sub
stance, motion shares with body the character of having quantity and of
being divisible into parts. Even apart from the influence of earlier attempts
to quantify motus or impetus, it seems to me that the resemblance of motion
to body would have made the hydrostatic model attractive to Descartes, and
with it the identification of the quantity of motion in a particular part of
matter with the product of volume and speed.
The conservation of the total quantity of motion in the world thus reflects
the single continuous action, and the single eternal volition, by which God
both creates and sustains it in existence. When two parts of matter whose
motions are incompatible meet, God cannot execute that volition simply by
conserving the quantity of each part. Instead he concurs in the effect of
each upon the other so as to resolve the incompatibility, and, if need be, to
diminish the motion of one part and augment that of the other so as to
sustain the total quantity he first created. That condition, it should be
noted, in no way entails the rules of collision that Descartes actually devised.
Beeckman's rules, or Regis's, are equally compatible with the conservation
of the total quantity of motion. For that one must turn to the history of the
problem, some of which I have outlined above.94
4· Force and divine action. At the beginning of this subsection I noted that
there are essentially three views about the ontology of force in Descartes's
example, we will not say that God wills rain and fair weather by two particular volitions, but
instead we will think that rain and fair weather, whatever opposition there is between them, are
two effects ofone and the same volition, by which God wills that rain succeeds fair weather and
fair weather succeeds rain" (Regis Cours 1pt2c9, 1:331).
94· In n.91 I argued in response to a question raised by Garber that figure and size, unlike
motion, are entirely determined once matter and the distribution of the total quantity of
motion are given (though questions remain concerning the definition of the parts of matter,
and though Descartes is quite unclear about the circumstances under which collision will bring
about the breaking off ofa piece of one of the two bodies rather than absorption or reflection).
Number, also mentioned by Garber, is not a mode of extension but only a modus cogitandi (PP
1§58, AT 8/1:27). In any case the number of parts of matter will, with the reseiVations just
mentioned, also be determined once matter and motion are given. The only conceptually
arbitrary aspect of the argument of §36 is the identification of the quantity of motion in a
moving part of matter with the product of its volume and speed. Here the history of the
problem, and especially Beeckman's collision rules, provide the best explanation (see the
analysis by de Waard in Mersenne Corr. 3, appendix I, and Gabbey's remarks in Gabbey
1972 :38o0 "). For Beeckman, as de Waard notes, quantity of motion is "a primitive possession of
matter" (643), whose use he does not attempt to justifY.

Authenticated
Motion and Its Causes [33 1 1
physics: that it pertains solely to divine action; that it pertains both to divine
action and to bodies; that it is a ja(on de parler; a way of describing how bodies
will act according to the laws of motion. The conclusion I reach here is that
in the metaphysical setting of Descartes's arguments for the laws of motion,
the last two views can be reconciled; the first, on the other hand, does not
accord with the dominant view that action inheres in the patient. The
historically significant fact is not the predominance of one or the other, but
their equivocal co-presence, which can in part be traced to the Aristotelian
doctrines Descartes made use of. It was left to a later physics to make one or
the other predominant.
I will first show in what sense conservation and concurrence were con
ceived of as instances of efficient causation. There is, I think, a significant
difference in how God was thought to act. I will then briefly take up the
distinction between the static forces that Descartes appeals to in his discus
sions of machines and in his definition of what is now called "work," and the
dynamic force attributed to moving bodies and measured by quantity of
motion. The distinction parallels the distinction between the conserving
and the concurring action of God. Finally I will argue that the distinction I
have appealed to several times already between the senses in which a change
could be called 'natural' permits the reconciliation of the three views prom
ised above.
In Aristotelianism, creation, conservation, and concurrence are for the
most part treated as instances of efficient causation. 95 The efficient cause is
defined as "that from which comes the first principle of change [ mutatio] or
rest. "96 Suarez glosses the definition in terms of actio, which is the depen
dence of mutatio on whatever gives it being, per se rather than accidentally,
so that the efficient cause is the "principle from which the effect flows forth
or depends by action. "97 While the formal cause is the terminus of change,
and the material cause that out ofwhich the form is produced, and in which
it inheres, the efficient cause is that which alone gives existence to the
95· God was also said by some philosophers to concur with the material, formal, and final
causes (Fonseca In meta. 5c2q12, 2:17oll·). Suarez, treating that view rather harshly, argues that
secundum proprietatem "we do not recognize in God any other force of causing outside
himself except that of omnipotence, which is an effective power only" (Disp. 22§6'l[28, Opera
25:809). The danger in supposing that God could be a material cause is obvious (see the
remarks on David of Dinant in §4.1); the danger in supposing that God could be a formal
cause is the temptation to regard him as the anima mundi. Although God, as the ultimate end,
does exercise final causality, there is no reason to suppose that he concurs in the "metaphorical
motion" of other final causes.
96. Aristotle Phys. 2c3text31, 194b29. Discussions of the definition and ratio causandi of the
efficient cause include Coimbra In Phys. 1:229 and 2c7q7, 1:25611'; Suarez Disp. 17§1, §to.
97· "Perinde enim est dicere efficientem causam esse primum principium uncle est actio,
ac dicere uncle est effectus, media actione, seu esse principium a quo effectus profluit seu
pendet per actionem" (Suarez Disp. 17§t'l[6, Opera 25:852).

Authenticated
Bodies in Motion
change itself. 98 Unlike the final cause, whose motio is only metaphorical, the
action of the efficient cause, which is, as we have seen, just the motus itself
with respect to its dependence on an agent (§2.2), is real.
The creation of substances is not, strictly speaking, a mutatio or motus, if
those terms presuppose a preexisting subject. Smirez writes: "This depen
dence [of creature on Creator] does not truly satisfy the definition of muta
tio [non habet veram rationem mutationis]; but it does truly satisfy the definition
of a way or becoming of the creature, and so it is called a passive creation; it
also truly satisfies the definition of an emanation from God, and insofar as it
is from him, it can truly and properly be called an action of God himself,
formally a transeunt action, by which the creature is produced" (Disp.
20§4~17, opera 25:774; cf. Thomas ST 1p45a3, opera [Parma] 1:185).
Against Thomas, who argues that creation, since it is not a mutatio, is not an
actio but only a relation between God and creature, Suarez holds that
mutatio can be more broadly used of "any action or effecting, which brings
some novelty" (SuarezDisp. 1 18), which in creation is actual existence itself.
In order for creation to be an actio it suffices that there be a "flux emanating
from God" (~21, 775).
In keeping with the identification of conservation and creation, Suarez
argues that although creation is in the broad sense a mutatio of the thing
created, it does not entail an actual succession from nonbeing to being, but
only what Suarez calls an "imaginary succession. "9 9 Since God is the only
necessary being, we can, for any other thing, at least conceive of it as not
existing, and then existing, even if it has in fact never not existed.
Creation sive conservation, then, is a peculiar sort of action. Suarez
distinguishes two ways in which things, insofar as they persist, depend on
God. The first is "permissive": things "remain in existence just so long as
they are permitted to remain by God, that is, so long as they are not deprived
of their existence, by him who can so deprive them." The other is "positive":
continued existence requires an actual influx of being from God to crea
tures. Suarez's primary argument for the positive way is that "every positive
action necessarily tends to some existence," and so if God required some
positive action to destroy things, he could only corrupt them, rather than
annihilate them. But since he can annihilate them, it follows that he does so
98. This is not to deny that a form can be an efficient as well as a formal cause, but rather
that the form attained in change is not numerically the same as that which initiates it.
99· "Even in eternal creation [i.e., the creation ofsomething that exists at every moment of
time] creation can be distinguished in reason from conseiVation. Insofar as such a creation
signifies existence absolutely in eternity, as a simple participation of the existence of the
created thing [in the existence of God], it satisfies the definition of 'creation'; but insofar as we
conceive in it an imaginary succession, then for any designated instant such a succession
satisfies the definition of 'conservation"' (Disp. 20§5'120, opera 25:784). It is in that manner
that the "generation" of the eternal Word from the divine essence is to be conceived.

Authenticated
by ending the positive action by which he sustains them in existence (Disp.

21§ 1'114, opera 25:789; the same argument is found in Descartes's reply to
Gassendi, AT 7:370).
Creation sive consexvation, in short, consists in the voluntary donation of
being by God to other things. All things exist in God as divine ideas; but only
some are permitted, and borne, as it were, into actuality through his power.
That permission and positive action are required at every moment of their
existence; so that "if for even a moment the concurrence of God in consexv
ing [them] were absent from created things, all would immediately return
to nothing" (Coimbra In Phys. 2c7q 10a2, 1:267; cf. Descartes To ***[Hyper
aspistes] Aug. 1641, AT 3:429.10-11). If there were a "force" associated with
it (as Gueroult seems to suggest), 100 it would be quite unlike the moving
force ascribed to parts of matter or to God himself (To More 30 Aug 1649, AT
5:4o3f). I will shortly compare it with the static force by which a body is held
in place when it is in equilibrium in a balance or suspended from a pulley.
Although it may be relevant to understanding the force of rest, it is, so far as
I can see, quite irrelevant to the force of motion.
Concurrence, on the other hand, is relevant. We have seen that, accord
ing to the central texts, the action of the first cause in concurring with that
of a second cause is not additional to the action of the second cause, but
rather identical to it. Concurrence does not consist in moving the second
cause to act, or in a merely instrumental use of creatures by God to accom
plish his ends (God can use creatures, but that is not what happens when
they act ordinarily). When God concurs with a second cause, the Aristo
telians argue, he does so immediately, both in the sense of not employing
intermediaries (like the celestial spheres) 101 and in the sense of being
present and exercising his power at the time and place where the second
cause acts. As Descartes says in his response to More, God is "everywhere"
not because he is an infinite extended thing coextensive with the world, but
because "his power is exerted, or could be exerted, in extended things"; he
is present per modum potentice, not per modum rei extensce (To More 30 Aug.
t6 49 , AT s:4o3; cf. 343).
100. In Gueroult's interpretation, the conseiVation of matter and the total quantity of
motion consists in their re-creation at each instant (Gueroult 1970:114£, 117). He therefore
believes that a force creatrice as well as a force mouvante is exercised by God. Since, for the reasons
given in n.SS, I don't think that in Descartes's view God re-creates the world at each instant, I
don't think that there is a force creatrice. The real question is whether there is a force conseroatrice
corresponding to the "force of resisting" of a body at rest distinct from the force that corre
sponds, in the manner described below, to the action of concurrence.
101. This would need to be qualified to distinguish among substances and various kinds of
accident (see Suarez Disp. 21 §3, Opera 25:79411 ). Suarez holds of motion in particular that it is
always conseiVed not only by God but by a secondary cause (1 14, 797). Many other kinds of
accident (e.g., quantity) and all substances are conseiVed immediately and solely by God. God's
action in concurrence is always immediate.

Authenticated
Bodies in Motion
Most important, the central texts hold, as I have already noted, that the
action of second causes is determined by them only quoad speciem, with
respect to kind; the divine contribution consists in a determination of the
action quoad singularitatem, with respect to their singularity:
The reason and manner of his pre-definition [prcefinitio, that is, God's
eternal, and therefore prior, volition that this secondary cause, acting here
and now, should produce this singular effect] can be conceived thus. God,
seeing that in the given circumstances he ought to concur with such-and
such a species of act according to the ordinary law, by which he has
decreed that a created agent shall not be destitute of his assistance, pre
defines the individual concurrence, with which he wishes to give being [to
the effect]. But because God and second causes give being to [injluunt]
the effect of second causes not by different but rather by the same concur
rence or action [ ...] , the concurrence even of created agents is deter
mined in individual degree, so that this rather than that effect of the same
species occurs, among those that could be produced in the same circum
stances. (Coimbra In Phys. 2qq15a2, 1:285)
The clotted syntax of the passage, which I have done my best not to relieve,
should not obscure the essential point. In every action of a second cause,
there is some indetermination with respect to its effect, an indetermination
that can only be supplemented by God acting with the second cause. An
archer, say, aims at the bull's-eye; but it is God who, in keeping with that aim,
determines the precise point at which the arrow strikes.
I suggested a moment ago that if there is a force associated with the
conserving action of God, it would be better compared with the forces
referred to in statics than to the forces of moving bodies. St. Bonaventura,
the Coimbrans remark, compared conservation by the first cause with "a
weight, that is held in the air by the hand" (Coimbra In Phys. 2c7q10a2,
1:267). Descartes seems to have thought of God's action as a concurrence in
their self-preseryation, as if each thing shared, however imperfectly, with
God the attribute of being causa sui (To *** [Hyperaspistes] Aug. 1641, AT
3:429). A certain divine action, in other words, is required merely to sustain
a thing in existence, even as the sustaining of a weight hung from a nail
requires a certain force to be exerted by the nail. What is more, the nail, at
least in principle, is capable of exerting whatever force is required to sustain
the weight. It resists, in other words, to whatever degree necessary the
downward force of the weight.
Descartes, as is well known, made a sharp distinction between the forces
defined in statics and the force of motion that he refers to in the third law.
In a letter to Mersenne, he reproves those who "make use of speed to
explain the force of the lever" and other similar machines: ''When you say

Authenticated
that a force that can lift a weight from A to Fin one moment, can also lift it
in one moment from A to G [a point twice as far from A as F] if it is doubled,
I in no way see the reason for that" (To Mersenne 2 Feb. 1643, AT 3:615).
When a scale has equal weights on both sides, the addition of only a little
weight to one side will make that side drop rapidly, and there is no obvious
relation between the additional weight and the speed of descent.
Perhaps for that very reason, if Descartes regarded the conseiVing power
of God as a static force, he would, in identifYing it with the force of rest, have
introduced into his theory of motion a heterogeneous element. That, it
seems to me, is just what he did. As Garber notes and as I argued earlier, the
force of rest comprises two elements: the natural inertia, present in every
body moving or not, and represented by the conversion factor MB/ M0 and
an additional element, present only when a resting body successfully resists
the action of a moving body according to Rule 4, and which is indefinitely
large. Rule 4, as we have seen, seems to have been introduced to account for
reflection from a body at rest (such as the earth), which would otherwise
always move off in the same direction as the moving body, joined with it. Just
as in statics, the fixed point from which a pulley is hung, or the fulcrum of a
lever, may be conceived of as being supplied with an indefinitely large
sustaining force, so a body at rest, in the situation of Rule 4, receives from
God whatever static force is needed for it to remain in place.
The force of a moving body, on the other hand, when it succeeds in
causing another body to change its state or its speed, can be referred to the
concurring action of God in that causing. It is, evidently, the efficient cause in
some sense of the change; but since Descartes ascribes no active power to
matter per se, the precise sense remains to be determined. I will return to
that question shortly, but first I want to follow out the consequences of
taking God to concur in the actions of moving bodies on others.
We have seen that in concurrence, the action of God and that of the
second cause are identical. God's action, moreover, is in accordance with his
ordained power; the ordinatio of his will which defines his ordained power
we have seen sometimes referred to as a law. Consider now the proof of the
second part of the third law: "it is clear that God [... ] not only moved its
diverse parts [i.e., of matter] in diverse ways, but at the same time also
brought it about that some should impel others and transfer their motions
to them: so that now, conseiVing [the world] by the same action, and with
the same laws [ ... ] , he conseiVes motion not always implanted in the same
parts of matter, but passing from some to others accordingly as they collide
with one another" (PP 2§42, AT 8/1:66). We have the following scheme.
When the modes of motion or rest of two bodies are incompatible, there
must be in some cases a transfer of motion from one to another. This is not a
literal passing of numerically the same motion from one subject to another;

Authenticated
Bodies in Motion
like the Aristotelians, Descartes denies that such a thing can naturally oc
cur.l02 Instead God introduces into the two bodies different degrees of
motion than those they previously had. Now the bodies themselves, qua
bodies, cannot occupy the same space; to that extent the fact that one or the
other changes its motion is determined by the nature of the bodies them
selves. But there are any number of ways in which their motions-their
successive rupturings from their vicinities-could be made compatible.
There are even any number of ways consistent with God's conseiVing the
total quantity of motion in the world. God alone determines the precise
amount transferred, which he does according to the third law and the
definition of quantity of motion.
Thus far God's action is that of a concurring cause in the manner spelled
out by the Aristotelians. But there is a joker in the Cartesian deck. To return
to the question postponed a moment ago: in what sense is the moving body
that changes the motion or rest of another the efficient cause of its motion?
Consider the situation of Rule 5, in which a moving body collides with a
body at rest and the two thereupon move off together in the direction of the
moving body. If God genuinely concurs with the moving body, rather than
simply employing it as an instrument to his ends, that body ought also to be
an efficient cause. Now there is an obvious sense in which it yields an
occasion for God's redistribution of the total quantity of motion in the
world consistent with the conseiVation of that quantity. Its volume and
speed, moreover, suffice to determine how God will do so (so long as the two
bodies must move as one after the collision). But is it "that from which
comes the principle of the beginning of motion"? It doesn't seem that it
could be. If we set aside God's concurring action and the laws governing it,
we are left with no re/ason at all why the body at rest should move-why it
should, in other words, be separated from the vicinity of its neighbors. The
other body's motion is no reason, since it could be conseiVed in other
ways-by a change of direction, for example.
Concurrence is, from the Aristotelian point of view, not quite the right
term for what is going on. Yet the moving body resembles an efficient cause
in many ways: it "acts" by propinquity, its motion is a sine qua non condition
of the effect, that effect is proportionate to it, and so forth. It is a perfect
simulation of an efficient cause, lacking only the active power that an effi
cient cause must have. The closest equivalent in Aristotelian natural phi
losophy is human generation. The human soul is not generated when a child
is conceived; only the body is generated. The soul is created by God on the
occasion of the fertilization of the matrix by the semen. That divine action is
102. To More 30 Aug. 1649, AT 5:4o4r. The less punctilious Descartes of Le Monde thought
differently: "The virtue or power of moving oneself that is found in a body can indeed pass in
whole or in part into another, and thus no longer exist in the first" (Monde 3, AT 11:11).

Authenticated
utterly natural in the sense of being entirely determined by its natural

concomitants (on the supposition, of course, that God acts according to his
ordained power). But the father has no more the power to create the soul of
his child than res extensa has to do anything at all.
The force of motion, then, is in God if it is anywhere. More precisely: the
power that brings about the genuine change that is change of motion, or
from rest to motion, can only be the power by which God concurs in the
"actions" of bodies as they collide with one another, the power by which he
alters the portion of the total quantity of motion allotted to each body. That
attribution is confirmed by the response to More: 'The moving force can be
of God himself conseiVing as much translation in matter, as he put into it
from the first moment of creation, or even of created substances, like our
mind or whatever other thing he has given the force to move body. And
indeed that force, in created substances, is one of their modes, but not in
God; but because that cannot be so easily understood by everyone, I did not
want to deal with this matter in my writings, lest I should seem to favor the
opinion of those who consider God to be the soul of the world, united with
matter" (To More 30 Aug. I64g, AT s:404). The "moving force" is not a
mode in God for the simple reason that God has no modes.l 03 It is the
volition with which God created and with which he conseiVes the total
quantity, an eternal volition that is unchanging, as we have seen, even
though its effects in the temporally ordered world vary.
But for that very reason it cannot be the vis ad agendum referred to in
Principles 2§43. Nor indeed can the vis ad resistendum also mentioned there
be the power by which God conseiVes a body. God's power cannot be said to
vary, but these vires do. 104 In fact that section of the Principles tells us some
thing on the face of it quite different: both the force of acting and the force
of resisting consist "in this alone, that each thing tends, quantum in se est, to
remain in the same state in which it exists, according to the first law" (AT
8! 1:66). That each thing so tends is, of course, brought about by God
conseiVing it. But that in itself does not tells us why forces should have
quantities (which they must if they are to be compared as in §36).
The gloss supplied for 'force' resembles the gloss on conatus later on: to
say that bodies have a conatus to recede from the center they resolve around
is just to say that they are situated and moved in such a way "that they would
103. The "created substances" referred to in the second sentence are the human or angelic
minds to which God has given the force to move body, not the bodies themselves (contrary to
Prendergast 1975:g6). Descartes's point is that they, being mutable, can be said to have modes,
but God cannot.
104. In the striking passage I discussed in §6.4, Toletus does speak of God as concurring
more strongly with the stronger of two causes, like an emperor who supports one of his kings
more strongly than another (Toletus In Phys. 2cgq 13, opera 4:76rb). It would be interesting to
know how widespread that way of thinking was.

Authenticated
Bodies in Motion
in fact go in that direction, if they were not impeded by another cause"

(3§46, AT 8/I: 108). In both cases it is tempting to say, as Garber does, that
the "forces [...] can be regarded simply as ways of talking about how God
acts, resulting in the lawlike behavior of bodies; force for proceeding and
force of resisting are ways of talking about how, on the impact-contest
model, God balances the persistence of the state of one body with that of
another" (1992:2g8). The view is, it seems to me, especially attractive in
talking about the force of rest, which (as Gabbey and Garber both point
out) reveals itself only when the situation of Rule 4 comes about, and then
somehow manages to vary itself with the motion of the other body.
But that solution I find not entirely satisfying. Earlier Garber himself has
favored what he calls the "divine impulse" view of God's action in creating
motion. One selling point of that view is that it makes "better sense of the
argument for the conservation principle": "what God directly conserves is
not the quantity of motion itself, but its cause, the impulsion he introduced
into the world at its creation." A bit later he adds that "the conservation law,"
according to the divine impulse view, "is a direct consequence of God's
commitment to keep pushing, as it were, and to keep pushing as hard as he
did when he set the world into motion" (283). 105 The phrase 'as hard'
seems to imply a relation between some quantity associated with the impulse
and the quantity of motion in the world: less push, less motion. On the same
grounds as before one could argue that this isjust a way of saying that God's
push produces the same quantity at all times. But then it is difficult to see
how one could produce an argument for the conservation principle out of
the divine impulse view.
What one must preserve first of all, I think, is the absolute simplicity and
continuity of the divine volition. The moving force, Descartes said to More,
was of "God himself conserving as much translation in matter, as he put into
it from the first moment of creation." It is, in other words, the execution of
that unchanging volition which can have, from the point of view of our
conception, different effects at different times. Considered with respect to
particular bodies, the execution of that volition will sometimes yield a
change of motion, or from rest to motion. In the Aristotelian world, the
105. I'm not sure, in fact, that I understand Garber's view. Garber says that God directly
conserves impulse. Either that impulse is an active power that God has imparted to matter itself
(as the Aristotelians thought) or it is God's own action, that is, the dependence of the modes of
motus on him as their principle of existence. I doubt, for reasons that will become clear, that it
could be an active power imparted to matter. But then what God "conserves" is his own action
(he "keeps pushing"). The situation is exactly parallel to that regarding the quantity of matter.
It would be an unmotivated precision to deny that God conserves matter "directly" because he
instead conserves-i.e., continues-the act of creation, and likewise to deny that God con
serves motion directly because he conserves the act of creating it.

Authenticated
agent of such a change can be a corporeal substance, or an active quality like

weight. One customary name for its action is indeed impulsus.
Now God can, from our mundane point of view, be said to act on particu
lar things, so long as one remembers that this does not imply any change in
God. (In this respect, the God of the Aristotelians resembles a Leibnizian
individual substance, all of whose properties are essential, yet "unfold" in
succession). Insofar as those actions result in change of speed or direction
or in rest becoming motion, they are analogous to Aristotelian impulsus. But
that is, like the apparent temporality of God's volitions, a fact not about God
but about the termini of his actions. Even the creation of motion is nothing
other than the fact that the instant of creation was for some bodies an
instant in which they were separating from the vicinity oftheir neighbors. It
is only a "shove" in the same improper or broad sense that creation of
substance is a mutatio.
The Cartesian picture thus involves a strange displacement of predicates,
which may account for the difficulty of interpreting it. The quantity as
signed to the moving force of God that "concurs" with bodies to change the
motions of other bodies does not properly belong to the force except
derivatively, by way of its terminus. On the other hand, the force that is
measured by quantity of motion, and whose effects are specified by it, be
longs not to the bodies that determine the quantity, but to God, in whom
there is neither quantity nor variation in quantity.
An illuminating parallel can be drawn between the problem of force and
the problem, mentioned in the discussion of divine immutability, of the
temporality of divine action. The effect of divine action in the world is, as
Fonseca puts it, conceived by us to be "terminated temporally" (n.78). Yet
the principle of that action, a divine volition, is not temporal at all, being
part of the divine essence. The action itself, the causal dependence of the
effect on the volition, or on God's power to execute his volitions, though it
inheres, like any action, in the patient, is a bridge between the eternal and
the temporal. If one asks when the action occurs, there are two plausible
answers: with respect to the creature in which it inheres, it occurs, say,
yesterday morning or tomorrow afternoon; while with respect to its divine
cause, it can hardly be said to occur at all-instead it exists eternally as a
relation between God and the effect that is contained eminently within
him.l 06
The same may be said of force. With respect to the body with which God
concurs in moving or resisting another body, we must conceive God's con
1o6. The same applies to the spatial location of divine action. When Descartes says that the
"impelling force" (presumably of God) "applies itself now to some parts of matter, and now to
others" according to the rules of collision (To More 30 Aug. 1649, AT 5:405), that can only be
with respect to the divine action, not to the principle of that action.

Authenticated
Bodies in Motion
currence at each instant to be terminated in that body's instantaneous size

and speed jointly; with respect to God as its cause, concurrence exists eter
nally, being nothing other than a particular expression of the general voli
tion by which matter and motion were created together. Ontologically that
action, as Gueroult and Gabbey have argued, is two-faced: a pure mode of
the divine essence, and a mode of the objects upon which God acts, a mode
implied in the nature of matter itself only on the supposition that God has
given to various parts of matter motions that cannot jointly persist if God
also conseiVes the quantity of matter itself. 10 7 The second mode can, all the
same, be designated in terms of quantity of motion, which determines how
God, while conseiVing the motion he put in the world, will "singularize" the
degree to which each body in a collision is caused to persist in its state. To
that extent Garber's nominalism is warranted. Ockham argues that the
proposition 'motion is in time' means that 'when something moves it does
not acquire or lose everything at once but part after part' (Ockham Q. in
phys. q2o, opera philos. 448). So too the proposition "the body A of size M
has six degrees of motion' would mean that 'the body A after a collision will
do such and such' depending on the circumstances of the collision.
In short: if by 'force' one means, following the loose usage of Descartes
and his contemporaries, God's action, then it is either the motus itself that is
conseiVed or changed by that action or the relation of that motus to God. But
if by 'force' one means a mode whose intensity would be measured by the
quantity Mv (in a moving body), then there is no such thing. The motus itself
has a speed (and a direction). The body has a volume. That, in a way, is all
given that God redistributes motion according to the third law.
The difficulty in understanding Descartes's conception of the vis ad agen
dum is a reflex of a much more general and profound difficulty in under
standing divine action, especially those actions that are, as the Aristotelians
put it, "owed" to nature. God, though absolutely free in all he does, can be
thought of as determined by the features of the things on which he acts
(specifically, at least, if not singularly). But when the features of the father
determine God to create a soul for the newly conceived child, or when the
quantity of motion in a moving body determines God to set another body in
motion, the father or the moving body are not the efficient cause of the soul
or the setting in motion. God alone is the efficient cause. Yet the features
that determine the specific effect are in things, not in God, if we set aside
the "contract" by which God has bound himself to be determined by those
features.
107. In a letter to More, Descartes holds that the interpenetration of bodies would require
the annihilation of matter (and would therefore not be "interpenetration" at all, but a kind of
melding; To Mure 15 Apr. 1649, AT 5:342). Since God conserves matter, he does not resolve
incompatibilities of motion by allowing bodies to meld.

Authenticated
Aristotelianism, because it imputed active powers even to inanimate

things, did not have to confront the problem in its full generality. Even in
human generation, the semen does not altogether lack power to organize
the material in the womb into a body suitable to receive a human soul. Less
perfect beings can, in some instances, effect their reproduction themselves
(though God must, of course, concur with their action). Descartes, on the
other hand, defines matter as res extensa in partjust to exclude active powers.
Indeed in his physics, if not his psychology, the very notion of concurrence
begins to lost its grip. Concurrence is co-action, not action simpliciter. But if
bodies do not act, there is no concurrence; there is only the outright effect
ing of change in the physical world by God alone.
Descartes in fact sometimes speaks of concurrence and conservation as if
they were the same action-as do, occasionally, the Aristotelians. If God is
the only genuine agent, distributing a fixed quantity of motion among the
parts of matter, then the only difference lies in the things to which motion is
given, not in God's action. Gueroult's thesis that force is nothing other than
existence is so far justified. 108 More precisely: the static force of resisting that
a body at rest exhibits in the situation of the fourth and sixth collision rules,
a force that is additional to its "natural inertia," does look to be a denomina
tion of the divine action ofconservation with respect to its effects. That force
is, moreover, the conservation not merely of the mode of rest, but of the
union of the resting body with its neighbors, and thus of the existence tout
court of some larger whole. The force of persevering that a body in motion
exhibits, on the other hand, cannot simply be "the power that from within
posits each [body] in its duration, and that consequently cannot be
distinguished from [its] existence" (Gueroult 1970: 117). The existence in
question cannot be that of the body; it can only be the existence of the
instantaneous modes of motus: its speed and direction. The action of con
serving them is distinct from that of conserving the body they inhere in (see
the critical remarks at Garber 1992:2g6f). Descartes can therefore preserve
some sort of distinction between conservation and concurrence. But the
distinction has been refashioned to suit a physical world without active
powers. That would not be the first time a coincidence of terms belied a
revolution in concepts.
108. "En realitt\ force, duree, existence, sont une seule et meme chose (le conatus) sous
trois aspects differents [Gueroult cites Principles 1§55-57], et les trois notions s'identifient
dans !'action instantanee par laquelle Ia substance corporelie existe, dure, c' est-a-dire possede
Ia force qui Ia pose dans !'existence ou duree [1§62]" (Gueroult 1970:87). The sections of the
Principles cited, however, argue that duration is distinct only in reason from substance itself;
they say nothing about force.

Authenticated
[g]
Parts of Matter
N ear the end of the Principles, Descartes settles accounts with

rivals to his physics. The only serious rivals are the Aristotelia
nism of the Schools and the newly revived atomism of Gassendi
and others. Descartes compares all three with the "commonest
and most ancient" philosophy of all, to which nearly every philosopher
through the centuries has adhered. No Aristotelian, no atomist, and cer
tainly not Descartes, ever doubted that "bodies move, and have various sizes
and figures, according to whose diversity their motions also vary, and that by
their mutual collision large bodies are divided into smaller bodies, and
change in figure" (PP 4§200, AT 8/1:323). But that common truth is ob
scured in Aristotelianism by otiose complications and unfounded distinc
tions, in atomism by incoherence and the suspicion of heresy. Only in
Cartesian physics does it find a clear and correctly argued expression.
It is signific~nt that Descartes chooses the nature of body as the topic on
which to rest his closing statement for the advantages of Cartesian physics
over its rivals. The early seventeenth century was one of the few times in
which fundamental questions about the nature of body were genuinely
open. Among the rivals of Aristotelianism, there were numerous variants of
atomism, and theories in which matter was endowed with some sort of vital
power or spiritus. There were even theories-proposed by Kepler and
Fludd, among others-in which light, as the generating principle of space,
was the forma corporeitatis. 1 By the end of the seventeenth century the alter
1. On this tradition, see Lindberg 1g86:2gn· (Kepler), Hedwig 1g8o:130, 163, 177, 218
(Witelo). On Grosseteste, see McEvoy 1982:151rr, 173n·; on Patrizi, see n.2o below. Descartes
would undoubtedly have known of Kepler's view; he may also have read Witelo (or Vitellio),
whose works were reprinted in Risner's Opticll! thesaurus ( 1572), a standard collection of optical
treatises.
Brought to you by | Stockholm University Library

Authenticated
Parts of Matter
natives had narrowed to one or another version of corpuscularian mecha

nism; but at the beginning there was, outside the Schools, no predominant
view, and even within the Schools adherence to Aristotle did not preclude
deep disagreement.
Descartes's treatment of body in Le Monde and the Principles, reticent as
ever about sources and external motivations, yields little sense of the con
temporary debate. Nevertheless, when he defines the object of physics to be
"a certain matter extended in length, width, and depth" (PP 2§ 1), a great
deal is packed into the assertion. The opening sections of the second part of
the Principles repudiate both the predominant Aristotelian view and the
views of those novatores who distinguished corporeal from incorporeal
space.
That is not to say that Descartes did not draw on his rivals. Like Aristotle,
Descartes holds that whatever has the dimensions of body is sufficiently like
body to exclude others from its place. Space considered generically as the
receptacle of matter is not independent of the singular spaces or "internal
places" of the bodies contained in it. The conception, which in classical
physics became second nature, of bodies as immersed in an ambient preex
isting space, is foreign not only to Aristotle's but to Descartes's treatment of
body.
Unlike the Aristotelians, however, Descartes holds not only that the na
ture of matter is quantity, but that the quantity in question is actual spatial
extension, and that it is not really distinct from the quantified substance
itself. Extension, in his words, constitutes corporeal substance; from that
Descartes concludes that the only real qualities in it are the modes of
extension. Substantial form-or at least the specific form argued for by the
Aristotelians (§3.2)-active powers, and qualities other than the modes of
extension are chimerical entities whose basis is a misplaced analogy be
tween body and soul.
With some but not all of the novatores, Descartes holds that space, rather
than being an accident of bodies as the Aristotelians had thought, is sub
stance. His response to the novel treatment of space in philosophers like
Telesio, Patrizi, and Campanella is clarified in the correspondence with
Henry More that followed the publication of the Principles. There it becomes
clear that although he shares with them their view that space is substance or
akin to substance-and the difficulties presented by that view for an Aristo
telian ontology of substance-he differs in taking spatial extension to be
sufficient for corporeality. Where the novatores attempt to distinguish space
from body on the grounds that body alone is tangible and impenetrable,
Descartes attempts to derive those properties from extension itself. What
came to be seen as the failure of that attempt hastened, I believe, the demise
of Cartesianism in the latter part of the seventeenth century.

Authenticated
Bodies in Motion
The discussion of body to follow has three parts. 2 In §g. 1 I take up the
argument of Principles 2§4-g, whose vocabulary and topic show that they are
addressed primarily to Aristotelians. Descartes attempts to show that the
nature of matter is quantity, by which he understands actual extension in
space. The heart of his argument is an explanation of rarefaction and
condensation, whose purpose is to exclude from matter any quantity of
mass distinct from quantity of extension. That exclusion, together with the
rejection of intensive quantities, illustrates that in Cartesian physics it is not
enough for a physical property to be representable quantitatively. Physical
properties must satisfy the more stringent criterion of being literally pre
sented, or immediately derived from literally presented properties.
I conclude §g.1 with Descartes's proof that the quantity of corporeal
substance is distinct only in reason from the quantified substance itself.
That thesis, with its nominalist overtones, is the pivot around which the
argument of the second part of the Principles turns from Aristotelian quan
titas to Cartesian extensio, and from opposing Aristotelian positions to oppos
ing the novatores.
In §g.2 I examine the delicate question of the relation between substance
and space. From a survey of representative figures in the late sixteenth and
early seventeenth centuries it becomes apparent that the novatores found it
difficult to encompass substantial space within the Aristotelian ontology of
substance; that they sought to distinguish substantial space from body on
the grounds that body alone is tangible and impenetrable; and that
Descartes had to argue against the Aristotelians that space is indeed sub
stance, while rejecting the novatores 'invention of a noncorporeal substantial
space. Like Patrizi and Gassendi, however, Descartes, having come to believe
that space is substance, found it necessary to revise the Aristotelian notion of
substance to accommodate his new intuition. The ontology of Principles
1§51-56 is an attempt to do so, motivated in part by opposition to other
philosophers' suggestions that substantial space is not corporeal, or even
2. Largely untouched in this chapter are two questions whose consideration would be
essential to a full treatment of Descartes's philosophy of corporeal substance. Except for some
remarks on representation and intensive quantities, I leave epistemological matters aside.
Such matters have been amply discussed by others (in, for example, Marion 1981 and Garber
1992:75-93). I also set aside the distinction of body and mind, in part because I think that
although securing that distinction was one of Descartes's motives in holding that the nature of
body is extension, and although he was always ready to diagnose his opponents' errors by
referring to their failure adequately to make the distinction, much of his argument on the
nature of body is independent of his views about mind. My belief is reinforced by the obseiVa
tion that seventeenth-century critics of Descartes like Cordemoy, Leibniz, and Newton at
tempted to refute Descartes's opinion on physical grounds. The now traditional interpretation
of modern philosophy as the history oftheories of knowledge has obscured the extent to which
Descartes and his near successors do not subordinate ontological and physical questions to
epistemological ones. This chapter is in part a corrective to the excessive "Kantification" of
seventeenth-century philosophy.

Authenticated
Parts of Matter
that it is somehow spiritual; here too one finds the best argument for identi
fying corporeal substance and quantity.
The arguments by which Descartes shows that an individual body is just a
determinate region of space say nothing about the criterion by which one
such body is distinguished from another. Descartes notoriously attempts to
define 'individual body' in terms of motion-or, more precisely, rupture
(§8.1). Although objections to the definition can be answered, the suspi
cion behind them is not unfounded. Descartes's conception is perched
precariously between two views of extension, the first ofwhich takes it to be a
"form of body" singular in each individual, the second a continuous under
lying stuff of which individuals are parts.
In §g.3 I consider Descartes's efforts to show that spatial extension is
sufficient to constitute body. In the Principles and the letters, three objections
emerge: the independence of impenetrability from extension; the existence
of vacuum; and the existence of so-called imaginary spaces. Descartes's
answers, especially in his correspondence with More, not only enlarged
upon some of the arguments sketched in the Principles, but also reveal some
of the theological underpinnings of the debate about separate space.
g.1. Extensive Quantity and the Nature of Matter

1. Quantity and extension. Descartes uses the term extensio rather than
quantitas to denote the essential property of corporeal substance. The
choice is significant. In Cartesian physics, Aristotelian quantitas is stripped
down, leaving only the extensio representable by geometric objects. From the
Aristotelian standpoint this amounts to a twofold reduction (see §4.2). The
distinction between potential and actual extension disappears; and inten
sive quantities-degrees of heat and the like-are banished. What remains
is a concept even more austere in certain respects than that of present-day
science.
Already in the Regulce Descartes writes that we should "diligently abstain
from the term 'quantity', because some philosophers have been so subtle as
to distinguish it also from extension," as well as from the quantified thing
itself (Reg. 14, AT 10:44 7). 3 In the Principles, he notes again that some
philosophers are so "subtle that they distinguish the substance of body from
its quantity and quantity itself from extension" (2§5, 8/1:42). Suarez was
one such philosopher. Against the nominalists he argued not only that
quantity is really distinct from that which has quantity, but that a material
3· See Marion's note to AT 10:447 (Reg. [Fr.] 26g); cf. Monde6, AT 11:35. Marion quotes
from the Suarez and Toletus passages discussed below; the Toletus passage can also be found in
Gilson 1912:257, no.399·

Authenticated
Bodies in Motion
substance can have quantity without being extended (see §4.2). It is not
essential to a body having quantity that it actually be extended, but only that
its parts should be "ordered among themselves so that if they are not super
naturally impeded, they must also have extension in place" (Disp. 40§4! 14,
opera 26:547). Contrary to what Ockham and others thought, to have exten
sion is not merely to have nonidentical parts. 4 It is to occupy a place and to
resist the entry of other extended things into that place.
Descartes will have none of this. He recognizes only actually extended
quantity. Determinate quantity can be potential only in the mathematical
sense mentioned earlier (see§8, n.48): in a square there are "potential"
triangles that would come to be if the square were divided. That is clearly
not what the Aristotelians have in mind. For them an extended thing always
would occupy a place and keep others out of it, but it may be inhibited from
doing so by God. Descartes, on the other hand, takes it to be obvious that
even God cannot bring it about that two extended things should occupy the
same place. Only when More calls that obvious truth into question does he
attempt to prove that impenetrability can be derived from extension.
Intensive quantity too is ruled out in the Reg;ulce. Of the varieties of quan
tity, the most easily and distinctly imagined is "the real extension of a body
abstracted from all else save that it is figured" (Reg. 14, AT 10:440-441).
The imagination or phantasia is, after all, itself a "genuine real body, ex
tended and figured," to which the mind applies itself when it imagines.
Intensive quantities, because they cannot be directly represented, will not do:
"for although one thing can be called more or less white than another, or
one sound more or less acute in pitch, and so for the rest, still we cannot
exactly determine whether such an excess consists in a duple or triple
proportion, etc., except by a certain analogy to the extension of figured
bodies" (ib. 441). Though Descartes later abandoned the line of argument
presented here, his conclusion became, if anything, stronger: sensible
qualities, represented as they are by obscure and confused ideas, do not
admit even of being classified into positive qualities and privations (Med. 3,
AT 7:44). 5 We have in fact no reason to believe that sensible qualities exist in
bodies. At best we may conclude only that something, we know not what, is
their cause (PP 1§70, AT 8/1 :34). But insofar as they exist in the mind, they
4· On the partes extra partes argument in Ockham, see §4.2. The phrase partes extra partes
occurs in a 1649 letter of Descartes to More: see n.5 below.
5· One indication that the earlier view persists is found in a letter to More in 1649.
Descartes writes that the predicate 'extended', though it can be used loosely of qualities like
weight or to spiritual substances, applies strictly only to actually extended things: "only that
which is imaginable as having parts outside of parts which are of a determinate magnitude and
figure I call extended, although other things are by analogy also said to be extended" (To More
5 Feb. 1649, AT 5:270). Significant here is the invoking of imagination. Descartes takes
imagination to be the "application" of the soul to certain parts of the brain, which are extended
things and thus literal representations of other extended things.

Authenticated
Parts of Matter
cannot have quantity: quantity implies divisibility, but the soul is not divis
ible. The analogy noted in the Regult1! is set aside; it was left to others to
begin learning how to measure and reason about intensive quantities. 6
There are exceptions. Although instantaneous change of place, or rup
ture, can be regarded as modes of extension, velocity cannot (see §8.1). It
can indeed be represented by line segments-as could any intensive
quantity-but it is itself not such a segment. Quantity of motion, the prod
uct of velocity and volume, is likewise not itself an extensive quantity. I have
already considered the question in §8.1, but one further point comes out
here. If average velocity is represented in terms of the proportions of mo
tions, or more precisely of distances traveled in equal times, then it is a ratio
of line segments (of the distances, not of distance to time), and so im
mediately derived from admissible quantities. The same cannot be said
of heat or weight, since their relation to those quantities is much more
indirect.
Descartes's conception is austere by the standards of both Aristotelian
and present-day physics. Its austerity does not consist merely in a restriction
to what can be quantified or represented geometrically: degrees ofweight
or temperature are eminently quantifiable and representable. All Descartes
will admit by way of quantities is line segments themselves, and quantities
immediately derived from them. His austere conception is not only main
tained in the ontology of material substance; it delimits to some degree the
admissible experimental determinations of quantity. Unlike Mersenne,
Descartes does not gather measurements ofweight, temperature, or specific
density, or include them (except qualitatively) among the phenomena. In
the Dioptrics, the only work that does offer a numerically precise explanation
of obseiVed quantities, the quantities in question are the angles oflight rays.
The neglect of measurement suggests that in Cartesian physics the repre
sentation of bodies must be notjust mathematical, notjust geometrical, but
peculiarly literal. If bodies are nothing other than res extenst1!, and their
properties modes of extension, the only quantities that occur in them ex
actly as they are represented to us in sense or imagination are figure, size,
and-derivatively-motion. A literal representation of heat could only be a
representation of the configuration of the particles of fire and of their
motion. Once we are apprised of the true nature of heat, it is idle to interject
an analogical representation of its degrees. To do so is to revert to a repre
6. Descartes takes no notice, for example, of the various "thermoscopes" or thermometers,

some using air, others water, with which Beeckman, Mersenne, and others measured degrees
of heat (see Rey aMersenne 1Jan 1632, Corr. 3:244· 248, ComieraMersenne 16Jan 1626, 1:33lll,
Beeckman a Mersenne Jun. 162g, 2:232). Since the accuracy of air thermometers was com
promised by variations in barometric pressure, Descartes had some reason to neglect them;
even though Rey had by 1632 invented a water thermometer that avoided the problem, liquid
thermometers came to be widely used only after 1654 ( Corr. 3:248).

Authenticated
Bodies in Motion
sentation that, though mathematical in form, is subject to the pitfalls of the

sensory representation that Cartesian physics seeks to replace.
2. Quantity and matter. It was generally agreed that matter cannot natu
rally exist without quantity. Whether quantity is part of the nature of matter
was, on the other hand, disputed. Thomists deny that it is; others are not so
sure (§4.2). Everyone at least acknowledges the existence of dissenters,
notably Averroes, according to whom prime matter is essentially endowed
with indeterminate quantity, quantitas interminata. One sixteenth-century
proponent of the view was Zabarella. In De prima rerum materia, Zabarella,
having disposed of the Thomist pura potentia and the Scotist forma cor
poreitatis, turns to his own view. "Prime matter," he writes, citing Averroes,
"according to itself, and prior to the reception ofform, has quantity, and has
three dimensions, length, breath, and depth, as an internal and inseparable
accident, although [its dimensions] are not circumscribed by any limits
[terminis], but are indeterminate [interminatas], and afterwards receive
various limits from diverse forms, accordingly as the various natures of
natural bodies require them" (Zabarella De prima rerum materia 2c6, De rebus
nat. 191D; cf. Averroes In Phys. 1com63). Qy,antitas interminata, like matter
itself, can neither be generated nor corrupted. Form cannot be its cause,
because form in itself is not divisible; nor can quality, for similar reasons;
nor, finally, can quantity itself be its own cause, and so only matter remains
(Zabarella ib. c7, 19¢). Zabarella adds that quantity can be "abstracted by
the mind from every natural form" and is therefore prior to form in thought
as well as in things.
With the important qualification that quantity can be potentially but not
actually extended, Suarez holds a similar view. In an argument showing that
the rational soul has neither matter nor material cause, he holds that "mat
ter and quantity are inseparable and reciprocal," and that no other property
of matter is. Like matter itself, quantity is "most apt for receiving and being
acted upon, and of itself is ordered to no particular action" (Suarez Disp.
13§ 11: 15, opera 25:460). Later he cites with approval the same passages
from Averroes that Zabarella cites; and, as we have seen (§5.3), he argues
not only that quantity precedes form in the composition of substance, but
that certain other accidents that dispose matter to receive form inhere in
matter alone by way of quantity (Disp. 14§31 1 off). Quantified matter is in
Suarez's metaphysics close to being an independent substance in which
sensible qualities inhere immediately rather than by way of form.
The argument of Principles 2§4-23 begins, as I have noted, by taking on
the Aristotelians and then shifts to the novatores. In 2§4 Descartes starts with
the Aristotelian terms materia and quantitas, only later switching to his own
terms corpus and extensio. The nature of matter, he holds, is quantity, a
quantity that will turn out to be nothing other than spatial extension: "we

Authenticated
Parts of Matter
perceive that the nature of matter, or of body universally regarded [corporis

in universum spectati], does not consist in its being a hard or weighty or
colored thing, or in any other mode affecting the senses: but only in its
being a thing extended in length, width, and depth" (AT 8/1:42). The
implied interchangeability of materia and corpus is worth noting. Descartes
uses materia, corpus, and substantia (qualified as corporea or extensa) to desig
nate the thing whose existence and properties are being considered in the
opening sections of Principles 2. The shift from one term to another was
sanctioned by usage. Goclenius, for example, writes that corpus has two
acceptations, one in the category of substance, the other in that of quantity.
Of the first he writes: 'That corpus which is Substance, is the Subject of
threefold dimensions, length, breadth, and depth; it is nothing other than
either corporeal Substance, and then it is a genus, or the Material part of
corporated substance. "7 Corpus as substance is material substance con
sidered under the aspect of the spatial quantity proper to it. Corpus as
quantity is simply the three dimensions of corporeal substance. Toletus
occasionally uses the term corpus mathematicum to denote corpus as quantity,
contrasting it with corpus naturale or corpus physicum. The first is quantity
terminated by figure, considered apart from matter, sensible accidents, and
natural change; the second is the substance that not only has quantity, but
all the other accidents as well. 8
Corpus, then, does not in the passage at hand denote an individual sub
stance, as it sometimes does elsewhere (see 2§25, where corpus alternates
with pars materice). It denotes 'body regarded universally' or what is common
to all individual bodies qua bodies, and alone persists through every possible
natural change~in short, what the Aristotelian calls matter. But it denotes
matter under the specific aspect of having the three spatial dimensions.
Since for Descartes that aspect constitutes matter, it is not surprising that in
his writings corpus took over from materia the function of denoting the
substrate of change.
The circulation in Descartes's texts of the terms materia, corpus, and sub
stantia, though it did not offend against usage, and was unlikely to have
confused his readers, does procure a certain rhetorical advantage. In Aristo
telian physics, the natures of corporeal substances are jointly determined by
7· 'Corpus, quod est Substantia, est Subiectum triplicis dimensionis, longitudinis, lati
tudinis, & profunditatis. Estque nihil aliud, quam vel Substantia corporea, ac tunc Genus est:
vel Materialis pars substantia:: corporata::" (Goclenius Lexicon, s.v. "Corpus," 481).
8. 'Body is twofold. One is 'natural', the other 'mathematical'. Mathematical body [cmpus
mathematicum] is quantity itself, having the three dimensions of length, breadth, and depth,
insofar as it is considered per se, stripped of all other sensible accidents, but only terminated by
some figure; [mathematical] body, they say, is of the category of quantity. [ ... ] Physical body
[physicum corpus] is the substance itself that has such dimensions, but with sensible qualities,
and subject to changes; and this body is in things, and is a substance" (Toletus In Phys. 4c2q 1,
opera 4:106rb-1o7va).

Authenticated
Bodies in Motion
matter and form (see §7.2). Individual and specific differences can be re
ferred to either. Zabarella agreed that the nature of matter is quantity; but
he did not deny that the natures of individual substances include sensible
qualities and active powers. Descartes, however, means to show not only that
the nature of body in general is extension, but that all corporeal accidents
can be reduced to the divisibility and mobility of extension. Corpus physicum
has no property not possessed by corpus mathematicum. For that it does not
suffice to show that the nature of matter is extension, or even that body and
space are one. In §g.2 I will examine whether the arguments of the first part
of the Principles support the conclusion; here it is enough to notice the gap
in the argument of the second part, to which I now return.
Taking hardness to be exemplary of the properties that others have mis
takenly imputed to matter, Descartes writes: "Nothing else about it is indi
cated to our senses than that the parts of hard bodies resist the motion of
our hands when they run into them. For if whenever our hands moved in
some direction all bodies existing there receded with the same speed as our
hands approached them, we would never feel hardness" (AT 8/1:42). Yet
mere motion cannot change the nature of matter; so it "does not consist in
hardness."
The argument misses its target. Those who argued that bodies are impen
etrable certainly did not take impenetrability to consist merely in the fact
that bodies cause us to have sensations of hardness. Being impenetrable
gives bodies the power to resist our touch, a power they cannot lose merely
because God never allows them to manifest it. Just as the bowels of the earth,
"though they have never been exposed to the sun, nor has anyone des
cended to them with lamp or torch," are still visible in themselves (More To
Descartes 5 Mar. 1649; AT 5:300), so too bodies never touched remain
tangible.
Descartes takes what is clear and distinct in sensations of hardness to be
the fact of resistance. To infer from that fact a power of resistance is a step
toward mystery, one that Descartes refuses to take. But then he has no
explanation of the fact. Only when More compels him to admit that re
sistance is a genuine property of bodies, one that could exist even if the
sensations characteristically caused by it did not, does Descartes bestir him
self to show that it indeed follows from being extended. I will return to this
question in §g.3.
3· There is but one quantity in matter. In 2§5 Descartes finds "two causes
[...] why it might be doubted that the true nature of body consists in
extension alone" (AT 8/1:42). One is that some people explain rarefaction
by supposing that one and the same body can, without having any matter
added to or taken away from it, have more or less extension. They infer that
matter has not only a quantity of extension but another, independent,

Authenticated
Parts of Matter
quantity as well-quantity of mass. The other is that some people think that
certain regions of space, which they call "empty," contain no bodies.
Descartes therefore divides the remainder of his task into two parts. He
must show that matter (or "corporeal substance") and quantity are distinct
in reason only-and thus that spatial extension is the only quantity in
matter; and that extension, internal place, and space differ in reason only.
The first is carried out in 2§4-9, the second in 2§10-19 (see §9.2-9.3).
For the Aristotelians, rarefaction and condensation were changes in the
intensive quantity associated with rarity and density. The standard case is the
change of one element into another-the same quantity of water, when
transmuted by heat into air, is said to increase tenfold in volume. Following
Aristotelian usage (§4.2), I distinguish quantity ofmass, which is preserved in
such transformations, from quantity of extension, which changes (we would
now speak of mass and volume). It was controversial whether rarefaction
and condensation are distinct from augmentation and diminution, and
quantity of mass from quantity of extension. Toletus concludes that in
rarefaction and condensation there is always change in quantity of exten
sion (In Phys. 4c9q11, opera 4:132vb). Thomists, on the other hand, held
that "rarity and density is something pertaining to the parts of a body as
ordered among themselves [in ordine ad se], and not as ordered in place"
(John of St. Thomas Nat. phil. 3q7a1, Cursus 2:709). One of their arguments
is that if quantity of extension were the sole quantity in matter, in rarefac
tion new matter would have to be generated. But experience tells us that the
matter of the generated element is derived entirely from that of the cor
rupted ·element. Transmutation conserves quantity of mass while augment
ing or diminishing quantity of extension. Hence the two quantities are
distinct.
Descartes's response is to show how the appearance of rarefaction or
condensation can be saved without admitting any quantity other than quan
tity of extension, and while maintaining the conservation of matter.9 His
explanation was not new. 10 What we ordinarily call a body is typically a
composite, like a sponge soaked in water, of a porous body with other
g. The discussion that follows is indebted to Miles 1983. Atomists explained rarefaction by
supposing that atoms moved apart; other philosophers thought that there could be "interstitial
vacua" within bodies. In 2§6 the target is Aristotelian explanations; in 2§19 Descartes applies
his proof of the impossibility of the vacuum to refute the others.
10. John of St. Thomas ascribes similar views to Scotus, Marsili us oflnghen, Herv<eus and
others ( Cursus 2:7o8; his source is Rubio). Descartes himself had affirmed a similar explana
tion as early as 1629 (ToMersenneS Oct. 162g, AT 1:25), his immediate source being Sebastian
Basso (Philos. nat. 332r; cf. Mersenne Corr. 2:302, 307~"). Beeckman briefly entertained the
hypothesis of spongelike particles in 1620 before shifting to a full-blown atomist explanation
(foumal2:157• 230; cf. Mersenne Carr. 2:297). By 1627 he had returned to an explanation
according to which subtle matter entering the pores of gross matter caused it to expand
(foumal3:127).

Authenticated
Bodies in Motion
bodies-assumed to be fluid-filling the pores. The porous body becomes

denser when "the inteiVals between its parts, as those parts approach one
another, diminish or even disappear altogether" (2§6, 8/1:43). Its outward
dimension decreases, thereby giving the appearance of a decrease in quan
tity of mass. But what really happens is a volume-preseiVing transformation
of the body into one having smaller outward dimension. That transforma
tion will, of course, also conseiVe mass. Anyone, then, who "attends to his
thoughts, and will admit only what he clearly perceives," will find that to
explain rarefaction and condensation he needs only to imagine "a mutation
of figure" and not the increase or decrease of some quantity distinct from
extension.
That much would justify Descartes in holding that the Aristotelian ac
count is not forced upon us. But he was not satisfied with so modest a
conclusion. Instead he holds that "we perceive that rarefaction can quite
easily occur in this way, but not in any other" (PP 2§7, AT 8/1 :44). It is, in
fact, "plainly repugnant that something should be augmented by a new
quantity or new extension, unless at the same time new extended substance,
that is, new body comes to it." Ifwe attend to our thoughts, we will see that it
is not "as agreeable to reason to concoct something which is not intelligible
[...] as to conclude, from the fact that [bodies] are rarefied, that there are
pores in them," pores filled with other bodies that we do not perceive.
The unintelligible something is either the quality of density or a quantity
of mass distinct from extension. Physics has done without the quality. But
quantity of mass has proved to be essential. Recent authors have harped on
Descartes's "obvious difficulties in explaining the relative density of
different bodies. "11 He himself asserts what is taken to be the problem: ''We
easily understand that it cannot be [...] that there should be more matter
or corporeal substance in a vessel when it is full of lead or gold or some
other heavy and hard body than when it contains only air [... ] : since the
quantity of parts of matter does not depend on their heaviness or hardness,
but on their extension alone, which in the same vessel is always equal
(Descartes PP 2§ 19, AT 8/1:51). In short, "the quantity of matter in equal
volumes is always equal" (Clarke 1989:77), and the density of every body will
be 1.
But Descartes knew that.l2 Why didn't he see the problem? The answer is
that 'mass per unit volume', or density as it is now defined, is not a Cartesian
notion. The phenomena that the Aristotelians attempted to explain by
reference to rarity and density were of three sorts: the evident change of
11. Clarke 1989:76; cf. Funkenstein 1986:74·

12. 'Tous les Cors estant de mesme matiere, deux parties de cette matiere, de mesme
grosseur & figure, ne peuuent estre plus pesantes l'une que !'autre" (To Mersenne 30Jul. 1640,
AT 3:135).

Authenticated
Parts of Matter
outward dimension which accompanies changes of phase, as in melting or

boiling; the more subtle change that accompanies warming and cooling
without change of phase; the fact that equal volumes of different substances
have different weights. For the first and second, the explanandum is the
change of outer dimension that is one effect of certain physical processes.
That is what Descartes wants to explain in 2§5-6. For the third, one would
have to refer to his theory of weight. The force by which a body is pushed
toward the center of the earth is not easily calculated. For bodies of fixed
outward dimension, it varies directly, though not linearly, with the amount
of matter they contain (see 4§2otr, esp. 24-25; AT 8/ 1:212ff). What is clear
is that the Newtonian equivalence of inertial and gravitational mass is absent
from Descartes's physics. With it goes the direct comparison of masses by
their weights, and the computation of density by weight and volume.
The explanation of rarefaction and condensation shows that physics can
do without the qualities of rarity and density. More significantly, it shows that
the only quantity one needs to associate with bodies is extension. Quantity
of mass is a perfectly respectable physical dimension. Descartes's refusal to
admit it, like his proscription of intensive quantities, must be put down to
his insistence that all continuous quantities considered in physics should
not merely be representable by determinate extension, but should them
selves be determinate spatial extensions.
4- Qyantity and substance. Having disposed of the argument from rarefac
tion and condensation, Descartes applies the same reasoning to show that
substance and quantity differ only in reason: "It cannot happen that even
the smallest amount should be removed from this quantity or extension,
unless the same amount of substance is taken away; nor conversely, that the
slightest bit of substance should be taken away without the same amount of
quantity being removed" (PP 2§8, AT 8/1 :45). This, together with a com
parison of quantity with number, which likewise differs only in reason from
the things enumerated, is Descartes's entire argument.
It is, unfortunately, irrelevant. In the Aristotelian conception, quantity is
an accident of corporeal substance. When a substance is deprived of quan
tity, as in the Eucharist, that quantity is not deducted; it is removed. What
remains is neither bigger nor smaller than it was. Having no quantity, it
admits no comparison, any more than souls do.
There are, I think, two reasons why Descartes did not offer better argu
ments at this point in the Principles. The first is that the strongest reasons for
a real distinction between substance and quantity are drawn from the Eu
charist (§4.2). To answer them, Descartes would have had to develop his
own explanation of transubstantiation, which would have been both impru
dent and out of place in a physics textbook. The second reason is that the
identification of substance and quantity is an instance of the general truth,

Authenticated
Bodies in Motion
asserted in 1§62, that no substance is really distinct from its principal at
tribute. I will examine that claim in §g.2.
However inadequately argued, the thesis of 2§8 is pivotal in the argument
of the second part of the Principles. In 2§4 the subject is Aristotelian matter;
in 2§11 it would, to judge from the examples used there, have shifted to
corporeal substance-in Aristotelian physics the composite of matter and
form. But to show that the nature of matter is to be extended (2§4) and to
show that quantity and substance are distinct only in reason is not quite to
show that the nature of corporeal substance is to be extended. One might
well agree with the nominalist thesis of 2§8, and even the conclusion of 2§4,
while denying that the nature-or rather the natures-of corporeal sub
stances consisted merely in their being extended. The Aristotelian believes
that the natures of corporeal substances are their forms more than their
matter; but here the possibility of a contribution by something other than
matter has been silently elided.
g.2. Substance and Space

If there is a characteristically Cartesian thesis on body, it is not that the
nature of matter is extension or that substance and quantity are one. It is
that space and body differ only in reason. That thesis sets Descartes apart
both from the Aristotelians, for whom there is a real distinction between
matter and quantity, and from the atomists, some of whom agreed that all
accidents of body are modes of extension, but denied that corporeal sub
stance and space differ only in reason. Descartes alone holds that space and
corporeal substance are one.
In this section I first survey the scene in which that claim was put forward,
a scene of dissension and instability. Aristotelians stuck for the most part
with Aristotle's rejection of a "separate" space independent of body. Atom
ists believed as a matter of course in separate space, but were at some pains
to explain what sort of entity it is. Other philosophers argued that space is a
substance, but unlike Descartes they held that it is distinct from body. They
too found it difficult to locate substantial space within Aristotelian ontology,
and were therefore inclined to tamper with the category of substance and to
contest the analysis of corporeal substance into matter and form.
I then examine Descartes's central argument for identifying corporeal
substance with spatial extension, one version of which is found in Principles
2§11. That argument is directed not only against the Aristotelians, but
against the novatores, many of whom held that there are two kinds of exten
sion, the corporeal extension of bodies and the incorporeal extension of
space. Here Descartes could justifiably regard himself as the true inheritor

Authenticated
Parts of Matter
of Aristotle: Aristotle too had argued that whatever has spatial "dimensions"
will exhibit many of the properties of body, including the crucial property of
impenetrability.
But to save Aristotle on that point was to forsake him at another. The
ontology of substance, attribute, and mode proposed in the first part of
Principles departs significantly from that of the Categories and from that of
material substance in the Physics. Among the motivations for the new ontol
ogy was that of explaining how body, conceived as extension, could nev
ertheless be regarded as a substance. I will examine how Descartes's em
phasis on existence in his definition of substance, and the notion of
constitution, which supplies the ground upon which Descartes rests his exclu
sion of sensible qualities from body.
I conclude with a look at the individuation of bodies. Extension, though it
fulfills some of the metaphysical functions of substantial form, cannot, it
would seem, supply a principle by which to distinguish one body from
another, as form was said to do for complete substances in Aristotelianism.
Descartes's definition of 'individual body' or 'part of matter' has met with
severe criticism, and indeed there are profound ambiguities in his treat
ment of space, which seiVes both as the materia common to all bodies and as
the theater within which they move and act on one another. But I think that
the definition can be vindicated against some of the criticisms made of it,
and that in doing so one may see that, contrary to what is often said, physical
space is for Descartes not simply the space of geometry made actual.
1. Background. Two arguments against the existence of a separate space
in the fourth book of Aristotle's Physics provided the starting point for later
discussions.l 3 Aristotle argues that if there were a place and a body in that
place, there would be two bodies in one place, which is impossible. The
Coimbrans paraphrase the argument thus: "If there were a place of some
sort, it would very much seem to be body, if in fact it had the threefold
dimensions, length, width, and depth, by which the nature of body is
defined: but it cannot be a body, since when that place, and that which is
held in place, were together, it would follow that two bodies existed at once
and permeated one another" (Coimbra In Phys. 40, explanatio, 2:6). The
presupposition of the argument is that whatever has the three dimensions
of bodies is itself body. The argument could be turned aside either by
denying the presupposition, as Patrizi, Gassendi, and More do, or by deny
ing that the place of a body is distinct from the body itself, as Descartes
does.
The second argument occurs a bit later. Aristotle has brought forth an
array of reasons to show that motion in a void would be incommensurable
13. On these arguments, see Grant 1981, c.1-2; on Aristotle, see Sorabji 1g88:761r.

Authenticated
Bodies in Motion
with motion in a resisting medium; the ultimate conclusion, of course, is

that there can be no void. He then shifts ground: "But even if we consider
what is called 'the void' by itself, it will appear to be truly void"-an empty
name (4c8, 216a16ff; Coimbra In Phys. 2:55):
If someone (he says) throws a cube into water, the water will immediately
cede the cup to it according to the size of the cube; the same holds for the
air, even if it escapes our view. But this, which is conceded by all, cannot
occur in empty space, if indeed only bodies cede their place. Yet those
who affirm the vacuum suppose that there is an inteiVal extended in three
dimensions, from which sensible bodies have been removed; they are
[thus] compelled to say that there is a penetration of dimensions, which is
no less absurd than if the cube were penetrated by water. Nor can they
escape [this consequence] if they answer that the dimensions of the vac
uum are not affected with sensible qualities. For the reason why several
bodies cannot be in the same place at once is not the admixture of
sensible qualities, but their threefold dimension. For if the cube also is
stripped of all accidents, it will resist the penetration of water no less than
it does now. (ib. explanatio, 2:56)
Anything, body or not, sensible or not, that has "dimensions" is subject to

the principle that two such things cannot occupy one place simultaneously.
The supposed void is either sufficiently like body to obey that principle, or
else it is nothing at all. Two keys to Descartes's argument are found here: the
thought experiment of stripping away accidents, and the claim that impen
etrability depends only on having dimensions. In adapting Aristotle's argu
ments, Descartes in effect defends the genuine Aristotle not only against the
novatores but also against the Aristotelians themselves.
Dissent from Aristotle's refusal to admit separate space came not only
from anti-Aristotelians, but from within Aristotelianism. Among Aristotelian
dissenters, I take Fonseca and Suarez as representatives, among anti
Aristotelians, Patrizi and Gassendi. My aim is a sketch of the intellectual
landscape within which Descartes's solution should be located.l 4
In a question on whether locus is a genuine quantity, we find Fonseca
grappling with Philoponus's reply to the arguments of Aristotle just men
tioned. Philoponus held (so Fonseca says) that "there exists a space of three
dimensions, fixed and immobile, in which together and, so to speak, pen
etratively every body whatever is contained, and which is called 'space' and
'plenum' insofar as body is contained in it, and 'vacuum' if every body has
14. Grant 1981 is the indispensable starting point for any such history. His discussion makes
it clear that virtually every position one could imagine remained a live option in the early
seventeenth century.

Authenticated
Parts of Matter [3571
left it" (Fonseca In meta. 5e13q7§I, 2:7ooE). 15 Mter rehearsing the argu
ments of Aristotle, Fonseca presents Philoponus's counterargument: "Phi
loponus, the defender of the being of space [ entitatis spatii defensor] answers
that three-dimensional quantity is of two sorts: one is material and requires
space and cannot exist together with another material quantity; the other
[is] immaterial, and is space [itself], which because it is immaterial allows
an adequate body [to coexist] with it" (ib. 702C). Space is not a real and
genuine quantity, but rather what Fonseca calls imaginary. It is imaginary not
because it depends on our imagination but because "space, which after its
fashion truly is, always was, and will be, is not a genuine but a fictive [ficta]
quantity" ( 703C). It is the capacity or aptitude for receiving each body into a
space of the right volume, a non repugnantia ad ea capienda. That capacity is
neither in any body nor in any place; Fonseca calls it a. "pure negation."
The negation is this: it is not the case that God could not have created
quantified matter, or more than he did create, or infinitely much. What
makes that true is not an accident or mode of matter. To use Fonseca's
earlier example: the truth of the proposition 'a human is not a stone' does
not rest on our having the "negative property" of not-being-a-stone, but on
our having certain positive properties-being animate, say (see scsq l'
2:323f). Likewise the "existence" of space, or the capacity to receive body, is
merely the reverse of the coin whose obverse is God's power to create
quantified matter. 1 6 Space so understood is not only infinite, but eternal. Its
eternity does not contradict the doctrine of creation because the capacity to
receive body is not a capacity belonging to any thing: it presupposes nothing
apart from God, and is therefore no exception to the proposition that all
things, save God, are created.l7
Patrizi, unlike Fonseca, unequivocally accepts the distinction between
material and immaterial space. According to his doctrine, which deserves to
15. On Philoponus, see Sorabji 1988, c.2 and pp.200-201; on his fortuna in the sixteenth
century, see Schmitt 1987.
16. The crucial point, according to Grant, is that Fonseca takes even pure negations to
exist. The intended contrast is with philosophers (Grant mentions Gabriel Vasquez) who held
that the void, being the mere negation of body, is therefore nothing at all (Grant 1981:157r). I
am not sure that the difference between Fonseca's view and Vasquez's is as "radical" as Grant
takes it to be. Fonseca does use the word 'existere' in his treatment of pure negation. But he
explicitly defines their "existence" to be the nonexistence of what would render them false:
"This pure negation-that man is not stone-is said to exist independently of the operation of
the intellect for the sole reason that there exists no man who is a stone" (In meta. 5c5q1§4,
2:323C). His intention is to deny existential import to certain "necessary connections" so that
their being necessarily and eternally true does not entail the necessary and eternal existence of
anything except God.
17. Goclenius sums up the point thus: 'imaginary space' is "sometimes taken also for that
which is not absolutely nothing, as for the capacity of the thing there is given this space [ ... ]
Fonseca holds that the connections of things are not genuine things per se, and not inventions,
but only identities of things, purely negative [ ... ] and so it is no wonder that they should
neither be created nor uncreated" (Goclenius Lexicon s.v. 'Spatium', 1068).

Authenticated
Bodies in Motion
be better known among Descartes scholars, 18 everything is in space, and for

that reason space precedes all else, save God. Locus, or space insofar as it is
occupied by body, is ·~ust as prior to body as body to corporeal qualities. For
that without which nothing else can exist, while it can exist without anything
else, is necessarily prior. "19 It is prior even to the world itself, "brought forth
by the First One before all other things, as if breathed out by the breath of
his mouth" (227[61c]).
Space is neither a body nor a property of bodies, but it is not non-ens. Like
body, it has three dimensions, but unlike body it has no figure, nor does it
offer resistance to movement (227-228[61c-6n]). Thus Patrizi answers
Aristotle's second argument: there can be mutual penetration of dimen
sions when at least one of the dimensions does not belong to a body
(230[62c]).
Space "is not embraced by any of the categories, and is prior to and
outside them all." It is a substance in the sense of id quod per se substat [that
which subsists by itself], of quce aliis substat [that which underlies others],
and of quce nulla aliarum rerum eget ad esse [that which needs no other in
order to be] (241 [6sbc]). Here Patrizi's position is much stronger than
Fonseca's: space has a positive, even substantial, existence. But it is not a
complete corporeal substance, since it is not composed of matter and form.
It is a "mean" between the corporeal and the incorporeal, "an incorporeal
body and a corporeal non-body," because it is not sensible nor resistant, and
yet unlike spirits it has dimensions (241[6sc]).20
On the face of it, little of this would be congenial to Descartes. But
Descartes too gives up matter and form; he too defines 'substance' as that
which needs no other to exist (PP 1§51, AT 8/1:24). What separates him
from Patrizi is his insistence that space and body coincide, or rather that every
18. Garber 1992 is one exception (see p.128). The work I draw on is Patrizi's De spacio
physico, first published in 1587 and later incorporated into the Nova de universis philosophia.
Not only was the latter placed on the Index in 1592, but the animus of the Church against
Patrizi's Platonism was strong enough that the Holy Office recommended after his death that
the chair established at Rome for the teaching of Platonism be suppressed. See Brickman's
introduction to De spacio physico, Henry 1979, Grant 1981:1991\ Copenhaver and Schmitt
1992.
Patrizi is mentioned unfavorably in two ofMersenne's early works, the Qucestiones celeberrimce
in Genesim (col.738-742) and the Vbitedes sciences (1:109), once in Beeckman's]oumalin 1633
(3:289), and at some length by Gassendi (Syntagma 2, Opera 1:246). Though his name does not
occur in Descartes's correspondence, it is unlikely that Descartes would not have known of
him. He does mention the other Italian philosophers with whom Patrizi is commonly associ
ated (see Descartes To Beeckman 17 Oct. 1630, AT 1:158, where Telesio, Campanella, and
Bruno are included among the novatores; on the association of Patrizi with these figures, see
Henry 1979:550 and the references cited there).
19. Patrizi De spacio phys. 231 [62c], 239[64d]; Patrizi is quoting-to his own purposes
Aristotle Phys. 4c1, 2o8b34 11 ; cf. Coimbra In Phys. 2:5.
20. In another part of the Pancosmia, the De p1imcevo lumine, Patrizi argues that space is filled
first with light, which like space is between the corporeal and the incorporeal. Cf. Henry
1979:556; Grant 1981:203 and n.142; Mersenne Corr. 1:334

Authenticated
Parts of Matter
delimited region of space coincides with the three-dimensional place ofone

or more bodies. Impenetrability and having effects on the senses are, for
Descartes, necessary consequences of "having dimensions." That is no trivial
difference. But neither that nor Descartes's repugnance toward any sugges
tion ofa "mean" between the corporeal and the incorporeal should obscure
the structural similarities between Patrizian space and Cartesian matter.
Gassendi, who knew Patrizi's work well,2 1 agrees that space, although it is
an existing thing, is not comprehended within the Aristotelian categories:
it is commonly agreed that Place and Time are corporeal accidents; so

that if none of the bodies on which they depend existed, there would be
neither Place nor Time. To us, however, because it seems that even if there
were no bodies, still constant Place and fleeting Time would remain, Time
and Place seem not to depend on bodies, and so not to be corporeal
accidents. But neither are they on that account incorporeal accidents
[...] Hence it happens that Being in its most general acceptation is not
adequately divided into Substance and Accident; rather Place and Time
should be added as two members to the division. (Gassendi Syntagma,
Physica>§21iblCl, opera 1:182a)
Even if all sublunary matter were annihilated by God, still the space en
closed within the lunar sphere would not cease to be (182b, 184a). 22 The
supposition, though it may seem absurd to some, is no more so than the
supposition that matter should exist without form: yet the Aristotelians
allow the second but not the first. The space that is thus proved to exist
apart from body Gassendi calls "incorporeal" (183b), not because it is a
thinking thing, but because it exhibits no repugnance in the face of pen
etration by bodies (see Syntagma, Logica> lib1c7, opera 1:55a).23
21. The Syntagma includes a precis ofPatrizi's magnum opus (Opera 1:246ab). Although the
Syntagma was not published until1658, much of it was written between 1635 and 1645. A draft
of the section on space and time is dated by Bloch to 1635-1636 (Bloch 1971:xxix). Bloch
obseiVes that the earlier version also affirms the Epicurean definition of time and space as
"accidents of accidents," which is inconsistent with the position sketched above. The inconsis
tency is resolved completely only in the Syntagma itself (Bloch 1971:186ff). On Gassendi's
theory of space and time, see Bloch 1971, c.6, and Grant 1981 :2o6rr.
22. A similar argument is made in a letter ofGassendi to Samuel Sorbiere in 1644. Sorbiere
wants to know "what can be said against the Cartesian dogma that There exists no vacuum."
Gassendi replies with a brief version of the annihilation argument (To Sorbiere 30 Apr. 1644,
Opera 6:187; see AT 4:108-109; Sorbiere's letter is in Gassendi Opera 6:469, and cf. AT 4:108
and Mersenne Carr. i3:11o). The annihilation argument was used by Suarez to show that
distances among bodies do not presuppose the existence of inteiVening bodies (Disp. 30§7'134,
Opera 26:106).
23. Applied to souls or to God, Gassendi says, the term 'incorporeal' denotes a "true and
full substance, and a true and full nature, with which their faculties and actions are consonant"
(syntagma 183b). Applied to space and time it denotes merely the negation of body. We thus
have a partial coincidence with Fonseca's characterization of space as a pure negation. But
Fonseca does not take the more radical step of enlarging the categories to include space and
time as distinct ways of being alongside substance and accident.

Authenticated
Bodies in Motion
Like Patrizi, Gassendi holds that space exists and is not an accident of
matter (or of spirit). He holds, moreover, that the mark of a corporeal thing is
that it should be impenetrable and affect the senses. But he denies that
space is a substance. The disagreement is not so great as it may seem: where
Patrizi enlarges the notion of substance, Gassendi denies that 'substance'
and 'accident'-in their Aristotelian acceptation-exhaust the ways that a
thing can be said to be. Common to both is dissatisfaction with the categori
cal conception of substance.
From this survey I draw out four themes to help in understanding
Descartes's position:
(i) Aristotle's claim that whatever has dimensions is body, or at least

sufficiently similar to body that it is subject to the constraint of not
existing in the same place with body;
(ii) the general agreement among Aristotle's critics that whatever is cor
poreal will indeed resist penetration by other corporeal things and
will be capable of affecting the senses;
(iii) the general agreement among the critics that space, although it has
dimensions, lacks the characteristics in (ii), and thus that (i) is false;
(iv) the general agreement that space is not an accident of bodies but a
substance in its own right, that it is not a composite of matter and
form, and that it cannot be subsumed under any of the categories,
together with a marked lack of agreement concerning the kind of
thing space is.
The bones of contention, then, are whether extension alone suffices to

make a thing a body; whether space is a substance; and how to characterize
the kind of entity space is without using the Aristotelian apparatus. We have
seen that quantity, or actual extension, is, in Descartes's view, distinct only in
reason from quantified substance. What remains is to show that, contrary to
what the novatores thought, bodily extension and spatial extension are one
and the same; and to revise the metaphysics of form and matter to accom
modate the new entity.
2. Descartes against the novatores. Principles 2§ 10-11 attempts to do two
things at once: to accommodate Aristotle's thesis that, since interpenetra
tion of dimensions is impossible, there is no space apart from bodies, and to
refute the novatores' claim that corporeal extension and spatial extension
are different. The first is accomplished simply by identifying "space, or
internal place" with the "corporeal substance contained in it"; the second by
showing that "the extension in length, width, and depth that constitutes
space is evidently the same as that which constitutes body."
We have, in fact, already been told that "extension [...] constitutes the
nature of corporeal substance" (PP 1§53, AT 8/1:25), that quantity or

Authenticated
Parts of Matter
bodily extension is the nature of matter, and that it is identical to the

substance that has it. What remains to be shown is merely that quantity is
spatial extension:
And indeed we would easily recognize that the extension which con
stitutes the nature of body and the nature of space is the same [...], if in
attending to the idea we have of a certain body, e.g. a stone, we reject from
it all that we recognize is not required for the nature of body: namely, we
reject first of all hardness, because it would be lost if the stone were
liquefied or divided into exceedingly minute bits of dust, and yet the stone
would not cease to be body; we reject also color, since we often see stones
so pellucid that there is no color in them; we reject heaviness, since fire,
although it is very light, is nonetheless held to be body; and finally we
reject cold and heat, and all other qualities, since either they are not
considered in the stone, or because if they are changed the stone is not
thereby judged to lose the nature of body. Then we would realize that
plainly nothing remains in the idea of it, except that it should be some
thing extended in length, breadth, and depth: and the same is contained
in the idea of space, not only space filled with bodies, but also that which is
called 'vacuum' (2§11, 8/1:46).
Although we are asked here to consider an individual body, the object of

Descartes's argument is body in general, the materia whose nature he has
earlier shown to be quantity. What is common to all bodies must persist
through every natural change; what does not persist is not part of the nature
of body. But reference to change is only a pedagogical device. What counts
is whatever belongs to all bodies, whether they can be changed into one
another or not. The result of the experiment, of course, is that only
extension-being endowed with the three spatial dimensions-belongs to
all bodies. But extension also belongs to all spaces, filled or (apparently)
not. So the novatores are mistaken: space is body. Out go the hesitations of
Fonseca, the incorporeals of Gassendi, the promiscuous mingling of matter
and spirit in Patrizi.
Many of Descartes's contemporaries rejected his conclusion, holding that
the nature of body includes impenetrability, or at least that space without
body is possible. Even if extension suffices to constitute substance, it does
not-so they believed-suffice to constitute body. The question ofthe suffi
ciency of extension is complicated enough that I will treat it separately in
§g.3. In the remainder of this section I consider the metaphysical problems
that arise from the identification of body in general with extension.
3· Revising the concept ofsubstance. In the standard Aristotelian conception,
figure and size are accidents of individual concrete substances, resulting
from the determination of matter by substantial form. Quantity too is an

Authenticated
Bodies in Motion
accident, whether of matter alone or of the composite of matter and form.

Space, so far as it can be distinguished from quantity, is the abstracted
dimensions of individual substances and thus exists only in our conception;
so far as it can be said to exist outside the intellect, it is simply quantity
under another name.
The novatores, believing that space and body are separate, recognized that
space could not easily be subsumed under the Aristotelian categories; nor
could it be analyzed into matter and form. Patrizi can designate it only by
oxymorons. Gassendi rejects the highest division of being into substance
and accident. But what, an Aristotelian might ask, could a thing be which is
neither substance nor accident, and yet real?
It was in this troubled context that Descartes, rewriting the logic of the
textbooks, devised the classification of things into substance, attribute, and
mode. Although there are adumbrations of it in earlier works, only in the
Principles does Descartes systematize his vocabulary. Here I deal mainly with
his endeavor to maintain the Aristotelian rejection of a space independent
of bodies, while foregoing the Aristotelian analysis of corporeal substance
into matter and form. Descartes had, of course, further reasons to introduce
the new classification; in particular, it is crucial to proving the real distinc
tion between mind and body. Such questions, which have been thoroughly
treated by others, I leave aside.
Descartes begins by noting that what "falls under our perception" we
consider either as "a thing or an affection of a thing, or as an eternal truth"
( 1§48, 8 I 1:2 2). Of things, the most general are "substance, duration, order,
number' and the like. These terms apply "to all kinds of things." In particular,
"by substance we can understand nothing other than a thing which so exists
as to need no other thing to exist" ( 1§51, 8/I :24). Strictly speaking, God is
the only substance. All others exist by the concurrence of God. But some
require only that concurrence in order to exist, and for that reason
Descartes calls them substances also (1§52, 8/1:25), while the modes of
created substances require not only divine concurrence but substance in
order to exist.'24
The existence of a thing-Descartes repeats an Aristotelian common
place-by itself can have no effect on us. Substances are known only
through one or another mode. From the existence of a mode, established in
perception, we infer, according to the "common notion" that nothing is
attributed to nothing, the existence of a substance to which it belongs. The
24- Descartes uses both 'attribute' and 'mode' in the general sense of 'created thing whose
existence is not independent of the existence of all other things save God'; but he also
occasionally restricts the sense of the tenus, so that 'attribute' denotes only those dependent
beings that do not vary so long as the substance they depend on exists, while 'mode' denotes
dependent beings that can vary while the substance they depend on continues to exist. To
avoid ambiguity I use 'attribute' only in the more restricted sense.

Authenticated
Parts of Matter
inference can be made from any mode whatsoever: we are no less certain,
Descartes later observes, that a body exists by virtue of its color than by virtue
of its figure (1§6g, 8/1:34). But for each substance there is one "principal
property," or attribute, which "constitutes its nature and essence, and to
which all others are referred." By reference to principal attributes we define
the two highest kinds [summa genera] of created things: "intellectual or
cogitative things, that is, those pertaining to mind or to thinking substance,"
and "material things, or those that pertain to extended substance, that is, to
body" (1§48, 8/1:23).25
Principal attributes are at first said to be modes: Descartes even uses the
term of art inesse, which in Aristotelianism denotes the relation between
accident and substance, to describe the relation between attribute and
substance. But after defining a "distinction of reason" to be that which
obtains "between substance and one of its attributes without which it cannot
be understood" (or between two such attributes), Descartes writes:
Thought and extension can be regarded as constituting the natures of

intellectual and corporeal substance; and then they should not be con
ceived otheiWise than as thinking and extended substance itself, that is,
mind and body; in that way they are understood most clearly and dis
tinctly. Indeed we more easily understand extended substance or thinking
substance than substance alone, omitting the fact that it thinks or is
extended, since there is some difficulty in abstracting the notion of sub
stance from the notions of extension or thought, which are diverse from it
only in reason. (1§63, 8/1:31)
The indistinctness that Descartes later affirms of substance and quantity

(2§8, 8/1:44), here holds between principal attributes and substances gen
erally. As I said in §g.1, that may be one reason why the later argument is
perfunctory.
Extension, when we take it to "constitute the nature" of body, is nothing
other than extended substance itself. But it can also be taken for a mode of
substance when we consider that "one and the same body, retaining its same
quantity, can be extended in several different ways"-by taking on various
figures or shapes ( 1§64, 8/1:31 ). We can conceive res extensa without con
ceiving it to be spherically extended. But we cannot conceive of being
25. There is a third class of things that, although they are certainly modes of thought,
cannot be fully understood without reference to body: sensations and passions. The French
translation comes close to suggesting that they are properties of a third substance-the "inti
mate union" of body and mind (g/2:45; cf. Cottingham Ig86:127-132). But the word at
tribuees is a poor translation of the Latin referri (earlier occurrences of which in the same
paragraph are translated by rapporter). To refer X to Yis not to regard X as an attribute of Y. It is
to denominate or conceive of X by reference to its cause (or, as in "referred pain," what we
think is its cause), in any of the Aristotelian senses.

Authenticated
Bodies in Motion
spherically extended without conceiving of the thing that is thus extended.

Spherical extension and extended substance are, as Descartes puts it,
modally distinct. Extension simpliciter, on the other hand, and extended
substance are not even modally distinct, but only in reason.
The crux, from an Aristotelian standpoint, is the claim that extension
could "constitute the nature" of body. The constituents of an Aristotelian
body are matter and form; the accidents of body, including quantity, de
pend on them, and cannot by themselves constitute substance. But that
quantity or extension does suffice to give being to a thing is just what
Descartes means to say. Created substance is defined as that which needs
nothing save the concurrence of God to exist. Nothing, however, merely
exists: everything exists somehow. Among created things, there are two ways
or modes of existing: the way of extended things and the way of thinking
things. Being extended is an "attribute" in the strict sense, which is to say, an
invariant mode, because it is, for bodies, nothing other than existence itself.
A comparison will help make this clear. Duration applies to all things; since
a substance (of whatever kind), "if it ceases to endure, ceases also to be,"
duration differs only in reason from the substance itself, and for that matter,
from its existence (1§62, 8/1:30; To *** 1645 or 1646, 4:349). So too, a
body, if it ceases to "extend," ceases to be.
One immediate consequence is that extension, considered as that which
constitutes the nature of certain created substances, is not an accident of
bodies. 26 Only when we consider that the determinate extension a thing
now has could be determined otherwise do we regard extension as a mode
( 1§64, 8 I 1:31). Extension as mode is a determination of extension as sub
stance, and cannot exist apart from it; but since determinations of extension
can vary while extension itself remains, extension as substance can exist
apart from each of its modes. Those modes are, suitably enough, only
modally distinct from it (1§61, 8/I:2g).
Descartes's terminology, however, varies significantly. Sometimes, as in
1§64, he speaks of extension itself as a mode. More often, he speaks of
figure and motion as modes of extension, reserving the term 'extension' to
denote extension as substance, as the principal attribute of body.2 7 When
extension is taken to be the attribute of body, its function is akin to that of
Aristotelian form. Together with thought, it serves as the basis for classifying
created substances; it is the ground from which flow the properties of
body-divisibility, figurability, mobility, and impenetrability. When, on the
26. My thinking on this question has benefited from reading Gloubennan 1978, a revised
version of which is in Gloubennan 1986, c.6.
27. 'Thus the extension of a body can indeed admit in itself various modes: its mode is
different if the body is spherical than if it is square; but the extension which is the subject of
these modes, regarded in itself, is not a mode of corporeal substance, but rather an attribute
which constitutes its essence and nature" (NottE in programma, 8/2:348-349).

Authenticated
Parts of Matter
other hand, extension is regarded as the subjectum of figure-and especially

the "configuration," as Malebranche called it, which characterizes natural
kinds like water or oil-its function is akin to that of matter, and the charac
teristic figure plays the role of form.
Extension is at once the differentia of body, resembling form in its tax
onomic role, and the substrate of change, resembling matter. Odd as that
might seem from an Aristotelian standpoint, Descartes's conception does
yield decisive answers to the questions that troubled Fonseca and other
Aristotelians, without risking the distinction of mind and body or sinking
into atomism. Space is indeed substance, though not a composite of matter
and form. It is the very same substance we call body; it therefore affects the
senses and can be known. Because it is body, there is no "penetration of
dimensions"-here Descartes preserves what is true in Aristotle-and no
possibility of space devoid of body. Atomism is not merely empirically but
necessarily false.
Descartes's conception has the further advantage of yielding, or appear
ing to yield, an a priori destruction of Aristotelian sensible qualities. Here
the notion of constitution comes into play. Extension is said to constitute
the nature of body, just as thought constitutes the nature of mind. In the
Aristotelian vocabulary, the term 'constitute' is applied most properly to the
relation between a substance and its intrinsic causes-matter and form.
Form is the principal constituent of substance, because (in the common
phrase) it "gives being" to substance.28 To constitute the nature of a thing,
then, is to stand to that nature as form to substance, to be the ground of the
properties that follow from having that nature. Extension by itself is not the
nature of corporeal substance. That nature also includes being divisible,
mobile, capable of figure. But all those properties follow, or so Descartes
believes, from extension.
Just as importantly only those properties follow from extension. Earlier, in
considering the argument of Principles 2§ 11, I asked whether Descartes had
shown that color or hardness could not be modes of extension, even if they
were not part of the nature of body. What is shown there seems to be at most
that body can exist without color; but if color cannot be conceived without
body, then it would seem to be a mode of extension. If it isn't, it must differ
somehow from the acknowledged modes.
One answer is that the idea of color is obscure and confused when re
ferred to something outside of us ( 1 §68). But that smacks of dogma, and
would establish at best that we cannot be certain whether color is a mode of
extension. It fails, moreover, to address the phenomenology of color. Colors
seem always to be areas of color; unextended color is as hard to imagine as
28. Goclenius Lexicon s.v. 'Constituo', 456.

Authenticated
Bodies in Motion
unextended figure. Colors seem to "presuppose" extension just as much as

figure and motion do ( 1§53), and the Aristotelians attributed, as a matter of
course, not only intensive but extensive quantity to them.
The arguments of 2§4 and 2§11 are often taken to address the question.
But 2§4 seems to take as proved that colors exist only as modes of thought,
while 2§ 11 presupposes that extension constitutes the nature of body, the
only question being whether corporeal extension is spatial extension. It
does not show why color or sound could not be modes of spatially extended
substance, even if not part of its nature qua body.
I would like to offer an answer that, although not explicitly put forward by
Descartes, can be devised with the means at his disposal. I recall one of the
arguments used by the Nominalists to show that quantity and substance are
one (§4.2). To have quantity is just to have "parts outside of parts" (partes
extra partes); but material substance of itself has such parts, so that supposing
in addition to the substance itself a quantity distinct from it adds nothing.
Descartes, though he does not use the argument, does regard extended
substance (or space) as having "distinguishable parts" of "determinate mag
nitude and figure" (To More 5 Feb. 1649, 5:270-271; cf. 305); the absence
of such parts suffices to show that a thing is not extended (ib. 2 70; cf. Med. 6,
7:86). But if to be extended is to have parts, the possibility of figure-of
boundaries between parts-is implied in the very idea of extension.
The possibility of color is not-not if one supposes that the putative
colors in things resemble our ideas of color (PP 1§66, 8/1:32). There is no
evident relation between extension and what an idea of color represents: "it
is the same," Descartes writes, ''when someone says he sees color in some
body or feels pain in some limb, as if he had said he sees or feels in it
something quite unknown to him-as if, that is, he had said he does not
know what he sees or feels" ( 1§68, 8/1 :33). Not knowing how it is possible
for an extended thing to have a mode resembling what is represented in a
sensation of color, we have no reason to include color in the nature even of
an individual body.
In explaining the nature of corporeal substance Descartes had to steer

between two alternatives: Aristotelianism, with its baroque apparatus and its
unwarranted distinctions; and the theories of the novatores, whose common
failing was to abandon the easily and completely understood conception of
bodily extension, and to propose instead various confused notions of some
thing neither material nor spiritual-notions, moreover, which could lead
their proponents onto dangerous theological ground, as we will see in
Descartes's correspondence with More. Descartes's doctrine offers a clear
cut resolution of Aristotelian questions about the relation of matter and
quantity, and about the nature of space; at the same time it avoids the

Authenticated
Parts of Matter
spiritualization or divinization of space that one finds in authors like Patrizi.

It preserved the basic Aristotelian tenet that the only independently existing
things are singular individuals, though not without a profound revision of
the ontological and physical analysis of substance. Substantial form in par
ticular is, so to speak, fragmented: its role in defining essence is taken over
by extension as substance; its role in providing a ground for the properties
of natural kinds is taken over by configuration. Extension as substance also
takes over the role of matter; the fusion of the two roles contributes, as we
will see, to the difficulties that Cartesian physics has in characterizing indi
vidual bodies.
4· Individual bodies. The argument of the second part of the Principles
proceeds through a series of identifications: matter is quantity, quantity is
substance, the extension that constitutes body is that which constitutes
space. In Aristotelian physics prime matter cannot naturally exist without
form; so too, we may suppose, extension in general cannot exist except in
the determinate extensions of individuals, in regions of definite size and
figure.
But what distinguishes one region from another? The question is impor
tant not only in its own right, but because it illuminates the relations of
physical and geometrical space in Cartesian physics. Descartes is frequently
said to have identified physical and geometrical space; their relation is, I will
argue, not that simple. One must be wary, moreover, of attributing to
Descartes notions of space-geometric or physical-that were not his, how
ever familiar they are to us. Once the relation of physical and geometrical
space is understood, the definition begins to make sense. Nevertheless, the
lack of genuine unity in bodies as Descartes defines them leaves his physics
with inadequate means to explain why bodies sometimes fragment in colli
sions and sometimes not, or under what circumstances a body will be
deformed rather than resisting the intrusion of another body.
In Aristotelian questions on individuation, the first issue was whether a
principle of individuation is needed. Taking Aristotle's side against Plato,
the Aristotelians agreed that all actual substances are "singular and individ
ual" (Suarez Disp. 5§1'1{, opera 25:146). But Ockham and his followers
went further. In their view, even species and genera, though they are said of
more than one thing, are "by themselves and first of all individual" (Fonseca
In meta. sc6q 1§ 1, 2:35 7C). The nature Man, in their view, "is nothing other
than this man or that," the only difference being that the common noun
'man' refers vaguely to all men, while the singular term 'Peter', say, refers to
one.
Though Nominalism was not without its effects on sixteenth-century Aris
totelianism, its central thesis was rejected. "Those natures to which we give
the names 'universal' and 'common', are real, and truly exist in things

Authenticated
Bodies in Motion
themselves," prior to any operation of the mind (Suarez Disp. 6§2t 1, opera
25:206). The question therefore arises: if common natures are "of them
selves indifferent" to the distinction between one and many, by what princi
ple are they "individuated" in singular beings? If Peter's and Paul's human
ness are one, then evidently something else must account for their being
two.
It is not my purpose here to descend into the labyrinthine details of
Aristotelian disputes over the principle of individuation. I simply note two
rejected alternatives, and then briefly characterize the kind of principles
typically adopted.
The first rejected alternative was that some collection of accidents could
serve as the principle (Fonseca In meta. sc6q 1§2-4, 2:359-361 ). Those who
favored it noted that although Peter and Paul may share humanness, there
are many other accidents they do not share. It might well happen that no
other human shares all the accidents of Peter, and so that collection could
be the principle by which Peter is distinguished from other humans. To this
the Aristotelians objected, among other things, that even a collection of
accidents is in principle communicable: although in fact there may be no
individual of exactly Peter's height and weight, there certainly could be (ib.
§3, 2:3sgE-F).
The second rejected alternative was the celebrated thesis of Thomas: the
principle of individuation among corporeal substances is materia signata (or
quanta), a "designated" portion of quantified matter.29 Thomas's thesis re
sembles Descartes's solution to his problem closely enough that it is worth
examining the reasons for and against it. The strongest reason for it is that
"matter is the principle of multiplication and distinction among individuals
of the same species"; since matter is incommunicable, Thomas's solution
overcomes the objection to individuation by accidents (Suarez Disp. 5§3'13,
opera 25: 162). Suarez replies that determinate quantity, like the other
dispositions of matter, is consequent upon form. Form, not matter or quan
tity, is the "principle of distinction" among complete substances ('15, 2 5: 163).
A corporeal substance can, moreover, be deprived of quantity altogether;
yet it will be no less an individual (t16, 25:167). Those objections lapse, of
course, if like Descartes one believes that quantity itself is the only form in
bodies, and that body cannot be deprived even by God of its quantity.
The kind of principle typically adopted is not so much defined as desig
nated. Fonseca writes that the principle of individuation is "a certain posi
29. What materia signata quantitate signifies was itself disputed. Suarez considers two inter
pretations: that materia signata is matter actually endowed with determinate quantity; and that
it is the capacity of prime matter to receive determinate quantity and other dispositions that
enable it to receive form (see §5.3; Disp. 5§3t8, g, 18; 25:164, 168). The essential point is that
bodies are individuated by virtue of having (or of including matter capable of having) determi
nate quantity-quantity endowed with size and figure.

Authenticated
Parts of Matter
tive difference," intrinsic and nonderivatively [primo] incommunicable;

added to the species, it "constitutes the individual" and distinguishes it from
others "prior to any discrimination by accidents" (ib. q5§ 1, 2:3381C). Since
the positive difference that constitutes an individual cannot be removed or
altered without destroying the individual, it belongs to the essence of the
individual. From this, Fonseca concludes in Leibnizian fashion that "no
power can bring it about that two individuals should be positively similar in
everything that pertains to their essence" (ib. §3, 2:385B). It does not,
however, contradict his conclusion, so far as I can see, to suppose that two
individuals could share all their accidents: positive differences are not
accidents.
Suarez's view is more elusive, in part because, as a subtle variant of the
nominalists', it comes near to dismissing the question.30 The principle by
which a thing is individuated is nothing other than "its being [entitas], or
[...] the intrinsic principle by which its being is established" (Disp. 6§6, 1,
opera 25:180). He rests his conclusion on the proposition, argued in an
earlier disputation, that "the ground of unity cannot be distinguished from
being itself." To be and to be one differ only in reason: 'being one' connotes
the negation of numerical division, while 'being' does not. The same is true
of 'being an individual', which merely connotes the incommunicability of
entitas. Socrates' soul is not something that Plato can share; but humanity
can exist in both. The difference between Suarez and Ockham is that Suarez
does not agree that Socrates' humanity is essentially individual, or that all
that is common to Socrates' humanity and Plato's is their name. 31
In particular matter is not individuated by quantity or any accident of
quantity: 'The matter which underlies this form of wood is numerically
distinct from that which underlies the form of water or man; it is therefore
individual and singular in itself. The ground of such a unity is not the
substantial form, nor being ordered to this or that form [...] , because
when any form is varied, there always remains numerically the same matter
[...] Quantity too cannot be the ground of the individual unity of matter,
as the same argument shows, if it is true that matter can lose and acquire
various quantities as substantial forms vary" (§6'12, 25:18of). Indeed, God
can remove all quantity from the matter of a thing. Even so-contrary to
what Ockham believed (§4.1) -its parts will remain distinct in being ( entita
tive distinctce), and the matter itself will remain distinct in being from other
matters (§3, 14, 25:167). Suarez grants that ''with respect to location
[situm] one matter is distinguished from another by quantity." But that is
not the essential difference: as an existing thing, "it is truly and really
30. On Suarez's position, see the lucid and thorough exposition in Gracia 1994
31. See Gracia 1994:500.

Authenticated
Bodies in Motion
distinguished by its being [per suam entitatem]" (§6'14, 25:181), and by that
alone.
Descartes takes the short way with species and genera. Like all universals,
they exist only in our conception, and are only modes of thought (PP 1 §59,
8/1:27). For him, as for the nominalists, there is no problem of individua
tion because there is no individuation: "When I say 'Peter is a man', the
thought by which I think of Peter indeed differs modally from that by which
I think of man, but in Peter himself being man is nothing other than being
Peter" (To*** 1645 or 1646, 4:350). So too the distinction between exten
sion in general and the extension of this body is only a distinction in reason:
singular extensions alone are real.
The traditional problem is consigned to the Scholastic shades. But new
ones pop up. Extension, as we have seen, can be regarded as constituting
the nature of body, and as the substrate of natural change, and thus as
substance. It can also be regarded as the determinate extension of this or
that body, and thus as a mode. The relation between the two aspects of
extension is summed up in Descartes's phrase 'part of matter' (pars materim).
That phrase has, it would seem, two senses. When Descartes speaks of
matter as partibilis, or of a simple body having many parts, the parts in
question are not actual individual bodies. Yet they are, according to
Descartes, really distinct, no less than the sun and the moon. How then are
they distinct? Since we are not yet thinking of bodies moving, but only of the
parts of one body, the question is one of the static individuation of potential
parts of matter.
Suppose that the static problem has been solved. Potential parts can be
distinguished. But some of those parts are actual, and correspond to what we
call individual bodies, some are not. The potential parts of a vase full of
water remain the same whether any part moves or not, so long as the vase
does not change: But the actual parts will differ a great deal if the water
freezes. This is true even according to common sense, and all the more so
for Descartes, for whom, as we have seen, motion is defined as the rupture of
a body from its neighbors. A second problem, then, is that of the dynamic
individuation of the actual parts of matter.32
Of the two problems, the static problem is simpler. But even it has pitfalls.
Descartes, in arguing that the potential parts of an extended substance are
really distinct, speaks of each part as being "delimited by us in thought [d
32. I make two simplifications. First, only simple bodies will be considered. Descartes calls a
vase full of water one body; but that "one" body is made up of many smaller actual bodies; the
attempt to include compound bodies also in the definition of 'part' raises further difficulties
(see Grosholz 1991 :68). Second, I omit sliding motions. Descartes's definition cannot straight
forwardly be applied to sliding parts. Since it will, if it fails in straightforward applications, no
doubt fail too in the rest, I will assume that the simple bodies being defined are rupturing from
their neighbors.

Authenticated
Parts of Matter
nobis cogitatione definitam]" (PP 1§6o, 8/1 :28). But how, one might ask, is this
to be done? As Emily Grosholz puts it, "in quiescent res extensa there is no
physical analogue of a boundary"; there are "no articulating parts which
themselves have the integrity ofshape" ( Grosholz 1991:68). Though she has
in mind the "great monolith" with which, in an at least conceivable instant
before God gave motion to the world, the creation began, her worry applies
equally to the interior of any body. A grain of sand, too, is a monolith.
Or again: Descartes writes as if motion were given to parts-as if there some
how were parts in motionless matter already, at least in God's understanding
of it. But in his official doctrine, motion and parts come into existence
simultaneously, just as when a girl is born, a mother and daughter come into
existence together. So matter, or "space in general," has no diverse parts to
which diverse motions could be given. Like Hegelian being, it threatens to
collapse into nothing, a "surd" as Grosholz calls it (1991:70, 66).
Whatever Ockham may have thought, entitative difference does not of
itself yield partes extra partes. Extra signifies not mere difference in being (as
one soul might be said to be "outside" another), but existence in distinct
locations. It is characteristic of quantity that difference in being among its
parts entails difference in location, at least naturally. An inch of extension is
such that "another inch of extension cannot be added to it without their
making together a quantity of two inches, because if they penetrated one
another, they would occupy no more space than one alone occupied when
they were separated" (Regis Cours 1ptl c 1, 1:282). Were they to occupy the
space of one, were they in exactly the same location, they would be one. Parts
of extension, in short, are individuated by their locations.33 The exteriority
proper to spatial parts must, it would seem, rest on a "topology," a system of
places prior in definition to the parts themselves.
Some of those places will be known to us because they coincide with the
places actually occupied by individual bodies. The rest must be grasped by
the imagination, which will impose on them locally the metric of the Geome
try. In the Principles Descartes lays out a series of distinctions that amount to a
construction of the topology of space in general from the "singular" spaces
of bodies:
(i) The singular space of an individual body differs only in reason from
the body itself; it can also be called the place of that body.
(ii) Generic space is the place of a bodyatagiven time which we imagine to
remain even when the body moves-the hollow in the sculptor's
mold, for example, where the wax was and the bronze is.
33· So Suarez, in passages discussed in §4.2, concludes that "having parts that occupy
extended and divisible space" or "extension of parts in order to place" is the "ratio and formal
effect" of quantity; or, more precisely, having the potentia to occupy space (Suarez Disp.
4o§4115, opera26:547).

Authenticated
Bodies in Motion
(iii) Finally there is space in general. At each instant this coincides with the
concatenation of all singular spaces; within it all generic spaces are
contained.
The construction is incomplete as it stands, since the set of generic spaces

depends on the places of actual bodies; we cannot be certain that those will
include every place a body could be, every part of space in general. Here
again the imagination must be put to work: once space in general has been
constructed, we can imagine within it arbitrary concatenations and divisions
of the bodies that now constitute it. The generic spaces of those imagined
bodies I will call generic in the broad sense, or locations.
Once a complete set of generic spaces has been constructed within space
in general, the problem of static individuation can be addressed. Within the
singular space of an actual individual body are contained infinitely many
generic spaces in the broad sense. A potential part is any such space. Like any
space, it is a body; it is distinct from other potential parts by virtue of its
place. Insofar as it is a substance in its own right, really distinct from poten
tial parts disjoint from it, it is no less a body than the whole. But insofar as its
boundaries are not marked, so to speak, by rupture, it is only a potential
individual body; our knowledge of it can come only from the imagination,
not from the senses.
Potential parts of matter are individuated by reference to the topology
implicit in the idea of extension. Motion enters in only by way of exhibiting
to us the singular, and thus the generic, spaces through which we come to
conceive space in general. Actual parts of matter, on the other hand, are
defined by Descartes in terms of motion, or, more precisely, in terms of the
fundamental mode of rupture.
Descartes's principle of the dynamic individuation ofbodies appears first
in Le Monde; it is repeated, without significant alteration, in the Principles. In
the earlier work it looks like this: "Observe in passing that I take here and
will always take hereafter as a single part all that which isjoined together and
which is not in the act of separating [qui n 'est point en action pour se separer]
[...]:thus a grain of sand, a stone, a boulder, and even the whole Earth may
hereafter be taken for a single part, insofar as we consider in it only one
entirely simple and uniform movement" (Monde 3, AT 11: 15). In the Princi
ples, the definition of 'part' is incorporated into the definition of motus: 'We
can say that [ motus] is that translation ofone part of matter, or one body, from the
vicinity of those bodies that immediately touch it and which are regarded as being at
rest, into the vicinity of others. Where by one body, or one part of matter, I under
stand all that which is transferred at once, even if this part may be itself
composed from many parts, which have other motions in them" (PP 2§25,
AT 8/1 :53-54). Descartes devotes little argument to the definition; he does
nothing to dispel the appearance of circularity when he juxtaposes the

Authenticated
Parts of Matter
definition of 'part' with that of 'motion'. Small wonder that it has irritated
critics from Cordemoy and Leibniz to the present.34
What follows is not a vindication of Descartes's principle but an explica
tion that reveals not only certain features of his conceptions of body and
space, but also some preconceptions about space that seem to impinge on
its interpretation by recent critics.
Space in general is, first of all, not Euclidean space as that is now under
stood. Descartes did not tamper with the Aristotelian conception of space:
that innovation should be credited to others, like Gassendi and More. What
he transformed was the Aristotelian conception of body, making actualized
corpus mathematicum the object of physics rather than corpus physicus. Doing
so did not preclude his retaining in large part Aristotelian conceptions of
extension, quantity, and space. Cartesian space, like Aristotelian space, has
no existence prior to or independent of the existence of bodies; it is, on the
other hand, independent of our thought. We have no way of knowing that
the world has the structure we expect it to have by virtue of the ideas
implanted in us by God, except through interaction with actual bodies. But
though our knowledge of the topology of space is derived from our knowl
edge of actual bodies, and though the existence of space depends on that of
actual bodies, space in general has the structure it has independently of our
thoughts and of the forms of the actual bodies it contains.
We have, then, space in general, abstracted from the motions of bodies,
and thus from their particular shapes. Space in general contains infinitely
many generic spaces in the broad sense, of all shapes and sizes. As Newton
put it in an early manuscript: "In all directions, space can be distinguished
into parts whose common limits we usually call surfaces; and these surfaces
can be distinguished in all directions into parts whose common limits we
usually call lines; and again these lines can be distinguished in all directions
into parts which we call points. [...] And hence there are everywhere all
kinds of figures" (Newton Unpub. 100/132; trans. Hall and Hall). Like
Gassendi, and Henry More in his later works, Newton believes that space is
immobile and indivisible. It is the scene within which bodies move, and by
reference to which motion is measured; it cannot itself be said to move.
With that I think most philosophers now would agree. Descartes does not.
Individual bodies, he claims, are parts of extension; the extension of a body
and the space it occupies are one and the same substance; and bodies move.
It follows that parts of space can move, and that space is divisible.
Those are strange claims. Their strangeness is a clue that any attempt to
situate the Cartesian definitions of motion and individual body within the
framework of classical physics is likely to result in misunderstanding. For
Descartes there is no space apart from the singular spaces of individual
34· One recent example is Funkenstein, who writes, "obviously, Descartes lacks a principle
of individuation for single bodies as physical entities" ( 1986:73).

Authenticated
Bodies in Motion
bodies; space in general is, so to speak, the mass composed of all such spaces
when one abstracts from their motion and therefore their boundaries. It is
tempting to think ofthat mass as being like Newtonian absolute space, and
of bodies as resulting from its partition. Descartes's description of creation
lends itself to that interpretation. But to interpret Descartes thus is to over
look the fact that only the singular spaces really exist.
Call two singular spaces apart if the boundary of each includes no part of
the boundary of the other; call them together otherwise. Unlike regions of
Newtonian space, singular spaces can be apart at one time and together at
another. Hence one can define the rnpture of singular spaces: two bodies
have ruptured if at a certain time they are apart, having for some immedi
ately prior inteiVal been together. Motion, as we have seen, is successive
rupture.
That might suffice were it not that Descartes then defines one body as "all
that which is transferred at once." One might well think that the definition
is intended to supply a means of determining which of the infinitely many
potential parts of matter-of the whole mass of extended substance that God
created-are actual. But potential parts, it would seem, are defined in terms
of the topology of space in general; parts so defined are either always apart
or always together. They cannot move: motion is defined only for singular
spaces, and thus only for actual parts. Though the definitions of part and of
motion taken singly might be acceptable, taken together they seem be
devilled by an equivocation on part.
Perhaps the best answer is this. At each moment the whole of space
consists of many, perhaps infinitely many, singular spaces. Each of those
spaces has potential parts; each part can be distinguished from others ac
cording to the static principle of individuation. Like the entire singular
space of a body, we must suppose, the parts of that space can be designated
and reidentified at subsequent moments-as if they too were actual individ
ual bodies. It makes sense, then, to ask at a subsequent moment whether any
parts of the body we began with were together and now are not-any that
have, in other words, ruptured. If so, then our body will not have been one,
since at least two of its parts will have been "in the act of separating," to
quote the definition from Le Monde.
In De ipsa natura ( 16g8), Leibniz argues that there must be, in addition to
motion and the figures that result from motion, intrinsic differences among
.bodies. The alternative is to be deprived not only of any means of
discriminating bodies, but of discerning their motion, if by motion one
means the "successive existence of the thing moved in diverse places" (Leib
niz Ph. Schr. 4:512). Supposing that two parts of matter congruent in shape
do not differ, and that motion preseiVes shape, "it is manifest that from the
perpetual substitution of indistinguishables it follows that the states of
diverse moments in the corporeal world can in no way be discriminated. For

Authenticated
Parts of Matter
it would be only an extrinsic denomination by which one part of matter was

distinguished from another, namely from the future, in other words, that it
would later be in one place and then another; but of things in the present
there is no discrimination" (513). Daniel Garber concludes from this that
"motion can be used to individuate bodies only if there is some way of
reidentifying bits of material substance across time, and this can only hap
pen [...] if there is something in body over and above extension"
(1992:181). That something, it turns out, is the active power or vis insita
that Descartes intended to get rid of for good when he took the essence of
body to be extension.
Garber is right, I think, in holding that there must be some way to reiden
tify bits of matter from one moment to another. But that there must be
something in body "over and above extension," I am not so sure. Among the
Aristotelians, Fonseca appears to have argued for a version of the identity of
indiscernibles (or, more precisely, the discernibility of nonidenticals). But
Suarez argues that the entitas that in his view is the principle of individuation
is distinct only in reason from the individual substance itself. Two bodies,
alike in all their modes, could still be genuinely two; God, at least, could
discriminate successive states of the world even in the face of the "perpetual
substitution of indistinguishables."
Even so, the definition faces serious empirical difficulties. First of all, Des
cartes has no way ofdetermining what will happen when bodies collide, once
he admits that real bodies sometimes fall apart in collisions. Suppose that a
body B collides with a body C that is twice its size and at rest (see Figure 20).
According to the fourth rule of collision, B will be reflected by C and return
whence it came. But it is clear that in the third part of the Principles Descartes
holds that some collisions result in the fragmenting ofone or both bodies. In
2§63 we find a reason: each of the two halves C1 and G.! of the larger body
"may be considered to be an individual body," and so, since B and C1 are
equal in size, the sixth rule will apply. Bwill be reflected, giving up some of its
motion to C1 , which will separate from G.! and move off to the right. But
Descartes gives us no criterion for deciding when we should "consider" the
body as two half-bodies, and when we should consider it as a one whole. What
will happen in particular instances remains, despite this putative explana
tion, "indeterminate and cannot be specified" (Grosholz 1991:95).35
Similarly, even though Descartes regards some simple bodies as deform
able, he has no principled way of determining when a body will bend when
35· The example is borrowed from Leibniz (Animadversiones ad 2§54-55, Ph. Schr. 4:386
387). Leibniz notes that the outcome of a collision between one body and a part of another
(i.e., a collision where contact is made only at part of the surface of the other, as in Fig. 20)
ought to depend, if Descartes were right, only on the size of that part. If C1 remains attached to
C2 (if, for example, it were larger than B), it does so not by virtue of resisting separation from
C2 , but per accidens, as Leibniz says. It will stay where it is even if C2 does not exist. The whole
body Cis unum only per accidens, not per se.

Authenticated
Bodies in Motion
Fig. 20. Collision and division
pushed by another, and when it will break. Squeeze a sponge and its holes
get smaller, just as Descartes says in his account of condensation; but try the
same experiment with a piece of pumice and the result is a heap of frag
ments. Descent to the microlevel will not help, since there too we find
particles that are supposed to be flexible, while others are supposed to be
rigid. At the urging of Mersenne, Descartes does attempt to show why some
bodies are resilient. But the explanation presupposes that they can be
deformed without breaking; what Descartes explains is at best how through
the action of subtle matter an already deformed body is forced back to its
original shape.
The two problems are symptoms of a general malaise. Descartes insists, as
we have seen, that the only "glue" holding the parts of a body together is
their mutual rest. The absence of rupture, he says, is the strongest glue of
all; nothing else is needed to explain why bodies hang together. But experi
ence tells us that a die sitting on a table can easily be knocked off, while a
protuberant part of the table top of the same size and shape cannot.
Descartes is not unaware of the difference. In a description of the making of
glass from ash he argues that the particles ofash are separated by particles of
the second element, while those of glass are joined along much of their
surfaces (see Figure 21; the left-hand figure represents ash, the right-hand
figure glass). One could argue perhaps that only processes like melting
result in contiguity among particles; setting one body atop another results
not in true contact, but in the kind of contact tires make with rain-soaked
pavement. Descartes seems to have thought that the force required to sepa-
Ash Glass
Fig. 21.joining of particles (after PP4§125, AT 8/1:271)

Authenticated
Parts of Matter [377]
rate two potential parts of one particle varies directly with the cross section
of the particle in the cutting plane.36
The rule gives us a sort of ersatz unity-ersatz because it admits degrees.
But it does not answer the objections. The point about collisions would
stand even if the body C were full of holes, so that its two halves met along
only part of the line a in Figure 20. And although it may be easier to deform
a porous body than to deform a solid body, porosity alone cannot account
for deformation (or fragmentation). Some solid parts of matter, including
the particles of the first element, are deformable; and some porous bodies
are rigid.
Clarifying the relation between geometrical and physical space, while it
brings to light significant differences between Cartesian physical space and
the space of classical physics, does not in the end resolve the difficulties that
Descarte's principle of individuation brought with it. At best one can hold
the conceptual objections of Leibniz at bay; but the empirical insufficiency
of the principle remains. Elasticity eludes Cartesian physics, and with it any
hope of explaining the outcome of collisions.
9·3· Physical Questions: the Sufficiency of Extension

To establish that the nature of body is constituted by extension, one must
show that everything in the nature of body follows from extension in the
manner suggested above, and that what follows from extension is sufficient
to yield substances of the sort we call bodies. To the first point, More
objected that impenetrability belongs to the nature of bodies, and yet it does
not follow from extension alone. To the second, More and Arnauld in
correspondence with Descartes, and Gassendi in later writings, argued that
the existence of spaces that neither are nor contain bodies is possible.
1. Impenetrability. To show that extension is sufficient not merely to con
stitute a substance, but to constitute it as body, one must show that in bodies
as we commonly understand them, there is no property that does not follow
from extension. In particular, impenetrability must follow. Descartes seems
to have believed that this was obvious. 37
36. 'Nevertheless glass is very fragile, because the surfaces by which its particles touch one
another are extremely sparse and small" (PP 4§128, 8/1:272). It should be added that fragility
in general is also affected by the inteiWeaving of the branches of the particles in certain
substances (ib.).
37. The Aristotelians believed that two bodies could be in the same place at the same time,
on scriptural grounds (the virginitas intacta of Mary, Christ's exit from the tomb, his ascension;
see Coimbra In Phys. 4c5q4, 2:40), on doctrinal grounds (the cmpora gloriosa of Christ and the
saints are said to penetrate other bodies), and on the basis of argument. Since "quantity is not
formally contrary to quantity," the exclusion of one by another is not necessary, and can be
inhibited by God. Quantity has only the "virtue" of filling space; it is therefore possible for one

Authenticated
Bodies in Motion
He was wrong. In Henry More he found an opponent willing to grant that

space is substance, but not that space and body are one. More holds not only
that empty space is absolutely possible, but that bodies have at least one
property that space does not: they are impenetrable. He begins with tan
gibility, one of the two marks by which Patrizi and Gassendi had distin
guished space from body:
Although matter need not be soft, or hard, or hot, or cold, it is neverthe

less most necessary that it should be sensible, or, if you disagree with that,
tangible, as Lucretius so well defines it:
Nothing can touch and be touched, unless it is body.
That notion ought to be all the less abhorrent to you, since your Philoso
phy most clearly establishes that all sensation [...] is touch" (More To
Descartes 11 Dec. 1648, AT 5:239, quoting De rerum natura 1:304) 38
Tangibility could be regarded as following upon impenetrability, the other

distinguishing mark of body. So the more fundamental objection is that
bodies cannot be identical with regions of space because such regions are
not impenetrable: "[Tangibility] signifies the mutual contact and the poten
tia of touching, among whatever sort of bodies, animate or not, supposing
the immediate juxtaposition of the surfaces of two or more bodies. Which
suggests another condition of matter or body, which one could call impen
etrability, namely that [a body] can neither penetrate other bodies nor be
penetrated by them. From that there is a most obvious difference between
divine and corporeal Nature, since [divine nature] can penetrate [cor
poreal nature], but [corporeal nature] cannot penetrate itself' (AT 5:240).
Extension, therefore, being an attribute of both divine and corporeal na
ture, cannot confer impenetrability on a thing.
In the Principles, as in earlier works, Descartes assumes, without proof,
that extended substances are impenetrable. 39 That may seem odd. He un-
body to be in another's place provided that only one of the two actually fills space (ib. 41). Only
if both are actually extended is it impossible for both to be in the same place.
38. To shorten citations, I here give the dates and Adam-Tannery page numbers of the
Descartes-More correspondence, all ofwhich is included in AT 5: More To Descartes 11 Dec. 16
48, 235-246; Descartes To More 5 Feb. 16 49, 267-279; More To Descartes 5 Mar. 16 49, 298-
317; Descartes To More 15 Apr. 16 49, 340-348; More To Descartes 23 Jul. 16 49, 376-390;
Descartes To More Aug. 16 49· 401-405; More "Responsio," Aug. 16 55, 643-647.
The last letter from Descartes to More is a fragment found at Descartes's death; More's
"Responsio" is contained in a letter to Clerselier, who had sent him the fragment. The corre
spondence was included, with scholia by More, in his opera (II.2:227-271). On the exchange,
see Gabbey 198o:299n.27 and Garber 1992:145fT.
39· Impenetrability is explicitly ascribed to parts of matter in Le Monde: "But let us conceive
[matter] as a true body, perfectly solid [ ... ] ; so that each of its parts always occupies a part of
this space, so proportioned to its size that it could not fill a larger [space] nor shrink into a
smaller, nor allow another, so long as it remains there, to enter [that part of space]" (Monde6,
AT 11:33). Cf. also 6 Resp. no.9, AT 7=442).

Authenticated
Parts of Matter
doubtedly knew that some philosophers distinguished corporeal-impene

trable and tangible-extension from incorporeal extension. There are, I
think, two reasons why he did not devote more attention to the problem.
The first is that the Aristotelians, the primary intended readers of the Princi
ples, believed that an actually extended thing must exclude other such things
from its place. Descartes merely removes the qualification: interpenetration
is not just naturally, but absolutely, impossible.
The second reason is that Descartes may have overlooked impenetrability
in his efforts to explain hardness and solidity. 40 One indication is that
although durus occurs both in the argument that the nature of matter is
quantity and in the statement of the third law, 41 the hardness of the bodies
considered in the laws of motion cannot be the hardness that is rejected
from the essence of matter in 2§4; for then the outcome of collisions would
depend on how God allowed matter to interact with our senses. Descartes
appears not to have noticed the equivocation. It is not surprising, therefore,
that in beginning his response to More, he should reiterate that sensible
qualities are not in bodies (AT 5:268).
More had emphasized that "tangible" denotes the powerto affect the sense
of touch, and not the qualities we sense (see AT 5:2gg), and "impenetrable"
the power bodies have to exclude one another from their respective places.
Ultimately Descartes concedes the point. He turns to what looks like a
logical quibble. Impenetrability, he says, is not a defining feature of matter. A
body could be extended but not impenetrable, at least in our conception:
Tangibility and impenetrability in body is merely, as risibility is in man,

proper in the fourth mode, according to the common laws oflogic, [and]
not a true and essential differentia, which I contend consists in extension;
and on that account, just as 'man' is not defined by 'risible animal', but by
'rational [animal]', so body (I contend) is not defined by impenetrability,
but by extension. This is confirmed by the fact that tangibility and impen
etrability have a relation to parts, and presuppose the concept of division
or termination; but we can conceive of a continuous body of indetermi
40. On hardness ( durities), see PP 2§4, 11 ,54, AT 8/ I :42,48, 70. Impenetrability is a proper
ty of every continuous single part of matter; it consists in a body's not allowing others to enter
or occupy its internal place. Impenetrability is distinguished from hardness, the power of some
bodies to resist the movements of our own and to produce a characteristic sensation; and from
solidity (soliditas), the power of some bodies to resist division into parts. Hardness and solidity
are typically attributed to sensible, "macroscopic," bodies, and explained in terms of the
cohesion or connection of their parts; impenetrability is a property of the smallest naturally
existing parts of matter and is taken to be primitive or, in Cartesian physics, derived from
extension.
4I. 'Ita experimur dura qu.elibet corpora projecta, ciim in aliud durum corpus impingunt',
(AT 8/ I :65); "si duo tantiim corpora sibi mutuo occurrerent, eaque essent perfecte dura" (ib.
67).

Authenticated
nate or indefinite size, in which nothing except extension is considered.

(AT 5:26g) 42
To consider in an actual finite body only its extension is to abstract from

whatever boundaries the body actually has and from the larger space in
which it is contained. But the idea of impenetrability requires that we con
ceive at least two bodies, and thus not only their extension, but their being
"terminated," as actually divided parts of matter. Though the possibility of
division is clearly contained in the idea of extension, res extensa need not be
divided. God could have created one continuous body. Division and termi
nation are in that sense superadded to extension. Impenetrability, too, fol
lows only on the assumption that there are distinct bodies. But bodies in the
plural come about only when various motions are impressed on the parts of
matter. If, leaving motion out, we contemplate only extension itself, we
abstract from impenetrability as well. Extension alone defines body.
If, on the other hand, there are distinct bodies, they will indeed be impen
etrable. The mutual penetration of two parts of space ''would imply a con
tradiction" if no part were destroyed (AT 5:271 ). More is, quite reasonably,
not persuaded. He cannot imagine how one "part of space" could be trans
ferred to the place of another without mutual penetration: "[I cannot] con
ceive that if they were translated [from one place to another] some parts of
empty space would not absorb others, and coincide with them in their
depths, and penetrate one another" (AT 5:302).43 Descartes's answer
· merely elaborates his point: "if they are absorbed, then the middle part of
space is destroyed & ceases to be; but what ceases to be, does not penetrate
the other. "44
An individual body has its space or "internal place," which is part of the
space in general composed of the spaces of all bodies together. If we try to
imagine one body B penetrating another body C, we are trying to imagine
B's space overlapping Cs. But the putative region of overlap is one part of
space in general, one location. There can be nothing to indicate that it is, so
to speak, duplicated within itself. God could produce the appearance of
42. On the modes of prapria, see Coimbra In Log. 1:264. An accident proper in the fourth
mode belongs to all members of a species at all times and to them alone. Risibility in humans is
the stock example.
43· More's reply to Descartes's next letter shows what "absorption" might consist in: if one
body invaded another, the second body could immediately condense so as to give way to the first.
Condensation, which More distinguishes explicitly from diminution, is decrease in volume
without loss ofsubstance or quantity (AT 5:378). In later works More concluded that space, which
he took to be incorporeal and distinct from body, is immobile and indivisible (Ench. met. 1c8§9,
opera 11.1: 167f). The problem of explaining how one part of space could penetrate or absorb
another no longer arises.
44· AT 5:342. By 'middle part' Descartes means the part of one or the other of the two
bodies that would have overlapped the other: see Gabbey's note on the passage, AT 5:676, and
Garber 1992:147.

Authenticated
Parts of Matter
penetration by annihilating some ofB's space. But even he cannot bring it

about that one part of space in general should be the internal place of two
bodies.
Bernard Williams objects that "it is [...] legitimate to think of two geo
metrical solids, occupying (in one sense) the same space, or parts of them
doing so, as when one conceives of two polyhedra constructed on the same
base. "45 But suppose that from each polyhedron the part that does not
belong to the other is truncated. What we have left is two polyhedra whose
internal places coincide. But why just two? Why not infinitely many? 4 6
The rejoinder may seem to prove too much. Could one not show by
similar means that sets cannot intersect? In any case, even if bodies are
individuated by location as that term was defined above, locations can over
lap. An answer of sorts can be wrested, I think, from the discussion of
Descartes's principle of individuation in §g.2. If extension, divided or not,
has by nature a "topology" of locations within it, locations that do not exist
apart from extension altogether, but which are independent of the particu
lar bodies that compose it, then it is tempting to suppose that "parts" of
extension are individuated by their locations. But that opens the way to
Williams's objection. We must, despite Descartes's deceptive language, not
take bodies to be, or to be solely individuated by, their locations; or at least
not without adding the additional condition that the locations of distinct
bodies are disjoint.
But why should that be? Location is, in Descartes's terms, generic; bodies
are singular. The mode of existence of bodies is to have parts outside ofparts,
where 'outside' means not merely distinct in reason, but really distinct,
capable of subsisting separately by the absolute power of God. If we take
Williams's polyhedra to be locations, there is no difficulty in supposing that
they overlap; the points common to them can indeed be distinguished in
reason as belonging first to one, and then to the other. But suppose we take
them to be bodies, and consider the larger body that would result from
joining them. That body would have at least two real-and not just
imagined-parts that were not outside one another, and that could not
subsist separately.
That this does not again beg the question, I am not sure. The claim is that
bodies have just that mode of existence which entails that one body is really
distinct from another only if their locations are disjoint. That our concep
tion of location arises from our experience of bodies (both real and imag
45· Williams 1978:229; cf. Funkenstein 1986:74, Shea 1991:262.

46. See Aristotle's argument against separate space: "How, therefore, will tlle body ofa cube
and an equal void or place differ? And if two such things [could occupy one place], why should
any number of things not also be togetller?" (Phys. 4c8, 216b9ff; Coimbra In Phys. 4c9tex76,
2:56).

Authenticated
Bodies in Motion
ined) is no objection to defining, once we have the conception, a principle

of individuation for them in terms of it. But now the burden rests on the
claim about the mode of existence of bodies. It may be that Descartes, like
Ockham, took it to be transparently true. Later philosophers disagreed:
rather than try to derive impenetrability from extension, they added it to the
list of primitive properties of body, and from that concluded that the nature
of body does not consist solely in its being extended.
2. Vacuum. If the extension that constitutes space is identical to that
which constitutes body, there can be no space apart from body, no "sepa
rate" space. That is the short way with the vacuum. In the Principles Descartes
takes a more circuitous route. In 2§ 10-15 he establishes a vocabulary for
place and space, answering along the way an argument occasionally offered
on behalf of the existence of void space: that movement would be impossi
ble in a plenum. In 2§16 he argues that there can be no void space, and
after telling one of his genealogical tales to explain why some people think
there is, he tackles a version of the annihilation argument found in, among
others, Gassendi.47
The extension that constitutes body is, Descartes tells us, the same as that
which constitutes what we call "space," except in the way we conceive it.
When it is conceived as the extension of a particular body, it changes when
ever and however the body changes. When it is designated as space, or as
place, we do notjudge it to change when the body that first filled it changes.
Instead "it remains one and the same, so long as it remains of the same size
and figure, and preserves the same situation [situm] among certain bodies
outside it," bodies by which we "determine" it (2§10, repeated in 2§12). The
traditional example is the space inside a vessel, which whether filled with
water, air, or some other substance, we call the same space.
Descartes insists, therefore, on the continual reoccupation of the same
generic space by successive bodies: "But meanwhile we judge the extension
of the place, in which the stone was, to remain and to be the same, although
now the place of the stone is occupied by wood or water or air or whatever
other body" (AT 8/1:46). And thus that all motion is circular: "From what
was noted above, that all places are filled with bodies, and that the same
parts of matter are always proportioned exactly to equal places, it follows
that no body can be moved except through a circle, so that one body expels
another from the place it enters, and that body another, and so on until the
last [body] which enters the place left by the first at the very moment the
first leaves it" (2§33, AT 8/1:58). 48 The necessary succession ofbodies and
47· On the history of arguments about the vacuum, see Grant 1981.
48. Descartes's argument holds only for a finite universe constant in volume. Even then the
most one can infer is that for each particle C that enters the space previously occupied by the
first particle A, there will be a continuous path entirely contained within the volumes of

Authenticated
Parts of Matter
the circularity of motion yield a reply to an argument often urged on behalf

of the independence of space. Epicurean in origin, it is found in Gassendi,
among others. A body removed from its place "must extrude another body
out of its place, or become joint tenant with it and possess one and the same
place. Extrude a body out of its possession it cannot, because the Extruded
must want a room to be received into; nor can the Extruded dispossess a
third, that third expel a fourth, that fourth eject a fifth, &c. since the
difficulty sits equally heavy on all." Hence there can be "no beginning of
Motion. "49 Descartes has no trouble answering: the "beginning of Motion"
and its end are one.
More challenging is the argument from annihilation cited earlier. It will
be useful to preface Descartes's treatment of it with a passage from Patrizi.
The located [locatum] body, on entering or leaving a locus [i.e., an Aristo

telian place, defined as the inward surface of the bodies surrounding the
body whose place it is], always takes its volume with it. But the locating
[locans] body first received within its volume the volume of the located
body, and then released that body so that it could take its volume with it.
But whither had the volume of the locating body withdrawn while it was
receiving the located body? This cannot even be conceived. It is therefore
necessary that the locating body, while receiving the located body, depart
thence entirely and leave the space, which is fixed there, empty of itself so
as to be filled by the entering body. When the latter in turn withdraws
completely that space itself, which is immobile, must receive another
entering body. 5°
Patrizi is not arguing for the existence of a vacuum, but for the indepen
dence and immobility of space. The "volume," or what Descartes would call
the internal place, of a body that receives another must be withdrawn so that
the entering body can bring its volume. So too that body must withdraw its
volume when a second enters. The space that the second body enters can
not be the internal place of the first body, since that is gone; its own internal
place isn't there yet. So there must be an immobile space, belonging to
neither, which admits the volume of the two in succession.
Descartes admits the commonsense way of putting things: the space oc
cupied by the first body "remains" after it leaves. It is distinct somehow from
that body. But not from every body, as Patrizi would have it. There is "indeed
contiguous displaced particles from C to A. [-P Spinoza, ad loc.]

49· Charleton Physiologia 1q§1, p24; cf. Gassendi Philosaphia E.picuri syntagma pt2§1c1,
opera 3:11; Regis Cours 1pt2c11, 1:329, where the connection is made explicit.
50. Patrizi De spacio phys. 230 [62b-c]; I have substituted 'located' and 'locating' for Brick
man's 'bounded' and 'bounding'. On the role of this passage in Patrizi's argument against
Aristotle, see Henry 1979:562.

Authenticated
Bodies in Motion
no connection between the vessel and this or that particular body contained
in it," and so we designate the "generic extension [ extensio in gencre sumpta]"
within the vase by reference to the figure of the vase. Thus is commonsense
accommodated: the generic designation is what an Aristotelian would call a
denomination of the concatenated singular extensions of the bodies in the
vase by reference to their container. What cannot be designated is a space
apart from substance, "since, as has often been said, there can be no exten
sion of nothing [nihil nulla potest esse extensio]" (2§18, AT 8/1 :so).
That maxim is an instance of the "common notion" that "there are no
attributes of nothing" (1§52,AT 8/1:25). Armed with this notion-which is
not so common that it cannot be overlooked51 -Descartes answers the
argument from annihilation: "If it is asked what would happen if God re
moved every body that was contained in a vessel, and permitted no other to
enter the emptied-out place, the answer is that the sides of the vessel would
by that act become contiguous. For since nothing lies between the two
bodies, necessarily they touch each other; it is manifestly repugnant that
they should be distant, or that between them there should be distance, and
yet this distance be nothing: because all distance is a mode of extension, and
so cannot exist without substance" (2§18, AT 8/I:so). If there is no body
between the sides of the vase, there is no subjectum in which the distance
between them could inhere, and so there is no distance between them, just
as I have no hair color if I have no hair.
More calls the argument "slight," Gassendi a "paralogism."52 More ac
cepts the argu~ent that distance entails an intervening substance; but he
thinks that the substance is God. 53 Gassendi, more radical in his departure
from the logic of substance and accident, denies that one need attach every
accident to a substance; the space measured by the distance between the
sides of the vase is "incorporeal." Since More's objection rests on his views
about divine extension, I will consider it later. Here I note simply that for all
51. 'The difficulty in recognizing the impossibility of the vacuum seems to originate first of
all in the fact that we do not sufficiently realize that there can be no properties of nothing" (To
Arnauld 2gjul. 1648, AT 5:223).
52. AT 5:241; Gassendi Syntagma, Pars Physica~§Ibk2c1, opera 1:184b. Gassendi adds that it
is false that "every quantity, and so every dimension is a corporeal Accident," since the void has
"incorporeal" dimension and quantity: "insofar as a greater or lesser portion of it can be
designated; it admits of being measured, and ofall the comparisons that body itself has, insofar
as it is quantified."
53· AT 5:302. In the same passage, however, More begins with an argument to show that
extension could be of a "nonexistent" (thus rejecting the common notion applied by
Descartes). Suppose that God annihilated this world and much later created a new one from
nothing. "That between-worlds [ intermundium], or absence ofworlds, would have its duration."
A nonexistent, therefore, could have duration, "which is a sort of extension." So too a vacuum
could be "measured in ells or fathoms'. Nevertheless More concedes the point, since he thinks
that distance does have a subjectum, namely God.

Authenticated
Parts of Matter
three, distance is a mode not of the two sides of the vase jointly, but of some
one thing. 54
3· Divine extension and imaginary space. The question of God's extension,
or more generally of the manner by which he is said to be everywhere
present in the world, was bound up in Aristotelianism with that of the
existence of a void space, usually called "imaginary," outside the cosmos. 5 5
The link between the two was the immensitas of God, attributed on the basis
of scriptural and patristic texts. 56 Immensity, according to Suarez, denotes
that mode of existence of the divine essence by which God "can be inti
mately present to all things and bodies, not only present [i.e., presently
existing] but possible, even if they should be increased to infinity in multi
tude and magnitude. "57 That God is immense would seem to entail only
that space be potentially infinite, or that the capacity for receiving matter; as
Fonseca called it, should have no limit. But in the view of the Coimbrans, the
space in which God exists, though it is not a real or positive being, is actually
infinite: "In this imaginary space we assert that God exists in actu, not as in
some real being but by his immensity, which because the entire world can
not receive it necessarily exists also beyond the heavens in infinite spaces"
(In Phys. 8ooq2<4, 2:370; cf. Grant Ig81:16off). Those infinite spaces are
not "factitious, or dependent solely on a notion of the mind"; we call them
"imaginary" only because ''we imagine them in space by a certain proportion
to the real and positive dimensions of bodies." But neither are they real.
They are not "endowed with genuine three-dimensional quantity," since if
they were they could not receive bodies-which was, of course, Aristotle's
argument against a space distinct from bodies.
There is, in Aristotelian discussions, a decided reluctance to grant to
space any reality distinct from that of the bodies contained in it; and yet also
54· Arnauld suggests that the empty space could be measured, even though, being nothing,
it has no properties: "I ask therefore whether, while I considered the winejar separately, I could
not measure the cavity within and find out the distance in feet from the bottom, and what the
diameter of the cylindrical cavity was, and so forth?" (Arnauld To Descarles 3 Jun. 1648, AT
5: 191). Descartes answers only that it is wrong to refer the cavity of thejar to the sides of the jar,
"as if it were not diverse from them" (ib. 194).
55· Mersenne uses the term to denote the "privation of all the bodies that exist in the
world," which would occur if God ceased to conseiVe them, in which case "nothing would
remain but the space in which they exist, which is ordinarily called imaginary" (Mersenne Harm.
univ. 1pr4, 1:8). On divine immensity and imaginary space, see Grant 1981, pt.II.
56. See the list given by Suarez (De div. subst., opera 1:48r). Augustine's reply to the Man
ichaeans that before creation God was "in himself and not in another" is frequently cited; so
too a passage from DecivitateDei ( 11c5), in which God'sexistence in infinite space is compared
to his existence throughout eternity (cf. Grant 1981:113 and 328n.51; Coimbra In Phys.
8cwq2a2, 2:366; Suarez Disp. 30§7'!128, opera 26:104). Also frequently cited is the well-known
saying of Hermes Trismegistus that God is a circle "whose center is everywhere and whose
circumference is nowhere" (cf., e.g., Fonseca In meta. 5C15q9§4, 2:897A).
57· Suarez De div. subst. 2c2§1, opera 1:48; cf. Disp. 30§711 and 46, opera 26:g5, wg.

Authenticated
a marked desire not to limit God's spatial existence to the finite world of
Aristotelian cosmology. The mildest expression of the desire was to hold
that God is wherever any body is and could be. But the 'could be' implies a
where that is not now the place of any body. That implication puts pressure
on the identification of space with the concatenation of bodily places, and
thus on the conception of space as an accident of bodies rather than as a
container existing independent of them. Little by little, resistance to the
separation of space and body was diminishing even in the Schools.
Descartes, needless to say, did not countenance the existence of an extra
cosmic void. In Le Monde imaginary spaces are mere machinery to bring
onstage his own fabulous world. 58 In the Principles he alludes to them in
arguing that the world has no limits: "for wherever we suppose (fingamus]
those limits to exist, indefinitely extended spaces beyond those limits can
not only be imagined by us but also be perceived as truly imaginable, that is,
real; and so a corporeal substance, indefinitely extended, is contained in
them also" (PP 2§52, AT 8/1:52). The word 'indefinite' here points to a
problem: if space is infinite, and if it is not only substance but body, then an
actually infinite thing, the world, will have real and positive being. But the
existence of actually infinite quantity was thought by many to involve para
dox, and the assertion of an infinite world had brought upon more than one
philosopher the censure of theologians. The Aristotelians avoided the prob
lem by denying reality to infinite extracosmic space; More by supposing
infinite space to be God; Descartes by holding that the world is indefinite in
size. We can, he agrees, imagine no boundaries to space (and thus to body).
But we have no grounds for affirming that any substance other than God is
infinite. The idea of God entails infiniteness; the idea of extension does
not. 59 Prudence dictates that we say merely that there is no finite space such
that a larger space cannot be imagined. That larger space is not void but
corporeal. It is riot "imaginary" in the Aristotelian sense, nor indeed in the
more usual sense.6o
58. 'The philosophers tell us that these spaces are infinite, and we may as well believe them,
since it is they themselves who have made them. But in order that this infinity should not
hinder or bother us, let us not attempt to go to the end of it; let us enter only so far that we may
lose sight of all the creatures God has made five or six thousand years ago" (Descartes Monde6,
AT ll:32). On this passage, see Cavaille 1991:212-218.
59· On the distinction between the indefinite and the infinite, see McGuire 1983:88, g6r.
Descartes's view is in keeping with the Aristotelian prohibition against actually infinite created
things (cf., e.g., Suarez Disp. 35§3, 26, opera 26:44 7, and especially Coimbra In Phys. 3c8q 1 &
2, 1:37411).
6o. 'Such a space [beyond the supposed limits of the world] is, according to me, a true
body. Nor do I care that by others it is called 'imaginary', and thus that the world isjudged to be
finite; since I know from which pr(J!judicia the error comes" (To More 15 Apr. 1649, AT 5:345).
The praijudicium in question is that according to which a space in which nothing affects the
senses is judged to be void (see n.51). Descartes evades the issue raised by More (312), which is
that by excluded middle anything that cannot be infinite must be finite, as More exasperatedly

Authenticated
Parts of Matter
The force behind the doctrine of imaginary space is the view that God is
everywhere, as the Coimbrans put it, "by his essence, presence, and power."
Everywhere includes all imaginable places, in which he is present "according
to his real and infinite existence" (Coimbra In Phys. 8ooq2a4, 2:370), and
imaginable space is not merely boundless but infinite in extent. Only such a
space agrees with God's immensity. The keys, then, are divine presence and its
relation to space.
Suarez, subtle as always, distinguishes between "intrinsic" presence, which
is not a relation but a mode akin to intrinsic Ubi or "whereness" (Disp.
51§ 1113ff, see §4.2), and the "bearing" or position of bodies relative to one
another:
when a thing is said to be somewhere, or in some place, two [properties]

can be included in that locution: one is the intrinsic and absolute pres
ence [prmsentia] that the thing has here; the other is a bearing to some
thing else, which touches or is nearby, whether by virtue of quantity of
mass, or quantity of active virtue, or the relative presence of its substance
or quantity. Both [relations] are necessary in bodies, which are sur
rounded by a containing body, since [a body] is said to be located among
the bodies that circumscribe it, by way of quantitative contact, and has
moreover in itself its intrinsic presence, which we explain by its bearing
toward the center and poles of the world, and which remains the same
even if the circumscribing bodies change or flow, and indeed even if all of
them perish by divine power. 01
God's presence in imaginary space is intrinsic only; he cannot be related to

it as to something apart from him, since imaginary space is "nothing" (~ 3 7,
26:107). 6 2 The effect of Suarez's distinction, then, is to pry apart the con
nection between presence and extension. God can be present in every
actual and possible Ubi not only by his effects, but even by his substance, and
yet not be an extended thing.
For More the connection cannot be broken. God "seems to be an ex
tended thing; so too the Angels [... ] And indeed, that God in his way is
extended, I judge to be obvious from the fact that he is omnipresent and
points out (304).

61. Suarez DisjJ. 3o§7'l[35, Opera 26:106; cf. s•§4'lr33· 26:gg8. I use 'bearing' to translate
Suarez's habitudo, a term one of whose uses is to denote the relation of a body to what
surrounds it (Goclenius Lexicon 623).
62. Because of its importance in discussions of the Eucharist, prmsentia receives a detailed
entry in Goclenius's lexicon. He seems not, however, to notice the distinction Suarez makes
(Lexicon 855), and indeed takes prmsentia always to be a relation: "est existentia rei cum re, a
qua non distat" (857). But he does list the standard distinction appealed to by Descartes
between "effective" or "virtual" presence, which consists in having effects at a place, and
"substantial" presence, or spatial nearness (856).

Authenticated
occupies intimately the whole machine of the world and all its singular
parts. "63 Hence even if all the bodies contained in a vessel were removed,
and all others prevented from entering, it is false that there would be nothing
between the sides of the vessel: "divine extension would be interposed"
between them (241). God would be there, just as he is everywhere.
Divine extension, moreover, exists in the same way above the heavens as
below: "if we suppose the world to be enclosed within the visible stellar
heavens, the center of the divine essence and of his whole presence would
be repeated outside the starry heaven in the same way that we clearly con
ceive it to be repeated and reiterated within" (305). Descartes has earlier
granted that the "amplitude" of divine substance is infinite, if not its exten
sion (which is, "properly speaking," null) (275). More's argument is that
since the "amplitude" of God is to that of finite spirits as an infinite to a
finite sphere, we do have reason to believe that space, the sphere of God's
presence, is not merely indefinite but infinite (cf. 304.15-20), and that
body, if finite, is not space.
Descartes, in his response, first sets the terms of the debate. Extension in
the usual sense is found only in bodies: "By 'extended being' everyone
commonly understands something imaginable [...] and in that being one
can distinguish by imagination various parts of determinate magnitude and
figure, ofwhich each is not [the same as] the others in anyway" (270). But
God cannot be imagined at all, let alone imagined to have "parts with
determinate magnitudes and figures." If God is extended, it is not in the
usual sense, but merely by virtue of his power, which can be exercised
everywhere.Just as fire, "even in a glowing iron, is not ferrous," God's power,
even when it is applied to extended things, is not extended-nor is God
himself (270; cf. 343, 403).
Descartes later distinguishes "extension in power" ( extensio potentice) from
"extension in substance" ( extensio substantice). Extension in power does not
belong to the substance that has the power. Only extension in substance
does. To prove this, Descartes uses the standard test of separability: "if there
were no body, I would understand there to be no space with which the Angel
or God were coextensive" (342). Or again: 'The extension attributed to
incorporeal things is merely extension in power, not in substance; that
power, since it is only a mode in the thing to which it is applied, cannot be
understood, if the extended thing that coexists with it be destroyed, to be
extended" (343). The argument can, by analogy with similar arguments
concerning force (§8.3), be construed thus: for God's power, or God him
63. AT 5:238, 305. In More's view thatjudgment in no way entails that God is or has a body,
since body is tangible and impenetrable (240, 301).

Authenticated
Parts of Matter
self, to be extended, it would have to have really distinct parts, admit of

determination in figure and size, and movement in parts. But such modes
belong only to the bodies on which God acts, not to his power.
More takes up the last point. Descartes treats divine power as if it were
"situated outside of God." But any real mode is always "intimately within the
thing of which it is a mode." So divine power and essence, in order to act on
a thing, must "by some real mode be united with it" (379). What is united
with extended substances must be itself extended. God, the Angels, and the
human soul are all res extensce. 64 There is, in short, but one way to be at a
place, and that is to be extended there. Where for Descartes the distinction
is between extension in power, which is extension in name only, and exten
sion in substance, which entails the impenetrability, figurability, and
divisibility proper to bodies, for More the distinction is between two kinds of
genuine extension-the very proposition against which the argument of
Principles 2 § 11 is directed.
Even for More, however, incorpareal extension is physically inert, except
insofar as it designates a site of operation for spirits and a possible place for
bodies; only corporeal extension has all the properties that Descartes as
signs to extension simpliciter. Thus put, the difference looks slight enough
that More can at last agree that "on God's [... ] omnipresence there re
mains no disagreement between us" (AT 5:643).
But perhaps he was merely being conciliatory. Descartes, after all, retains
in large part the Aristotelian view. Like Suarez, he distinguishes between
prcesentia, which suffices to answer the question 'Where is it?', and occupatio,
the invariable concomitant of prcesentia for bodies. Only occupatio entails
having parts and thus divisibility, impenetrability, and tangibility. God's pres
ence, even in an extended substance, need not of itself entail that he is
extended. For Descartes as for most Aristotelians extension brings with it
the other properties of body; to admit that God is extended would be,
therefore, to admit that he is or has a body-the long-suppressed heresy of
David of Dinant.
More, like the other novatores, denies the inference from extended to cor
poreal. In doing so, he points toward the absolute space of Newtonian
physics. Amos Funkenstein has documented a line of thought leading from
certain pantheistic tendencies in the Renaissance to More and Newton. At
the end of that line God has a body-Newtonian space.65 One should
64. In the Enchiridium More offers many pages of argument against the Cartesians or
Nullibistm (Nowhere-Men). As in the correspondence with Descartes, the "fulcrum" of the
debate between him and the Nullibista! is the contention that every extended thing is material
(see More Ench. met. 1c27§s-6, opera II.1:309r).
65. Funkenstein 1986, c.2; see especially pp.g6, 116.

Authenticated
Bodies in Motion
note, however, that while More's God has spatial dimensions, space itself is
neither body (as Descartes thought) nor an accident of body (as Aristotle
thought). So if God can be said to acquire a "body" in the work of More, it is
a peculiar sort of body, penetrable and indivisible; its descendant might best
be regarded not as matter, but as the ether.

Authenticated
[ 10]
World without Ends
T he Cartesian transformation of physics is of the sort that used to

excite talk of incommensurability. There can be little doubt that it
was profound. Yet when one descends into details, one finds no
point at which an Aristotelian would not at least have understood
what Descartes was up to, however repugnant his results may have seemed.
In this concluding chapter I consider what becomes of finality in the new
world of res extensce.
The brief answer is: nothing. The Cartesian world is the world momen
tarily envisaged by Suarez, in which God simply allows natural agents to act
independent of him. In such a world "still the stone would fall down, fire
would generate its like," and so forth; there would be no final causality but
only "mere natural necessity" (Disp. 23§ 10~ 8, Opera 25:888). For Suarez the
strongest of reasons argue against that world; for Descartes the strongest of
reasons favor it. Not just final causes, but the directedness essential to the
Aristotelian concept of change, are absent-or rather they are positively
excluded. To hold that the nature of corporeal substance is constituted by
extension is to deny that corporeal substance could have active powers; to
hold that motion is rupture is, as we have seen, to refuse the Aristotelian
definition of motion as the actus of potentia. All this the Aristotelian would
have understood well: the truth of Cartesianism, like that of the more famil
iar atomistic physics of Democritus, Epicurus, and Lucretius, would entail
the falsity-the necessary falsity, according to Descartes-of what was most
basic to Aristotelian physics.
Of Descartes's arguments against the appeal to ends in natural philoso
phy, I will examine the two that attack that appeal directly. 1 The first is that
1. Descartes also argues that it is vain, even impious, to inquire into God's ends, which are

Authenticated
in attributing ends to the actions of natural agents the Aristotelian is treat

ing them as if they had souls. I have already considered to what extent the
accusation isjust (§6.3). The Aristotelians, far from promiscuously allotting
souls even to the rocks that fall, are quite guarded even in attributing to
higher animals recognition of their ends. Though finality-being directed
toward ends-occurs everywhere in nature, final causes need be adduced
only in explaining the actions of rational agents. What remains to be con
sidered in Descartes's argument, then, is why, if not through ignorance or
malevolence, he came to misrepresent the Aristotelians' position.
The second argument is, from the standpoint oflater natural philosophy,
the more interesting. Descartes claims he has no need of final causes or of
ends in his physics. The directedness of natural change, which so struck the
Aristotelians, is illusory; the stability of nature, the cooperation of efficient
causes in producing a single result, the fitness of the parts of nature to one
another and especially to human needs, are all just the working out of mere
natural necessity. The cosmogony of the Principles is an implicit polemic
against any claim that the present state of the world needs any other ac
count than that provided by bodies moving, dividing, and uniting according
to law. From the Aristotelian point of view, that would be disastrous. But
proponents of what came to be called "design" could accommodate natural
necessity so long as the natural world-however it came about-could be
interpreted as fulfilling the intentions of a benevolent creator. If the arrow
strikes the bull's-eye, the necessity with which it is impelled is no argument
against its having been aimed there; it testifies, rather, to the ingenuity and
power of the archer.
But the role of design in Descartes's physics must be restricted still fur
ther. The laws of nature themselves are founded not, as Leibniz believed, on
God's benevolent choice of the best possible world, but on his immutability.
One cannot argue, therefore, that God has chosen those laws so that natural
agents, acting in accordance with them, will produce the ends envisioned by
him.
The implication, perhaps not fully recognized by Descartes, is that the
course of natural events has no end. The most shocking expression of this
to Leibniz, at least-is the hypothesis entertained in the third part of the
Principles: the world, Descartes says, may be supposed to proceed through all
physically possible states. None of those states, one may infer, can be re
garded as a state toward which the world tends. But that, in Leibniz's view,
would "destroy providence."
incomprehensible and beyond our competence to pass judgment on (see PP 1§28, AT

8/1:15). That criticism, which was not unknown to the Aristotelians, challenges the utility of
teleological reasoning. But a useless project need not be an incoherent project. The two
arguments I discuss seem to me more profound.

Authenticated
World without Ends
Here the novelty, or the modernity, of Descartes's natural philosophy

reveals itself most fully. The vast enterprise of natural theology that emerged
near the end of the seventeenth century, continues into the nineteenth
century, and even now reverberates in traditional accounts of the "rise of
man," can from this standpoint be regarded as an endeavor to repair the
damage done by Descartes, or if not by him then by those who, like Spinoza,
were taken to have worked out the hidden implications of his philosophy.
1. Little souls. Several times in his writings Descartes accuses the Aristo
telians of attributing souls to corporeal substances. Sometimes the basis of
the accusation is what he takes to be imaginary distinctions; in Principles 2§g,
for example, the Aristotelian claim that quantity and substance are really
distinct is said to rest on the false attribution of a confused idea of incor
poreal substance to bodies. 2 Sometimes the basis is the attribution of ends
to the actions of bodies. In the sixth Replies Descartes recounts how he had
come to think ofweight as a ''real quality" of bodies, only to realize later that
he was thereby mistakenly conceiving of it as soul: "But that the idea of
weight was taken partly from the idea I had of mind, is shown principally by
the fact that I thought gravity carried bodies toward the center of the earth,
as if it contained in itself some knowledge of the center. For that clearly
could not occur without thinking, and there can be no thought but in a
mind. Nevertheless I attributed certain things to weight that could not be
understood in the same way as mind-that it is divisible, measurable, and so
forth" (6 Resp., AT 7:442). Whether this is a true confession is unimportant.
The significant point is that Descartes believes that to act toward an end
(here the attainment of the natural place of heavy bodies) presupposes the
capacity to recognize that end, and so also a "little soul," as he says to
Mersenne (AT 3:648).
We have seen that the Aristotelians believed that final causes act only on
rational agents (§6.3). Not even animals can be said genuinely to recognize
the ends toward which, nevertheless, they undoubtedly act. They, and all
inanimate things, are acted upon by final causes only by virtue of being the
instruments of rational. agents. Where the Aristotelians differ from
Descartes is in refusing to restrict finality to the actions of agents who recog
nize the ends toward which they act. To understand the downward motion
of a heavy body as a natural change is, as we have seen, to subsume it under a
schema in which it can be seen as the step-by-step actualization ofthe body's
potentia, whose completion or actus consists in arriving at its natural place.
The presence in a stone of heaviness links the stone with something else
its natural place. As I argued earlier, the natures of things in the Aristotelian
world are bound together through relations of finality: in the nature of the
2. On this argument, see Gilson 1984:168-173 and Garber 1992:97-101.

Authenticated
Bodies in Motion
seed, that of the plant is implied; in the natures of lower animals, those of
higher animals are implicated if the lower exist for the sake of the higher.
One reason, I think, that Descartes finds the "little souls" story so persua
sive is that he can conceive such relations only by way of the one relation
that can connect the natures of distinct things: intentionality. His idea of the
sun, he writes in the second &plies, "is the Sun itself existing in thought
not, indeed, formally, as it exists in the heavens, but objectively, i.e. in the
way that things usually exist in thought" (AT 7: 102). Having the idea of the
sun establishes a relation between me, in whom that idea is a mode, and the
thing in the heavens-between two things, in other words, that othetwise
are really distinct.
When a stone falls, its relation to the terminus of its motion could only be
of two sorts. Either the terminus is the efficient cause of the motion, which is
evidently false, or else the stone is capable of having the terminus exist
objectively in it-which is to say, is capable of thinking of the terminus. But
that too is false. So there is no relation between the stone's motion and the
terminus (now a terminus in name only) of its motion, save that it happens to
be pushed in the direction of the terminus by subtle matter-which cannot,
of course, be thinking of the terminus either. Despite what the senses seem to
show us, we must not think of the actions of inanimate things as directed.
Even the dog that leaps toward food is pushed by its animal spirits, which are
in turn caused to move by light reflected into its eyes.
2. Doing without finality. The replacement of final by efficient causes
ideas where rational agents are concerned, blind pushes where natural
agents are concerned, was already occurring among the Aristotelians. Jean
Buridan, notably, had argued that in animals the efficient cause suffices;
Suarez, as we have seen, has to exercise his subtlety in order to show that
ends envisioned by rational agents are not merely efficient causes.
Descartes's views here are rather the culmination of a trend than a radical
departure.
One further example is worth noting. The Coimbrans instanced the sta
bility of the elemental composition of the world as evidence for what in §6.1
I called "collective" ends. The processes by which the elements are trans
formed into one another do not yield a bleak and uninhabitable world of
fire alone, or water alone. Instead we have a world in which all four elements
are present in due proportion and in suitable places, a world in equilibrium.
Descartes's cosmogony yields, so he asserts, both the right proportions of
his three elements-enough to form worlds like the earth on whose surface
water is found and around which circulates air like the air we breathe. In
particular-this is my example-the motions of the particles originally
given motion by God, which Descartes supposes for simplicity's sake to have
been uniformly sized small cubes, eventually produce the spherical particles

Authenticated
World without Ends
of the second element. Second-element particles are stable in the sense that
their size and shape is, on the whole, maintained. They are of a size such
that collisions with other particles are no more likely to increase than to
decrease that size; and because they have no knobs or branches that could
easily be broken off, their shape is also unlikely to be altered. They are in a
kind of mechanical equilibrium with their surroundings.
To the naive eye, it might seem as if the motions of those and other
particles were conspiring, as it were, to keep the particles of the second
element spherical. The collective end of those motions, one might say, is to
preserve the shape of the second elements. But-Descartes would reply
there is no need to suppose that the particles move on account of the shapes
that result from their motion. The preservation of that shape can be ex
plained entirely in terms of the possible actions of efficient causes. One
could, of course, say that God has given the parts of matter just the proper
ties that will bring about the preservation of second-element particles. But
that claim does no work in physics.
It is when one turns to nature in the large that the thoroughness with
which Descartes rejected finality in physics becomes most apparent.
There is, first of all, the hypothesis that so incensed Leibniz. In the third
part of the Principles Descartes attempts to build the universe from the
ground up. Though he casts his account into hypothetical form so as to
avoid controverting the book of Genesis, it is, I think, seriously meant. The
story begins, as origin myths often do, with chaos-an "entire confusion of
all the parts of the universe" ( PP 3§47). Referring his reader to the un
published Le Monde, he says that he has shown elsewhere how the order that
exists at present in the world could have arisen from chaos. In the Principles,
however, because "it does not agree so well with the sovereign perfection of
God to make him the author of confusion rather than order, and because
the notion we have of chaos is less distinct," Descartes starts not from chaos
but from an initial state in which "all the parts of matter are equal among
themselves, both in size and in movement."
That assumption may seem arbitrary. Descartes defends it in the following
terms: "It matters very little how I suppose matter to have been disposed at
the beginning, since its disposition must afterwards be changed, according
to the laws of nature; and one can scarcely imagine any [disposition] from
which one could not prove that by these laws it must continually change,
until finally it composes a world entirely similar to this one [ ... ] For since
these laws cause matter to take on successively all the forms it is capable of, if
one considers all those forms in order, one will be able finally to arrive at
[the form] that exists at present in the world" (PP 3§47). The universe
proceeds, in other words, indifferently through all the states consistent with
the quantity of motion initially given to it by God.

Authenticated
Bodies in Motion
Leibniz was horrified. In his polemic against Cartesianism he returns to

the passage I have quoted several times. If the world moves successively
through all the states it is capable of, then Descartes's God cannot be said to
will the good, or to be, as the Aristotelians argued, the ultimate end of the
world. Descartes's God is instead nothing other than the "necessity or prin
ciple of necessity acting in matter" (Leibniz Ph. Schr. 4:2gg)-in short, the
God of Spinoza.
Leibniz was almost right. What the passage from the Principles precludes is
any appeal to divine goodness to explain either the succession of states of
the world or why we find the world in one state rather than another. If God
brings about, or permits, every possible state indifferently, then the only
object of which we could say that God willed it because it was good is the
entire temporal succession. Perhaps it was better for God to have created
the world than not; but beyond that there is nothing more to be explained
by appealing to his goodness.
There are other passages, notably in the fourth Meditation, where the
goodness of God seems to have a more substantial role. Mter explaining
that our susceptibility to error arises from the disproportion between our
will and our understanding, Descartes notes that God could nevertheless
have prevented us from erring if he had enlarged our understanding so as
to be sufficient to our needs, or else impressed upon our will a law forbid
ding us to judge when we do not have clear and distinct ideas. Why, then,
didn't God do so? It may be, Descartes answers, that the imperfection of a
part of the world is necessary to the greater perfection of the whole.
Though the passage is sometimes taken to be an anticipation of Leil:>
nizian optimism, there are reasons not to do so. The first is that Descartes
insists that the good is dependent on God's will:
Ifwe attend to God's immensity, it is manifest that nothing whatsoever can

exist that does not depend on him: not only no subsistent thing, but also
no order, no law, nor any reason of truth or goodness; since otherwise [
... ] he would not have been entirely indifferent in creating what he
created. If a reason of goodness [ratio bonz1 had preceded his pre-ordering
[of the world], it would have determined him to do what is best; but the
contrary is true, because he determined himself to make the things that
now exist, they were for that reason, as it says in Genesis, "exceedingly
good'-the reason for their goodness, that is, depended on his willing
that they be made thus (6 &sp., AT 7=435-436; cf. 7:432 and To Mersenne
15 Apr 1630, AT 1:145 and subsequent letters to Mersenne in 1630)
Although one may then say that God, since his will does not change, will
maintain the world according to the standard that was created with it, the
argument of the fourth Meditation has to do with a world-not the actual

Authenticated
World without Ends [397]
world-in which human nature is other than it is. To borrow the analogy
with royal legislation which Descartes draws on: if justice depends on the
decrees of the head of state, then comparison of the laws of one state to
those of another is idle, since there is no independent standard. So it is vain
for me to complain about my error-prone condition in this world, and to
wish that I had a more perfect nature, not because my condition is necessary
to the greater perfection of the whole, but because such comparisons have
no basis. When Descartes says, therefore, that "we should take no reasons,
where natural things are concerned, from the end which God or nature
proposed to himself in making them, because we should presume that we
are privy to his counsels" (PP1§28, AT 8/I:15), the point is not just that
God's ways are mysterious, but that the goodness of the world in no way
precedes its existence; it therefore explains nothing.
The second reason not to take Descartes to anticipate Leibniz is that the
laws of nature are derived not from any "reason of the good" but from the
immutability of the divine will. Immutability, like the steadfastness that
Descartes recommends in the morale provisoire of the Discourse, is a formal
condition on God's will; it is compatible with any content. Unlike Leibniz's
God, who orders the world so that the simplest means will yield the greatest
variety of beings, Descarte;5's God can order the world as he pleases, so long
as he conseiVes that order afteiWard. Although it is true that for Descartes
the simplicity of a hypothesis is a reason to favor it over others that explain
the same phenomena, hypotheses chosen according to that criterion can be
only morally certain. They are such that, in other words, we know that we
shall never go wrong in adopting them. But we cannot argue that God must
operate according to that criterion.
In Aristotelianism, the ordering of the actions of creatures toward their
own good, toward ours, and toward that of nature as a whole provides the
basis for a natural morality. Suarez, as we have seen, argues from the propo
sition that woman is ordered to man to the proposition that a husband has
dominion over his wife (§6.2). By their nature humans have, in fact, domi
nion over the entire corporeal world, since they are its most perfect mem
bers. The usus of the world is ours; it would be a kind of disruption of the
natural order for us not to exercise our dominion.
In the Cartesian world, there is no such order, and therefore no such law.
There is indeed a kind of fitness of the human body to the human soul; that
the body is such as to be useful to us is obvious. Perhaps by extension one
could say that there is a kind of fitness of the natural world to the needs of
beings like us. But one cannot truly say of the human body, or of any other
corporeal things, that it has us as its end. Bodies have no ends; what governs
their behavior is not individual, collective, or cosmic ends but a set of laws
whose ground is the purely formal condition that whatever God wills he wills

Authenticated
Bodies in Motion
always and everywhere. The relation of our nature-that we are thinking,

willing things-to the natural world amounts to nothing more than that we
have the power to initiate motion in it, a power no corporeal thing has, and
that God, by acting on corporeal things, can act on us. The Aristotelian
hierarchy of created things collapses into a single step: from soul to body. ·
The only morality, it would seem, to be gleaned from the natural world so
understood consists in the unique admonition: do what you will. An Aristo
telian who genuinely conceived such a world would not find it incompre
hensible. But he would find it unheimlich.

Authenticated
Bibliography
Note: Essays in collective volumes are cross-referenced to the volume in which they
appear (e.g., Filosojia della natura).
Primary Sources
Abrade Raconis, C. F. Tertia pars philosophia!, seu physica. Lyons: Iremei Barlet, 1651.
(Orig. pub. 1617, with twelve subsequent editions, mostly in Paris, through 1651.
Also in Manuscripta list 101 reel 3 pt.6.)
.tEgidius Romanus. Commentaria in octo libros phisicorum Aristotelis. Venice, 1502. (Facs.
repr. Frankfurt: MineiVa, 1968.)
Agrippa, Cornelius. De occulta philosophia. In his opera , v.1. Lyon: Beringi, s.d.
[l6oo]. (Facs. repr. with intro. by Richard Popkin. Hildesheim: Olms, 1970.)
Alain de Litle (Alanus de Insulis). Anticlaudianus. Ed. R. Bossuat. Paris:]. Vrin, 1955.
(Written ca. 1150.)
Albertus Magnus. De ca!lo.
Albertus Magnus. opera omnia. Aschendorff: Monasterii Westfalorum, 1951-.
Albertus Magnus. Physica. Ed. P. Hossfeld. Aschendorff: Monasterii Westfalorum,
1987. (In opera [Inst.] 4, pt. I.)
Alembert, Jean le Rond d'. Discours preliminaire. In Diderot Encyclopedie 1:i-xlv.
Aristotle. De anima. Trans. and notes by R. D. Hicks. Cambridge: Cambridge Univer
sity Press, 1907. (Repr. Amsterdam: Hakkert, 1965.)
Aristotle. De ca!lo et De mundo. Trans. J. Tricot. Paris: Vrin, 1949.
Aristotle. Categories. Trans. and notes by J. Tricot. Nouvelle edition. Paris: Vrin, 1977.
(Aristotle's Organon 1.)
Aristotle. De generatione et corrnptione. Trans. and notes by C. J. F. Williams. Oxford:
Clarendon, 1982.
Aristotle. De la generation et de la corrnption. Trans. and notes by J. Tricot. Paris: Vrin,
31971.
Aristotle. Organon Graece. Ed. and comm. by Theodore Waitz. Leipzig: Hahn, 1896.
Aristotle. Physics. Trans. and intro. by Philip H. Wicksteed and Francis M. Cornford.
Cambridge: HaiVard, 21957. (Loeb Classical Library.)

Authenticated
Bibliography
Aristotle. Physics, Books Ill and N. Trans. and notes by Edward Hussey. Oxford:
Clarendon, 1983.
Aristotle. Physique. Ed. and trans. Henri Carteron. Paris: Les Belles Lettres, 1926.
Arnauld, Antoine, and Pierre Nicole. La logique ou !'art de penser. Ed. Pierre Clair and
Fran~ois Girbal. Paris: Presses Universitaires de France, 1965. (Based on the 5th
ed., Paris, 1683. Orig. pub. Paris, 1662.)
Arriaga, Rodericus de. Cursus philosophicus. Antwerp: Balthasar Moretus, 1632. (Ten
subsequent editions, including four in Paris, through 1669. Also in Manuscripta
list 84 reel 7.)
Augustine. Confessionum libri XIII. Ed. Lucas Verheijen. Turnhout: Brepols, 1981.
(Corpus Christianorum, Series Latina 27.)
Augustine. De genesi ad litteram. Ed. Joseph Zycha. Prague: Academia Litterarum
C;esare;e, 1894. (Corpus scriptorum ecclesiasticorum latinorum 28.)
Augustine. De trinitate. Ed. W. J. Mountain. Turnholt: Brepols, 1968.(Corpus
Christianorum, Series Latina so.)
Averroes [Ibn Rushd]. Aristotelis de physico auditu [. . .] cum Averrois Cordubensis variis
in eosdem commentariis. In opera V-{-
Averroes [Ibn Rushd]. Aristotelis metaphysicorum [. . . ] cum Averrois Cordubensis in
eosdem commentariis, et epitome. In Operav.8. (Includes Theophrastus, Metaphysicorum
tiber; M.A. Zimara, Solutiones contradictionum.)
Averroes [Ibn Rushd]. Aristotelis opera cum Averrois commentariis. Venice, 1562. (Facs.
rep., Frankfurt am Main: MineiVa, 1962.)
Avicenna [Ibn Sina]. Opera in lucem redacta. Venice, 1508. (Facs. repr. Frankfurt am
Main: MineiVa, 1961.) (Sufficientia is the Latin title given to a collection of eight
treatises on natural philosophy known in Arabic as al-Shifa '.)
Bacon, Francis. De augmentis scientiarum. 1623. In Works 1:415-837·
Bacon, Francis. Works. Ed. James Spedding, Robert Leslie Ellis, and Douglas Denon
Heath. London: Longman, 1858. (Facs. repr. Stuttgart-Bad Cannstatt: Friedrich
Frommann, 1963.)
Basso, Sebastian. Philosophice natura/is adversus Aristotelis libri xii. Geneva, 1621.
Bayle, Pierre. Dictionnaire historique et critique. Amsterdam, 5 17 40. (Selected articles
rep. in CEuvres diverses, Supplement 1, pt.1-2.)
Bayle, Pierre. CEuvres diverses. The Hague, 1727. (Facs. repr. with intro. by Elisabeth
Labrousse. Hildesheim: Olms, 1982.)
Beeckman, Isaac. journal tenu par Isaac Beeckman de r6o4 ii r634· The Hague: M.
Nijhoff, 1939.
Boethius. De duabus naturis. Paris: Garnier/Migne, 1877 (Patrologia latina, series
latina 64.)
Boyle, Robert. Selected Philosophical Papers of Robert Boyle. Ed. and intro. by M. A.
Stewart. Manchester: Manchester University Press, 1979. (Includes 'The origin of
forms and qualities" [ 1666], 1-96; "About the excellency and grounds of the
mechanical hypothesis" ( 1674), 138-154; "A free inquiry into the vulgarly re
ceived notion of nature" ( 1686), 176-191.)
Boyle, Robert. The Works of the Honourable Robert Boyle. Ed. Thomas Birch. London,
1744·
Bruno, Giordano. Opera latine conscripta. Ed. F. Tocco, H. Vitelli. Florence: Le Mon
nier, 1879-1891. (Facs. repr. Stuttgart-Bad Cannstadt: Friedrich Frommann,
1962.)
Buridan, Jean. Acutissimi philosophi reverendi Magistri johannis buridani subtillissime
questiones super octo phisicorum Iibras Aristotelis ... [Paris, 1509]. (Facs. repr. Frank

Authenticated
Bibliography
furt am main: MineiVa, 1964, under the title Kommentiir zur Aristotelischen Physik.
Composed ca. 1328.)
Buridan,Jean. Expositio de anima. In Traite de l'iime 5-163. (Patar dates the work to
1337-)
Buridan, Jean. In Metaphysicen Aristotelis Qy,O!stiones ... Paris, 1518. (Facs. repr.
Frankfurt am Main: MineiVa, 1964.)
Buridan,Jean. Le traite de l'iime de jean Buridan. (De prima lectura.) Ed. and intro. by
Benoit Patar. Louvain-La-Neuve: Editions de l'institut superieur de philosophie,
1991.
Calicidius. Timfl!Us a Calcidio translatus commentarioque instructus. Ed. J. H. Waszink.
London/Leiden:Warburg Institute/E.]. Brill, 1962.
Capreolus, Joannes. Commentaria in quattuor libros sententiarum ... Venice: Scotus,
1483.
Charleton, Walter. Physiologia Epicuro-Gassendo-Charltoniana: or a fa/nick of science
natural, upon the hypothesis ofatoms. London: Thomas Newcomb, 1654. (Facs. repr.
with intro. by Robert H. Kargon, New York: Johnson Reprint, 1966.)
Chauvin, Stephanus. Lexicon philosophicum. Leeuwarden: F. Halma, 1713. (Facs. repr.
with intro. by L. Geldsetzer. Dusseldorf: Stem-Verlag Janssen, 1967. The first
edition was published in 1692.)
Coimbra [Collegium Conimbricensis]. Commentarii Collegii Conimlnicensis e Societate
Iesu: in universam dialecticam Aristotelis Stagiritfl!. Cologne, 1607. (Facs. repr. with
intro. by W. Risse, Hildesheim:Olms, 1976.)
Coimbra [Collegium Conimbricensis]. Commentarii Collegii Conimlnicensis {. . .] in
octo libros physicorum Aristotelis. Coimbra, 1594. (Facs. repr. Hildesheim: Olms,
1984. N.B.: in this edition the page numbers 331 to 352 are used twice. I have
marked pages from the second series with an asterisk.)
Cordemoy, Gerauld de. CEuvres philosophiques. Paris: Presses Universitaires de France,
1968. (The Discernement du corps et de l'iime was first published in its entirety in
1666; the text in the CEuvres is that of the 4th edition [ 1704].)
Cusanus, Nicolaus. Idiota de mente. In Wemev.5. (Written in 1450.)
Cusanus, Nicolaus. Weme: Neuausgabe des Strassburger Drucks von I 488. Berlin: Wal~er
de Gruyter, 1967.
David ofDinant. Quaternulorumfragmenta. Ed. M. Kurdzialek. Warsaw, 1963. (Studia
Mediewistyczne 3).
Descartes, Rene. La Dioptrique. Published with the Discours, 1637.
Descartes, Rene. Discours de la methode. Leyden: jean Maire, 1637. In AT 6.
Descartes, Rene. Meditationes de prima philosophia, in qua Dei existentia et AnimO! immor
talitas demonstratur. Paris: Michel Soly, 1641. (The second edition has the subtitle:
In quibus Dei existentia, et animO! humanfl! a corpore distinctio, demonstrantur. Amster
dam: Louis Elzevier, 1642.)
Descartes, Rene. Le Monde. In AT 11. (This work, which is sometimes called the Traite
de la Lumiere, was first published in 1664 in Paris.)
Descartes, Rene. CEuvres. Ed. Ferdinand Alquie. Paris: Gamier, 1963-1973.
Descartes, Rene. CEuvres de Descartes. Ed. Charles Adam and Paul Tannery. Nouvelle
presentation. Paris: Vrin, 1964.
Descartes, Rene. Les principes de la philosophic. Trans. Claude Picot. Paris: Henri le
Gros, 164 7. In AT g. (For a discussion of the status of the translation as a Cartesian
text, see the "Avertissement" in this volume. Adam concludes that the passages
where the French adds substantially to the Latin are by Descartes himself.)

Authenticated
Bibliography
Descartes, Rene. Principia philosophill!. Amstedam: Elzevier, I644. In AT 8, pt. I.

Descartes, Rene. Rigles utiles et claires pour la direction de l'esprit en la recherche de la verite.
Trans.Jean-Luc Marion, mathematical notes by Pierre Costabel. (Annotated trans
lation, following the "lexicon of Descartes before I63 7," of the Crapulli edition.)
Descartes, Rene. &gulll! ad directionem ingenii. Ed. Giovanni Crapulli. The Hague:
Martinus Nijhoff, I966. (Latin text collated from the editio princeps [Amsterdam,
I 70 I], the Hanover manuscript used by Leibniz, and the first Dutch translation
[Amsterdam, I684].) See also AT IO. I have used Crapulli's text but the page
numbers are according to AT.
Diderot, Denis, ed. 1<-ncyclopedie ou Dictionnaire raisonnee des sciences, des arts et des
metiers. Paris, I 7 5 I .
Dionysius Areopagitica. De coelesti hierarchia. In opera II9-370.
Dionysius Areopagitica. De divinis nominibus. In opera 586-996.
Dionysius Areopagitica [pseudonym]. opera omnia qull! exstant. Latin and Greek
paraphrase by Georgius Pachymeres, notes Balthasar Corderius, intra. D. Le
Nourry. Paris: J.-P. Migne, I889. (Patrologia Gneca 3.)
Dupleix, Scipion. La physique, ou science des choses naturelles. Ed. Roger Ariew. Paris:
Fayard, I990. (Ariew's text is based on the edition of I64o [Rauen: Louys du
Mesnil]. The first edition was published in I603.)
Du Plessis d'Argentre, Charles. Collectio judiciorum de novis erroribus ... Paris: An
dneas Cailleau, 1736. (Facs. repr. Brussels: Culture et civilisation, I963.)
Durandus, de Sancto Porciano [Durand de St. Poun;:ain]. In libros Sententiarum.
Venice, I57L
Eustachius, a Sancto Paulo. Summa philosophica quadripartita ... Paris, I620. (I st ed.
I609.)
Fonseca, Petrus. Commentariorum Petri Fonsecll! Lusitani [. . .] in Libros metaphysicorum
Aristotelis . .. Cologne: Lazarus Zetzner, I615. (Facs. repr. Hildesheim: Olds, I964.)
Galilei, Galileo. Galileo's Early Notebooks. Trans. and comm. by William A. Wallace.
Notre Dame: University of Notre Dame Press, I977· (Wallace dates the text to
I590.)
Galilei, Galileo. Le opere di Galileo Galilei. Ed. Antonio Favaro. Florence: G. Barbera,
I890-I909.
Gassendi, Pierre. Exercitationes paradoxica adversus Aristoteleos. In opera 3· (Book I:
Grenoble, I624. Book 2 was published posthumously in the opera.)
Gassendi, Pierre. opera omnia . .. Lyon, I658.
Gassendi, Pierre. Syntagma philosophicum. In opera I-2.
Goclenius, Rudolphus. Lexicon philosophicum . .. Frankfurt: Matthias Becker, I6I3.
(Facs. repr. Hildesheim: Olms, I964.
Gregory of Rimini [Gregorius Ariminensis]. Lectura super primum et secundum Senten
tiarum. Ed. A. Damasus Trapp and Venicio Marcolino. Berlin: Walter de Gruyter.
(Spatmittelalter und Reformation, Texte und Untersuchungen 6-I2 .)
Hamilton, William. Lectures on Metaphysics and Logic. Vol. I: Metaphysics. Boston:
Gould and Lincoln, I859·
Hobbes, Thomas. De corpore. In opera I.
Hobbes, Thomas. Critique de De Mundo de Thomas White. Intra., ed., and notes by
Jean Jacquot and Harold Whitmore jones. Paris:]. Vrin, I973· (The editors date
the composition of this work to I643; see p.44-45.)
Hobbes, Thomas. opera philosophica qull! latine scripsit omnia ... Ed. W. Molesworth.
London:]. Bohn, 1839.

Authenticated
Bibliography
Huygens, Christiaan. De motu corporum ex percussione. In CEuv. 16:1-168. (The first

draft of the De motu was written in 1652; the definitive manuscript, completed
after 1673, was first published in the Dpuscula postuma in 1703.)
Huygens, Christiaan. CEuvres completes. The Hague: Martinus Nijhoff, 1888-.
John of St. Thomas. Cursus philosophicus Thomisticus. Ed. P. Beato Reiser. Turin:
Marietti, 1930-1937. (Orig. pub. Rome, 1637-1638, with numerous subsequent
editions.)
John of St. Thomas. Naturalis philosophi(£ . .. Ed. P. Beato Reiser. Turin: Marietti. Pt.1
and 3: 1933. Pq: 1937. (Part II of the Cursus. Orig. publ. separately, 1633-1635.)
Kepler,Johannes. Gesammelte Werke. Ed. W. von Dyck and Max Caspar. Munich: C. H.
Beck, 1937-.
Kepler, Johannes. Strena seu de Nive sexangula. Frankfurt, 16ll. In Werke 4:259-280.
Le Bossu, Rene. Parallile des principes de la physique d'Aristote et de celle de Rene Descartes.
Paris: Vrin, 1981. (Facs. of 1674 ed.)
Leibniz, Gottfried Wilhelm. Discours de metaphysique et correspondance avec Arnauld.
Intro. and notes by G. LeRoy. Paris: Vrin, 4 1984.
Leibniz, Gottfried Wilhelm. PhilosophischeSchriften. Ed. Gerhardt. Berlin, 1875-1889.
Leibniz, Gottfried Wilhelm. Siimtliche Schriften und Briefe. Ed. Preussischen Adademie
der Wissenschaften. Darmstadt: Otto Reichl, 1923-. (Reihe 6 is the Philosophische
Schriften.)
Locke, John. An Essay Concerning Human Understanding. Ed. P. H. Nidditch. Oxford:
Oxford University Press, 1975; repr. with corrections, 1979.
Lucretius. De rerum natura. Ed. Martin Ferguson Smith, trans. W. H. D. Rouse. Cam
bridge: HaiVard University Press, 1975.
Maignan, Emanuel. R P. Emanuelis Maignan Tolosatis ordinis Minimorum ... Cursus
philosophicus . .. Lyons:Joannes Gregoir, 1673.
Mersenne, Marin. Correspondance du P. Marin Mersenne, religieux minime. Ed. Comelis
de Waard etArmand Beaulieu. Paris: PUF and CNRS, 1932-1988.
Mersenne, Marin. Harmonie universelle contenant la theorie et la pratique de la musique.
Paris: Sebastien Cramoisy, 1636.
Mersenne, Marin. L'impiete des deistes, athees, et libertins de ce temps, combatui, & renversee
de point en point par raisons tirees de la Philosophie, & de la Theologie. Paris: Pierre
Bilaine, 1624. (Repr. Stuttgart-Bad Cannstatt, 1975. The second part, published
under the title L'impiete des Deistes, et des plus subtils Libertins decouverte . .. , is not
included in the reprint.)
Mersenne, Marin. Les nouvelles pensees de Galilee. Ed. and notes by Pierre Costabel and
Michel-Pierre Lerner. Preface by Bernard Rochot. Paris:]. Vrin, 1973. (A partial
translation and reworking of Galileo's Discorsi.)
Micraelius, Johannes. Lexicon philosophicum terminorum philosophis usitatorum. Intro.
Lutz Geldsetzer. Dusseldorf: Stem-Verlag Janssen, 1966. (Facs. repr. of the sec
ond edition, 1662; orig. pub. 1653.) (Instrumenta philosophica, Series Lexica 1.)
More, Henry. Enchiridion metaphysicum. In opera 11.1:131-334.
More, Henry. opera omnia. London:]. Macock, 1679. (Part I is the opera theologica,
Part II the opera philosophia.; facs. repr. Hildesheim: Olms, 1966.)
Newton, Isaac. Unpublished Scientific Papers ofIsaac Newton. Ed. A. Rupert Hall & Marie
Boas Hall. Cambridge: Cambridge University Press, 1962.
Nifo, Augustino [Niphus]. Expositiones in Aristotelis libros metaphysices. Venice: Hiero
nymus Scotus, 1559· (Facs. repr. Frankfurt am Main: MineiVa, 1967.) Bound with
the following.

Authenticated
Bibliography
Nifo, Augustino [Niphus]. Metaphysicarum disputationum in Aristotelis Decem & quatuor

Libros Metaphysicorum. Venice: Hieronymus Scotus, 1559· (Facs. repr. Frankfurt am·
Main: Minerva, 1967.)
Ockham, William of. Brevis summa libri physicorum. In opera philos. 6:1-134.
Ockham, William of. opera philosophica. St. Bonaventure, N.Y: St. Bonaventure Uni
versity, 1967-.
Ockham, William of. Qucestiones in libros physicorum Aristoteles. Ed. Stephen Brown. In
opera philos. 6:395-813. (Written1323-1324; see Brown's Introduction, 41 *.)
Ockham, William of. Summulce philosophice naturalis. Ed. Stephen Brown. In opera
philos. 6:135-394. (Written 1319-1321; see Brown's Introduction, 29*.)
Oresme, Nicole. Nicole Oresme and the Marvels ofNature: The De causis mirabilium. Ed.,
trans., and comm. by Bert Hansen. Toronto: Pontifical Institute of Medi<eval
Studies, 1985.
Oresme, Nicole. Nicole Oresme and the Medieval Geometry of Q:-talities and Motions [Trac
tatus de configurationibus qualitatum et motuum]. Ed. Marshall Clagett. Madison:
University of Wisconsin, 1968.
Pare, Ambroise. Des monstres et prodiges. Ed. and comm. by Jean Ceard. Geneva, 1971.
(Travaux d'humanisme et renaissance 115.)
Pascal, Blaise. CEuvres completes. Ed., intro. and notes by Jacques Chevalier. Paris:
Gallimard, 1954.
Pascal, Blaise. Pensees. Ed. Louis Lafuma. Paris: Seuil, 1962. Also in CEuvres.
Patrizi, Francesco. "On Physical Space." trans. Benjamin Brickman. Journal of the
History of Ideas 4(1943):224-245. (Translation of book I [De spacio physico] and
parts of book II [De spacio mathematico] of Pancosmia, the fourth section of the Nova
de universis philosophia (Ferrara, 1591; the edition used by Brickman is Venice,
1593). I have given page numbers in the original in brackets after the page
numbers of the translation.
Paulus Venetus. Summa philosophice natura/is ... Venice, 1503. (Facs. repr. Hi
ldesheim: Olms, 1974.)
Ramus, Petrus [Pierre de Ia Ramee]. Scholarum physicarum libri octo ... Frankfurt: A.
Wecheli, 1683. (Facs. repr. Frankurt am Main: Minerva, 1967. Orig. pub. 1562.)
Regis, Pierre Sylvain. Cours entier de philosophie ou systeme general seton les principes de M.
Descartes. Amsterdam: Huguetan, 1691. (Facs. repr. with intro. by Richard Watson.
New York: Johnson Reprints, 1970. Orig. pub. 1690.)
Reid, Thomas. The Works of Thomas Reid, D.D. With preface, notes, and supplemen
tary dissertations by Sir William Hamilton. Edinburgh, 71872.
Reisch, Gregor. Margarita philosophica. Intro. by L. Geldsetzer. Dusseldorf: Stern
Verlag Janssen, 1973. (Facs. repr. of the Basel, 1517 edition, the last in Reisch's
lifetime.)
Rohault, Jacques. System of Physics. Trans. and notes by Samuel Clarke. London:
James Knapton, 1723. (Repr. with intra. by L. Laudan. New York: Johnson Re
prints, 1969.)
Rubius, Antonius. Commentarii in universam Aristotelis dialecticam una cum dubiis et
quceestionibus hac tempestate agitari solitis. Alcala, 1603. (Many subsequent editions
up to 1641; see Risse 1964, 1:399 and Lohr 1974-1982, 1980:703.)
Scaliger,Julius Caesar. E.xotericarum exercitationum tiber xv de subtilitate ad Hieronymum
Cardanum. Paris, 1557·
Soncinas, Paulus. Q:-tcestiones metaphysicales acutissimce. Frankfurt: Minerva, 1967.
(Orig. pub. 1588.)

Authenticated
Bibliography
Soto, Domingo de. De natura et gratia. Paris: Ioannis Foucher, 1549. (Facs. repr.
Ridgewood, NJ.: Gregg, 1965.)
So to, Domingo de. Super octo libros physicorum Aristotelis qucestiones. Salamanca, 2 1555.
Suarez, Franciscus. Deangelis. In opera v.2. (Orig. pub. under the title Pars secunda
Summce Theologice de Deo rerum omnium creatore: In tres prcecipuos tractatus distributa,
Quorum primus de Angelis hoc volumine continetur. Lyons: Jacques Cardon, 1620.)
Suarez, Franciscus. De anima. In opera v.3. (Orig. pub. under the title Partis secundce
Summce theologice Tomus alter, complectens tractatum secundum de opere sex dierum ac
tertium de Anima. Lyons:Jacques Cardon and Pierre Cavellat, 1621.)
Suarez, Franciscus. De arcana Missce sacrificii mysterio. In opera v.20-21.
Suarez, Franciscus. Disputationes metaphysicce. Hildesheim: Olms, 1965. (Rep. ofv.25
26 of the Opera.)
Suarez, Franciscus. De divina substantia, eiusque attributis ... In opera v.1.
Suarez, Franciscus. opera omnia. Ed. D. M. Andre. Paris: L. Vives, 1856.
Suarez, Franciscus. De opere sex dierum. In opera v.g. (Orig. pub. with De anima.)
Thomas Aquinas. In Aristotelis libros de Ccelo. In opera Parma 19.
Thomas Aquinas. In Aristotelis libros de Generatione et corrnption. In opera Parma 19.
Thomas Aquinas. In Aristotelis libros de sensu et sensato. Ed. R. M. Spiazzi. Turin:
Marietti, 1949. Also in opera Leonina 45, pt.L
Thomas Aquinas. In Aristotelis libros physicorum. In opera Leonina 2.
Thomas Aquinas. In Aristotelis librum de Anima commentarium. Ed. A. M. Pirotta. Turin:
Marietti, 1936.
Thomas Aquinas. In Metaphysicam Aristotelis commentaria. Ed. M.-R. Cathala. Turin:
Marietti, 1926.
Thomas Aquinas. In Sententiarum libris quatuor. In Opera Leonina 17-20.
Thomas Aquinas. opera omnia. Rome: Commissio Leonina, 1882-.
Thomas Aquinas. opera omnia. Parma: Petrus Fiaccodorus, 1852-1873· Facs. repr.
New York: Musurgia, •949·
Thomas Aquinas. De potentia Dei. Turin: Marietti, 1942. (Qucestiones disputatce 1.)
Thomas Aquinas. Sentencia libri de Anima. Rome/Paris: Commissio Leonina/Vrin,
1984. In opera Leonina 45, pt.l.
Thomas Aquinas. Summa contra gentiles. In opera Parma 5; Leonina 13-15.
Thomas Aquinas. Summa theologice. In opera Parma 1-3; Leonina v-4-12.
Toletus, Franciscus. Commentaria una cum Qucestionibus in libros Aristotelis de Genera
tione et Corruptione. In opera v.5.
Toletus, Franciscus. Commentaria una cum Qucestionibus in octo libros Aristotelis de Physica
auscultatione. In opera V+ (First printed 1572, with numerous subsequent
editions.)
Toletus, Franciscus. Commentaria una cum Qucestionibus in tres libros Aristotelis de Anima.
In opera v.g.
Toletus, Franciscus. Commentaria [. . .} in universam Aristotelis Logicam. In opera v.2.
Toletus, Franciscus. opera omnia Philosophica. Cologne: Birckmann, 1615-1616.
(Fasc. repr. with intro. by Wilhelm Risse. Hildesheim: Olms, 1985.)
Zabarella,Jacobus. Commentarii [. . .} in III: Aristot[elis] Libras de Anima. Frankfurt: L.
Zetzner, 16o6.
Zabarella, Jacobus. opera logica. Frankfurt, 16o8. (Facs. repr. Frankfurt: Minerva,
1966.)
Zabarella, Jacobus. De rebus naturalibus libri XXX. Frankfurt: Zetzner, 2 1607. Orig.
pub. 1597. (Facs. repr. Frankfurt: Minerva, 1966.)

Authenticated
Bibliography
Secondary Sources
Adams, Marilyn. William Ockham. Notre Dame: University of Notre Dame Press,
1987. (Publications in Medieval Studies 26).
Alexander, Peter. Ideas, Q;talities, and Corpuscles: Locke and Boyle on the External World.
Cambridge: Cambridge University Press, 1985.
Andresen, Carl. Handbuch der Dogmen- und Theologiegeschichte. Gottingen: Van
denhoeck and Ruprecht, 1982.
Ariew, Roger. "Damned if You Do: Cartesians and Censorship, 1663-1706." Perspec
tives on Science 2 (1994):255-274.
Ariew, Roger. "Descartes and Scholasticism: The Intellectual Background to
Descartes' Thought." In Cottingham 1992:58-go.
Armogathe,Jean-Robert. Theologia cartesiana: L'explication physique de l'Eucharistie chez.
Descartes et dom Desgabets. The Hague: Martinus Nijhoff, 1977. (Archives interna
tionales d'histoire des idees 84.)
Arthur, Richard. "Continuous Creation, Continuous Time: A Refutation of the Al
leged Discontinuity of Cartesian Time." journal of the history of philosophy
26(1g88):349-375·
Auer,Johann. "Der Wandel des Begriffes 'supernaturalis'." In Filosofia della natura:
331-349·
Balme, David. "Teleology and Necessity." In Gotthelf and Lennox 1987:275-285.
Barlow, H. B. and]. D. Mollon, eds. The Senses. Cambridge: Cambridge University
Press, 1982.
Belgioioso, G., et al., eds. Descartes: Il metoda e i saggi. Rome: Istituto della En
ciclopedia Italiana, 1990.
Bergson, Henri. CEuvres. Intro. by Henri Gouhier, notes by Andre Robinet. Paris:
Presses Universitaires de France, 1959. (L'Evolution creatricewas first published in
1907.)
Berti, Enrico. Aristotle on science: The "Posterior Analytics. "Padua: Editrice An tenore,
1981. (Studia aristotelica, g.)
Birkenmajer, A. "Decouverte de fragments manuscrits de David de Dinant." Revue
neoscolastique de philosophie 35(1933):220-229.
Bloch, Olivier Rene. La philosophie de Gassendi: Nominalisme, materialisme, et metaphysi
que. The Hague Martinus Nijhoff, 1971. (Archives internationales d'histoire des
idees 38.)
Blum, Paul Richard. "Des Standardkursus der katholischen Schulphilosophie im 17.
Jahrhundert." In KeBler et al. 1988:127-148.
Blumenberg, Hans. Die Legitimitiit der Neuzeit. Frankfurt am Main: Suhrkamp, 21g88.
Blumenberg, Hans. "'Nachahmung der Natur': Zur Vorgeschichte der Idee des
schopferischen Menschen." Studium generate 10 ( 195 7): 266-2 8 3.
Bonitz, Hermann. Index aristotelicus. In Aristotelis opera, ed. Academia Regia Borussica,
v. 5· Berlin: G. Reimer, 1870.
Brehier, Emile. La theorie des incorporels dans l'ancien stoi'cisme. Paris: Vrin, 21928.
Brockliss, L. W. B. "Aristotle, Descartes, and the New Science: Natural Philosophy at
the University of Paris, 1600-1740." Archives of Science 38( 1981) :33-69.
Brockliss, L. W. B. French Higher t.aucation in the Seventeenth and Eighteenth Centuries: A
Cultural History. Oxford: Clarendon, 1987.
Brockliss, L. W. B. "Philosophical Teaching in France, 1600-1740." History ofUniver
sities 1(1981):131-168.
Burnyeat, Miles. "Aristotle on Understanding Knowledge." In Berti 1981:97-139.

Authenticated
Bibliography
Carteron, Henri. "L'idee de la force mecanique dans le systeme de Descartes." Revue

philosophique 94( 1922):243-277, 483-511.
Cartwright, Nancy. Nature's Capacities and Their Measurement. Oxford: Oxford Univer
sity Press, 1989.
Cavaille, Jean-Pierre. Descartes: La fable du monde. Paris: J. Vrin, 1991.
Ceard,Jean. La nature et les prodiges. Geneva: Librairie Droz, 1977.
Charlton, William. "Aristotle on the Place of Mind in Nature." In Gotthelf and
Lennox 1987:408-423.
Chenu, M.-D. La theologie au douzieme siecle. Paris:]. Vrin, 1957.
Clarke, Desmond. Descartes' Philosophy of Science. University Park: Pennsylvania State
University Press, 1982. ,
Clarke, Desmond. Occult Powers and Hypotheses: Cartesian Natural Philosophy under
Louis XIv. Oxford: Clarendon, 1989.
Clavelin, Maurice. La philosophie naturelle de Galilee: t:Ssai sur les origines et la formation
de la mecanique classique. Paris: Armand Colin, 1968.
Close, A.J. "Commonplace Theories ofArt and Nature in Classical Antiquity and the
Renaissance." Journal of the history of ideas 30(1969):467-486.
Cohen, H. F. Qy,antifying Music: The Science of Music at the First Stage of the Scientific
Revolution. Dordrecht: D. Reidel, 1984.
Cohen, I. B. "'Quantum in se est': Newton's Concept of Inertia in Relation to
Descartes and Lucretius." Notes and Records of the Royal Society 19( 1964): 131-155.
Cooper, John M. "Hypothetical Necessity and Natural Teleology." In Gotthelf and
Lennox 1987:243-274.
Copenhaver, Brian, and Charles Schmitt. Renaissance Philosophy. Oxford: Oxford
University Press, 1992. (A History of Western Philosophy 3.)
Costabel, Pierre. "Essai critique sur la mecanique cartesienne." Archives intema
tionales de l'histoire des sciences 20(1967):235-252.
Cottingham, John. Descartes. Oxford: Basil Blackwell, 1986.
Cottingham,John, ed. The Cambridge Companion to Descartes. Cambridge: Cambridge
University Press, 1992.
Crewe, A. V. "Atoms and Irony" (Letter to the Editors). American Scientist
81 (1993):207·
Dainville, Fran.;:ois. L'education desjesuites. Paris: Minuit, 1978.
Darmon, Pierre. Le mythe de la procreation a l'age baroque. Paris: J.:J. Pauvert, 1977.
Daston, Lorraine, and Katherine Park. "Unnatural Conceptions: The Study of Mon
sters in Sixteenth- and Seventeenth-Century France and England." Past and Present
92 ( 1981) :20-54·
Dear, Peter. 'jesuit Mathematical Science and the Reconstitution of Experience in
the Early Seventeenth Century." Studies in the History and Philosophy of Science
18( 1987): 133-175·
Dear, Peter. Mersenne and the Learning of the Schools. Ithaca: Cornell University Press,
1988.
Dear, Peter. "Totius in verba: Rhetoric and Authority in the Early Royal Society." Isis
76( 198 5 ): 145 -161.
Denifle, Henri, and A. Chatelain. Chartularium Universitatis Parisiensis. Paris, 1889
1891.
Denzinger, Heinrich Joseph, and Adolf Schonmetzer. Enchiridion symbolorum: Dejini
tionum et declarationum de rebusfidei et morum. Freiburg im Breisgau: Herder, 36 1976.
Dijksterhuis, E. J. The Mechanization ofthe World Picture: Pythagoras to Newton. Trans. C.
Dikshoorn. Foreword by D.J. Struik. Princeton: Princeton University Press, 1961

Authenticated
Bibliography
(repr. I g69). (English translation of De Mechaniseringvan het wereldbeeld. Orig. pub.

Amsterdam, I959·)
Duhem, Pierre. Etudes sur Leonard da Vinci. Paris, Igo6-I9I3.
Duhem, Pierre. Origines de la statique. Paris, I904-I905.
Emerton, Norma E. The Scientific Reinterpretation ofForm. Ithaca: Cornell University
Press, I984.
Fiedler, Wilfried. Analogiemodelle bei Aristoteles. Untersuchungen zu den Vergleichen
zwischen den einzelnen Wissenschaften und Kiinsten. Amsterdam: B. R. Gruner, I978.
(Studien zur antiken Philosophie g.)
La jilosojia della natura net Medioevo. Milan, I g66. (Atti del Terzo Congresso Internazionale
di jilosojia medioevale, Passo della Mendola, I964.)
Freddoso, Alfred]. "Medieval Aristotelianism and the Case against Secondary Causa
tion in Nature." In Morris I988:74-II8.
Frede, Michael. "Substance in Aristotle's Metaphysics." In Gotthelf I985:I7-26.
Freeland, Cynthia A. "Aristotle on Possibilities and Capacities." Ancient Philosophy
6 (I g86) :6g-8g.
Funkenstein, Amos. Theology and the Scientific Imagination from the Middle Ages to the
Seventeenth Century. Princeton: Princeton University Press, Ig86.
Furth, Montgomery. Substance, Form and Psyche: An Aristotelean Metaphysics. Cam
bridge: Cambridge University Press, Ig88.
Gabbey, Alan. "Explanatory Structures and Models in Descartes' Physics." In
Belgioioso Iggo, I:273-286.
Gabbey, Alan. "Force and Inertia in the Seventeenth Century: Descartes and New
ton." In Gaukroger Ig8o:230-320. (A revised and expanded version of Gabbey
I97I.)
Gabbey, Alan. "Force and Inertia in Seventeenth-Century Dynamics." Studies in His
tory and Philosophy of Science 2 (I 97I): I-67.
Gabbey, Alan. [Review of Scott 1970.]. Studies in History and Philosophy of Science
3(I972):373-385.
Garber, Daniel. "Descartes, les aristoteliciens et Ia revolution qui n'eut pas lieu en
I637·" In Mechoulan 1988:I99-2I2.
Garber, Daniel. Descartes' Metaphysical Physics. Chicago: University of Chicago, 1992.
Garber, Daniel. "Science and Certainty in Descartes." In Hooker I978:II4-I5I.
Gaukroger, Stephen. "Descartes' Project for a Mathematical physics." In Gaukroger
I98o:g7-I40.
Gaukroger, Stephen, ed. Descartes: Philosophy, Mathematics, and Physics. Sussex: Harves
ter, Ig8o.
Gilbert, Neal. Renaissance Concepts of Method. New York: Columbia University Press,
I96o.
Gill, Mary Louise. "Aristotle on Self-Motion." In judson Iggi:243-265.
Gill, Mary Louise. Aristotle on Substance: The Paradox of Unity. Princeton: Princeton
University Press, I989.
Gilson, Etienne. Etudes sur le role de la pensee medievale dans la formation du systeme
cartesien. Paris: Vrin, 5 Ig84. (Orig. pub. I930.)
Gilson, Etienne. Index Scholastico-cartesien. Paris: Alcan, I g I3. (Facs. repr. New York:
Burt Franklin, I964.)
Gilson, Etienne. Introduction d l'etude de Saint Augustin. Paris: Vrin, 2 I943·
Gilson, Etienne. Laliberte chez Descartes et la theologie. Paris: Vrin, I9I3. (Facs. repr.
Paris: Vrin, I982.)

Authenticated
Bibliography
Glouberman, Mark. "Cartesian Substances as Modal Totalities." Dialogue

17(1978):320-343. Repr. in Moyal1991, 3:363-384.
Glouberman, Mark. Descartes: The Probable and the Certain. Wiirzburg/Amsterdam:
Konigshausen und Neumann/Rodopi, 1986. (Elementa 41.)
Gotthelf, Allan. Aristotle on Nature and Living Things. Pittsburgh/Bristol:Mathesis/
Bristol Classical Press, 1985.
Gotthelf, Allan. "Aristotle's Conception of Final Causality." In Gotthelf and Lennox
1987=204-242. (Reprint, with additional notes and postscript, of a paper first
published in 1976.)
Gotthelf, Allan, and James G. Lennox, eds. Philosophical Issues in Aristotle's Biology.
Gouhier, Henri. La pensee religieuse de Descartes. Paris: J. Vrin, 192 1. (Etudes de phi
losophic medievale 6.)
Grabmann, Martin. I divieti ecclesiastici di Aristotele sotto Innocenzo III e Gregorio IX.
Rome, 1941. (Miscellanea histori;e pontifici;e 5: I Papi del duecento e l'aristo
telismo, fasc. 1.)
Gracia,Jorge. "Francis Suarez." In: Gracia 1994:475-510.
Gracia, Jorge J. E. Individuation in Scholasticism: The Later Middle Ages and the Counter
Reformation, I I50-I6JO. Albany, N.Y: State University of New York Press, 1994·
Grafton, Anthony. "Teacher, Text, and Pupil in the Renaissance Class-room: A Case
Study from a Parisian College." History of Uuniversities 1 ( 1981) :3 7-70.
Graham, D. W. "The Paradox of Prime Matter." journal of the History of Philosophy
25(1987>=475-490.
Grant, Edward. "Cosmology." In Lindberg 1978:265-302.
Grant, Edward. "Medieval Explanation and Interpretations of the Dictum That 'Na
ture Abhors a Vacuum'." Traditio 29(1973):237-355.
Grant, Edward. Much Ado about Nothing: Theories of Space and Vacuum from the Middle
Ages to the Scientifc Revolution. Cambridge: Cambridge University Press, 1981.
Grant, Edward. "Ways to Interpret the Terms 'Aristotelian' and 'Aristotelianism' in
Medieval and Renaissance Natural Philosophy." History of Science 25(1987):335
358.
Gregory, Tullio. Anima mundi: La filosojia di Guglielmo di Conches e la scuola di Chartres.
Florence: G. C. Sansoni, 1955.
Grosholz, Emily. Cartesian Method and the Problem of Reduction. Oxford: Clarendon
Press, 1991.
Gueroult, Martial. Dynamique et metaphysique leibniziennes. Sui vi d 'une Note sur le principe
de la moindre action chez Maupertuis. Paris: Les Belles Lettres, 1934. (Publications de
la Faculte des Lettres de l'Universite de Strasbourg 68).
Gueroult, Martial. Etudes sur Descartes, Spinoza, Malebranche, et Leibniz. Hildesheim:
Georg Olms, 1970. ("Physique et metaphysique de Ia force chez Descartes et chez
Malebranche" was first published in 1954.)
Gueroult, Martial. "Physique et metaphysique de la force chez Descartes et chez
Malebranche." Revue de metaphysique et de morale 59(1954):1-37, 113-134. In
Gueroult 1970:85-143.
Hatfield, Gary. "Force (God) in Descartes's Physics." Studies in History and Philosophy of
Science ro(I979):IIJ-I40. AlSo in Moyal1991:123-152.
Hedwig, Klaus. Sphtera lucis: Studien zur Intelligibilitiit des Seiendenim Kontext der mit
telalterlichen Lichtspekulation. Munster: Aschendorff, 1980. (Beitrage zur
Geschichte der Philosophic und Theologie des Mittelalters n.F. 18).

Authenticated
Bibliography
Henry, John. "Francesco Patrzi da Cherso's Concept of Space and Its Later Influ
ence." Annals of Science 36(1979):549-575·
Herivel, John. The Background to Newton's Principia: A Study of Newton's Dynamical
Researches in the Years r664-r684. Oxford: Oxford University Press, 1965.
Hickey,]. S. Summula philosophim scholasticm in usum adolescentium. Dublin: M. H. Gill
and Son, 1919.
Hissette, Roland. Enquete sur les 219 articles condamnes aParis le 7 mars r 2 77· Louvain:
Publications Universitaires, 1977.
Hoffman, Paul. "The Unity of Descartes's Man." Philosophical Review 95 ( 1986) :339
370. Also in Moyal1991:168-192.
Hooker, Michael. Descartes: Critical and Interpretive l!.ssays. Baltimore: Johns Hopkins
Hussey, Edward. See under Aristotle, Physics.
Hussey, Edward. "Aristotle's Mathematical Physics: A Reconstruction." In Judson
1991:213-242·
Hutchison, Keith. "Supernaturalism and the Mechanical Philosophy." History of Sci
ence 21 ( 1983) :297-333.
Ingegno, Alfonso. 'The New Philosophy of Nature." In Schmitt and Skinner
1988:236-263.
Jammer, Max. 'The Program and Principles of Descartes' Physics." In Belgioioso
1990:303-334·
Judovitz, Dalia. Subjectivity and Representation in Descartes: The Origins of Modernity.
Judson, Lindsay. "Chance and 'Always or for the Most Part' in Aristotle." In Judson
1991:73-100.
Judson, Lindsay, ed. Aristotle's Physics: A Collection ofEssays. Oxford: Oxford University
Press, 1991.
Kalivoda, Robert. "Zur Genesis der natiirlichen Naturphilosophie." In Filosofia della
natura : 397-406.
KeBler, Eckhard, Charles Lohr, and Walter Sparn. Aristotelismus und Renaissance: In
memoriam Charles B. Schmitt. Wiesbaden: Otto Harrassowitz, 1988. (Wolfenbiitteler
Forschungen 40.)
Kosman, L.A. "Aristotle's Definition of Motion." Phronesis 14(1969):4off.
Koyre, A. Etudes galiteennes. Paris: Hermann, 1939. (Actualites scientifiques et indus
trielles 852-854.)
Kraye, J., W. F. Ryan, and C. B. Schmitt, eds. Pseudo-Aristotle in the Middle Ages: The
Theology and Other Texts. London: Warburg Institute, 1986.
Kretzmann, Norman, Anthony Kenny, and Jan Pinborg, eds. The Cambridge History of
Later Medieval Philosophy. Cambridge: Cambridge University Press, 1982.
Kuhn, Thomas. "Reflections on My Critics." In Lakatos and Musgrave 1970:213
278.
Kullman, Wolfgang. "Different Concepts of the Final Cause in Aristotle." In Gotthelf
19 8 5 :169 -175.
Kurdzialek, M. "David von Dinant und die Anfange der Aristotelischen
Naturphilosophie." In Filosofia della natura, 407-416.
Kurdzialeck, M., ed. See David of Dinant.
Lakatos, Imre, and Alan Musgrave. Criticism and the Growth ofKnowledge. Cambridge:
Cambridge University Press, 1970.
Lang, Helen S. "Aristotelian Physics: Teleological Procedure in Aristotle, Thomas,
and Buridan." Review of Metaphysics 42(1989):569-91.

Authenticated
Bibliography [411]
Leff, Gordon. The Dissolution of the Medieval Outlook. New York: NYU Press, 1976.
Lenoble, Robert. Mersenne ou la naissance du mecanisme. Paris: Vrin, 1943.
Lerner, Michel-Pierre. Recherches sur la notion de finalite chez. Aristote. Paris: Presses
Universitaires de France, 1969.
Lindberg, David C. ''The Genesis of Kepler's Theory of Light: Light Metaphysics
from Plotinus to Kepler." Osirisser. 2, 2(1986):5-42.
Lindberg, David C. john Pecham and the Science of optics. Madison: University of
Wisconsin Press, 1970.
Lindberg, David C. Science in the Middle Ages. Chicago: University of Chicago Press,
1978.
Lloyd, G. E. R. "The Invention of Nature." In Lloyd 1991:417-434. Also in Torrance
1992:1-24·
Lloyd, G. E. R. Methods and Problems in Greek Science. Cambridge: Cambridge Univer
sity Press, 1991.
Lockwood, Michael. Mind, Brain, and the Quantum: The Compound 'I'. Oxford/
Cambridge: Blackwell, 1989.
Lohr, Charles. "Renaissance Latin Aristotle Commentaries." Studies in the Renaissance
21(1974):228-289; Renaissance Quarterly 28(1975):689-741; 29(1976):714
745; 30(1977):681-741; 31(1978):532-603; 32(1979):529-580; 33(198o):
623-734; 35(1982):164-256.
Lubac, Henri. Sumaturel: Etudes historiques. Paris: Aubier/Montaigne, 1946.
MacClintock, Stuart. Peruersity and Error: Studies on the "Averroist" john of]andun.
Bloomington: Indiana University Press, 1956.
Maier, Anneliese. An der Grenze von Scholastik und Naturwissenschaft. Rome: Storia e
letteratura, 1952. (Maier's Studien zur Naturphilosophie der spiitscholastik 3; Storia e
letteratura 41 .)
Maier, Anneliese. Ausgehendes Mittelalter: Gesammelte Aufsiitze zur Geistesgeschichte des
I4.]ahrhunderts. v.2. Rome: Storia e letteratura, 1967. (Storia e letteratura 105.)
Maier, Anneliese. Metaphysische Hintergrilnde der spiitscholastischen Naturphilosophie.
Rome: Storia e letteratura, 1955. (Maier's Studien zur Naturphilosophie der
spiitscholastik 4; Storia e letteratura 52.)
Maier, Anneliese. On the Threshold ofExact Science: Seclected Writings ofAnneliese Maier
on Late Medieval Natural Philosophy. Ed. and trans. Steven D. Sargent. Philadelphia:
University of Pennsylvania Press, 1982.
Maier, Anneliese. Die Vorliiufer Galileis. Rome: Storia e letteratura, 1949. (Maier's
Studien zur Naturphilosophie der spiitscholastik 1; Storia e letteratura 2 2.)
Maier, Anneliese. Zwei Grundprobleme der scholastischen Naturphilosophie. Rome: Storia
e letteratura, 2 1951. (Maier's Studien zur Naturphilosophie der spiitscholastik 2; Storia
e letteratura 3 7.)
Maier, Anneliese. Zwischen Philosophie und Mechanik. Rome: Storia e letteratura,
1958.(Maier's Studien zur Naturphilosophie der spiitscholastik s; Storia e letteratura
69.)
Marenbon,John. Later Medieval Philosophy (I I50-I350). London: Routledge, 1987.
Marion, Jean-Luc. Sur la theologie blanche de Descartes. Analogie, creation des verites eter
nelles etfondement. Paris: Presses Universitaires de France, 1981.
McEvoy, James. The Philosophy of Robert Grosseteste. Oxford: Clarendon, 1982.
McGuire,]. E. "Space, Geometrical Objects and Infinity: Newton and Descartes on
Extension." In Shea 1983:69-112.
Mechoulan, Henri, ed. Probtematique et reception du Discours de Ia methode et des
Essais. Paris: Vrin, 1988.

Authenticated
Bibliography
Mercer, Christia. 1'he Vitality and Importance of Early Modem Aristotelianism." In

Sorrell 1993:33-67. ·
Michaud-Quantin, Pierre. Etudes sur le vocabulaire philosophique du moyen age. Rome:
Edizioni dell'Ateneo, 1971.
Michel, Paul-Henry. La cosmologie de Giordano Bruno. Paris: Hermann, 1962.
Mignucci, Mario. '"Oa enl TO noA.u et necessaire dans la conception aristotelicienne
de la science." In Berti 1981:173-203.
Miles, Murray Lewis. "Condensation and Rarefaction in Descartes' Analysis of Mat
ter." Nature and system 5(1983):169-180.
Milton, John. ''The Origin and Development of the Concept of the 'Laws of Na
ture'." Archives europeennes de sociologie 22(1981):173-195.
Moravcsik,Julius. ''What Makes Reality Intelligible?" In Judson 1991:31-47.
Morris, Thomas V. Divine and Human Action: Essays in the Metaphysics ofTheism. Ithaca:
Cornell University Press, 1988.
Moyal, Georges]. D. Rene Descartes: Critical Assessments. London: Routledge, 1991.
Murdoch, John E., and Edith D. Sylla. ''The Science of Motion." In Lindberg
1978:206-264.
Murray, Alexander. "Nature and Man in the Middle Ages." In Torrance 1992:25-62.
Nadler, Steven, ed. Causation in Early Modem Philosophy. University Park: Pennsylvani
University State Press, 1993.
Nardi, A. "Descartes 'presque' galileen: 18 fevrier 1643." Revue de l'histoire des sciences
39(1986):3-16.
Nardi, Bruno. Studi difilosofia medievale. Rome: Storia e letteratura, 1960. (Storia e
letteratura 78.)
Newton-Smith, W. H. The Structure of Time. London: Routledge and Kegan Paul,
1980.
Oakley, Francis. Omnipotence, Covenant, and Order: An Excursion in the History ofIdeas
from Abelard to Leibniz. Ithaca: Cornell University Press, 1984.
O'Neill, Eileen. "Injluxus physicus." In Nadler 1993:27-55.
Owen, Joseph. The Doctrine ofBeing in the Aristotelian Metaphysics. Toronto: Pontifical
Institute of Mediaeval Studies, 3 1978. (First ed. 1951.)
Pellegrin, Pierre. "Logical Difference and Biological Difference: the Unity of Aris
totle's Thought." Trans. A. Preus. In Gotthelfand Lennox 1987:313-338.
Perez-Ramos, Antonio. Francis Bacon's Idea ofScience and the Maker's Knowledge Tradi
tion. OxfoFd: Clarendon, 1988.
Prendergast, Thomas L. "Descartes and the Relativity of Motion." Modern Schoolman
50(1972):64-72. Repr. in Moyal1991, 4:101-109.
Prendergast, Thomas L. "Motion, Action, and Tendency in Descartes' Physics." Jour
nal of the History of Philosophy 13(1975):453-462. Repr. in Moyal 1991, 4:89
100.
Redondi, Pietro. Galileo Heretic. Trans. Raymond Rosenthal. Princeton: Princeton
Rindler, Wolfgang. Essential Relativity: Special, General, and Cosmological. Rev. 2 ed.
New York: Springer, 1979.
Risse, Wilhelm. Logik der Neuzeit. Stuttgart-Bad Cannstatt: Friedrich Frommann,
1964. (Bd.1: 1500-1640; Bd.2: 1640-1780.)
Rochemonteix, Camille. Un college dejesuites aux pni• et xviiie siecles: Le College Henri N
de la Fteche. Le Mans: Leguicheux, 1899.
Rossi, Clement. L'Anti-nature: Eliments pour une philosophie tragique. Paris: Presses
Universitaires de France, 1973.

Authenticated
Bibliography
Russell, J. L. "Action and Reaction before Newton." British journal for the History of
Science 9( 1976) :25-38.
. Salmon, Wesley. Explanations and the Causal Structure of the World. Princeton: Prince
ton University Press, 1984.
Sarnowsky,Jiirgen. Die aristotelisch-scholastische Theorie der Bewegung: Studien zum Kom
mentar Alberts von Sachsen zur Physik des Aristoteles. Munster: Aschendorff, 1989.
(Beitrage zur Geschichte der Philosophie und Theologie des Mittelalters n.f. 32).
Schmitt, Charles. "Experimental Evidence For and Against a Void: The Sixteenth
Century Arguments." Isis 58(1967):352-366. Also in Schmitt 1989, c.7.
Schmitt, Charles. john Case and Aristotelianism in Renaissance England. Kongston/
Montreal: MeGill-Queen's University Press, 1983. (MeGill-Queen's Studies in the
History of Ideas 5.)
Schmitt, Charles. "Philoponus' Commentary on Aristotle's Physics in the Sixteenth
Century." In Sorabji 1987.
Schmitt, Charles. "Pseudo-Aristotle in the Latin Midle Ages." In Kraye eta!. 1986:3
14; Schmitt 1989, c. I.
Schmitt, Charles. Reappraisals in Renaissance Thought. London: Variorum Reprints,
1989.
Schmitt, Charles. "The Rise of the Philosophical Textbook." In Schmitt and Skinner
1988:792-893·
Schmitt, Charles. Studies in Renaissance Philosophy and Science. London: Variorum
Reprints, 198 1.
Schmitt, Charles, and Quentin Skinner, eds. The Cambridge History of Renaissance
Philosophy. Cambridge: Cambridge University Press, 1988.
Schonberger, Rolf. "Eigenrecht und Relativitiit des Natiirlichen bei Johannes Bur
idanus." In Zimmermann and Speer 1991:216-233.
Schuster, John. Descartes and the Scientific Revolution, r6r8-r6J4· Ph.D. Diss., Prince
ton University, 1977.
Scott, J. F. The Scientific Work of Rene Descartes (I596-r65o). London: Taylor and
Francis, 1952. (Facs. repr. New York: Garland, 1987.)
Scott, Wilson L. The Conflict between Atomism and Conseroation Theory r644-r86o. New
York: Elsevier, 1970.
Shea, William R. The Magic ofNumbers and Motion: The Scientific Career ofRene Descartes.
Canton, Mass.: Science History Publications, 1991.
Shea, William R. Nature Mathematized: Historical and Philosophical Case Studies in Classi
cal Modem Natural Philosophy. Dordrecht: D. Reidel, 1983. (University of Western
Ontario Series in Philosophy of Science 20.)
SiiVen, J. Les annees d'apprentissage de Descartes (I596-I 628). Albi: Imprimerie Coop
erative du Sud-Ouest, 1928.
Smith, A. Mark. "Getting the Big Picture in Perspectivist Optics." Isis 72 ( 1981) :568
589.
Sorabji, Richard. Matter, Space, and Motion. Ithaca: Cornell University Press, 1988.
Sorabji, Richard. Philoponus and the Rejection of Aristotelian Science. Ithaca: Cornell
Sorrell, Tom, ed. The Rise ofModem Philosophy: The Tension between the New and Tradi
tional Philosophies from Machiavelli to Leibniz. Oxford: Clarendon, 1993·
Speer, Andreas. "Kosmisches Prinzip und MaB menschlichen Handelns: Natura bei
Alanus ab Insulis." In Zimmermann and Speer 1991:107-128.
Steenberghen, Fernand van. Introduction al'etude de la philosophie medievale. Preface by
Georges van Riet. Louvain: Publications Universitaires, 1974.

Authenticated
Bibliography
Steenberghen, Fernand van. La philosophie au xiii' siecle. Louvain-la-Neuve:Editions

de l'Institut Superieur de Philosophie, 2 1991.
Thery, G. Autour du decret de I2Io. I. David de Dinant. Paris, 1925. (Bibliotheque
thomiste 6.)
Tocanne, Bernard. L'ldee de nature en France dans la seconde moitie du xvii• siecle:
Contribution al'histoire de la pensee classique. Paris: Klincksieck, 1978. (Bibliotheque
fran~aise et romane, Ser. C: 67.)
Torrance,John, ed. The Concept ofNature: The Herberl Spencer Lectures. Oxford: Claren
don, 1992.
Verbeek, Theo. Descartes and the Dutch: Early Reactions to Cartesian Philosophy, I637
I65o. Carbondale: Southern Illinois University Press, 1992.
Wallace, William A. Causality and Scientific l!.xplanation. Ann Arbor: University of
Michigan Press, 1972.
Wallace, William A. Prelude to Galileo: Essays on Medieval and Sixteenth-Century Sources of
Galileo's Thought. Dordrecht: D. Reidel, 1981.
Waterlow, Sarah. Nature, Change, and Agency in Aristotle's Physics. Oxford: Oxford
Weinberg, Julius R. A Short History ofMedieval Philosophy. Princeton: Princeton Univer
sity Press, 1964.
Weisheipl, James A. Nature and Motion in the Middle Ages. Ed. William E. Carroll.
Washington, D.C.: Catholic University of America, 1985. (Studies in Philosophy
and the History of Philosophy 11.)
Westfall, Richard S. Force in Newton's Physics. New York: American Elsevier, 1971.
Wild, John. 'The Cartesian Deformation of the Structure of Change and Its Influ
ence on Modern Thought." Philosophical Review 50(1941):36-59· In Moyal1991,
4:20-42·
Williams, Bernard. Descartes: The Project ofPure Enquiry. Harmondsworth, Middlesex:
Penguin, 1978.
Williams, Raymond. Keywords. New York: Oxford University Press, 2 1985. (1st ed.
1976.)
Wolff, Michael. Geschichte der Impetustheorie: Untersuchungen zum Ursprung der
klassischen Mechanik. Frankfurt am Main: Suhrkamp, 1978.
Wright, Larry. Teleological Explanations. Berkeley: University of California Press, 1976.
Zimmermann, Albert, and Andreas Speer. Mensch und Natur im Mittelalter. Berlin:
Walter de Gruyter, 1991. 2 1, pt. I and 2.)

Authenticated
Index
Abra de Raconis, C. F., 10, 30;

JEgidius Romanus, 143; on condensation
distinguishes creation and conservation,

and rarefaction, 107; on matter and
325; instances of ends in nature, 168;

darkness, 83
motion does not persist, 275-276;

agent, not affected in acting, 44
senses of natura, 214

agent and patient, 43-44, 47-48, 170
absorption, 288-290, 300

Alain de Lille, 144, 216
accidental form, 73
Albertus Magnus, 94, 206; inchoate forms,
accidents: are caused by substance mate

140
rially, formally, and efficiently (Suarez),

Alembert,Jean le Rand d', 168-169
16o; by definition cannot exist sepa

Alexander, Peter, 75
rately from their subject (Boyle,

alteration, 61; definition of, 25; irrevers
Chauvin), 132; do not persist through

ible, 73
corruption (Thomists), 144-145; em

'always or for the most part', 208
bellishing, 144; empirical arguments for

angels, 230, 238
persistence through corruption, 146

animals, 61; have natural appetites only
148; existence apart from substance,

(Toletus), 202; sagacity and industry of,
129-13 1; id cuius esse est inesse, 131 ;
181; self-movement of, 230
identified with modes in seventeenth

anima mundi, 125, 181
century authors, 133-134; incapable of

annihilation: argument from, 359;
producing forms, 162; inhere in the

Descartes' response, 384; is the cessa
union of matter and form, 145; "pri

tion of a positive divine action, i.e. con
mary" and "adverbial," 113; real, 113,

servation, 332-333
127; as signs (signa), 66

apart, 374
action: immanent, 41-43; transeunt, 41

appetite: adventitious, 236; distinguished
42 from propensity, 235; innate and elic
action and passion, 40-46; not really dis

ited, 236; intellectual, sensitive, and nat
tinct from motus, 44-46; are identical

ural (Toletus), 202; natural, 235-236;
(Descartes), 259
necessary quoad exercitiuim and quoad spe
actus, 29, 57; complete and incomplete,

ciem, 236
93; of existence distinct from that of in

Ariew, Roger, 1, 12
forming (Fonseca, Suarez), 126; meta

Aristotelianism, scope of the term, 7
physicus, 93; physicus, 93; terminus ad

Aristotle: Categaries, 29, 87, 110, 148, 151,
quem of change, 63
381; definition of 'nature', 227; inher
Adams, Marilyn, 82
ence, 130-132; matter, 82; matter,
Brought to you by | Cambridge University Library

Authenticated
Index
Aristotle (cont.) Boethius, 214

form, and privation, 56-57; Metaphysics, Bourdin, Pierre, 301, 310
56,148, 214;Phyncs,82, 227, 235; Boyle, Robert, 54, 64, 213; education of
qualities, 110; separate space, 355-356; form from matter, 140, 142; form, 59;
use of Aristotle's works in teaching, 8 inherence, 130
10 Brockliss, L. W. B., 7, 8
Arnauld, Antoine, 24, 79; empty space Bruno, Giordano, 125
could be measured, 385 Buridan,Jean, 35, 37, 264; the actions of
Arriaga, Rodericus de, 2; artificial forms, animals depend only on efficient
245 causes, 198-199; against final causes,
art: Aristotelian analogies of nature to, 186-187; 'nature' a relative term, 218
243-244; compared with divine pro Bumyeat, Miles, 22
duction, 247-251; divine art imitates
the divine intellect, 250; imitates things
that "ought to have existed," 242; imita Calcidius, 124
tion of nature, 240-244; local motion Capreolus,Joannes, 37, 100
alone required to produce, 245-246 Carteron, Henri, 328
Arthur, Richard, 326 causalitas, 189
artifacts, 230-231, 240, 243; have neither cause: causality of final, 189; coordination
figure nor form (Ockham), 111; have of causes, 179; creation, conservation,
no natures, 229; idea precedes produc and concurrence the three modes of
tion,246,247 divine efficient causation, 327; defini
artificial forms, 244-247 tion of (Suarez), 189; efficacy of second
atomism, 82, 83, 86, 115, 365-366 causes, 248, 251; efficient, 62, 178
attribute, 362; principal, 363 179, 331; efficient causes are blind,
Auer, Johann, 215 178; extrinsic and intrinsic, 128-129;
augmentation, 107; definition of, 25 final cause replaced by efficient cause,
Augustine, 91, 124, 385; on matter, 83 394-395; instrumental, 162, 249;
automata, 244 material, formal, efficient, and final
Averroes, 9, 66, 81, 87, 126, 230, 348; ed causes defined, 189; of motion, general
ucation of form from matter, 141; the and particular, 272; object of final,
essence of matter is not potentia, 92; 194-200; particular and universal, 143;
motus, 36 in potentia and in actu, 317; principal,
Avicenna, 82, 88; on the dator formarum, 249; second causes, efficacy of, 163,
142; on the esse intentionale of ends, 191 166
Ceard,Jean, 206, 207, 221
Bacon, Francis, 168 celestial causes, the "refuge of the
Balme, David, 187 wretched in philosophy" (Petrus Au
beauty: attracts the soul, 201; as an end, reolus, Oresme), 165
176-177; of the world enhanced by the chance, 210
existence of monsters (Coimbra), 208 change: acting on another need not entail
Beeckman, Isaac, 275; condensation and change in the agent, 318-319; exis
rarefaction, 351; law of persistence of tence of, 35; inception of, 22; natural,
motion, 276-277 21; transitive, 312; transitive and intran
beneficiary, 17 2 sitive, 23, 41-42, 63. See also motus, nat
Bergson, Henri, 35 ural change
Blum, Paul Richard, 153 Charleton, Walter: form without matter,
Blumenberg, Hans, 34, 91, 124 127; independence of space, 383; mat
body: distinguished from space, 378; one ter, 92; prime matter, 81
really distinct from another only if their Chauvin, Stephanus, 100; defines acci
locations are disjoint, 3 81. See also dents as modes, 132; on the definition
corpus; extension; matter; space of disposition, 151

Authenticated
Index
circular motion, is preternatural, 225

ond causes, 321; with material, formal,
Clarke, Desmond, 6, 286-287, 295, 300,
and final causes, 331
352
Condemnationsof1277,38, 131
Claves, Etienne de, 64

condensation and rarefaction: in argu
Cohen, I. B., 273

ments about quantity, 107; Descartes'
Coimbra [Collegium Cimbricensis], 2, 10,

explanation of, 351-352
355; animals do not recognize the ends configuration, 64; efficacy of, according to
toward which they act, 199; arguments Oresme, 116
against including matter in essence, conservation: of accidents and incomplete
233; art cannot supplant nature, 244; substances, 326; analogy between con
artifacts have no nature, 231; artificial seiVation and illumination, 327; analogy
forms have no effective power, 247; art with static force, 334-335; divine con
never forms true things, 241; concur servation of total motion, 315; inclina
rence, 321-322; conservation not dis tion of agents to self-conversation, 226;
tinct from creation, 326; creation is an actual influx of being (Suarez),
entails no change in God, 318, 323; the 332; of matter, 315; of motion is a sin
dictum 'matter desires form', 202; the gle continuous act, 330; of motus, 326
ends by which monsters are explained, 327; not distinct from creation, 325
207-208; eternity of the world, 125; 327
generation, 143; God's ordained power, constitution, 365
320; hurror vacui, 175-176; imaginary contest model of collision, 290, 310
space, 264; matter is part of essence, contiguity, the most natural disposition of
234; metaphorical motion, 189; motus bodies, 175
and terminus, 38-40; natural limits to contranatural, 222-224
size, 203; persistence of things in their contraries: in Descartes' rules of collision,
natural states, 274; role of accidents in 290-291; and natural change, 56-58
generation, 162-163; senses of natura, Cooper,John, 181, 238
214; successiveness of motus, 269; union Copernicanism, 271-272
of matter and form, 135, 137 corpus: material substance under the as
collision: has no terminus, 170; outcome
pect of quantity, 349; mathematicum,
not determinate in Cartesian physics,
349; physicum, 349
375-376; transfer of motion in, 305

corruption, 33, 96; definition of, 25; is
307. See also rules of collision

only an incidental effect of generation,
color, why it is not a mode of extension,

133; is not violent (Aristotle), 222. See
365-366
also generation, substantial change
commentaries, distribution of, over the

Costabel, Pierre, 293, 309
Aristotelian canon, 153

Cottingham,John, 363
composition of motions, 310

counterfactuals, 277
conatus, 259-260, 281, 337-338; not a vis

creation: continual, 280; entails no
or power, 277
change in god, 317-318; is not a muta
concurrence: of bodies moving other

tio or motus, 332
bodies is referred to, 335; determina

crocodile, 203
tion of singular causes, 321-322, 334;

curriculum, role of Aristotle's works, 8-10
difficulties of applying the concept in

cursus: changes in, 153, 186; character of,
Cartesian physics, 341; effects deter

10-11
mined according to the nature of the
agents concurred with, 219-220, 323;

Daston, Lorraine, 222
God concurs with each cause according

dator (datrix) formarum, 142
to its strength (Toletus), 208; is a free

David ofDinant, 94
act of God, 322; is to action as conseiVa

definition: in argument, 24; and essence,
tion to existence, 320; ordinary, 319;
235
whether distinct from the action of sec

deformation,375-377

Authenticated
Index
density, 120; mass per unit volume not a Emerton, Nancy, 54· 59
Cartesian notion, 352. See also conden Encyclopedie: senses of 'nature', 214
sation and rarefaction ends: act according to their esse reale, 191;
Denzinger, Heinrichjoseph, 8, 98, 125,

act only on rational agents, and perhaps
215
animals, 194-200; are causes with re
dependentia, 1 2 8
spect to esse intentionale, 191; collective,
Descartes: actus and potentia, 78; coding of 174-176; collective, 174-176; collective
physical properties into shapes, 116; can outweigh individual, 176; cosmic,
definition of motus, 260; independence 176-1 77, 396-398; definition of, 171;
of matter from God, 91; "little souls" ar divine, 391; explain existence and com
gument, 393; nature is like art, 240. See presence, 180-181; and goods, 184;
also extension; laws of motion; matter; how ends act on things that lack cogni
rules of collision; space tion, 190; impossibilia as, 193-194; indi
determination, 309-310, 322

vidual, 171-174,393-394;and
diagrams: depict "topological" relations,

individual natures, 235-239; meta
106-107; Oresmian diagrams in

phorical motion, 189-191; move the
Descartes' works, 304

will through being cognized (Suarez),
differentia, 68, 87
192; must be cognized (Descartes), 394;
diminution, 87; defined, 25; and the

and natural agents, 195; purely formal,
definition of motus, 33
235; and regularity, 177-179; whether
Dionysius Areopagiticus, 126

means share the ends of their users,
direction, 281, 283, 290

196-198. See also cause; finality
disposition, 142; definition of, 151;

ens: per se and per accidens, 135-136; per ac
dispositions inhere in matter, 151; in
cidens, 234
ference to, 154; and proportion, 152; as

entelechia, 2 5
a quality, 151-152; reduced to micro

esse intentionale, 191
structure, 157-158; role of dispositions

essence, 25; and matter, 233-235; and
in the education ofform, 149, 153;

unity, 135
whatever exhibits order can be called

esse objectivum, 191
dispositio (Smirez), 152

esse reale, 191, 193
disputatio, as a kind of writing, 7

Eucharist, 145; Descartes' doctrines con
distance, 384-385
demened, 3; the Eucharistic worm, 233;
distinction, real, 101-102, 127

existence of real accidents without sub
divisibility, 99
stance in, 129; quantity in the doctrine
dominion, 397; of men over the rest of

of, 98-99, 103-104
creation, 183
Eustachius, 1o-11, 26; distinction be
dormitive virtue, 24, 156

tween matter and form, 127; prime
Duns Scotus, 82, 84

matter, 93; substantial form, 76
Dupleix, Scipion: prime matter, 84; salti

experimenta: combustion, 59, 62; crocodiles
ness of the sea, 168

never stop growing, 203; death, 146; ex
DuPlessis d'Argentre, Charles, 3

pansion of air on being heated, 107;
duration: differs only in reason from sub

falling bodies, 231, 393; fire produced
stance itself, 364

by friction or radiation, 164; heating
and cooling, 29; role of, in argument,
eduction, of form from matter, 139-144

155-156; rusting of iron, 71; spon
elasticity, 3 77
taneous cooling of hot water, 73-74;
elements, 167; equilibrium of, 181, 395;

water turning to steam, 71; wax and
transmutation of, 69, 85-86

stamp, 62
emanation: in Neoplatonist and Gnostic

explanatory autonomy, 217-218, 220,223
cosmologies, 91; of active powers from
extension, 99; Aristotelian quantitas re
form, 158-161
duced to, 345; constitutes the nature of
emergence, 141
corporeal substance (Descartes), 360,

Authenticated
Index
extension (cont.) 368; quantity distinct from substance,

363-364; and distinctness of parts, 100; 103; space is a "pure negation", 357;
generic, 384; identified by Descartes why matter is called Tenelna, 83
with prime matter, 86; incorporeal, 389; force: interpretations of, 314, 330-331;
is imaginable (Descartes), 387; is the measured by quantity of motion, 339
sole essential differentia of body 340; of motion, and concurrence, 333;
(Descartes), 380; as mode and as sub of motion, is in God, 337; problem of,
stance, 364; not an accident of bodies 313; of rest, 307; of rest, and conseiVa
(Descartes), 364; singular, 383; spatial tion,335
extension and corporeal extension are form: actus of matter, 6o, 65; artificial,
identical (Descartes), 361. See also cannot add to natural powers, 244; ar
quantity tificial, not educed from matter
(Suarez), 246; as the cause of active
fate, 209 powers, 158-161; and definition, 58
felicity, human, consists in the contempla 59; educed by the agent in the patient,
tion of God, 236 62; educed from matter, 139-144; for
figure, 57, 64; analogy with form, 109; mative cause or semen, 59, 64; hierarchy
in the Categories, 110; defmition of, 110; offorms, 166-168, 183, 185; higher
efficacy of, in divination, 116; follows forms require an "exquisite and pecu
upon form, 111, 241; index of substan liar mode of generation • (Suarez), 165;
tial form, 109; inertness of, 246; a inchoate, 141; latent, 140; in mathe
mode of quantity, 111-112; in math matics, 11 7; and matter are unum per se,
ematics, analogous to form, 117; mod 135-138; plurality of forms, 88; really
ulates effects of active powers, 114; distinct from matter, 127; "received in
never the terminus ad quem of natural the patient from the agent," 62. See also
change, 115; not alterable per se, 109; substantial form
not intended per se in natural change, forma artificialis, 2 3 g
247; passivity of, 113-114; and simili forma cadaveris, 147, 153, 233-234
tude, 241; use of figure in art, 241 forma wrporeitatis, 64, 84, 87-92, 121, 153,
finality: connects things otherwise distinct, 342·348
184-185; fading away of the question forma jluens, 30-32
An natura agat propter finem, 186. See also Freddoso, Alfred J., 248
ends; cause Freeland, Cynthia A., 27
finis cui, 171-172 Funkenstein, Amos, 125, 389
finis cuius, 171-172; distinguished from Furth, Montgomery, 68; on differentia, 56
the means (Suarez), 173
finis generationis, 172 Gabbey, Alan, 273, 287, 290, 293, 300,
finis rei genitlli, 1 7 2 310,322,340
fire, 144, 154; produced by friction or ra Galileo, 271; circular motion persists, 275;
diation, 164 resistance, 51; resistance is not distinct
Fludd, Robert, 228; on Creation, 124 from existence, 274
jluxus, fr:mntli, 3o-32 Garber, Daniel, 12, 262, 266, 271, 288,
Fonseca, Petrus, 10, 150, 272, 321, 339· 310,318,326,328,338.358.375.393
368, 375; classification of appetites, Gassendi, Pierre, g, 25, 373, 383; not all
236; creation entails no change in God, quantities are corporeal accidents, 384;
318; esse reate and final causes, 193; finis space is neither substance nor accident,
cuius, 172; genus and species, 8g; imag 359generation and corruption: defini
inary space, 357; immutability of God, tion of, 25; not motus in strict sense, 25;
317; inherence, 130; matter without reduced to alteration in Descartes'
form, 126, 126; motus as the instanta physics, 6g. See also substantial change
neous succession of forms, 268-26g; po generation, 140-144, 161; accidents in,
tentia obedientialis, 28; principle of 161-167; action of Cartesian bodies
individuation is an intrinsic difference,

Authenticated
[420] Index
generation ( cont.)
Herding, Georg, 13
analogous to human generation in Aris

hierarchy: of agents, 2 26; in Cartesian ism,
totelianism, 336-337; of animals, re

collapses to soul and nature, 398; of
quires celestial or divine assistance, 163,

forms, 166-168
164-165; distinguished from creation,

Hobbes, 281
g6, 142; equivocal, 143; final causes in,

h= vacui, 16g, 175-176
179-180; the generation of one thing is

Hurtado de Mendoza, 169
the corruption of another, 144; the

Hussey, Edward, 32
"most natural" of the soul's powers (Ar

Hutchison, Keith, 218
istotle), 157, 174; the most noble opera

Huygens, Christian, 255, 287
tion of Nature, 144

hydrostatic model, 306-311, 330
genus, 87; has no corresponding form, go

Gerson,Jean, 199
identity of indiscemibles, 375
Gill, Mary Louise, 75, 161

images, occult, have no powers of their
Gilson, Etienne, 1, 13, 393; on substantial

own, 244
form, 65
imagination: gives birth to monsters, 2o6;
glass, 376
and the representation of extension,
Glouberman, Mark, 364
346,388
Gnosticism, 125
immensitas, 385, 396
Goclenius: accidens, 132; actio, used im

immutability, 339, 397; continual change
properly for motus, 259; ars, 239; curpus,

an argument for, 316-317; ensures law
349; monsters, 205; natura, 214; spa

likeness of the laws of motion, 316; of
tium, 357
God's will, 315-319; proofs of, 317
God: absolute power of, 131; actus purus,

impenetrability, 350; not a true and essen
125; the coauthor of souls, 165; divine

tial differentia of body (Descartes), 379;
presence does not entail extension

of bodies, denied by Aristotelians 105,
(Descartes), 388; has intrinsic presence,

377
but not extension (Suarez), 387; his

in actu, 26
body, 389-390; immensity of, 385; im

inclinatio, 2 81
portance of divine goodness in Carte

incorporeal: negation of body (Gassendi),
sian physics, 396; mode of action in

359
supporting real accidents or matter

indefinite, the world is not infinite, but
without form, 128; presence, 102; pres

(Descartes), 386
ence per modum potentitl!, 333; whether

individuation: of bodies, 367, 382; princi
creation is divine art, 247-251

ples of, 367-370; static and dynamic,
grace: and potentia obedientialis, 28; re 370
quires a supernatural act of God, 220

inertia, 335; contrasted with self
Gracia,Jorge, 369
conservation of forms, so; Descartes' ex
Grafton, Anthony, 7
planation of "natural inertia", 307
Grant, Edward, ~5· 176; 357

infinite: existence of actual infinite
Grosholz, Emily, 371, 375

denied, 386
Gueroult, Martial, 313, 315, 325, 340

injluere, 18g, 249
341 inherence: in the definition of substance,
79; only potential, not actual, is essen
habitus, 6o-61 tial to accidents (Fonseca, Suarez),
Hansen, Bert, 165, 205

130-131; a "positive, real, and intrinsic
hardness, 350, 379

mode" (Suarez), 129
Hatfield, Gary, 314

in potentia, 26, 29-30
Henry,John, 383
instant, boundary between intervals, 28o
Henry of Ghent, 84
instinct, 198
Heraclides, 124
instrument, 196-197, 249

Authenticated
Index [421]
intelligences, celestial, 142-143

Major,John, 163
intended state, 172; subordinate to the

Manich<eism, 91, 92, 125
beneficiary, 173
marques d'envie, 207
intensity, 112. See also quantity

materia, interchangeable with carpus
(Descartes), 349
Janimer, Max, 295

materia signata, 368
John of St. Thomas, 26, 150, 351; against

matter: cannot naturally exist apart from
the persistence of accidents through

form, 133-134; celestial and terrestrial,
corruption, 145, 147; emanation of
86; and darkness, 124; dependence of,
powers from form, 159-16o; quies, 278;
on form is extrinsic (Suarez), 128; dis
violence and the preternatural, 223,
tinct from form, 59-60; eternity of,
225
125; and form are unum per se, 135
Judson, Lindsay, 208
138; formal effect of substantial form,
126; functions of, in Aristotelian
Kantification, 344
physics, 150; granted a more indepen
knowledge, as a human end, 235-236
dent role by late Aristotelians, 126; has
Kosman, Aryeh, 278
but one kind of quantity (Descartes),
Kurdzialek, M., 95
350; how it can exist without form, 129;
individuated by neither quantity nor
Lang, Helen, 186-187
form (Suarez), 369; inertia of, 72; intel
law: and concurrence, 323; and ordinatio,
ligible and sensible, 116; 'matter desires
335; and potentia ordinata, 320
form', how interpreted, 201-202; mat
laws of motion: first, 273-279; first,
ter without form is contradictory (Son
derivation of, 314-316; and potentia re
cinas), 126; matter without form would
sistendi, 50; second, 279, 311; third,
be otiose, 133; potentia of form, 65;
286-290, 315, 319, 335-336; third,
prime and proximate, 139; proximate,
derivation of, 328-330
237; pura potentia, 145; and quantity,
laws of nature, 201
97-109; will take on all forms
least modal change, principle of, 310-311
(Descartes), 395· See also body; corpus;
Le Bossu, Rene: active and passive po
prime matter
tentiOI, 48-49; figure and form, 111
means, useful or not only by "extrinsic
112
denomination," 197
Leibniz, Gottfried, 185, 395; absence of

mechanism, of the world, 178
providence in Cartesian physics, 396;

Melissus, 35
Cartesian motus is "nothing real," 261;

Mercer, Christia, 17
continuity condition, 297, 301; final

Mersenne, Marin, 124, 358, 385; objec
causes, 170; intrinsic differences an1ong

tions to Descartes's rules of collision,
bodies, 374-375
299
· light-metaphysics, 342
metaphorical motion, 189-191
Lindberg, David C., 17, 106

metaphysics, is desirable to man insofar as
Lloyd, G. E. R., 214

he is man (Suarez), 235
local motion, 25, 26o; never yields new

middle mathematics, 117, 12 1;
forms, 245; whether distinct from the

distinguished from pure mathematics,
mobile or the terminus, 40

ll9
location, 371-372,381
Mignucci, Mario, 208
locus, 262-263; definition of (Descartes),

Miles, Murray Lewis, 351
264; is a mode of substance, 264

Milton,John, 125
Lubac, Henri, 219

minima, 154
mobile, 23
Maier, Anneliese, 17, 30, 82, 186

mode: and (real) accident, 132; definition
Maignan, Emmanuel, 140

of (Descartes), 362; has the san1e

Authenticated
Index
mode: and (real) accident (cont.)

natural change, 217; varieties of, 25
denotation as 'quality', 302; inference

naturalism, 217
from mode to substance, 362-363

natural kinds, 21, 56, 67, 232
moles (bulk), 104, 107

nature: the concept of nature presupposes
moment, 325
a demarcation of natural change, 218;
monsters, 22; in art there are none

definition of individual nature, 227
(Toletus), 205; definition of, 205; ex

231; individual, 216-217; individual na
planation of, 205-210; material and

ture and ends, 235-239; and Nature
efficient causes of, 206; whether in

distinguished, 216; "owed to nature,"
tended by Nature, 207-210

219, 322, 324, 340; senses of 'nature',
More, Henry, 266, 373; God is an ex

213; universal principle, 228;
tended thing, 387; motus as actio or

necessity, hypothetical, 238
force, 259; space is God, 387; space is

Newton, parts of space, 373
not impenetrable, 378

Nifo, Augustino [Niphus], 140
Morin,Jean-Baptiste, 258
nihil ex nihilo, 140-142
motion: circular, persistence of, 273-275;

Nominalists: individuation, 367; quantity
composition of, 310; infinitesimal, 282;

and substance, 100
is a quality (Descartes, Monde), 302; like

nothing, has no attributes, 384
matter, created and conseiVed imme

Nullibistte (Nowhere-men), 389
diately by God, 328; necessity of circu
lar, 383; quantity of, 287, 330; quantity

Oakley, Francis, 322
of, determines only the transfer of, 294;

occupation, 389
simple, 284-285. See alm laws of mo

Ockham, William of, 82, 264, 367, 371;
tion; motus
condensation and rarefaction, 1o8;
motus: category of, 33, 36; classification of,

motus, 35· 37; quantity and distinctness
25-26; defined as translation

of parts, 101
(Descartes), 260; definition of, 25-32;

O'Neill, Eileen, 189
divisibility of (Toletus), 31; God as the

optics, 118-119
cause of, 272; of the heavens, 38; in

order, 152, 179; mundane and divine,
stantaneous succession of forms (Fon

323
seca), 268-269; is in the mobile

Oresme, Nicole, 83, 165, 205; on figure,
(Descartes), 259; is in the patient, 43

us; geometric representation of inten
44; is not action (Descartes), 258; is not

sive quantity, 303-304
the actus of a being in potentia

outcome (of collision), 293
(Descartes), 260; metaphorical, in final
causation, 189-191; as passio, 36; per

Pare, Ambroise, 207
sistence of (Coimbra, Abrade Raconis),

Park, Katherine, 222
274-276; reciprocal (Descartes), 260

Parmenides, 35
261; relativity of (Descartes), 261-262,

part of matter, 370; actual, individuated
265; scheme of, 23; tends to quies only

by motion, 372-373; potential, 372
per accidens, 278-279; uniqueness of

partes extra partes, 1oo, 3 71
(Descartes), 261, 267; whether distinct

Pascal, Blaise, So
from action and passion, 44-46;

Patrizi, Francesco, 228; existence of space,
whether distinct from the mobile or the

358; independence of space, 383; space
terminus, 35-40. See also change; mo

is a "mean" between body and spirit,
tion; natural change

358
music, 117-118
Paulus Venetus, 37
mutatio, 25
peccata naturte, 205
Pellegrin, Pierre, 56
Nagel, Thomas, 141

Perez-Ramos, Antonio, 6o
Nardi, Bruno, 140

perfection, in acting and being acted on,
natural agents, 195

46

Authenticated
Index
persistence, 273-278, 3u; positive (re

same denotation as 'mode' (Descartes),
sistance), 274
302; only passive qualities are alterable
Peter ofTarantasia (Innocent V), 91

per se, 57; species of, 110
Philoponus, existence of space, 356-357

quantitas interminata, 348
philosophy of nature, 1
quantity: changes in condensation and
physico-mathematics, 120
rarefaction (Toletus), 108; Descartes's
place, 262-263; extrinsic, 101; internal,

austere conception of, 347-348;
380; intrinsic, 102

disposition for the reception of form,
Plato, 166, 187, 201; Timt1!Us, 92, 124

148, 154; and distinctness of parts, 101;
plenitude, 239
extensive and intensive, 99; and form
Port-Royal Logique, 24, 79; on accident

are unum per accidens, 148; inheres in
and mode, 132

matter alone, 148; the intelligible mat
potentia, 63, 297-298, 393; active and pas

ter of mathematics, u7; intensive, 82;
sive, 46-51, 61; coexistence of active

intensive, not admitted by Descartes,
powers, 18o; coordination of active

346-347; intensive and extensive, 303;
powers in generation, 179; "dormitive

is potential, not actual extension, 104,
virtue," 24; logica, 27, 318; Morin ac

106; is to matter as quality to form, 146;
cuses Descartes of equivocation on, 258;

of mass, distinct from extension, 351;
neutra, 28; obedientialis, 28, 29; ordinata,

persists through substantial change,
ordinaria, 2 21, 320; resistendi, 49-51,

148; potential, not admitted by
274
Descartes, 346; preserved in condensa
power. See potentia tion and rarefaction (Ockham), 108; to
prreternatural, 2 22
tal, 303; uniqueness of the quantity in
predication, and the definition of sub

matter (Descartes), 350
stance, 79
quantum in se est, 213, 273, 324
preferred states, 73
Quercetanus (Joseph Duchesne), 65
Prendergast, Thomas, 259, 265

quies, 278, 295· See also motus, rest
presence: contrasted with occupation,
102, 389; intrinsic, 387

Ramirez, Ioannes, 225
prime matter: actuality of, 91; compared

Ramus, Petrus, 25
to darkness, 83; has a proper actus

reason, favored over the senses (Fonseca),
(Toletus, Suarez), 93-95; identified
147
with God, 93, 95-96; independence

reciprocal motion, 260-261
from God, 91; indifferently in potentia to

reflection, 288-290,300,304,312,335
all forms, 86; is pura potentia (Thomas,
refraction, 289-290
Coimbra), 91-96, 125; paradox of, 84;

Regis, Sylvain, 383; Descartes's second law
pura potentia, 150; as pura potentia
of motion, 281-283; movement con
(Thomas), 85; seventeenth-century crit
served in a single divine act, 329; op
icisms of, 81-82; uniqueness of, 85
position to final causes, 170; quantity
principle of least modal change, 310-311
individuated by location, 371; reflected
privation, 6o; cannot be intended as such,
motion, 288-289
133
Regius, Henricus, 78, 259
pro-dea, 144
regularity, 21; Cartesian and Aristotelian
prodigies, 221
explanations contrasted, 21 o-211
proportion, 152
reidentification, 375
providence, 209
Reisch, Gregor, on monsters, 205
proximity, vitalis or animalis, 192

relations, rest on nonrelational fonda
menta, 43
qumstio, structure of, 7

relics, 145
quality: can have both intensive and ex representation, of figure and quantity,
tensive quantity, 303; definition of, 120
uo-111; extraneous, 225-226; has the resistance, 49-51, 289, 298

Authenticated
Index
rest, 297; force of, 307; the "glue" that space: distinguished from body by its in
holds bodies together, 298, 3 76; not v = tangibility and penetrability, 378; in
o, 245-246 general, 372-373; generic, 371; identi
resultantia, resultatio, 159 fied with God (More), 387; imaginary,
resurrection, 233 263, 385-387; is substance (Descartes),
Rohault,Jacques, and Descartes's second 365; neither substance nor accident, yet
law of motion, 281-283 real (Gassendi), 359; "pure negation"
Rosset, Cli~ment, 243 (Fonseca), 357; separate, Aristotle's ar
rules of collision, 290-311; Rule 1, 294; guments against, 355-356; singular,
Rule 2, 294; Rule 3• 294; Rule 4• 256, 37 1
293· 295· 297·300, 304,308-309, 311; species infima, go
Rule 5, 293, 295, 305-306, 311; Rule species, preseiVation of, 174-1 7 5
6, 293, 295; Rule 7, 294-295, 304-305; speed, 281; in Aristotelianism, the inten
1639 Rule, 2gg, 3o6; statement of, in sity of impetus, 303; average, 302-303;
the Principles, 291-292 degree of, 287; instantaneous, 303
rupture, 270, 28o; definition of, 374; is a Spinoza, 93· 95· 97
mode, 302 stability: of natural features, explained by
reference to ends, 181; of second
satisfaction, 202 element particles, 394-395
Scaliger,Julius Caesar, 276 state, 290
scheme, general form of explanation, 57 Stoicism, 125
Schmitt, Charles, 7, 17, 175 Suarez, Franciscus, 10, 21, 26, 66, 371,
Schuster,John, 288 375; absolute and ordained power, 320;
Scotists: on the emanation of powers from active and passive potentit1!, 47-48; all
form, 159; on form, 66; on the role of accidents inhere in substance by way of
accidents in generation, 161-162 quantity, 97; animals recognize the ends
sea, 168 toward which they act, but only "mate
seed, cannot produce the soul by itself, rially," not "formally," 199; appetite for
165 knowledge, 236; concurrence, 321; con
seminal spirit, 64 currence of God with second causes,
senses, teleological definition of, 237 2 19; contraries, 56; creation and con
simplicity: of circular motion, 285; of rec seiVation are one continuous action,
tilinear motion, 284; unity as, 285 325; creation is not a motus, but it is a
situs, 263-264 transeunt action, 332; emanation of
size, limits on, 203-204 powers from form, 16o-161; ends and
skill: as artifact, 242; is not nature, 228 cognition, 191-192; essence includes
229 individual, not specific, form, 234; Eu
Smith, A. Mark, 51, 185 charist, 129-131; every "connate neces
solidity, 379 sity" arises from an end, 18o; existence
Soncinas, Paulus, 86-87, 259; ends act ac of geometric points, 297; finis cuius and
cording to their esse intentionale, 191; es finis cui, 172; generation of fire and
sence and matter, 233; matter without minerals, 164; God's presence, 387; im
form, 126; potentia resistendi, 49-51; manent action, 41-42, 46; intrinsic
quantity and matter, 100 presence distinguished from relative
Soto, Domingo, 216; concurrence of God position, 387; matter and form, 59;
with second causes, 219; Eucharist, 129 matter and quantity are inseparable and
soul: confusedly attributed to body, 393; reciprocal, 348; matter is not God, 95;
exists for the sake of thinking, 236; hu matter without form, 126; natura, sup
man soul in the hierarchy of kinds, 231; positum, and subsistentia, 215; per
multiplicity in one individual, 88; why sistence of accidents through
joined with matter, 237· See also anima COfl"\lption, 145-147; plurality offorms,
mundi 88; potentia obedientialis, 28; potentia re

Authenticated
Index
Suarez, Franciscus (cont.) local motion has none (Descartes), 285.

sistendi, 49-51; against preexisting See also ends
forms, 141; preferred states, 7g-74; Terreni, Guido, 191
principle of individuation is entitas, g6g, thermometers, g47
g75; privation, 6o; refutes the Nominal Thomas Aquinas, 26, gg2; cited by Suarez
ist position on quantity, 102-10g; reg on the Eucharist, 129; emanation of
ularity would result from efficient powers from form, 158-159; Eucharist,
causes alone, 178; subjectum or substrate g8; intelligible and sensible matter, 116;
of form, 150; 'substantial form' not con materia signata is the principle of indi
tradictory, 76; union of matter and viduation of bodies, g68; nothing oc
form, 1g5; unity and order of accidents, curs outside divine governance, 207;
7g prime matter, 82, 91; sense of 'violent',
substance: "abstractness" of the Aristo 22g
telian view, 106; as self-subsistent, 78, Thomists: concurrence is distinct from the
g62; changes in the definition of, 78; action of second causes, g 21; matter
classification of substances according to without form, 126; quantity and matter,
activity and passivity, 2g1; created sub 100, 105
stances act through accidents, 16g, 166; Tocanne,Bemard, 170,186
first, 149; first and second, 78, 87; in together, g74
complete, 77, go; not distinct from Toletus, Franciscus, 10, 92, gog, g15, g51;
quantity (Descartes), g5g; only inciden appetites, 202; art and imitation, 241;
tial to optics and astronomy, 117; recon corpus mathematicum, physicum, g4g; the
ciliation of the Categories and the Physics, dictum 'matter desires form', 202; edu
87; structure of material substance, 146, cation of form from matter, 14g; eter
148-151 nity of the world, 125; figure, 110;
substantial change, 61, 6g; distinguished h= vacui, 175; immutability of God,
from accidental change, 71; and the g18; inanimates have only a passive
forma corporeitatis, 87 principle of movement, 2go; intelligible
substantial form, 61; creation of, 142 and sensible matter, 116; motus and ter
144; definition of, 65; empirical argu minus, gg; natural limits to size, 20g
ments on behalf of, 70-74; essential 204; persistence of accidents through
qualities are not substantial forms, 167; corruption, 147; quies as the terminus of
and matter are unum per se, 77; no motus, 278; reception of form, 62; spon
determinate contrary to, 67; not a con taneous cooling of hot water, 74; trans
tradiction in terms, 76-77; origin of ac mutation, 85
cidents, 72; specificity of (Suarez), 8g; trajectory, 270
uniqueness of, 73, 88; used to explain translation: in the definition of motus,
the coordination of motus in generation, 260-261; is a symmetric relation, 265;
75; used to explain spontaneous cool logic of, 266; temporality of, 267
ing, 7g-74; used to explain unity of Trent, Council of, g8
powers, 71 truth, double, doctrines of, 131
substrate argument, 5g-6o, 82
succession, 268-271 Ubi, 262-263; is superfluous (Descartes),
sun, its role in generation, 164 264; a real intrinsic mode of bodies,
supernatural: extraordinariness, 2 2 1; im 102; terminus oflocal motion
mediacy of divine action, 220; porten (Coimbra), 26g
tousness, 2 2 2 ubicatio, 245
suppositum, 1g6 union: how mediated by accidents, 146
15g; in Descartes's rules ofcollision, 2go
tangibility, g78-g7g 291;ofaccidentandsubstance, 1g7; of
tendency, 259 form and matter, 1g5-1g8;ofmatterand
terminus: ad quem, 23, 6g; a quo, 2g, 63; quantity, 148-150. See also unum

Authenticated
Index
unity: of continuation, 234; intrinsic and Weinberg, Julius, 124

adventitious, 234, of powers, 71 Weisheipl,James A., So, 200
universals, go, 187 Westfall, Richard, 313
unum: per accidens, 135; per se, 77, 135 Wild, John, 279
137, 234· 237 will, metaphorically moved by ends, 190
Williams, Bernard, 381
vacuum, explanations of the horror vacui, Williams, C.J. F., 24
175 Williams, Raymond, 214
Verbeek, Theo, 12
vicinity, 2 6 5 Zabarella, 66, 82, 350; definition of
violent change, 222 disposition, 151; distinction between
vis ad agendum, 312, 337 matter and form, 124; education of
vis initiativa, 22 form from matter, 140; elemental muta
Vitelleschi, Mutius, 275 tion, 68; figure, 110; matter, 86-87;
vortices, 267, 272 plurality offorms, 88, 91; potentia re
sistendi, 50, 274; preternatural move
Wallace, William, 17, 275 ments, 224; quantitas interminata, 348;
Waterlow, Sarah, 22, 23, 41, 229, 278 reception of form, 62
weight, 393 Zimara, Marcantonio, 9, 224

Authenticated
Authenticated
,!7IA8A1-eigihg!

Authenticated

Des Chene, Dennis - Physiologia - Natural Philosophy in Late Aristotelian and Cartesian Thought-Cornell University Press (2000)

Uploaded by

Copyright:

Available Formats

Des Chene, Dennis - Physiologia - Natural Philosophy in Late Aristotelian and Cartesian Thought-Cornell University Press (2000)

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

Des Chene, Dennis - Physiologia - Natural Philosophy in Late Aristotelian and Cartesian Thought-Cornell University Press (2000)

Uploaded by

Copyright:

Available Formats

PHYSIOLOGIA

Cornell University Press

Ithaca and London

Copyright© I996 by Cornell University

First published I996 by Cornell University Press

First printing, Cornell Paperbacks, 2000

Library of Congress Cataloging-in-Publication Data

Des Chene, Dennis.

Physiologia : natural philosophy in late Aristotelian and

Cartesian thought I Dennis Des Chene.

Includes bibliographical references and index.

ISBN o-8oi4-3072-o (cloth: alk. paper)

ISBN o-8oi4-8687-4 (pbk. : alk. paper)

1. Physics-Philosophy. 2. Aristotle-Influence. 3· Descartes,

Rene, I596-I65o-Influence. 4· Philosophy, Medieval. 5· Philosophy

Printed in the United States of America

Abbreviations and Orthographical Conventions xi

PART 1 Vicaria Dei

2 .1. Potentia and Actus 24

2.2. Independent Existence of Motus 34

2.3. Action and Passion 40

2.4. Active and Passive Potentim 46

3· Form, Privation, and Substance 53

3.1. Principles of Change 55

3.2. Substantial Form and Prime Matter 64

3·3· Form as Substance 76

4· Matter, Quantity, and Figure

4.2. Quantity and Prime Matter 97

4·3· Figure and Other Qualities 109

5· The Structure of Physical Substance 122

5·3· Conditions for the Reception of Form: Dispositions 138

5-4- Substantial Form and Active Powers 157

6. Finality and Final Causes 168

7· Nature and Counternature 212

7·3· Artifacts, Human and Divine 239

PART n: Bodies in Motion

g. Parts of Matter 342

9·3· Physical Questions: The Sufficiency of Extension 377

10. World without Ends 39 1

During that time I received support from the philosophy department; a

a, b first, second column

A,B,C ... top-to-bottom divisions of the page, when marked

his Coimbra In Phys.

f ff as usual, but superscripted to avoid confusion

Orthography and Translations

preseiVed. I have also kept the accentuation occasionally used to distin­

Oxford Calculators, and to the cosmologies of Kepler and Copernicus. The

Brought to you by | University of Warwick

textbooks: the marking of territory in the intellectual landscape accom­

Brought to you by | University of Warwick

Brought to you by | University of Warwick

Brought to you by | University of Warwick

other, directed changes. Cartesian matter, moreover, is, from an Aristotelian

Brought to you by | University of Warwick

Brought to you by | University of Warwick

an author's systematic views will be parceled out among responses to dozens

Brought to you by | University of Warwick

What is contained in natural Philosophy is either about the principles or

Brought to you by | University of Warwick

preseiVed. I have also kept the accentuation occasionally used to distin

textbooks: the marking of territory in the intellectual landscape accom

F or the Aristotelian, common sense, authority, and reason con

The Physics would have change occur in three categories-quantity, qual