Descartes on Causation

Descartes on Causation

This page intentionally left blank


TAD M. SCHMALTZ

12008

1Oxford University Press, Inc., publishes works that further

Oxford University’s objective of excellencein research, scholarship, and education.

Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi

New Delhi Shanghai Taipei Toronto

With offices in Argentina Austria Brazil Chile Czech Republic France Greece

Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2008 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.198 Madison Avenue, New York, New York 10016

www.oup.com

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,

electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication DataSchmaltz, Tad M., 1960–

Descartes on causation / Tad M. Schmaltz.p. cm.

Includes bibliographical references and index.ISBN 978-0-19-532794-6

1. Descartes, René, 1596-1650. 2. Causation. I. Title.B1878.C3S26 2007

122.092–dc22 2007002235

2 4 6 8 9 7 5 3 1Printed in the United States of America

on acid-free paper

www.oup.com

Je ne puis pardonner à Descartes: il voudrait bien, dans toute la philoso-phie, se pouvoir passer de Dieu; mais il n’a pu s’empêcher de lui donnerune chiquenaude pour mettre le monde en mouvement; après cela, il n’aplus que faire de Dieu.

I cannot forgive Descartes: in his whole philosophy he would like to dowithout God; but he could not refrain from giving him a flick to set theworld in motion; after that, he had no more use for God.

—Propos attribués à Pascal


Acknowledgments

Work on this book was made possible by a research grant from the Arts and SciencesCommittee on Faculty Research at Duke University. I presented some of the materialin the book as a faculty member of the 2004 NEH Summer Institute at the Universityof Wisconsin–Madison, and also as speaker for colloquia or conferences at theCenter for Philosophy of Science at the University of Pittsburgh, Harvard University,Simon Frasier University, the University of Oxford, the University of Sydney, theUniversity of Washington, the University of Western Ontario, and WashingtonUniversity in St. Louis. I am grateful to the editors of Oxford Studies in EarlyModern Philosophy for permission to reprint material in chapter 2 that appeared inan earlier version in that journal, vol. 3 (2006).

It is a pleasure to acknowledge the assistance, both professional and personal, ofcolleagues, friends, and family. Special thanks to my Newtonian/Kantian friendsAndrew Janiak and Eric Watkins (both experts on the issue of causation in modernphilosophy) for reading and commenting insightfully on the entire penultimate ver-sion of the manuscript. I also received helpful comments on various chapters or relat-ed material from more colleagues than I can remember. With apologies to thoseI have forgotten, I would like to thank Roger Ariew, Ric Arthur, Andrew Chignell,Ken Clatterbaugh, Michael Della Rocca, Dennis Des Chene, Karen Detlefsen, AlanGabbey, Dan Garber, Geoff Gorham, Sean Greenberg, Dan Kaufman, Sukjae Lee,Tom Lennon, Peter Machamer, Ted McGuire, Steve Nadler, John Nicholas, EileenO’Neill, Andy Pessin, Gonzalo Rodriguez-Pereyra, Michael Rosenthal, MarleenRozemond, Don Rutherford, Lisa Shapiro, Alison Simmons, James South, and thereferees for the press. Thanks also to my editor, Peter Ohlin, who expertly shep-herded the manuscript through to publication. On a more personal note, I owe a greatdebt to my wife, Louise, and to my children, Johanna and Sam, for providing thelove and encouragement that have sustained me through thick and thin. Finally,

viii ACKNOWLEDGMENTS

I would like to dedicate this book to the memory of the late Margaret Wilson, whonot only influenced but also supported my work on Descartes, as she has the workin early modern philosophy of so many other scholars of my generation.

T.M.S.Durham, North Carolina

November 2006

Contents

Abbreviations xi

Introduction 3

1 The Scholastic Context 9

1.1. Medieval Rejections of Occasionalism 12

1.2. Suárez on Efficient Causes and Concursus 24

1.3. From Suárez to Descartes 44

2 Two Causal Axioms 49

2.1. The Containment Axiom 51

2.2. The Conservation Axiom 71

2.3. From Axioms to Causation 84

3 Causation in Physics 87

3.1. God as Universal and Primary Cause 89

3.2. Laws as Particular and Secondary Causes 105

3.3. Descartes’s Conservationist Physics 125

4 Causation in Psychology 129

4.1. Mind–Body Interaction and Union 131

4.2. Body-to-Mind Action 145

x CONTENTS

4.3. Mind-to-Body Action 163

5 Causation and Freedom 178

5.1. Jesuit Freedom and Created Truth 180

5.2. Indifference and Human Freedom 192

5.3. Human Freedom and Divine Providence 208

Conclusion 217

Works Cited 221

Index 231

xi

Abbreviations

In the notes and text, I use the following abbreviations, keyed to the texts in WorksCited.

DESCARTES

AT Descartes 1964–74 (ed. C. Adam and P. Tannery), cited by volume:page;abridged English translations of RM, W, DM, PP, and PS are in Descartes1984–85, vol. 1, English translations of the Meditations and the accompa-nying Objections and Replies in Descartes 1984–85, vol. 2, and abridgedEnglish translations of Descartes’s correspondence in Descartes 1991.Translations in these texts are keyed to pagination in AT.

DM Discourse on the Method/Discours de la Methode, cited by part; in AT 6.

PP Principles of Philosophy/Principia Philosophiae, cited by part.article; orig-inal Latin edition in AT 8-1, French edition in AT 9-2.

PS Passions of the Soul/Passions de l’ame, cited by part.article; in AT 11.

RM Rules for the Direction of the Mind/Regulae ad Directionem Ingenii, citedby rule; in AT 10.

W The World/Le Monde, cited by chapter; in AT 11.

xii ABBREVIATIONS

SCHOLASTIC CONTEXT

DA Suárez, De Anima, cited by book.chapter, and paragraph; in Opera 3.

MD Suárez 1967 (Metaphysical Disputations/Disputationes Metaphysicae),cited by disputation.section, paragraph, and volume:page; thus, XVII.1, ¶6,1:582 = seventeenth disputation, first section, sixth paragraph, in the firstvolume, page 582. There is an English translation of disputation VII inSuárez 1947, of disputations XVII–XIX in Suárez 1994, and of disputa-tions XX–XXII in Suárez 2002.

Opera Suárez 1866 (Opera Omnia), cited by volume:page.

QPG Thomas Aquinas, Questions on the Power of God/Quaestiones de PotentiaDei, cited by question.article; in TA 13, with an English translation inThomas Aquinas 1952.

QT Thomas Aquinas, Questions on Truth/Quaestiones de Veritate, cited byarticle.section; in TA 14, with an English translation in Thomas Aquinas1987.

S Durandus 1964 (On the Theological Sentences/In Sententias Theologicas),cited by book.distinction.question, and paragraph, and volume:page; thus,II.1.5, ¶11, 1:131 = second book, first distinction, fifth question, first arti-cle, eleventh paragraph, in the first volume, page 131. Online English trans-lation of S II.1.5 in Durandus (n.d.).

ST Thomas Aquinas 1964–81 (Summa Theologiæ), cited by part.question.article, and response (ad); thus, I.104.1, ad 4 = first part, question 104, firstarticle, response to the third objection. This edition includes the Latin withfacing English translation.

TA Thomas Aquinas 1871–80, cited by volume:page.

Citations marked with an asterisk (*) do not have publicly available English transla-tions. Although I have consulted the translations cited above, all translations of pas-sages from the primary texts above are my own unless indicated otherwise.



3

Introduction

Margaret Wilson has observed that the issue of causation in the early modern period“presents the interpreter with a peculiar problem,” since on the one hand, “the notionof causality is central to the period’s major positions and disputes in metaphysics andepistemology,” whereas on the other hand “few of the most prominent figures of theperiod enter into detailed or precise accounts of the relation of causal dependence orcausal connection” (Wilson 1999e, 141).1 Though Wilson uses this observation toframe a discussion of Spinoza, Descartes would seem to provide a case in point aswell. In his argument for the existence of God in the Third Meditation, for instance,Descartes relies heavily on claims concerning causation for which he provides rela-tively little explication or defense. Most notably, there is his appeal in that text to theaxiom that an efficient cause contains “formally or eminently” all of the “reality”that it produces in its effect. There is also Descartes’s notorious claim in a letter toPrincess Elisabeth that we derive from the senses a “primitive notion” of the unionof our mind with a body that involves a conception of the “forces” of the united ele-ments to interact. It is largely left to the interpreter to determine the precise status ofthis notion and the precise nature of the forces involved in the union.

1. For some recent survey accounts of causation and causal explanation in the seventeenthand eighteenth centuries, see Yakira 1994 and Clatterbaugh 1999. For a more substantivetreatment of such accounts, see Carraud 2002. Carraud’s discussion of the issue of causationin Suárez and Descartes in particular has influenced the discussion in the present study.Whereas Carraud focuses on the connections of causation to the principle of sufficient reason(as reflected in his title, Causa sive ratio), however, I am more concerned here with the natureof causal connection or dependence.

Even so, we can at least start to understand Descartes’s theory of causation interms of its relation to an older account of causality deriving from the work ofAristotle. In the second book of his Physics, Aristotle lays out four different kindsof “cause” (aition): material, formal, efficient, and final. The material cause is that outof which something comes to be, such as the bronze of a statue, the formal cause theform of that which comes to be, such as the shape of the statue, the efficient causethe primary source of change, such as the sculptor in the case of the production of thestatue, and the final cause that for the sake of which there is a change, such as the goalof the sculptor in producing the statue.2 This account presupposes a broader conceptof causality than that with which we are now familiar.3 Yet it also is broader than theconcept that Descartes employed, since he tended to understand causality exclusivelyin terms of efficient causes.4 It is not too surprising that this shift to a focus onefficient causation has a history. What is perhaps surprising, though, is that the devel-opment of the Aristotelian theory of causation in early modern scholasticism preparedthe way for the shift reflected in Descartes’s writings. To understand this shiftadequately, we need to consider the scholastic context in which it occurred.

The general claim that Descartes’s views cannot be adequately understood in abstrac-tion from their relation to developments in scholasticism is of course not new. There isan established tradition in the French literature of emphasizing this relation, and severalrecent English-language studies have focused on Descartes’s debts to scholasticism withrespect to issues in metaphysics, natural philosophy, and psychology.5 What is new,however, is my sustained attempt here to use our knowledge of scholastic treatments ofcausality as a key for deciphering Descartes’s often-cryptic remarks concerning variouskinds of causal connections. Sometimes the relevance of the scholastic context is uncon-troversial, as in the case of the appeal in the Third Meditation to the formal or eminentcontainment of the effect in its efficient cause. However, the overarching thesis of thisstudy is that Descartes’s theory of causation is in fundamental respects similar to amedieval account of causality that many early modern scholastics rejected.

It might be thought that the medieval account to which I refer is occasionalism,the view that God is the only real cause and creatures merely “occasional causes” ofchanges in nature. After all, we will discover that such a view has medieval origins,and that it was almost universally rejected in early modern scholasticism. Furthermore,

4 DESCARTES ON CAUSATION

2. The characterizations of the causes, but not all of the particular examples, are drawnfrom Aristotle’s remarks in Physics II.3, 194b15–195a1 (Aristotle 1984, 1:332–33).

3. For a discussion that emphasizes the oddity of the Aristotelian concept of causalityfrom a more contemporary perspective, see Frede 1980. Frede draws attention to the narrow-ing of this concept in the work of the Stoics.

4. Though see the complications for Descartes’s restriction to efficient causality that I dis-cuss in §2.1.2 (ii).

5. The classic discussion in the French literature of Descartes’s connections to scholasti-cism is Gilson 1930; see also Gilson 1913a and 1925. For a critique of the understanding ofthese connections in Gilson, see Dalbriez 1929 (cf. chapter 2, note 42, on the Gilson-Dalbriezdebate). For examples of recent work in English on the scholastic context of Descartes’s work,see Des Chene 1996, 2000a, and 2001; Rozemond 1998; Ariew 1999; and Secada 2000.

several of Descartes’s Cartesian successors endorsed various versions of occasional-ism, the strongest form of which is present in the work of the French CartesianNicolas Malebranche.6 As the introductory remarks in chapter 1 document, the viewthat Descartes himself was an occasionalist, and indeed, the father of occasionalism,dates from the seventeenth century. More recently, commentators have continued todefend occasionalist readings of Descartes. Perhaps the classical source in theEnglish-language literature for such readings is Norman (later, Kemp) Smith’s 1902text, Studies in Cartesian Philosophy. In this work, Smith claims to find in Descartesthe view that God conserves bodies by re-creating them at each moment, and hetakes such a view to support the occasionalist conclusion that bodies “cannot becapable of causing changes in one another: not having sufficient reality to persist,they cannot have sufficient force to act” (Smith 1902, 73–74). In addition, Smithargues that the implication of Descartes’s dualism that mind and body have radicallydifferent natures leads inevitably to the denial of genuine mind–body interaction.Though he admits that Descartes sometimes asserted that there is such interaction,Smith insists that this evidence reveals merely that he “inconsistently and vainlyattempts to escape occasionalism,” concluding that “the inevitable consequences ofhis rationalism are one and all emphasized by his successor, Malebranche” (85).7

Smith’s emphasis on the problems for causal interaction deriving from Descartes’sdualism is linked to what Bernard Williams has called “the scandal of Cartesian inter-actionism,” which scandal derives from the fact that there is “something deeply mys-terious about the interaction which Descartes’s theory required between two items oftotally disparate natures, the immaterial soul, and the [pineal] gland or any other partof an extended body” (Williams 1978, 287). There is the claim in the literature that atcertain moments, at least, Descartes was led by his recognition of the scandalous prob-lem of mind–body interaction to deny that mind and body, as entities with distinctnatures, can be real efficient causes of changes in each other.8 But commentators also

Introduction 5

6. For a survey of the different forms of occasionalism in the work of Cartesians such asClauberg, Clerselier, Cordemoy, La Forge, and Malebranche, see Prost 1907; Gouhier 1926,ch. 3; Specht 1966, chs. 2 and 3; and Nadler 1997.

7. In his New Studies in the Philosophy of Descartes, (now, Kemp) Smith emphasizesmore Descartes’s distance from Malebranche; see Smith 1952, 212–17. Even in this text, how-ever, Smith claims that the occasionalism of Geulincx and Malebranche “may be explained”by the view in Descartes that “God, in His re-creation of things, has to be regarded as contin-uously modifying them in an orderly fashion” (218–19).

8. For the textbook view that early modern occasionalism is a response to the scandalousproblem, see the references in the works cited in chapter 1, note 1. An example of the viewthat Descartes took this problem to preclude mind–body interaction is provided by Keeling’sclaim: “The defining attributes of mind and body being wholly different and mutuallyexclusive, direct causal interaction between them, [Descartes] maintains, is necessarily impos-sible” (Keeling 1968, 153). For variations on this claim, see Radner 1971, 1985a, and 1985b;Mattern 1978; Broughton and Mattern 1978; Baker and Morris 1996, 138–62; and Gorham1999. In Broughton 1986, there is the more limited conclusion that Descartes was led by hisconception of the differences between mind and body to rule out genuine body-to-mind actionin sensation.

have revived Smith’s conclusion that Descartes was committed to occasionalism in hisphysics. There is, for instance, Daniel Garber’s claim that “it seems to me as clear asanything that, for Descartes, God is the only cause of motion in the inanimate worldof bodies, that bodies cannot be genuine causes of change in the physical world ofextended substance” (Garber 1993, 12).9 Despite their differences, the various inter-pretations connected to Smith’s early discussion agree in taking Descartes’s system todeviate from the standard scholastic position that both material and immaterial crea-tures make a genuine causal contribution to natural interactions.

Nevertheless, the deviant account of causation that I attribute to Descartes is notoccasionalism; in fact, the view I have in mind is, in the context of the medievalscholastic debate over causality, the antipode of occasionalism, namely, the view thatcreatures rather than God are the real causes of natural change. This “mere conser-vationism”—so called because God’s role in natural causation is limited to the cre-ation and conservation of the world10—was simply too radical for most scholastics.At the beginning of the early modern period, the received scholastic view was a“concurrentism” that allowed, against occasionalism, that creatures have real causalpower but that nonetheless held, against mere conservationism, that God contributesa causal “concursus” to every creaturely action. Descartes admittedly helped himselfto the language of concurrence in his discussions of causation. This fact servesto explain why recent critics of occasionalist readings of Descartes’s theory ofcausation have attempted to link this theory to a more standard sort of scholasticconcurrentism.11 However, I find reasons internal to Descartes’s system—drawnparticularly from his account of causation in physics—for the conclusion thatcreated entities rather than God are the true causes of natural change. Given this con-clusion, the challenge for Descartes is to reconcile the claim that creatures arecausally efficacious not only with the fundamental tenets of his ontology but alsowith the doctrine, which he inherited from the scholastics, that the created world canremain in existence only because God continually conserves it.

Scholasticism also turns out to be relevant to the debate in the literature over thepurported “scandal of Cartesian interaction.” I have mentioned that this scandal issupposed to derive from the fact that mind and body differ in nature. Critics have


9. There is a more detailed defense of this claim in Garber 1992, ch. 9. Garber’s occasion-alist reading of Descartes’s physics is anticipated not only in Smith 1902 but also in Machamer1976, 178–80, and Hatfield 1979, and it is embraced in Bennett 2001, 1:98–100, and Gorham2004, 400–403. See also Des Chene 1996, 312–41. In contrast to Smith, however, Garberemphasizes that his interpretation does not take Descartes to be an occasionalist in aMalebranchean sense, since it allows him to attribute causal powers to finite minds as well as toGod; see Garber 1992, 299–305.

10. This label was not used in the medieval or early modern periods but derives from thecontemporary literature; see chapter 1, note 4.

11. See, for instance, Della Rocca 1999 and forthcoming; Pessin 2003; and Hattab 2000,2003, and 2007. See also Clatterbaugh 1995 and 1999, ch. 3, though Clatterbaugh emphasizesthat the kind of concurrentism he attributes to Descartes differs fundamentally from the oldscholastic version of this doctrine.

charged that what renders this sort of difference incompatible with mind–body inter-action is Descartes’s scholastic axiom requiring the containment of the reality of aneffect in its cause.12 Over the past couple of decades, several apologists haveresponded on Descartes’s behalf that the axiom itself is perfectly consistent with hisown commitment to mind–body interaction.13 However, this line of defense has notalways been supplemented with a detailed consideration of the relevant scholasticcontext of this axiom.14 Here I make such a consideration the centerpiece of myargument that this axiom raises no general problem in Descartes for mind–bodyinteraction. Yet in contrast to much of the apologetic literature, I also claim that therewere in scholasticism specific difficulties for interaction between material andimmaterial entities that were more pressing for Descartes than the purportedly scan-dalous problem that has tended to preoccupy recent commentators.

As the foregoing remarks indicate, there is an emphasis in this study on the con-text, and more specifically the scholastic context, of Descartes’s theory of causation.Chapter 1 is devoted to a consideration of that context. I start with a brief account ofthe origins of occasionalism in medieval Islamic theology and then turn to the rejec-tion of this account of causation in later medieval thought. Though concurrentismwas the dominant alternative to occasionalism offered during this period, I also con-sider arguments for the more radical mere conservationist alternative. In addition tothe medieval responses to occasionalism, though, an important part of the scholasticcontext of Descartes’s theory of causation is provided by the distinctive metaphysi-cal framework for efficient causality in the work of the early modern scholasticFrancisco Suárez. I highlight in particular those features of this framework that pre-pare the way for the transition from a more traditional Aristotelian view of causalityto what we find in Descartes.

Then I turn in chapter 2 from the scholastic context to Descartes’s own views, start-ing with an endorsement of basic causal axioms that reveals most clearly his debts toscholasticism. Here I counter the suspicion that his discussion of these axioms does notreflect a particularly deep view of causality. Indeed, I argue that Descartes is best readas adapting abstract constraints that he inherited from the scholastics to fit his radicallyanti-scholastic ontology. Yet though I side with the apologists in taking Descartes’scausal axioms to allow for the causal efficacy of created beings, I also emphasize thenegative point that his remarks concerning these axioms do not settle the issue of hisfinal position regarding the three main accounts of causation that emerged from themedieval period, namely, occasionalism, concurrentism, and mere conservationism.To discern his true intentions with respect to these accounts, we need to considerDescartes’s treatments of concrete instances of causal interaction.

Chapter 3 concerns the treatment in Descartes’s physics of body–body interaction.We have already encountered the occasionalist interpretation of Descartes’s physics.

Introduction 7

12. Thus, the axiom is prominent in discussions in the literature cited in note 8.13. See, for instance, Richardson 1982 and 1985; Loeb 1981, 134–49, and 1985; Bedau

1986; O’Neill 1987; Jolley 1987; Schmaltz 1992b; and Wilson 1999b.14. An important exception to this is O’Neill 1987; I discuss her reading of Descartes’s

causal axiom further in §2.1.3 (ii).

Such an interpretation may seem to be confirmed by his own claim—central to themetaphysical foundations of his physics—that God is the “universal and primary cause”of motion.15 However, I argue that the particular account in Descartes of the “ordinaryconcursus” that God provides as primary cause of motion in fact supports the mere con-servationist position that such a concursus consists simply in the continued creation ofmatter in motion. Changes in motion are to be explained by appeal not to this concur-sus but rather to the features of bodies that correspond to the bodily “forces” thatDescartes posited in his physics.

In chapter 4, I consider causal interaction in the context of Descartes’s dualistic psy-chology. As even Smith acknowledges, Descartes explicitly allowed for genuine causesother than God in the case of both body-to-mind action and mind-to-body action. Ineach case, however, there are complications for his account of causation that differ fromthe more familiar scandal of Cartesian interactionism but that are linked to earlierscholastic discussions. Moreover, Descartes’s conservationist physics creates difficul-ties for his view of mind-to-body action that he never fully confronted. Even so, his sug-gestion throughout is that God’s contribution to mind–body interaction is exhausted byhis continued production of the natures that serve to explain such interaction.

Finally, in chapter 5 I take up the special considerations for Descartes’s theory ofcausation that arise from the case of our free action. In this case especially it isimportant not to succumb to the temptation of taking the easy way out by consider-ing his views in abstraction from their scholastic context. We can adequately under-stand what he has to say about human freedom only in relation to different viewswithin scholasticism concerning the “indifference” of our free action and of thecompatibility of that action with divine foreknowledge and providence. Descartes’sdiscussion of human freedom is distinguished from all other scholastic accounts bythe fact that it presupposes his idiosyncratic doctrine of the divine creation of eter-nal truths. However, it turns out that the implication of this doctrine that God is thecause even of truths concerning our free action does not compromise Descartes’sconsidered position that our undetermined will, rather than God, is the immediatecausal source of that action. Once again I find reason to distance Descartes not onlyfrom occasionalism but also from the concurrentism that was so prominent in earlymodern scholasticism.


15. See the view in Smith 1902 discussed above, as well as the literature cited in note 9.

9

1

The Scholastic Context

In his 1696 Doubts concerning the Physical System of Occasional Causes (Doutessur le systeme physique des causes occasionnelles), the then-future perpetual secre-tary of the Paris Académie des sciences, Bernard le Bovier de Fontenelle, offered anintroductory histoire des causes occasionnelles. There he claims that “occasionalcauses are not ancient,” but in fact derive from the dualism of Descartes. Descartes’sview that mind as thinking substance is really distinct from body as extended sub-stance introduced “an extreme disproportion between that which is extended and thatwhich thinks.” Given this disproportion, the question arose “how bodily motionscause thoughts in the soul” and “how thoughts in the soul cause motions in the body.”Recognizing that motion and thought “have no natural connection” and thereforecannot “be regarded as real causes,” Descartes “invented” the theory of occasionalcauses, according to which “God on the occasion of bodily motion, could imprint athought in the soul, or on the occasion of a thought of the soul, imprint a motion inbody” (Fontenelle 1989–2001, 1:529–30). Here is an early source for the old textbookview that occasionalism arose from the problem in Descartes of explaining how sub-stances as different in nature as mind and body could interact.

In fairness to Fontenelle, it must be said that he does not endorse the suggestionin the textbooks that occasionalism is merely an ad hoc solution to the Cartesianproblem of mind–body interaction.1 He notes in the Doubts, after all, that Descartesappealed to the “occasional causes that owed their birth to the system of the soul” inorder to provide an explanation of how motion can be communicated in collision.

1. For the textbook view, see also the English-language literature cited in Nadler 1997,75–76, n.1, and the German- and English-language literature cited in Perler and Rudolph 2000,15, n.1. The authors of the texts including these citations are themselves critical of this view.

According to Fontenelle, Descartes made “God the true cause that, on the occasionof the collision of two bodies, transported the motion of the one into the other”(1:530). To an extent, then, Fontenelle anticipates the recent objection that earlymodern occasionalism addressed problems concerning the physics of force, as wellas those concerning the metaphysics of dualism.2

Even so, the view common to Fontenelle and the textbooks that occasionalismoriginated in Descartes is, in a word, false. In fact, the theory was quite ancient bythe time of Descartes’s birth. Occasionalism owes its origins not to Cartesian meta-physics and physics, but rather to a view of divine omnipotence that was prominentwithin a certain group of medieval Islamic theologians. Islamic occasionalism wassubject to attack during the High Middle Ages, when a consensus was reached thatsettled on the position that God as “primary” cause communicates his power to “sec-ondary” causes in nature. Later thinkers proposed importantly different accounts ofsecondary causality, but by the beginning of the seventeenth century occasionalismwas all but a dead theory. So much so, in fact, that one early modern scholastic—tobe discussed below—could find no recent author who unequivocally endorsed theview that “created things do nothing but that God instead effects all things in theirpresence” (MD XVIII.1, ¶1, 1:593).

At best, then, problems in Descartes led to the revival of an old and, by the startof the early modern period, largely discredited theory of occasionalism. The ques-tion of whether Descartes himself endorsed a version of occasionalism is one that wewill address in due course. Before taking up his views concerning causation, how-ever, we need to consider the context in which these views were developed. Thiscontext is provided by Aristotelian scholasticism, which at the beginning of the earlymodern period was a dominant intellectual force in Europe. The importance ofscholasticism is particularly evident given Descartes’s own appeals in his discus-sions of causation in the Meditations and elsewhere to scholastic axioms such as thatthe effect must be contained “formally or eminently” in its cause and that the con-tinued existence of the world depends on a divine act not distinct from his creationof that world (see chapter 2). Closer consideration reveals that the axioms to whichDescartes appealed were in fact linked to profoundly anti-occasionalist theories ofcausation. By itself, this fact does not reveal that Descartes himself rejected occa-sionalism. There remains the possibility that he had a revolutionary understanding ofthe old scholastic concepts. But to see what he did in fact think, we need to considerhow he stood in relation to scholastic accounts of causation.

In §1.1, I begin my consideration of the scholastic context of Descartes’s theory ofcausation with the medieval rejection (or, better, rejections) of occasionalism. I turnfirst to the most prominent form of occasionalism in the medieval period, whichderives from the Islamic tradition. My somewhat selective survey of Islamic occa-sionalism focuses on the discussion of this position in two important medievalsources. Then I examine two different alternatives to this theory in the work ofThomas Aquinas (1225–74), the Dominican church father, and of his Dominican


2. For this objection, see, for instance, Nadler 1997. Cf. the similar anticipation in theviews in Smith 1902 considered in the introduction.

critic, the theologian and bishop Durandus of Saint Pourçain (†1334). Thomas heldagainst the occasionalists that though all operations in nature involve the operation ofthe divine will, nonetheless God acts with “secondary causes” to bring about naturaleffects. He concluded that the divine operation that results in such effects is compat-ible with the genuine efficacy of these secondary causes. Here Thomas offered a viewthat Dominik Perler and Ulrich Rudolph, in their comprehensive study of medievaland early-modern occasionalism, have labeled “causal compatibilism.”3 Duranduslater protested that Thomas’s response to the occasionalists deprives creatures of theircausal power, and claimed that occasionalism can be resisted only if the divine con-tribution to creaturely causality is limited to God’s creation and conservation of sec-ondary causes. Durandus’s “mere conservationism,” as Alfred Freddoso has called it,4

was widely rejected in later scholasticism, and it may seem to provide no more thana footnote to the story of the medieval rejection of occasionalism.5 However, we willdiscover that his position is surprisingly relevant to Descartes’s theory of causation.

In §1.2, though, I move to an account of causation in closer temporal proximity toDescartes, namely, the one in the work of the prominent early modern scholastic, theSpanish Jesuit Francisco Suárez (1548–1617). In singling out Suárez, I do not meanto suggest that his position is representative of scholasticism in general. In fact, it hasbecome increasingly clear to scholars that early modern scholasticism was not amonolithic doctrine, but involved different mixtures of nominalist, Ockhamist,Scotist, or hard-line Thomist positions with basic Aristotelian doctrines.6 But thoughSuárez was merely one scholastic among many, he is particularly important for ourpurposes, since he wrote what is perhaps the most comprehensive treatment of causal-ity in the early modern period. In this treatment, which he included in his massiveMetaphysical Disputations (Disputationes Metaphysicæ) (1657), Suárez followsAristotelian orthodoxy in distinguishing four main causes, namely, material, formal,efficient, and final.7 Yet he anticipates Descartes’s views in taking efficient causalityto provide the paradigmatic instance of causation. Moreover, as part of his treatmentof efficient causality, Suárez offers a sophisticated account of God’s causal contribu-tion to the course of nature. In particular, he develops positions in Thomas by argu-ing not only that divine conservation is required for the world to remain in existence,but also that this act of conservation does not differ from God’s initial act of creationex nihilo. He further articulates Thomas’s causal compatibilist alternative to occa-sionalism in terms of the position that God contributes a “concursus” to the action ofsecondary causes that is distinct from his act of conserving such causes in existence.

I close in §1.3 with a brief consideration of the path from the scholastic account ofcausality in Suárez to Descartes’s theory of causation. Descartes’s theory is coupled

The Scholastic Context 11

3. See Perler and Rudolph 2000, 154, which refers to Thomas’s position as“Kompatibilismus als Gegenmodell zum Occasionalismus.”

4. See Freddoso 1991.5. As it is, indeed, in Perler and Rudolph 2000, 245, n.1.6. For two recent discussions of early modern scholasticism that draw attention to its com-

plexity, see Des Chene 1996 and Menn 1997.7. But see note 31.

with a spare ontology that does away with many of the forms and qualities that areprominent in Suárez’s account. Nevertheless, I have indicated that Suárez’s emphasison efficient causality prepares the way for Descartes. Moreover, the discussions inSuárez of divine creation and conservation are linked to Descartes’s own treatment ofthese notions, which are central to his theory of causation. Far more than the positionof the medieval Islamic occasionalists, Suárez’s views provide an appropriate stan-dard against which to measure what Descartes has to say about causation.

1.1. MEDIEVAL REJECTIONS OF OCCASIONALISM

1.1.1. Medieval Islamic Occasionalism

Medieval Islamic occasionalism is an extraordinarily complex historical phenome-non, involving various debates among Islamic theologians and philosophers datingfrom the eighth century. I cannot hope in my brief survey here to provide an exhaus-tive discussion of its development.8 However, I would like to present some basicfeatures of the position by means of a consideration of two medieval sources, the firsta late-twelfth century discussion of Islamic occasionalism from an outsider, and thesecond an insider’s account of this position dating from the end of the eleventh cen-tury. The former is found in the Guide of the Perplexed (Dala-lat al-Ha-’irı-n)(c. 1190), a text of Rabbi Moshe ben Maimon, better known by his Greek name,Maimonides. Chapter 73 of the first part of the Guide concerns the views of theMutakallimu-n, a group of “dialectical theologians” within medieval Islam. At thetime Maimonides wrote, there were two main schools within this group, the firstthe Basrah School of Mu‘tazila, and the second the Ash‘arite School associated withthe former Mu‘tazilite, the tenth-century theologian As‘arı- (Abu- l’Hasan al-As‘arı-).There are various methodological and doctrinal differences between the twoschools,9 but the one most important difference for our purposes concerns the issueof causality. In particular, Maimonides notes that whereas most of the Mu‘tazilitesallowed that created powers can produce effects, the majority of the Ash‘aritesregarded as “abhorrent” the view that such powers displace God as the cause ofeffects in nature (Maimonides 1963, 1:203). The Ash‘arites were thus the mainmedieval proponents of the occasionalist doctrine that God is the only real cause.

In chapter 73, Maimonides offers twelve “premises” that he took to be commonto the Mutakallimu-n. Given the disagreement over occasionalism, he understandablydid not include this doctrine in the list. However, the issue of occasionalism is


8. But see the thorough treatment of medieval Islamic occasionalism in Perler and Rudoph2000, 23–124.

9. For instance, the Mu‘tazalites tended to emphasize more the power of natural humanreason to discover moral and political truths, whereas the Ash‘arites tended to emphasize itslimitations with respect to the grasp of such truths. Moreover, members of the former schooltended to emphasize the indeterministic freedom of the human will, whereas members of thelatter school tended to emphasize God’s predetermination of all events, including human action.

broached in Maimonides’ discussion of the sixth premise, namely, that an accidentcannot “endure for two units of time” (Maimonides 1963, 1:194). The third premiseattributed to the Mutakallimu-n has it that these units are discrete indivisible instantsthat compose time. Maimonides claims that it was a view of “the majority” that inorder for a certain type of accident to endure over time, God must create at differentinstants numerically distinct accidents of the same species (1:200).10 Moreover,when there is any change in accidents, it is God who brings about this change. Thus,“when we, as we think, dye a garment red, it is not we who are by any means thedyers; God rather creates the color in question in the garment when the latter is injuxtaposition with the red dye” (1:201). In general, the view attributed to theMutakallimu-n is that “God creates at every one of the instants—I mean the separateunits of time—an accident in every individual among the beings, whether thatindividual be an angel, a heavenly sphere, or something else” (1:203).

The evidence that the Mutakallimu-n endorsed an atomistic conception of timeseems to be rather thin. One commentator has claimed that the only clear endorsementof such a conception is found in a single text of one of Maimonides’ Islamic contem-poraries, Fakhr al’Dı-n al-Ra-zı-, who was in fact not a typical Mutakallim (see Schwartz1991, 177).11 However, Maimonides may well have thought that the conclusion thattime is atomistic simply follows from other doctrines that predominate in the writingsof the Mutakallimu-n. Indeed, in the Guide he claims that such a conception followsmerely from the first premise he attributed to these thinkers, according to which everybody is composed of indivisible atomic parts. If this conception does follow from thepremise, then there would be good reason to attribute it to the Mutakallimu-n, sincealmost all such thinkers accepted an atomistic account of matter (see Schwarz 1991,169). Maimonides’ argument that it does so follow appeals to the result in Aristotlethat distance, time, and local motion must be proportionate (Maimonides 1963,1:196).12 Given this result, if time were infinitely divisible, the particles that thesethinkers took to be atomic would have to be infinitely divisible as well.

One problem for this argument is that the infinite divisibility of time seems torequire the infinite divisibility not of the particles themselves, but only of the distancethey travel. However, another option for Maimonides would be to link the atomisticconception of time to the sixth premise that accidents cannot endure through time. Thisproposition can be found in Islamic texts dating back to ninth century, and was indeed,as Maimonides reports, a view popular among the Ash‘arites, who formed the major-ity of the Mutakallimu-n (see Schwarz 1991, 194).13 Given this opinion, it would seem


10. Maimonides mentions the minority view of the Mu’tazilites that certain accidents canendure through time.

11. Indeed, Schwarz’s conclusion is that of the twelve premises mentioned in the Guide,the evidence confirms a source in the writings of the Mutakalimu-n only for “Maimonides’premises 1, 6, 7, 8, 9, and 11. For the rest of the premises the evidence seems partial at best”(Schwarz 1991, 172).

12. This result follows in turn from the Aristotelian definition of time as the measure of motion.13. Schwarz notes that the Ash‘arite Ba-qilla-nı- (†1013) defined an accident as that which

cannot exist longer than an instant (Schwarz 1991, 185).

to follow that accidents cannot endure through any divisible portion of time, and so canexist only at an indivisible instant.

Of course, the atomistic conception of time does not itself entail that God alonecan be a cause of the accidents that exist at any given moment. Maimonides indicatesthat Islamic occasionalists attempted to rule out the claim that accidents can causeother accidents by appealing to the premise that “an accident does not go beyond itssubstratum” (Maimonides 1963, 1:202). This premise seems to derive from thethought that an accident is something that merely inheres in its substance, and thusthat is incapable of bringing about anything other than this inherence. Such a prem-ise still seems to leave open the possibility that the substance produces the accidentthat inheres in it. However, one reason to rule out the substance as a cause is sug-gested by the view, which Maimonides attributes to the Mutakallimu-n, that earlierand later accidents are linked by means of a “habit” that God imposes (1:201). Wecan understand this habit to consist in a lawlike correlation between the accidents.Even if a substance could produce its own accidents, it does not follow that it caninstitute the law that serves to connect these accidents to other accidents. Indeed, theassumption among the Mutakallimu-n is that only God could establish a lawlike cor-relation that holds for all of the relevant accidents. Thus in the case of a human agentmoving a pen, it must be that “God has instituted the habit that the motion of thehand is concomitant with the motion of the pen, without the hand exercising in anyrespect an influence on, or being causative in regard to, the motion of the pen”(1:202). God institutes the habit operative in this case by producing in successiveinstants the accidents that constitute the motion of the pen.

In the medieval Islamic philosophical tradition, there was an alternative to thisaccount of the lawlike habits that hold in nature. Drawing on a mixture of Aristotelianand Platonic (or Neoplatonic) positions, philosophers such as Fara-bı- (Abu- NasrMuhammad al-Fara-bı-), in the tenth century, and Avicenna (Abu- ‘Ali al-Husayn ibn Sı--na-),in the eleventh century, insisted that the natural course of events derives necessarilyfrom certain “forms” that though emanating ultimately from God through pure intel-ligences, nonetheless exist in created objects. Perhaps the most direct response amongthe Ash‘arites to this position in “the philosophers” was provided by Ghaza-lı- (Abu- Ha-

mid al-Ghaza-lı-). In his Incoherence of the Philosophers (Taha-fut al-Fala-sifah)(c. 1095)—our second medieval source for Islamic occasionalism—Ghaza-lı- offersrefutations of the purported demonstrations in the work of the Islamic philosophersof twenty propositions concerning metaphysics and the natural sciences. The discus-sion of propositions concerning the natural sciences begins with the seventeenthproposition, according to which any departure from the natural course of events isimpossible. To defend the possibility of miraculous events, Ghaza-lı- argues that therelations between natural causes and their effects are not absolutely necessary sincethey derive ultimately from the divine will. The sort of causes and effects of concernin the natural sciences “are connected as the result of the decree of God (holy be hisname), which preceded their existence” (Ghaza-lı- 1958, 185).

In contrast to what one might expect from Maimonides’ remarks in chapter 73 ofthe first part of the Guide, the section on causation in the Incoherence emphasizesneither the atomistic conception of time nor the restriction of accidents to a single


instant.14 Rather, this section opens with what we could call (following Nadler1996) the “no necessary connection argument.” In particular, the argument is that therelations between what we take to be causes and their effects cannot be necessarygiven that the affirmation of the existence of the one does not logically require theaffirmation of the existence of the other, nor the denial of the existence of the onethe denial of the existence of the other (Ghaza-lı- 1958, 185). It may seem a bit of aleap to Ghaza-lı-’s subsequent claim that cause and effect “are connected as a resultof the decree of God” (185). Why couldn’t there be some other source of the neces-sity? However, Ghaza-lı- argued earlier in the Incoherence that the action involved incausation can be attributed only to the will of an agent.15 Moreover, it seems that aneffect can follow necessarily only from the will of an omnipotent being. To be sure,Ghaza-lı-’s claim that God acts either directly or “through the intermediacy of angels”(186) seems to leave open the possibility that the wills of finite beings necessitateeffects. I return to this complication presently. However, it is significant that even inthe case in which angels serve as intermediaries, God is said to be the agent respon-sible for causal relations in nature.

Given the strong claim in the Incoherence that causal relations hold only becauseGod “has created them in that fashion, not because the connection in itself is necessaryand indissoluble” (Ghaza-lı- 1958, 185), it is not surprising that this text is standardlyread as a defense of occasionalism. However, there are some complications for such areading. I have just noted the complication deriving from the suggestion in the text thatGod can act through “the intermediacy of angels.” Yet there is the further complicationthat Ghaza-lı- presents in the Incoherence not one but two alternatives to an account ofcausation that precludes miraculous events. In addition to the suggestion that God pro-duces the causal correlations either directly or through the intermediacy of angels, heoffers a second position that concedes the point of the philosophers that objects havecertain attributes in virtue of which they “habitually” produce certain effects, andmerely insists that God can miraculously impede or change the speed of naturalprocesses (see Ghaza-lı- 1958, 190–91). In the Incoherence of the Incoherence (Taha-futal-Taha-fut) (c. 1180), Ghaza-lı-’s twelfth-century critic Averroes (Ibn Rushd) takes thefact that he offered this second account to indicate his abandonment of the strong occa-sionalist denial of real causal efficacy in nature, and others have insisted more recentlyon a non-occasionalist interpretation of his views on causation.16


14. It is unclear whether Maimonides read Ghaza-lı-’s work (as indicated in the editorialcomments in the introduction to Maimonides 1963, 1:cxxvii). Even if he did read it, however,the fact that he did not take note of Ghaza-lı-’s innovations may be explained by the fact that,as Perler and Rudolph have observed, what twelfth-century Ash‘arite theologians had to sayabout causality “kingt vielmehr ganz konventionell” than what is found in the Incoherence(Perler and Rudolph 2000, 109; cf. 115).

15. In particular, he argued for this conclusion in a discussion of the third proposition ofthe philosophers, according to which the created world follows necessarily from God’s nature(see Ghaza-lı- 1958, 63–64).

16. For this charge, see Averroes 1969, 1:316–33. For a more recent example of an inter-pretation that questions Ghaza-lı-’s commitment to occasionalism, see Frank 1992.

Nevertheless, there is some reason to think that Ghaza-lı- offers the second accountmerely for the sake of argument, and not to indicate a shift in his position. After all,his main concern in offering this account is to show that miracles are intelligibleeven given certain aspects of the view of causation offered by the philosophers.17

Admittedly, the reference to the intermediary action of angels cannot be explainedaway in this manner. However, I believe that Michael Marmura has shown that whenGhaza-lı- speaks of angels as intermediaries, he means to indicate only “that they arethe locus of divine action” (Marmura 1995, 99). This would seem to be in line withthe emphasis in the text—which I noted previously—on the fact that the agent ulti-mately responsible for causal connections in nature is God rather than the angels. Inany event, there is in the Incoherence a forceful statement of the occasionalist posi-tion that natural causes do not necessitate their effects, but are merely linked to themby divine decree. Moreover, the defense of this position is distinctive in the contextof medieval Islamic thought, since it focuses not on the nature of time or of quali-ties, but rather on the lack of a necessary connection between perceived causes andtheir effects and the need for a grounding of causal relations in the omnipotent willof God. On both points Ghaza-lı- anticipated the later argument for occasionalism inthe work of Nicolas Malebranche. Thus, in his discussion in the 1674/75 Searchafter Truth (Recherche de la vérité) of the “error of the philosophy of the ancients,”and particularly of the Aristotelian philosophy, regarding causation, Malebrancheinsists that a true cause by definition “is one such that the mind perceives a neces-sary connection [liaison nécessaire] between it and its effect,” and that the mindperceives such a connection “only between the will of a necessary being and itseffects” (bk. VI-2, ch. 3, Malebranche 1958–84, 2:316/Malebranche 1997, 450). Theoccasionalist challenge to causal realism that emerges from Ghaza-lı-’s Incoherenceis to explain how the doctrine that creatures have real causal power can be reconciledwith the result that all connections in nature that do not involve logical necessityderive from acts of the divine will that alone can suffice to establish their effects.A relevant question for us—broached by the remarks in Fontenelle’s Doubts thatI considered at the outset of this chapter—is whether Descartes joined Malebranchein issuing this sort of challenge.

1.1.2. Thomas’s Causal Compatibilism

In a thirteenth-century text, Questions on the Power of God (Quaestiones de PotentiaDei), Thomas Aquinas devotes an article to a defense of the claim that “God operatesin the operations of nature.” However, he is concerned there to distinguish his viewfrom the position reflected “in the law of the Moors, as Rabbi Moses [Maimonides]


17. For this argument, see Marmura 1981. Marmura also emphasizes that the point of theIncoherence is merely to refute the strong views of the philosophers, and not necessarily todefend the final truth on the matters discussed (see Marmura 1981, 98–99). Marmura alsoresponds to Frank’s more revisionist interpretation (see note 16) in Marmura 1995. But cf. thediscussion of the Marmura-Frank exchange in Perler and Rudolph 2000, 71–73, which is crit-ical of some features of Marmura’s position.

relates,” according to which all natural forms are mere accidents that God creates inobjects. Thomas’s initial response is that this position is “manifestly repugnant to thesenses,” since the senses merely passively receive the effects of sensible objects(QPG III.7, TA 13:58–59). Ghaza-lı- anticipated this response when he noted in theIncoherence that sensory effects “are observed to exist with some other conditions,”but we do not see that such effects “exist by them” (Ghaza-lı- 1958, 186). Yet Thomasadds that it is “repugnant to the divine goodness” that God does not communicateto creatures the power to produce effects. Thus, he insists that “the operations ofnature” follow from various created forms.18 Moreover, he responds to the view ofthe Moors that everything in substance is a mere accident by drawing on hisAristotelian ontology of material substance, according to which such substancespossess not only accidental forms, such as that of heat, that inhere in them, but alsosubstantial forms that unite with matter to constitute the substances (TA 13:59). Inthis view, both kinds of forms serve as principles of natural operations, and thus arenot merely passive effects of divine creation.19

Given Aquinas’s position that natural operations derive from accidental and sub-stantial forms, there may seem to be no room for his thesis that God operates in theseoperations. As we will discover, this was in fact the objection that Durandus laterleveled against this thesis. However, Aquinas responds to this line of objection in Onthe Power of God by insisting that the operation of God in producing effects innature is compatible with the operations of “secondary causes” in producing thosesame effects. He appeals here to the fact that we can understand a certain effect tobe produced both by an instrument and by an agent who uses that instrument. Anexample he commonly uses to illustrate the nature of instrumental causality is thatof an agent who uses a pen to write. The pen is a real cause of the written words, butis able to be efficacious in this way only because the agent uses it to produce thiseffect. Similarly, Aquinas holds that though contrary to the view of the Islamic occa-sionalists, secondary causes can produce effects, nonetheless they cannot producethese effects through their own power ( per virtutem propriam), but must participatein the power of a “primary” or “principal” cause, namely, God. In this way, a sec-ondary cause is “the instrument of the divine power of operating” (instrumentumdivinæ virtutis operantis) (QPG III.7, TA 13:60).

To understand the nature of this “power of operating,” we need to compare it tothe other two divine powers of operation that are essential for the existence of theworld, namely, creation and conservation. In his Summa Theologiae, Thomas claimsthat any being that does not exist by its own nature and thus is a being only “by


18. This response assumes that God can communicate his power to creatures, and thusseems to beg the question against the die-hard occasionalist who insists that it is not possible forGod to so communicate, and thereby that his failure to do so does not detract from his goodness.Aquinas offers various other arguments against occasionalism, but these additional argumentsalso arguably (though I cannot argue here) employ premises the occasionalist would reject. Fora discussion of these other anti-occasionalist arguments, see Perler and Rudolph 2000, ch. 4.

19. I say more about the details of Suárez’s version of this position in §1.2.1. See also theremarks concerning Suárez’s metaphysics in §§1.3 and 3.2.1 (iii).

participation”—that is, any being other than God—can exist in the first place onlybecause of a creative act of God (ST I.44.1). Since all being by participation dependson this creative act, moreover, the act must involve the creation of being fromnothing, that is, creation ex nihilo (ST I.45.1).

Thomas also holds that just as all beings by participation depend on God’s act ofcreating ex nihilo to exist in the first place, so they depend on his act of conservationto continue to exist. In arguing for the need for this additional dependence on God,he appeals to the distinction—which, as we will discover in due course, later becameimportant for Descartes20—between causae secundum fieri, or causes of becoming,and causae secundum esse, or causes of being. Thomas notes that though a housecan continue to exist without its builder, this is only because the builder is a causasecundum fieri that directly produces not the being of the house and its material, butonly its coming to be a house through a certain arrangement of the material. Even incases of natural operations that involve more than mere arrangement, such as whenaccidental or substantial forms act to produce similar forms in matter, the former arenot causes of the very being of the latter. If they were, the forms would have to causetheir own being, which they share with the being of what they produce. Rather, theproducing forms merely “educe” produced forms similar to them that are containedpotentially in matter (ST I.104.1).

In contrast, Thomas claims that in cases where the cause is “more noble” than theeffect, it can be a cause secundum esse that produces the being of the form itself. Heappeals to the fact that the sun does not merely educe light from the air, but rathercreates a new form that “has no root” (non habet radicem) in the nature of air.21

Because the air alone cannot support the existence of this new form, light dependsessentially on the continuing action of the sun. Thomas claims that creatures dependon God in the same manner, and thus that without the continued action of God, crea-tures would cease to exist (ST I.104.1). It is important to emphasize the point herethat conservation involves merely the continuation of God’s act of creation. ForThomas himself responds to the objection that conservation cannot add anything tothe creature not already provided by creation by noting that God conserves creaturesin existence “not by a new action, but by a continuation of that action whereby hegives being” (ST I.104.1, ad 4).

In On the Power of God, Aquinas recognizes as an objection to his own positionthat the only operation of God involved in the operations of nature is that by whichhe “either makes or conserves in being a natural power” (QPG III.10, TA 13:57).However, he responds that “God is not only the cause of the operations of nature thatconserve natural powers in being, but in other modes, as has been said.” What wassaid in particular is that secondary causes depend on God not only for their initial


20. See the remarks concerning Descartes’s appeal to this distinction in §2.2.1. See alsothe discussion below of the relation of this distinction to views in Durandus (§1.1.3) andSuárez (§1.2.2 (ii)).

21. Thomas is assuming here that light is a quality that depends essentially on the actionof the substantial form of a body that is self-luminous (see ST I.67.4 and ad 1). It is becauseair is not a self-luminous body that light can have no “root” in it. Cf. chapter 2, note 66.

and continuing existence but also for the action by which they bring about theireffects. A secondary cause “acts as an instrument of a superior power; whence,exclusive of the superior power, the inferior power has no operation” (TA 13:61).

The appeal here to instrumental causality in explaining the divine power of oper-ating provides a further reason to distinguish the acts of this power from divine cre-ation and conservation. For when God creates or conserves a secondary cause, he isnot using that cause as an instrument, since that cause contributes nothing either tothe divine act of creation ex nihilo or to the continuation of that act in conservation.It is only when God is using an already existing secondary cause to bring about aneffect in nature that this cause contributes something to the action. Indeed, Aquinasholds that in this case the very same action derives both from God as primary agentand from secondary agents (see, e.g., ST I.105.5, ad 2). Admittedly, this claim wouldbe nonsense if an action were something in the agent. But in fact Aquinas claimed,in line with many later scholastics, that the action is something external to the agent,and so is distinct from the principle in the agent from which the action issues.22 Inthis scholastic view, action is not that which produces an effect, but rather the actu-alization of that effect, which actualization occurs in the patient. Thus, the fact thatGod as primary agent and secondary agents act by means of distinct principles neednot imply that the actions deriving from those principles are distinct.

The premise that a single action can proceed from both God and creatures is keyto Thomas’s causal compatibilism, since without this premise the causal activitywould have to be attributed either to God alone, thus resulting in occasionalism, orto the creature alone, thus overturning the conclusion in Thomas that God operatesin all operations of nature. We have seen already a willingness among the Mutakallimu-nto embrace the occasionalist horn of this dilemma. But there also was a member ofThomas’s own Dominican order who was willing to embrace the other horn, and soto restrict the divine contribution to natural operations to creation and conservation.In §1.2 we will explore the further development of Thomas’s causal compatibilismin the work of the early modern scholastic Suárez. Before making the transition tothe early modern period, however, we need to consider the more radical medievalrejection of occasionalism in a text of Thomas’s Dominican critic, Durandus of SaintPourçain.

1.1.3. Durandus’s Mere Conservationism

Durandus was a controversial fourteenth-century figure whose critique of certaintheological doctrines in Thomas earned him two censures from Dominican authori-ties eager to identify the order with Thomism.23 However, we are concerned herewith his challenge to Thomistic metaphysics, and in particular with the critique


22. But see note 58.23. In particular, Durandus was censured in 1314 and 1317 for rejecting more orthodox

Thomistic views on the nature of the distinctions among and the processions of the persons ofthe Trinity. For more on the theological dispute that Durandus triggered in the Dominicanorder, see Iribarren 2005.

of Thomas’s causal compatibilism that he offers in the second book of On theTheological Sentences of the Commentary of Peter Lombard in Four Books (InSententias Theologicas Petri Lonmbardi Commentariorum Libri Quattuor). The fifthquaestio of the first distinctio of this section of Durandus’s Sentences is devoted toThomas’s claim in the Summa Theologiae that “God acts immediately in all actionsof creatures” (S II.1.5, 1:130, citing ST I.103.6 and 105.5). Durandus there agreeswith Thomas that “the being of a secondary cause . . . is the immediate effect of theprimary cause, which is an immediate cause not only in bringing it into being, butalso in conserving it in being” (S II.1.5, ¶17, 1:131). He also concurs in Thomas’srejection of occasionalism, holding that “this view is now rejected by everyone asimprobable, because it denies of things their proper operations and also denies thesensory judgment by which we experience that created things act on one another”(¶4, 1:130). What he cannot accept, however, is Thomas’s view that the claim thatGod acts immediately in all actions of secondary causes is compatible with the attri-bution of real efficacy to those causes. For Durandus, the only acceptable alternativeto occasionalism is the “mere conservationist” position that God contributes only thecreation and continued conservation of a secondary cause to the production of aneffect by that cause.

Durandus’s initial point against Thomas is that one who holds that God actsimmediately in every action of a creature cannot say merely that God is responsiblefor a certain feature of an effect. In the case of the generation of a material substance,for instance, it would not be sufficient to claim that God produces the matter of thatsubstance, whereas the secondary cause produces its form. For then God would notbe acting immediately in the production of the form by the secondary cause (S II.1.5,¶6, 1:130). Thus, defenders of the Thomistic position must go further in claimingthat the effects of secondary causes “are from God as wholes and immediately, butnot totally, that is, not in every way” (¶7, 1:130). The effects must be from God “aswholes and immediately” to avoid what we could call “the problem of the dividedeffect,” according to which God is responsible for one part of the effect, the second-ary cause for another. But if the effects were from God “totally, in every way,” thenthe secondary cause would seem to be doing no work, just as the occasionalist con-tends.24 So there needs to be some sort of complementary contribution to the pro-duction of one and the same effect.

What Durandus cannot comprehend is how an effect could be from God as awhole and immediately but not totally. He considers the suggestion, deriving fromAristotle’s remarks in the Physics, that universal aspects of an effect can be tracedback to a universal cause, whereas particular aspects of the same effect can betraced back to a particular cause (S II.1.5, ¶8, 1:130). And Thomas himself suggested


24. One might think that there is the possibility of causal overdetermination. Durandus’sresponse to this possibility is that since actions are individuated by their effects, diverseactions cannot result in numerically the same effects (S II.1.5, ¶13, 1:131). Suárez later coun-tered that though an effect cannot have more than one total cause in a certain order, it can havedifferent total causes in different subordinated orders (MD XXVI.4, 1:929–35*). For more onthis response in Suárez, see §1.2.3 (ii).

this model of God’s contribution to natural operations when he claimed in On thePower of God that “instances of the causing of absolute being [entis absolute] aretraced back to the first universal cause, whereas the causing of the other things thatare superadded to the esse, or are that by which the esse is made specific, pertainsto the secondary causes” (QPG III.1, TA 13:38). However, Durandus insists that ina living thing, for instance, esse and the determination of that esse as somethinginvolving life differ only “by reason” (ratione), that is, merely conceptually and notin reality. Since the effects in this case differ only by reason, it would seem that thecauses differ only by reason as well (S II.1.5, ¶10, 1:130). By the same token,according to Durandus, “it is impossible for numerically the same action to be fromtwo or more agents in such a way that it is immediately and completely from each,unless numerically the same power is in them” (¶11, 1:131). Numerically the samepower cannot be present in God and creatures insofar as God’s power is infinite,whereas the power in creatures is limited in nature. Thus, for Durandus, it cannotbe said that the very same effect derives immediately and completely from bothGod and creatures. For God and creatures to cooperate in producing an effect, itmust be the case that the power in God is responsible for one feature of the effect,whereas the power in creatures is responsible for another feature of the effect thatis distinct in reality from the feature God produces.

It might be thought that this line of response simply begs the question againstAquinas in assuming that distinct powers cannot produce the same effect. Indeed,Aquinas offered instrumental causality as an example of a case where the sameeffect issues from both the instrument and the agent using that instrument. Whycouldn’t the same be true in the case of God’s action with secondary causes?25

Durandus objects to the comparison to instrumental causality by appealing to thepossibility that an action derive principally from a secondary cause. Since “an actionthat does not exceed the power [virtute] of the species of the agent is sufficientlyelicited by just the power of the species,” in this case “it would be superfluous toposit another immediate principle eliciting such an action” (S II.1.5, ¶11, 1:131). Wewill discover presently that Suárez attempted to address this objection by drawinga distinction between instrumental and principal causality that nonetheless providesroom for God’s “concursus” in the case of the action of principal secondary causes(see §1.2.3 (ii)). However, Durandus could argue that there is an additional problemwith the analogy to instrumental causality. In particular, he could point out that inthe case of the use of the pen, the fact that the words are black and the fact that thereare certain words rather than others pick out distinct effects. On the Aristotelian viewcommon to Thomas and Durandus, the color and the shape of the words are differ-ent accidental features of it. Durandus thus could argue that in this case these distincteffects derive from distinct causes.

This line of argument does not establish that there are no cases in which the sameeffect derives from different powers. Indeed, it is not clear to me that Durandus hasan argument for this conclusion that does not rely on the assumption that distinct


25. For this line of objection to Durandus, see Freddoso 1994, 148–50.

causal powers cannot issue in the very same effect. In the Summa Theologiae,Thomas allows there cannot be more than one complete cause of a single effect thatbelongs to the same causal order, thus ruling out a kind of causal overdetermination.However, he also claims that such an illicit overdetermination is not present in thecase of God’s cooperative action with secondary causes, since these causes belongto a different causal order than God (ST I.105.5, ad 2).26 Given that Durandus failedto address this distinction between total causes of the same order and total causes ofdifferent orders, his argument for the impossibility of Thomistic causal compatibil-ism cannot be regarded as conclusive.

Nevertheless, it will turn out that the question of whether Durandus succeeded inrefuting Thomas is less important for our purposes than the question of whether hismere conservationism is itself a tenable position.27 Thus, it is appropriate that wenow switch from offense to defense, as it were, and consider Suárez’s argumentagainst Durandus that the conclusion that God acts “immediately in every actionof the creature” simply follows from the claim, on which Durandus himself insisted,that all beings depend on God immediately for their conservation in existence oncethey are produced. For Suárez, creatures are no less dependent on God for their ini-tial production than they are for their subsequent conservation. As he expresses theargument,

[I]f it is not the case that all things come to exist immediately from God, thenneither is it the case that they are conserved immediately, since a thing is relatedto being [esse] in the same way it is related to becoming [fieri]. For the being of athing cannot depend more on an adequate cause after it has been made than it didwhen it was made. (MD XXII.1, ¶7, 1:803)

This argument is strengthened by Suárez’s doctrine that the act by which God con-serves a being in existence is merely a continuation of the act by which he causes thatbeing to exist in the first place (see §1.2.3 (i)). Given this doctrine, it would be natu-ral to conclude that if God is not the immediate source of the existence of an object,he cannot conserve that object by continuing the act by which he immediately pro-duced it. However, there is a way of expressing the point of Suárez’s argument thatdoes not rely on his particular account of divine conservation. For one basic objectionhere is that given that a secondary cause can immediately and completely produce theesse of an object on its own, there seems to be no reason to deny that it can immedi-ately and completely conserve that esse. Suárez notes that it is as obvious to thesenses that there are conserving secondary causes as it is that there are productive


26. Cf. the remarks in note 24.27. I would just mention, however, Freddoso’s proposal that one can make sense of the

fact that God and secondary causes make different contributions to the effect not by splittingthe effect, as Durandus requires, but rather by distinguishing different states of affairs thatconcern a unitary effect (Freddoso 1994, 150). As Freddoso himself admits, this proposal isin need of further articulation and defense, and it is not entirely clear that when these are pro-vided we will have a viable alternative to Durandus’s position. But it also is not clear fromwhat Durandus has said that no such alternative proposal could succeed.

secondary causes. To expand on an example he used, the senses reveal not only thatfire produces the quality of heat in water, but also that heated water itself conservesthis quality after the fire has ceased (MD XXII.1, ¶8, 1:804). Yet if the water were asecondary cause in Durandus’s sense, namely, one that produces its effects immedi-ately by itself, then it could not be the case that God immediately conserves the heatin the absence of the fire. It therefore would not be the case, contrary to whatDurandus claimed, that God immediately conserves all beings in existence.28

There may be a way around this objection that draws on the distinction in Thomasbetween causes secundum fieri and secundum esse (see §1.1.2). Durandus suggeststhe strong view that God cannot be in any way an immediate cause of the effects ofsecondary causes. But we could perhaps modify this view to say only that God can-not be the immediate cause both secundum fieri and secundum esse of such an effect.This modification would allow for the position that God is the sole cause secundumesse of an effect that secondary causes produce as its sole causes secundum fieri.Thus, for instance, God alone would be the cause of the esse of forms educed frommatter, whereas secondary causes alone would be the cause of the educing of formswith that esse. This proposal would clearly seem to allow for conservationism giventhat Thomas had introduced the distinction between the two causes in the first placeto defend the thesis that all creatures need to be kept in existence by God (ST I.104.1).

Durandus’s claim against Aquinas that the esse of a particular object is only con-ceptually distinct from the determinate form of its esse (see S II.1.5, ¶6, 1:130) per-haps requires that the secondary cause of the fieri of this determinate form also be theimmediate cause of its esse. If so, he could not in the end allow for the division ofcausal labor in the production of the object that my modification of his view requires.But whether or not Durandus could accept it, there seems to be at least some con-ceptual room for the position that God as primary cause is responsible for what isactual in causal interactions, namely, the esse of both cause and effect, whereas sec-ondary causes are responsible for changes in what is actual, namely, what Thomascalled the fieri of the effect. If the fact that God alone is the cause secundum esse ofall natural effects is compatible with the fact that secondary causes alone are causessecundum fieri of those same effects, we would seem to have a version of mere con-servationism that sidesteps one of Suárez’s main objections to Durandus. More to thepoint, given the topic of this book, we may well have a version of this position thatDescartes could accept.

To determine whether Descartes could accept this sort of mere conservationism, how-ever, we must settle the question of whether he even allowed that secondary causes canproduce changes in objects and, if so, whether he held that such causes can produce thesechanges immediately and completely by themselves, without any assistance from Godthat goes beyond his creating and conserving activity as the cause secundum esse of theworld. A positive answer to the first part of this question would reveal that he followedthe vast majority of scholastics in rejecting occasionalism. A positive answer to the sec-ond part would indicate that he deviated from most scholastics, and in particular from


28. I borrow here from the discussion of this Suárezian objection to Durandus in Freddoso1991, 566–69.

Suárez, in accepting a form of Durandus’s mere conservationism. I will be concerned toaddress these issues in the course of the discussion in the following chapters of variousaspects of Descartes’s theory of causation. Before turning to Descartes, though, I need toconsider Suárez’s account of causality, since such an account is a particularly importantpart of the scholastic context of that theory.

1.2. SUÁREZ ON EFFICIENT CAUSES AND CONCURSUS

Suárez’s Metaphysical Disputations includes a “mini treatise” on causality that spansdisputations XII through XXVII and covers a total of 590 pages in the Vivès edition, orabout a third of the total work. This treatise concerns the familiar quartet of Aristoteliancauses—material, formal, efficient, and final.29 However, disputations XVII throughXXII, which cover a total of 263 pages, or close to half of the treatise on causality, con-cern exclusively the case of efficient causes. This imbalance reflects Suárez’s conclu-sion at the start of his discussion of causation that “the whole definition of the cause ismost properly suited to efficient [causes]” (MD XII.3, ¶3, 1:389*). Such a conclusionin fact provides a bridge from a traditional Aristotelian account of the four causesto Descartes’s restriction of explanations in natural philosophy to efficient causes.30

Moreover, we will discover that Suárez’s discussion of efficient causes is particularlyrelevant to Descartes’s theory of causation, since the former includes his treatment ofthe nature of God’s efficient causality in creation, conservation, and concurrence.

Before turning to the particular features in Suárez that serve to link his accountof causality to what we find in Descartes, however, I pause to consider Suárez’s gen-eral project in the Disputations of renovating scholastic metaphysics. I provide asketch of the context of this project that, though very rough, hopefully serves to indi-cate the significance of Suárez’s contributions to scholastic metaphysics as well asthe relevance of these contributions for Descartes’s views (§1.2.1). Then I take upthe account of causality in the Disputations, beginning with a discussion of Suárez’streatment of the four main Aristotelian causes that highlights his view that efficientcauses have a special kind of priority (§1.2.2).31 Finally, I consider his account of the


29. Following an initial disputation, entitled De causis entis in communi, disputations XIIIand XIV are devoted to material causes, disputations XV and XVI to formal causes, disputa-tions XII to XXIV to final causes, and disputation XXV to exemplary causes (see note 31).The section on causation closes with disputation XXVI, concerning the relation betweencause and effect, and disputation XXVII, concerning the relation of causes among themselves.

30. But see the discussion in §2.1.2 of the complications for this view in Descartes.31. As indicated in note 29, Suárez also devotes a section of his treatise on causation to

exemplary causes, which involve the influence of ideas in the production of an effect, a cate-gory of cause that, as he emphasizes, derives from the Platonic rather than the Aristotelian tra-dition (see MD XXV.1, ¶1, 1:899*). We can set aside this non-Aristotelian category of causesgiven Suárez’s own endorsement of the view of “those who deny that the exemplary cause con-stitutes a proper genus of cause, but who say that it pertains to the efficient cause” (MD XXV.2,¶8, 1:913*). For a discussion of Suárez’s reasons for this endorsement, Carraud 2002, 150–52.

distinctive sort of efficient causality exhibited in the three main divine contributionsto causation in nature, namely, creation, conservation, and concurrence (§1.2.3).Drawing on views in Thomas that we have considered, Suárez not only denies thatdivine conservation is distinct in reality from God’s act of creation ex nihilo, but alsoconcludes that in addition to creation and conservation God contributes a distinct“concursus” to the action of secondary causes.

1.2.1. Renovating Scholastic Metaphysics

Suárez belonged to a metaphysical tradition that Stephen Menn has labeled “liberalJesuit scholasticism” (Menn 1997).32 As with most labels, this one requires someexplanation and qualification.33 An initial point is that though Iberian Jesuits weremost prominent in this tradition, one of its main pioneers during the mid–sixteenthcentury, more than a generation before Suárez, was the Dominican Domingo deSoto.34 Soto is distinguished from his hard-line Thomistic contemporary Cajetan byhis acceptance of the voluntarist axiom, deriving from the Paris Condemnation of1277, that God can produce any creature in separation from any other creature reallydistinct from it.35 The significance of this departure from orthodox Thomism is indi-cated by Descartes’s appeal in the course of his Sixth Meditation argument formind–body distinctness to the principle that God can create separately what we canclearly and distinctly understand apart from each other (AT 7:78).

What is “liberal” about the view of Soto and the later Jesuits is the way in whichthe voluntarist axiom that later appeared in Descartes led them to deny the more “con-servative” view that the Aristotelian categories faithfully reflect real distinctions inbeing. For the hard-line Thomists, the category of substance and the nine categoriesof the predicamental accidents (viz., quantity, quality, relation, action, passion, time,place, position, and having) pick out non-overlapping kinds of really distinct res. Incontrast, the liberal opponents of Thomistic orthodoxy held that the impossibility ofconceiving of members of certain categories as existing apart from members of othercategories reveals that the former are in fact not res really distinct from the latter.


32. Though I suggest some refinements of Menn’s characterization of this tradition, myremarks in this section are indebted to his exemplary discussion of it. For a general study ofSuárez’s metaphysics, see also Courtine 1990.

33. Menn himself notes some concerns about his label in Menn 2000, 120.34. Soto (1494–1560) influenced the work of such prominent Jesuits as Pedro da Fonseca

(1528–99), Francisco de Toletus (1534–96), Luis de Molina (1535–1600), and the Spanishschool of the Conimbricenses.

35. Though the axiom is not explicitly endorsed in the Condemnation, there is a repeatedcondemnation in this text of propositions that seek to limit divine power. Included are condem-nations of purported implications of the teachings of Aquinas, such as the claim that God cannotmultiply individuals of the same species without matter (see props. 42, 43, 110, and 116 in thereorganized and translated version of the Condemnation in Lerner and Mahdi 1995, 335–54). Asindicated in Menn 1997, 229, Cajetan rejected the axiom and held that any anti-Thomisticelements of the Condemnation were revoked when Thomas was made a saint (in 1323).

We can illustrate this difference in terms of what was, for the scholastics, the par-ticularly problematic case of shape. On the Aristotelian view that the scholastics inher-ited, shape belongs to the fourth species of the category of quality, a category that itselfis distinguished from the category of quantity.36 The “conservative” line of theThomistae was that the mere distinction of these categories suffices to reveal that shapeis a res distinct from quantity, even though not even God can create a shape apart fromquantity.37 However, many Jesuit scholastics took the voluntarist axiom to show thatshape cannot be a res distinct from quantity, and thus that it does not follow from thefact that shape and quantity belong to distinct categories that they are distinct beings.

In denying that shape and quantity are distinct res, Suárez and other “liberal Jesuitscholastics” agreed with the view of the nominalists that derives from the work of thefourteenth-century scholastic William of Ockham. For the nominalists accepted boththe voluntarist axiom and the claim that shape cannot exist apart from quantity.However, these thinkers also endorsed the Thomistic principle that the only alterna-tive to a real distinction is one drawn “in reason,” and so concluded that shape ismerely rationally distinct from quantity. Indeed, they radically reduced the number ofdistinct res in holding that only substance and its affective qualities (e.g., in materialsubstances, sensible qualities such as colors, sweetness and bitterness, heat and cold)are distinct in this way. The nominalist conclusion is that the other predicamentalaccidents are merely rationally distinct from substance and its qualities.

The Jesuit scholastics who followed Soto were concerned to provide a middle waybetween this sort of deflationary nominalism and the extreme form of realism in thework of the Thomists.38 So instead of speaking of their liberal scholasticism, perhapsit is better to refer to their metaphysical position as “moderate realism,” that is, a real-ism that accepts the limitations on distinctions in being that follow from the volun-tarist axiom but that attempts to avoid the extremes of nominalism. To forge thismiddle way, the (primarily, though not exclusively) Jesuit moderate realist scholasticsrequired metaphysical distinctions that stood between the real and rational distinc-tions that both Thomists and nominalists took to be exhaustive. Prior to Suárez, otherscholastics had proposed various possibilities. In the fourteenth century, for instance,the Franciscan John Duns Scotus introduced intermediate formal and modal distinc-tions. Scotus embraced the principle that “things one of which can remain without theother are really distinct,” but also held that even inseparable items may differ suffi-ciently to be more than merely rationally distinct.39 Thus inseparable items that havedifferent defining features are said to be “formally” distinct, whereas a certain quali-fication of a quality is said to be “modally” distinct from that quality. Though the


36. See Categories §8, 10a11–24, in Aristotle 1984, 1:16. The four species of quality are, first,habitus or dispositio, which assists the actualization of a potentia; second, potentia or impotentia(i.e., the privation of a potentia); third, the affective qualities; and fourth, shape or form.

37. Thus in a passage cited in Menn 1997, 243, n. 22, Cajetan offered the example of therelation of quantity to shape as a counterexample to the voluntarist axiom.

38. Though, as Suárez notes (MD VII.1, ¶9, 1:252–53), Soto was inconsistent on the ques-tion of whether there are intermediate distinctions between the real and the merely rational.

39. See the passage from Scotus cited in Menn 1997, 234, n. 13.

human intellect and will are inseparable, they are formally distinct insofar as what itis to have an intellect differs from what it is to have a will, and vice versa. And thougha particular degree of intensity of whiteness and the whiteness that has that degree ofintensity are inseparable, the former is modally distinct from the whiteness insofar asit must be understood through the nature of whiteness, but not the nature of whitenessthrough it.40

The Scotistic notion of formal and modal distinctions reappeared in the work oflater Iberian Jesuits such as Fonseca, who took them to be a means of accepting thevoluntarist axiom without falling into the nominalist trap.41 According to Suárez, how-ever, the Scotistic distinctions are unclear and in need of fundamental renovation. Asa first move away from Scotus, Suárez insisted that one-way separability is not suffi-cient for a distinctio realis, that is, a distinction of res from res. What is required,rather, is mutual separability.42 Moreover, he held that where there is mutual insepara-bility, there can be only a distinctio rationis, that is, a distinction merely in reason andnot in reality. In this case, there is simply a single res that is conceived in differentways. Finally, Suárez transformed Scotus’s modal distinction into a distinction of a resfrom a modus that cannot exist apart from it, though it can exist apart from the modus.In contrast to a distinction of reason, a modal distinction marks some distinction inreality, albeit not a distinction of res from res. Whereas those influenced by Scotustended to hold that shape is formally distinct from quantity, Suárez claimed that theformer is distinct in reality from the latter insofar as there is a modal distinctionbetween the two.43

It is clear that Descartes had some knowledge of Suárez’s Disputations, since atone point in the Fourth Replies he appealed to a passage from this text in support ofhis conception of “material falsity” (AT 7:235).44 However, this one relatively minorpoint of contact hardly exhausts the influence of Suárez’s views on Descartes’s sys-tem. I have already indicated that the voluntarist axiom that was central to Suárezand other Jesuit moderate realists reappears in Descartes. Moreover, Suárez’s spe-cific form of metaphysics is reflected in the theory of distinctions that Descartesoffers in his Principles of Philosophy. For following Suárez, Descartes holds in this


40. For discussion of Scotus’s account of formal and modal distinctions, see King 2003, 22–26.41. On Fonseca, see Menn 1997, 242–50.42. Menn shows that Scotus’s more permissive criterion for a real distinction lands him in

difficulties with respect to the transcendental relation of inherence. Given his view that anaccident can exist (if only miraculously) without inherence, it follows that the accident and itsinherence must be really distinct res. But since he was committed to the voluntarist axiom,Scotus must hold that the inherence can exist apart from the accident as well, and so withoutits inhering in the accident. By the same line of reasoning, however, the inherence’s inherencemust be really distinct from that inherence, and we are on our way to an infinite regress (seeMenn 1997, 234–35). In denying that one-way separability entails two-way separability,Suárez was able to avoid this regress.

43. For Suárez’s theory of distinctions, see MD VII, 1:250–74.44.44. Descartes cited MD IX.2, ¶4, 1:322*, in defense of his remarks concerning material fal-

sity in the Third Meditation, at AT 7:41. On material falsity in Descartes, see chapter 2, note 44.

text that there is a threefold distinction tied to the separability of the objects beingdistinguished. Thus for Descartes, as for Suárez, two-way separability results in adistinctio realis, one-way separability in a distinctio modalis,45 and mutual insepa-rability in a distinctio rationis (PP I.60–62, AT 8-1:28–30).46

Further evidence of a Suárezian influence is provided in a 1643 letter toMersenne, in which Descartes is concerned to deny the scholastic view that “thereare any real qualities in nature, which are attached to substance, as little souls totheir bodies, and which can be separated from it by divine power” (26 Apr. 1643, AT3:649).47 Here Descartes has in mind the scholastic claim—common to Thomisticextreme realists, Jesuit moderate realists, and nominalists—that sensible qualitiesare really distinct from the material substances in which they inhere. In contrast tosuch a view, he insists that there is “no more reality either in motion, or in all theseother variations of substance that one calls qualities, than the philosophers com-monly attribute to shape, which they call not qualitatem realem, but only modum”(To Mersenne, 26 Apr. 1643, AT 3:648–49). But the “philosophers” who call shapea mode rather than a real quality are not the scholastics in general, or even the Jesuitmoderate realists as a group, but Suárez in particular, who offered as an alternativeto various other scholastic views the technical concept of mode and the accompany-ing theory of distinctions that Descartes incorporated into his metaphysics.48

Even if he recognized this connection to Suárez’s renovated metaphysics, whichis perhaps questionable, Descartes did not draw attention to it. Nor did he acknowl-edge any specific debt to Suárez’s account of causation. But my brief considerationof the impact of Suárezian metaphysics on Descartes should warn us against takinghis indifference to the details of this account to indicate its irrelevance for his con-cerns. Indeed, it will turn out that the Suárezian account is distinguished by claimsconcerning efficient causality and God’s causal contribution to natural interactionsthat are directly relevant to Descartes’s theory of causation.


45. There is admittedly a complication in this case given Descartes’s admission in thePrinciples that there can be a modal distinction between different modes of the same substance,even though such modes can exist apart from each other (PP I.61, AT 8-1:29–30). However,Descartes indicates that this case counts as a modal distinction only because both modes areinseparable from the same substance. He notes that in the case where the modes belong toreally distinct substances, the distinction between them is more properly a real than a modalone (AT 8-1:30). Thanks to Eric Watkins for bringing this complication to my attention.

46. In the First Replies, Descartes follows the lead of his critic Caterus by invoking theview in Scotus that there is a formal distinction between any items that can be conceivedthrough different concepts (AT 7:120–21; cf. First Objections, AT 7:100). Whereas he identi-fies formal and modal distinctions in this text, however, Descartes notes in the Principles thatthe formal distinction between thoughts of attributes that are only rationally distinct is itself adistinctio rationis, not modalis (PP I.62, AT 8-1:30).

47. I return to the (mis-)characterization of the scholastics as positing tiny souls attachedto bodies in §2.1.2 (ii.b) and toward the end of §2.1.3 (ii).

48. For more on Suárez’s understanding of the Aristotelian categories and its relation toDescartes’s views, see the remarks in §1.3.

1.2.2. The Priority of Efficient Causes

Suárez begins his treatise on causation in the Disputations by addressing the ques-tion of whether there is any ratio common to all cases of causality. After consider-ing and rejecting various suggestions drawn from Aristotle’s texts, he settles on theclaim that “cause is a per se principle from which being flows into another” (causaest principum per se influens esse in aliud) (MD XII.2, ¶4, 1:384*). Practically everyterm in this sentence requires explanation. In saying that a cause is a principle,Suárez means to indicate that it is the thing that causes (res quae causat), as opposedto the causality itself (causalito ipsa) or the relation grounded in that causality (¶1,1:384*). Thus, it is the heat in the fire that produces heat, rather than its productionof heat or its relation to the heat it produces, that serves as the principle of this pro-duction. By holding that the principle is per se, Suárez means to exclude those thingsthat are not res properly speaking or that are res but are linked merely per accidensto the cause of an effect. Thus, the fact that fire is not cold or the fact that it is yel-low is linked only per accidens to its production of heat: in the first case, since thelack of cold is a privation and not a res at all, and in the second case, since the heatderives from the heat in the fire rather than from its color. Finally, the fact that thecause influit being into the effect indicates that it “communicates” or “gives” beingto another (dandi vel communicandi esse alteri), a being of a sort that the cause itselfsomehow “contains” (¶4, 25:384*).49

Suárez admits, however, that this definition does not apply equally to all membersof the Aristotelian quartet of material, formal, efficient, and final causes. The defini-tion applies least well to the first two, which he called “intrinsic causes,” since suchcauses communicate being to “another” only in an attenuated sense. It is only in thecase of the latter two, which he called “extrinsic causes,” that being is straightfor-wardly communicated to something external to the cause. However, even in the caseof the latter the definition applies in the strictest sense only to efficient causes insofaras most final causes communicate being not directly by means of “an action,” but onlyindirectly by means of a “metaphorical motion.”50 To fully understand these conclu-sions, we need to delve a bit into Suárez’s account of the metaphysics of causality.51

I consider first the case of material and formal causes, then efficient causes, which forSuárez provide the gold standard for causation, and finally the complicated case offinal causes.


49. As I indicate in §2.1.3, Suárez holds that this being is contained in its cause “formally”when it is the same kind of being as what produces it and is contained in its cause “eminently”when what produces it is “more noble.”

50. As we will see in §1.2.2 (iii), however, Suárez makes an exception for God’s finalcausality, since he held that this causality produces its effects by an action and so is not dis-tinguishable from God’s efficient causality.

51. For a more detailed consideration of Suárez’s account of the four causes that empha-sizes the priority of efficient causality, see Carraud 2002, 145–63 (in a section appropriatelytitled “La reduction des causes à l’efficience”). There is a complementary discussion ofSuárez’s account in Olivo 1997.

(i) Material and Formal Causes

Suárez’s account of intrinsic causes assumes the hylomorphic view basic to scholas-ticism, according to which the basic elements for composition of bodies are primematter and various substantial and accidental forms. Prime matter is a material causethat is the recipient of change, whereas forms are formal causes that are the activeprinciples of change. The distinction between substantial and accidental formsserves to distinguish the formal causes of the generation of composite material sub-stances (viz., substantial forms) from the formal causes of accidental changes in suchsubstances (viz., accidental forms).

On these points, most scholastics were agreed. However, the details of Suárez’saccount of material and formal causation were more controversial. For instance,orthodox Thomists held that matter, as pure potentiality, does not have any being ofits own apart from form. Such scholastics therefore could not accept Suarez’s viewthat the material cause fits the definition of a cause that “inflows” its being into theeffect. But Suárez insists that even though prime matter is merely potential, it has itsown essence apart from form, namely, the essence of a potential recipient of change.It is this essence that matter contributes to the effect (MD XIII.4, ¶9, 1:411*).52

There is no similar dispute over the status of formal causality, since Suárez agreeswith the Thomists that forms are principles of activity, and thus have their ownbeing. Nevertheless, Suárez’s view is distinguished from that of earlier scholasticsby his claim that a formal cause is not a cause in a full and proper sense, since it actsmerely by means of “a formal and intrinsic union” with matter (MD XV.6, ¶7,1:520). The “influx” of both the material and the formal cause thus involves merelyan “internal composition” to which matter contributes the “mode of potentiality” andform the “mode of activity” (MD XII.3, ¶9, 1:391*). Suárez’s conclusion is thatsince such an influx is not precisely the same as the influx that occurs when a causeproduces an effect external to and distinct from itself, material and formal causes canbe called causes only “by analogy.” The analogy, in particular, is to the efficientcause, which “most properly inflows being” (MD XXVII.1, ¶10, 1:952*).53

(ii) Efficient Causes

Suárez starts his discussion of efficient causality with a consideration of Aristotle’sdefinition of an efficient cause as that “whence there is a first beginning of changeor rest.” He rejects this definition on various grounds, including the fact that it doesnot exclude material and formal causes, which in some sense also provide a princi-ple for the beginning of change or rest, and the fact that it does not include divinecreation, which does not involve a beginning of change or rest in an already existing


52. Here Suárez was under the influence of the Scotist position that prime matter is a resreally distinct from substantial form. For discussion of this position, see Des Chene 1996, §5.1.

53. As we will discover in §2.1.2 (ii.a), Descartes also allowed for formal causes that aremerely analogous to efficient causes, though his account of formal causality differs substan-tially from the account the account in Suárez that I have just considered.

subject (MD XVII.1, ¶¶2–4, 1:581–82).54 The alternative definition he proposes isthat an efficient cause is “a principle from which the effect flows forth, or on whichit depends, by means of an action” ( principium a quo effectus profluit seu pendet peractionem) (¶6, 1:582).

The definition of an efficient cause as that which involves a “flowing forth” of theeffect may not seem to be less than entirely clear. Indeed, Leibniz complained in hispreface to a 1670 edition of Nizolius’s On the True Principles of Philosophy (Deveris principiis . . . philosophandi) that Suárez’s definition “is rather barbarous andobscure, . . . more obscure than what it defines: I would hope to define cause moreeasily than this term influxus taken so monstrously” (Leibniz 1978, 4:148). However,Suárez indicated that in a general sense influxus means simply “giving or communi-cating being to another” (dandi vel communicandi esse alteri) (MD XII.2, ¶4,1:384*). Leibniz also had difficulties with the notions of giving or communicatingbeing, most of which rested on the fact that he could not conceive of the literal trans-fer of some feature of the cause to the effect.55 Yet when Suárez speaks of the effi-cient cause as involving the flowing of an effect, he should not be understood toclaim that a feature of the cause is literally transferred to the effect. His view in factrequires that an efficient cause is extrinsic for the very reason that it does not com-municate “its own proper and (as I will put it) individual esse to the effect.” Rather,what occurs in the case of efficient causality is “some other [being] really flowingforth [profluens] and proceeding [manans] from [an efficient] cause by means of anaction” (MD XVII.1, ¶6, 1:582). In either creating or educing an effect, an efficientcause produces an esse that is distinct from, though in some manner similar to,56 theesse that it possesses.

Suárez claims that an efficient cause not only produces a new esse, but also pro-duces it by means of an “action,” where this consists in “the emanation or depend-ence of an effect on its extrinsic cause, from which it receives being.” Suárez himselfadmits that this definition may seem to be uninformative, since it makes action“almost the same” as an efficient cause (MD XVII.1, ¶5, 1:582). In his view, how-ever, the action is distinguished from the cause by the fact that the former constitutesthe causality of the efficient cause, whereas the latter is the principle of that causal-ity. Suárez follows other scholastic thinkers in taking the action to be something thatresides in the patient rather than the agent.57 But drawing on his renovated form ofscholastic metaphysics, he characterizes this action as a certain mode of the effect


54. For the point about creation, see §1.2.3 (i).55. In 1696 comments on his “New System of Nature” (“Système nouveau de la nature”),

for instance, Leibniz objected to “the way of influence” on the grounds that “we can conceiveneither material particles nor immaterial qualities or species that can pass from one of thesesubstances [viz., the soul and body] to the other” (Leibniz 1978, 4:499). For more on the back-ground to Leibniz’s conception of “the way of influence,” or what he also called, followingSuárez (see MD XVII.2, ¶6, 1:585), influxus physicus, see O’Neill 1993.

56. See note 49.57. However, Suárez mentions Cajetan and Scotus as the main dissenters from this posi-

tion; see MD XLVIII, ¶2, 2:888–89*.


that the cause produces, namely, the mode of depending on that cause (MDXVIII.10, ¶8, 1:682).58

In identifying the action with the causality of the cause, Suárez offers—characteristically enough—a middle way between the views of Thomistic extremerealists and nominalists. On the one hand, he holds against the nominalists that anaction is something distinct in reality from the agent, its power, and the effect in thepatient. On the other, he holds against the Thomists that causality is not somethingover and above the action of an agent, but is identical to this action, which itselfexists as a mode of the effect (MD XVIII.10, ¶5, 1:681).

Suárez’s theory of action is complicated by the fact that he recognizes two dif-ferent kinds of action that efficient causes can involve, namely, transeunt action,which “has an effect outside the agent itself,” and immanent action, which “has noeffect outside the agent” (MD XLVIII.2, ¶1, 2:874*). I have noted above the case ofthe eduction of substantial or accidental forms from matter, which for Suárez is anexample of transeunt efficient causation (i.e., efficient causation by means of atranseunt action). When the air causes the apple to become brown, the action is thedependence on the air that modifies the process in the apple of turning brown,whereas the terminus of the action is the qualified change involving the inherence ofthe accidental form of brownness. When worms cause the apple to decompose intoits elements, the dependence on the worms that modifies the process of decomposi-tion is the action, and the terminus of the action is the unqualified change involvingthe union of the matter of the apple with the new substantial forms of the elements.

The case of transeunt efficient causation is best suited to the definition of a causeas that from which being flows forth into another. The case of immanent efficientcausation (i.e., efficient causation by means of an immanent action) is more prob-lematic insofar as the distinction of the effect from the cause is less clear. Since heaccepted the Aristotelian principle that motion (in the broad sense of any change)requires an external mover in the case of material objects (see MD XVIII.7, ¶37,1:642), Suárez claims that the primary examples of immanent efficient causation arechanges that pure intellect or will causes in an intellectual (i.e., angelic) or rational(i.e., human) soul. In the case of the cognitive acts of pure intellect, however, Suáreznotes that the effect is an “intelligible species” that is really distinct from the facultythat produces this species. Here he is simply following the Thomistic position thatintellectual cognition involves the impressing of this species by the “agent intellect”in the “passive intellect.”59 This way of saving the distinction of the effect from thecause is not available in the case of the will, given Suárez’s position that no distinct

58. But see the discussion in Hattab 2003 of the dissenting view in the work of the earlymodern scholastic Charles François d’Abra de Raconis that the causality of the efficient causeis distinct from its action in the patient.

59. For more on this Thomistic view in Suárez, but also his disagreements with Thomasconcerning the production of the intelligible species, see §4.2.1. Suárez notes that thoughthere is no species involved in the angelic contemplation of its own substance, still its sub-stance as the object of the intellectual act can be distinguished from that substance as the prin-ciple of that act (MD XVIII.7, ¶48, 1:646).

species are involved in this case. Nevertheless, Suárez insists that there is a distinc-tion in the will between a “first act” involving the power of producing a certainimmanent effect, on the one hand, and the “second act” consisting in the immanenteffect, on the other. In producing a desire, the will in first act merely has the powerto produce the quality of desiring an object, whereas the exercise of that powerresults in the second act of the inherence of that quality in the will (¶51, 1:647).Since desire is itself a real quality, and so really distinct from the power of the willthat produces it,60 its production involves the flowing of being into something dis-tinct from the volitional power that serves as the principle of this effect.

In giving priority to efficient causes over material and formal causes, Suárez fol-lows the view of the medieval philosopher Avicenna that the requirement that theeffect be in some way distinct from the cause is central to the notion of causality.61

What is distinct, in particular, is the esse of the effect that “flows forth” from thecause. However, in claiming that an efficient cause produces the esse of its effect,Suárez need not hold that it is a cause secundum esse as Thomas understood thisnotion. For recall the view in Thomas that a cause secundum esse brings about notonly the presence of its effect, but also the fact that the effect has the nature that itdoes (see §1.1.2). There is no suggestion in Suárez that it is essential for somethingto be an efficient cause that it bring about the latter. What is essential is only that thecause produce some being, whether with the assistance of other causes (as in the caseof all actions of secondary efficient causes, which depend on God’s “concursus”) orentirely by itself (as in the case of divine creation and conservation).62 Before turn-ing to Suárez’s views concerning secondary efficient causality and its relation toGod’s causal activity, however, we need to complete our summary of his account ofcausation by considering his complex attitude toward what for Descartes, at least, isthe most problematic of the four kinds of Aristotelian causality, namely, the causalityof final causes.

(iii) Final Causes

According to Suárez, final causes are the second of the two kinds of extrinsic causes.Thus, as in the case of efficient causes, the general notion of causality (“the principlefrom which being flows into another”) fits final causes better than material or formalcauses. Indeed, at the start of his discussion of final causes in the Disputations, Suárezeven claims, in apparent conflict with his main thesis of the priority of efficient


60. Desire belongs to the third species of the predicamental category of quality, whereas thevolitional power that produces it belongs to the second species of that category; see note 36.

61. As indicated in Gilson 1986, Suárez also followed Avicenna and Peter of Auvergne(†c.1310) in combining efficient causes with “motive” causes that thinkers such as Aquinashad distinguished from them. It is because he held that the divine creation of being and theproduction of motion/change by secondary causes both involve an inflowing of being into aneffect that Suárez was able to treat both as instances of efficient causality.

62. In §1.2.3 (i), I discuss Suárez’s comments on the passage from the Summa Theologiaethat concerns the secundum fieri/secundum esse distinction.

causes, that of the four main causes, final causes “are in some manner the most prin-cipal of all, and also the first” (MD XXIII.1, 1:843*).63 His reasoning here is thatsince even the action of an efficient cause is directed toward a terminus as its end, effi-cient causality involves the causal efficacy of an end, and thus final causality (¶7,1:845*). However, Suárez admits that “the reason of the causing of [the final cause]is more obscure” than in the case of the other three kinds of cause (1:843*). Thisobscurity is due to the fact that there are very different kinds of causality dependingon whether the action involves (a) “an uncreated intellectual agent, which is Godalone,” (b) “created intellectual agents, among which humans are best known to us,”or (c) “agents that are natural, or lacking intellect” (¶7, 1:845*). What supportsSuárez’s thesis of the priority of efficient causes is both the fact that final causality incase (b) involves not genuine action but only “metaphysical motion,” and the fact thatwhen case (c) is considered in abstraction from God’s causal contribution, there is nogenuine final causality at all. It is only in case (a) where the final cause produces itseffect through an action, and in this case only because there is no real distinctionbetween God’s final and efficient causality. Let us consider these three cases, startingwith the case best known to us, namely, the one in which we as created intellectualagents act as final causes.

(b) As with other created intellectual agents, final causality enters into only theimmanent actions of our will. Earlier we noted the distinction in Suárez’s account ofsuch action between the first act involving the power to produce an internal effectand the second act identified with the attainment of this effect. The ends of actionthat we cognize are final causes insofar as they incline the will in first act to pursuethese ends as opposed to others. The “motion” associated with this inclination ismerely “metaphorical” insofar as we do not actually pursue the particular endstoward which we are inclined in first act.64 The pursuit is actual, and thus the endsare efficacious, only when our will produces by means of an immanent action thedesire for or love of those ends. Thus, even though the cognized ends as final causesare “in some manner the most principal” and “the first” insofar as they incline thewill to act in a particular manner, it is the will itself rather than these ends that is theefficient cause that directly produces the relevant second acts.

The insistence on the merely metaphorical nature of the motion involved in thefirst act is particularly important for Suárez in the case of our free actions, since hewas deeply committed to the position that such actions do not derive necessarilyfrom our will in first act.65 In his view, this first act can be free, and therebyelicit a second act that is free, only if it is “an active faculty that has control over itsown action in such a way that it has within its power to exercise that act and not to


63. In this subsection I am following the helpful treatment of Suárez’s views on finalcausality in Carraud 2002, 152–61. See also the more general discussion of the relevantscholastic background in Des Chene 1996, ch. 6.

64. Suárez cites Aristotle (On Generation and Corruption I.7, 324b14–15, Aristotle 1984,1:530) and Thomas (ST I-II.1.1) as sources for his account of metaphorical motion. See alsothe development of this notion in texts from Scotus cited in Carraud 2002, 158, n.1.

65. I return to this position in Suárez in §1.2.3 (ii) and then again in §5.1.1.

exercise it and, consequently, to elicit one action or another—that is, opposite—action” (MD XIX.2, ¶18, 1:698). Suárez identifies this lack of determination to aparticular action with the “indifference” of the free active faculty of will. He furtherdistinguishes two kinds of indifference, namely, indifference with respect to theexercise of an act, which is required for our freedom to act or not to act, and indif-ference with respect to the specification of an act, which is required for our freedomto elicit one action as opposed to other contrary actions (MD XIX.4, ¶9, 1:708–9).Given these kinds of indifference, the cognized end cannot be said to produce in thewill an actual motion (in the broad Aristotelian sense of a change) that, if unim-peded, necessarily terminates in a particular second act. Rather, it merely entices thewill to freely produce this act, that is to say, it serves only as a final and not as anefficient cause of that effect.66

(c) I have noted the passage from the 1643 letter to Mersenne in which Descartescaricatures the scholastics as holding that bodies have real qualities attached to themas little souls (AT 3:649). Around the same time, in the Sixth Replies, he reports thatin his youth he was under the sway of the scholastic view that the free fall of bodiesis explained by the fact that they possessed the real quality of heaviness (gravitas),one that “carried bodies toward the center of the earth as if it had some knowledgeof the center within itself” (AT 7:442). In effect, the proposal here is that the finalcausality of heaviness is to be understood on the model of the scholastic account ofthe final causality of created intelligent agents. For just as Suárez takes such agentsto be directed by cognized ends to act in a particular manner, so the real quality ofheaviness is supposed to be directed by its cognition of the center of the earth tocarry the body to which it is attached to that location.

The suggestion in Aristotle is that final causality is not restricted to cases involv-ing cognition, but rather derives in general from the forms of composite substances,including those substances that lack intellect.67 However, Suárez is concerned todeny that any natural being—that is, any being that does not act by means of will—can be a final cause at all. This is clear from his claim in the Disputations that in thecase of “those actions, that are from natural agents, there is properly no final causal-ity, but only an inclination toward a certain terminus” (MD XVIII.10, ¶6, 1:887).Even when created intellectual agents act by some means other than will, accordingto Suárez, they are merely natural agents, and so are not true final causes (MDXXIII.3, ¶18, 1:857*). The finality of the actions of natural agents thus cannot beexplained by appeal to the nature of these agents alone, who serve merely as efficientcauses. Rather, Suárez claims that “there is final causality in them only as they are


66. Whereas Suárez holds that there is no volitional act, in this life, at least, toward whichwe are not indifferent with respect to exercise, he allowed that we are not indifferent withrespect to specification toward volitional acts directed to ends proposed under the concept ofa universal good. In the case of such acts, Suárez’s conclusion is that we perform them vol-untarily but not freely; see MD XIX.8, 1:726–32. There is a further discussion of Suárez’sviews on these points in §5.1.1 (i) and 5.2.1.

67. In MD XXIII.10, ¶2, 1:886*, Suárez cites as the source of this view Aristotle’s dis-cussion in Physics II.7–8, 198a14–199b30 (Aristotle 1984, 1:338–40).

from God, as in God’s other external and transeunt actions” (MD XXIII.10, ¶6,1:887*). Thus, natural agents are directed to their ends by God, whose action isdirected by his cognition of these ends. We have here one reason for Suárez to con-clude that divine action is involved in the action of all natural agents. Such a con-clusion is reinforced by his concurrentist position that God acts by means of theaction of all secondary causes (see §1.2.3 (ii)).

(a) We have noted the position in Suárez that the final causality of created intellec-tual agents involves a metaphorical motion of the will. However, he holds that sincethe one uncreated intellectual agent is purely actual, and thus has no potentiality, thisagent, namely, God, can in no way possess metaphorical motion (MD XXIII.9, ¶6,1:883*). Indeed, Suárez denies that there is any final causality internal to God himself.Though God does love himself or others for the sake of his own goodness, his attrib-ute of goodness is not a final cause of this love. Rather, it is “only the reason (as it issaid) of the divine will” (rationem tantun (ut dixi) voluntatis divinae) ( ¶6, 1:883*).Final causality is involved only when God acts as a transeunt efficient cause, and thusproduces effects by means of an action external to him (¶12, 1:885*). However, thereis no real distinction here between God’s final and efficient causality insofar as “thefinal causality of God with respect to external effects consists in this, that God pro-duces the external effect by the intuition and love of his goodness.” Thus “one and thesame operation . . . pertains to God whether by reason of efficacy or by reason of end,since it is related to God both as omnipotent and as the greatest good” (¶9, 1:884). Theview here that God’s production of external effects involves both final and efficientcausality is reflected even in Descartes, who despite his disdain for appeals to divinefinal causes (see §2.1.2 (ii.b)), nonetheless told a correspondent that all creatures canbe said to exist for God’s sake insofar as “it is God alone who is the final cause as wellas the efficient cause of the universe” (To Chanut, 6 June 1647, AT 5:54).

As in the case of Descartes, however, Suárez’s discussion of God’s causal contri-bution to the world emphasizes more the relation to his power as efficient cause thanthe relation to his goodness as final cause. Thus, in the section of the treatise on causal-ity in the Disputations that concerns efficient causes, disputations XX through XXIIare devoted to God’s activity as primary efficient cause in creation, conservation, andconcurrence. There Suárez takes the first two kinds of activity to be intimately related,as shown by his thesis that divine conservation is not distinct in reality from God’s actof creation, but is merely the continuation of that act. However, he insists against crit-ics such as Durandus that there is a divine concurrence that involves a “concursus” thatis distinct from God’s act of creation and conservation. I have indicated that theseclaims are an essential part of the “causal compatibilism” that Thomas proposed sev-eral centuries prior to Suárez. However, Suárez moved beyond Thomas in explicatingthese claims in terms of a comprehensive theory of efficient causality.

1.2.3. Creation, Conservation, and Concursus

(i) Creation and Conservation

Suárez’s discussion in disputation XX opens with the stipulation that creation involvesthe production of an entity ex nihilo. Since prior to this action there is nothing on which


to act, creation must be distinguished from more mundane examples of efficient cau-sation that involve a change produced by action on an existing subject. Thus, creationdiffers from the eduction either of an accidental form from a material substance, as inthe case of accidental change, or of a substantial form from prime matter, as in the caseof substantival generation (MD XX.1, ¶1, 1:745). In both of these cases, the efficientcausality involves a change in a patient, whereas in the case of creation, there is nosuch change, since it is the existence of the patient itself that is produced. Nevertheless,Suárez insists that creation can be placed in the same category with the efficient cau-sation of accidental and substantival change, since they all fit his definition of efficientcausation, namely, the flowing forth of being into another by means of an action. Thedifference is merely that whereas change presupposes the existence of the patient thatreceives the new esse, creation does not. Moreover, in creation as well as in the othercases of efficient causation, the action is a mode of the effect, and thus is somethingthat is only modally and not really distinct from that effect. In particular, this action isthe dependence on the cause that modifies the effect that is produced.68

Suárez is concerned to distinguish creation ex nihilo from creation de novo, orcreation in time. He does defend the claim that creation ex nihilo is compatible withcreation de novo against the objection that what is created must be eternal insofar asthe divine act of creation is eternal. He responds by appealing to his position, men-tioned previously, that an action is in the patient rather than the agent. His conclu-sion is that reason is perfectly consistent with the dictate of faith that an eternal Godcreated the world with a starting point in time (MD XX.5, ¶¶5–10, 1:780–82).However, Suárez also holds that God could have created the world ab aeterno, andthus could have created a world that is eternal in the sense of having no beginning intime. Creation “out of nothing” thus could signify not that there was a point at whichthe creature did not exist, but only that the creature would not have existed were itnot for the fact that from eternity esse had been communicated to it from another(¶¶11–12, 1:782). Suárez claims that since neither matter (and the material formseduced from matter) nor finite immaterial entities exist a se, that is, from their ownnature, they can exist in the first place only because they have being from an efficientcause that does exist a se, namely, God (MD XX.1, ¶¶15–21, 1:783–85).69 Divinecreation would be necessary whether or not these entities were eternal.

Suárez accepts the traditional conclusion that God alone can create a being exnihilo. But though most scholastics followed Thomas in holding that natural reasoncan demonstrate this conclusion, Durandus argued against Thomas that there isnothing in the notion of creation as such that precludes a creature from creating


68. See MD XX.4, 1:769–79. In this section, Suárez offers this position as an alternativeboth to the Thomist position that the dependence of the creature on the Creator is a res dis-tinct from that creature, since such a dependence belongs to the category of relation, and thenominalist position that this dependence is only distinct in reason from the creature. His posi-tion is thus perfectly in line with his renovated metaphysics (see §1.2.1).

69. Suárez notes that even though the material substances are generated out of matterrather than directly created by God, still they depend on divine creation insofar as the matterout of which they are generated must be created (MD XX.1, ¶22, 1:751).

(S II.1, 4, 1:129–30*, responding to ST I.45.5). Suárez claims that there are moreconstraints on creation than Durandus allows. He concedes to Thomas, for instance,that only something with infinite power could have the unlimited ability to create anybeing whatsoever. Furthermore, Suárez’s concurrentist position that God must con-cur in all creaturely action (see §1.2.3 (ii)) precludes the possibility of any causeother than God creating without any divine assistance. Even so, Suárez notes that itstill seems possible that creatures could have a more limited power to create with thehelp of God’s concursus (MD XX.2, ¶39, 1:764, also responding to ST I.45.5). Hegrants that we can know by faith that God has not in fact created any being that hasthe power to create. He also argues that since something with the perfection of beingable to create would have added to the overall perfection of the universe if it existed,and since if it were possible God would have created this thing for that reason, thefact that no such thing exists provides grounds for thinking that no such thing is pos-sible. Suárez concludes, however, that though the conclusion of this argument is cer-tain for those who accept the tenets of faith, it is not evident on the basis of naturalreason alone (¶12, 1:755–56).

However it is established, the conclusion that God alone can produce creaturesex nihilo falls short of the thesis that creatures depend on God for the continuationof their existence subsequent to their creation ab aeterno or de novo. Suárez arguesfor this additional thesis in disputation XXI, where he claims not only that the con-tinuation of the existence of creatures depends on God’s efficient causality, but alsothat this continuation depends on the very same act by which God created them inthe first place. In arguing for the former point, Suárez starts with the thesis—drawnfrom the Thomistic distinction between causes secundum fieri and secundumesse—that when “an effect that depends on its cause directly and per se and prima-rily with respect to its esse, it depends on that cause not only in becoming [ fieri]but also in being conserved [conservandi]” (MD XXI.1, ¶6, 1:787). He notes thatThomas himself explained the distinction between being a cause of fieri and beinga cause of esse “somewhat obscurely” (¶8, 1:787). The obscurity here seems toderive from the fact that even a cause secundum fieri is a source of the esse of itseffect, and so is not clearly distinct from the cause secundum esse. However, Suárezproposes that an effect is from a cause secundum fieri insofar as “it does notabsolutely and unconditionally require that cause to exist, but instead requires itonly to exist through the action or production in question.” In contrast, the effect isfrom a cause secundum esse insofar as “it absolutely and unconditionally requiresthat cause in order to exist” (¶8, 1:787). Thus, Adam is a cause of Abel only secun-dum fieri insofar as Abel does not absolutely require Adam to exist; God could havecreated Abel without any causal input from Adam. In contrast, the cause secundumesse of Abel must be such that Abel could not exist without the activity of thatcause (¶8, 1:787).

Though offered as an analysis of Thomas’s distinction between causes secundumfieri and secundum esse, Suárez’s alternative definitions differ in at least one importantrespect from those that Thomas offered. Thus, whereas Thomas’s own example of thesun suggests that he allowed for causes secundum esse other than God, Suárez empha-sizes that given his definition God alone can be such a cause. For since God can pro-duce any effect by himself that he produces with secondary causes, God alone can be


absolutely and unconditionally required for the effect.70 Given this strong requirementfor being a cause secundum esse, it is perhaps clearer in Suárez than in Thomas whyan effect depends on such a cause not only at the first moment of its existence, but alsoat every moment it exists. For it follows from this requirement that the existence of theeffect, at whatever time it exists, depends absolutely and unconditionally on its causesecundum esse. Thus, insofar as creatures must be created by God to exist in the firstplace, they must also depend on God for their continued existence.

Suárez’s second point above is that God conserves creatures by means of thesame act by which he created them. Drawing again on Thomas’s discussion of thesecundum fieri/secundum esse distinction, Suárez claims that there is no more justi-fication for saying that God conserves by means of an act distinct from creation thanthat the sun continues to propagate light by means of an act distinct from that bywhich it first produced the light (MD XXI.2, ¶3, 1:791). In the case of the sun, thedifference between production and propagation is merely that the term for the for-mer connotes the prior absence of the light, whereas the latter connotes the priorpresence of that light. The difference here is only a difference in which one and thesame act is described, and so is a mere distinctio rationis, and not a distinction inreality. Likewise, in the case of God the difference between creation and conserva-tion consists in the fact that the term for the former connotes the denial of a previ-ously possessed esse, whereas the term for the latter connotes the prior possessionof esse. Here again, the difference is in the words used to describe the action ratherthan in the action itself.

In my discussion in §1.1.3, I noted the objection in Suárez that since there can besecondary conserving causes, a mere conservationist such as Durandus has no goodreason to hold that God must conserve all beings in existence. What requires expla-nation here is how Suárez himself conceived of the relation between conservingsecondary causes and divine conservation. In the case of secondary causes he dis-tinguishes between the per accidens conservation that involves the removal of animpediment to continued existence, as when an angel conserves a human being byturning away a rock, and the per se conservation that involves the contribution ofsomething needed for continued existence, as when the sun conserves life by givinglight. However, even the latter sort of conservation is distinct from divine conserva-tion insofar as it is merely “remote and mediate,” whereas divine conservation is“direct and immediate.” The contrast here derives from the fact that divine conser-vation alone involves “the persistent influx of that very esse that was communicatedthrough production” (MD XXI.3, ¶2, 1:794). In line with my remarks toward the endof §1.1.3, however, I would simply note the possibility of a mere conservationistposition according to which God alone is the direct and immediate per se conserv-ing cause of objects, but secondary causes act alone as per accidens or per se remoteand mediate conserving causes of those objects.


70. For the relevance of this difference between Thomas and Suárez to Descartes’s under-standing of the Thomistic distinction between causes secundum fieri and secundum esse, see§2.2.1, at note 65.

(ii) Concursus and Secondary Causes

At the start of disputation XXII, Suárez observes that “of the concursus of the primarycause with secondary [causes] as regards their actions, one finds that little is said byAristotle and other philosophers” (MD XXII, 1:802). His conclusion in this sectionthat God as primary cause concurs per se and immediately in all actions of second-ary causes of course recalls the thesis of Thomas’s causal compatibilism that Godoperates in all operations in nature. However, Suárez does not follow Thomas inequating secondary causes with instrumental ones. Moreover, he offers an alternativeto an account of God’s concursus with free human action in the work of some fol-lowers of Thomas that appeals to the relation of agents to instrumental causes.Though Suárez’s account of divine concurrence clearly is indebted to Thomas, it alsodiffers on important points of detail from the views of Thomas and later Thomists.71

We have seen that Thomas’s defense of causal compatibilism relies on the anal-ogy to instrumental causality. God’s action with creatures is compared to an agent’suse of an instrument. In both cases, there is a single effect that two subordinatedagents produce by the same action. However, we have also seen the objection inDurandus that secondary causes are not mere instruments when they elicit theireffects by means of a power that is proportioned to those effects. Thus, even thougha pen does not have the power to produce words unless moved by an agent, it seemsthat fire has the power to heat on its own. Insofar as the fire has such a power, therewould be no need to appeal to another principle of the effects of this power in God.

Suárez discusses various scholastic attempts to respond to this line of objection bydistinguishing instrumental causes from “principal” efficient causes (MD XVII.2,¶¶7–19, 1:585–91). Most notable is Scotus’s proposal (considered in ¶¶10–12,1:587–88) that instrumental causes merely dispose a patient to receive a form fromthe principal efficient cause. Scotus insisted that though even secondary principalcauses must be subordinated to God, the subordination in this case differs from thesubordination of an instrumental cause to a principal cause. For whereas the subordi-nated instrumental cause does not produce the ultimate effect, the subordinated prin-cipal cause is enabled to produce this effect by the activity of the primary cause.72

Suárez rejects Scotus’s proposal on the grounds that some instrumental causes pro-duce the ultimate effects directly, as when certain accidents immediately educe a sub-stantial form (MD XVII.2, ¶11, 1:588).73 However, he shares Scotus’s view that the


71. Suárez’s theory of divine concursus is, however, close to the position that his fellowSpanish Jesuit Luis de Molina offered in his 1588 Concordia, the full title of which is Liberiarbitrii cum gratiæ donis, divina præscientia, providential, prædestinatione et reprobationeConcordia (The Compatibility of Free Will with the Gifts of Grace, Divine Foreknowledge,Providence, Predestination and Reprobation). For a comparison of the views in Molina andSuárez, see Perler and Rudolph 2000, 201–13. As we will see in chapter 5, the theory of “mid-dle knowledge” that Molina offered in his Concordia is an important part of the scholasticcontext of Descartes’s discussions of human freedom.

72. For further discussion of Scotus’s proposal and Suárez’s response, see Menn 2000, 131–33.73. Suárez also provides as an example the immediate effecting of an intelligible species

by a phantasm. As I indicate in §4.2.1, however, this example is problematic for him.

subordination of a secondary cause to God need not be the same as the sort of subordi-nation involved in instrumental causality. To capture the difference between these twokinds of subordination, he offers the view that an instrumental cause is one that “concursin, or is elevated to, the production of something more noble than itself, that is, some-thing beyond the measure of its own proper perfection and action” (MD XVII.2, ¶17,590). In the case of a secondary principal cause, the effect is not more noble than itself,and thus its subordination to God does not result in the conclusion that it is a mere divineinstrument. Nonetheless, Suárez insists that this cause is subordinated to God, since it canproduce the effect proportionate to it only with the help of the divine concursus.

Durandus’s question, of course, is why this further assistance is needed given that theeffect is proportionate to the secondary cause. The answer in Suárez, broached in §1.1.3,is that the effect has an esse that requires God’s immediate and per se causality as muchfor its production as for its conservation. However, it might be possible to develop fur-ther the response on behalf of the mere conservationist that I offered earlier. We have con-sidered the distinction in Suárez’s metaphysics between a res and a modus of that res (see§1.2.1). Though there is some distinction in reality here between a mode and its res, theesse of the mode is not independent of the esse of the res, but is a mere determination ofthe latter. Thus, it could perhaps be said that God produces the esse of a mode just inso-far as he creates and conserves the esse of the res that mode modifies. And such a claimseems to leave open the possibility that secondary causes alone produce modifications inan already-existing res. Of course, Suárez would protest that secondary causes can pro-duce substantial and accidental forms that are not mere modes but res distinct from mat-ter. But for someone, like Descartes, who rejected such qualities (see §1.3), a version ofmere conservationism that allows for such a possibility would appear to be a live option.74

As we know, however, Durandus concluded not only that his mere conservationismis an acceptable position, but also that Thomas’s causal compatibilism is an unaccept-able alternative. One of his main arguments for this conclusion is that since God mustproduce the effect of a secondary cause by means of an action that differs from thatcause, either God’s action produces the entire effect, thus rendering the action of thesecondary cause superfluous, or brings about only part of the effect, in which case theaction of the secondary cause produces the other part without divine assistance. Thisdilemma is possible given Durandus’s claim that God cannot produce an effect bymeans of the same action as that of the secondary cause, since “it is impossible fornumerically the same action to be from two or more agents in such a way that it isimmediately and completely from each, unless numerically the same power is in them”(S II.1.5, ¶12, 1:131). However, Suárez simply endorses the Thomistic line, consideredabove, that even though the same action cannot derive entirely from two differentcauses of the same order, it does not follow that it cannot so derive from causes in dif-


74. Cf. Philip Quinn’s suggestion on Durandus’s behalf, as reported in Freddoso 1991,583, n.26, that God is a per se and immediate conserver just of substances and not of acci-dents (see also Quinn 1988). Freddoso objects to this suggestion on the grounds that “no full-bodied naturalist will dispute the claim that secondary causes are capable of effectingsubstances as well as accidents” (Freddoso 1991, 583, n.26). As I indicate in §1.3, however,Descartes at least is not a full-bodied naturalist.

ferent causal orders. Following the development of the Thomistic position in Scotus,Suárez claims that these causes are compatible in the case where one is essentiallysubordinated to the other. Given that the activity of secondary efficient causes is sub-ordinated to God’s activity as primary cause, a single action in the patient can derivefrom causes of both kinds (MD XXII.3, ¶4, 1:826–27, citing ST I.105.5, ad 2).

Suárez’s identification here of the actions of secondary causes with God’s con-cursus with those actions reveals one important difference between divine concur-rence and conservation. We have noted the view in Suárez that God’s per se andimmediate conservation of an object at different times occurs by means of the sameaction, which itself is merely the continuation of his act of creating that object. Incontrast, Suárez emphasizes the distinctness of the acts by which God concurs withsecondary causes. Thus, he argues that since “the concursus external to God is noth-ing other than the action itself ” by which the secondary cause acts, “the concursuswill vary according to the variety of the actions” (MD XXII.4, ¶8, 1:831). WhereasGod immediately conserves an object at different times by means of the same act,then, he must concur by distinct acts in the different operations of that object.

In one sense, we should expect Suárez to distinguish concurrence from conserva-tion. After all, he is concerned to set himself apart from the mere conservationistwho holds that divine creation/conservation exhausts God’s contribution to second-ary causality. However, it will be important in the context of a later consideration ofDescartes’s own views concerning God’s activity as primary cause to remember thisimplication in Suárez that divine concurrence involves a kind of inconstancy in theeffect that is not present in the case of divine conservation.

There is one final objection to concurrentism in Durandus that we have not yet con-sidered. In his Sentences, Durandus appealed at one point to his mere conservationistposition in support of the conclusion that though God is the “universal and primarycause” of our sinful actions, their “proximate and immediate cause” is not God butrather our free will (S II.38.1, ¶4, 1:192*, citing II.1.5, 1:130–31). Suárez is sensitiveto this line of objection, offering as a reason to reject his concurrentism the claim thatin the case of sinful free action, “it is unseemly to attribute such actions to the primarycause insofar as it is operating per se and immediately” (MD XXII.1, ¶5, 1:803).

Suárez’s response to this claim depends on his account of the difference betweenGod’s concursus with “necessary” or “natural” causes, on the one hand, and his concur-sus with “free” causes, on the other. Necessary causes are such that, all the conditionsfor action being posited, the action itself follows necessarily (MD XIX.1, ¶1, 1:688).God’s concursus with a necessary secondary cause is determined to a particular effect.Whereas Suárez claims that all natural and nonrational beings are necessary causes, heholds that there are rational volitional agents that are free causes in the sense that theyare not determined to a particular action even when all the conditions for acting havebeen posited (MD XIX.2, ¶11, 1:696). As I have mentioned, his view is that free agentsare immanent causes that in “first act” are indifferent with respect to which “second acts”to elicit (see §1.2.2 (iii)). Suárez holds that though there is a divine concursus identicalto the second act that the free agent in fact elicits, the conditions for action include God’soffer of a concursus with refraining from eliciting the second act or with eliciting othersecond acts, and so the agent is able either to refrain from acting (and so has “freedomof exercise”) or to act differently (and so has “freedom of specification”) (MD XXII.4,


¶21, 1:834). Since God does not offer only one concursus in the case of free sinfulaction, he does not determine the agent to that action, and so it is the agent rather thanGod who is responsible for the sin.75

As I indicate in chapter 5, Suárez’s view that indifference is an essential element ofhuman freedom was standard among the Jesuits but also a source of controversy in theearly modern period. We also will discover in that chapter that this controversy is animportant part of the context for Descartes’s various discussions of human freedom anddivine providence. However, there is a further feature of Suárez’s account of freehuman action that is connected to his worries mentioned previously concerning theappeal to the case of instrumental causation in an explanation of the relation of God’sactivity as primary cause to the activity of secondary causes. As Suárez notes, certainsixteenth-century Thomists cited Thomas’s claim that God uses secondary causes as hisinstruments in support of the conclusion that God concurs with free human agents bymeans of a “physical premotion.” Just as the craftsman produces an effect by applyinga tool in a particular manner, so God concurs in a free action by “premoving” the willto act in a certain way (MD XXII.2, ¶11, 1:813). However, Suárez claims that Thomasin fact favored the less problematic position that God’s concursus with free humanaction is simultaneous with that action, and indeed is identical to it (¶¶16 and 49–50,1:814 and 823–24). We need not enter here into the dispute over the interpretation ofThomas.76 What is more relevant to our concerns is Suárez’s conclusion that his theoryof “simultaneous concurrence” (as it came to be called) avoids certain difficulties thatconfront the Thomistic theory of physical premotion.77 One crucial difficulty is that anyphysical predetermination through premotion precludes genuine human freedom. Forin Suárez’s view, such freedom requires that the will be indifferent to an action evengiven the presence of all of the prerequisites for that action. But if the predeterminationto a particular action is part of the set of prerequisites, then the will cannot be indiffer-ent to that action, and so not be free in eliciting that action (¶39, 1:821).78 Suárez admitsthat his theory of human freedom has implications for an interrelated set of theological


75. For further discussion of Suárez’s account of divine concurrence in the case of sinfulfree human action, see Freddoso 2001.

76. Suárez’s admitted that certain remarks in On the Power of God support the interpreta-tion of Thomas offered by the Thomistae, but claimed that the relevant discussion in theSumma Theologiae does not (MD XXII.2, ¶52, 1:824).

77. For an indication that these were the standard labels, see the 1704 Use of Reason andFaith (Usage de la raison et de la foi) of the French Cartesian Pierre-Sylvain Regis (or Régis)(1632–1707), which includes a chapter on the dispute over divine concours between defenders ofla prémotion Physique and le (now, la) concours Simultanée (I-2.32, Regis 1996, 383–87). Thelabels for these positions give the misleading impression, which Suárez in fact encourages, thatthe Thomists understood the divine moving of the human will to be temporally prior to the act ofthat will. In fact, they held that the priority is one of nature and not time, and they allowed thatthe premotion occurs at the very instant that the will acts. The difference from Suárez consistssimply in the fact that they distinguished this instantaneous premoving from the act of the will.

78. A typical Thomist response is that the divine predetermination is not to be included inthe set of prerequisites, since these include only what is required on the part of other second-ary causes. There is a sympathetic discussion of this response in Osborne 2006.

issues concerning divine providence, foreknowledge, predestination, and grace, butnotes that his main concern is to address the philosophical question of how God’s activ-ity as primary cause is related to the activity of free human agents (¶41, 1:821). ThoughDescartes was notoriously reticent to become entangled in theological disputes, he wasforced to confront this philosophical question. We will consider his response to it as thelast stage of our treatment of his theory of causation.

1.3. FROM SUÁREZ TO DESCARTES

Suárez inherited the traditional Aristotelian distinction among material, formal, effi-cient, and final causes. However, I have noted the view in Suárez that efficient causesbest reflect the definition of a cause as that which serves as “a per se principle fromwhich being flows into another” (see §1.2.2). Though we do not find in Descartesthis (or, indeed, any other) formal definition of cause, the focus on efficient causal-ity is reflected in his remarks on causal explanation. Thus, in the Principles ofPhilosophy he claims that in explaining natural events in terms of “God or nature,”we should consider God “as the efficient cause of all things” (PP I.28, AT 8-1:16).Admittedly, Descartes is rejecting here explanations in terms of God’s final causal-ity that he found in the scholastics, and that we have seen in Suárez (see §1.2.2 (iii);cf. §2.1.2 (ii.b)). However, even in Suárez there is a decided emphasis on God’scausal contribution as an efficient cause in his creation and conservation of the worldand in his concursus with the action of secondary causes (see §1.2.3).

In presenting Suárez as preparing the way for Descartes, I certainly do not meanto deny that they offered efficient causal explanations that differ in fundamentalrespects. After all, Descartes himself insists on the importance of the fact that hiscausal explanations of the material world do away with the sort of theoretical enti-ties found in scholastic explanations. Thus, in speaking of the schoolmen he chal-lenges a correspondent to “compare all their real qualities, their substantial forms,their elements and countless other such things with my single assumption that allbodies are composed of parts” (To Morin, 13 July 1638, AT 2:200). On the scholas-tic view in Suárez, prime matter and substantial forms are distinct res that composematerial substance, whereas accidental forms are res distinct from the material com-posite that inhere in it.79 In contrast, Descartes proposes that matter is nothing morethan divisible res extensa, and that bodily accidents are not res but rather modes ofthe parts that compose matter.80 While Descartes’s conception of a mode is drawn


79. An exception here is the case of the accidental form of quantity, which Suárez, inopposition to a more orthodox Thomistic position, takes to inhere in prime matter directlyrather than in the composite. In §3.1.2 (i), I indicate that the Suárezian account of quantity isin important respects closer than the Thomistic account to Descartes’s view of matter.

80. I am assuming here that Descartes takes the parts that serve as the subjects of themodes to be substantial. Cf. the alternative view, cited in chapter 2, note 9, that he is commit-ted to the conclusion that the only material substance is the whole of res extensa, and that theparts of this substance are modes rather than substances.

from Suárez (see §1.2.1), his view that all bodily accidents are merely modal fea-tures of res extensa most assuredly is not.81

The differences here make a difference with respect to the particular accounts ofefficient causality in the material world that Suárez and Descartes offer. ThoughSuárez posited the substantial form as a formal cause of a material substance (see§1.2.2 (i)), he also held in the section of the Metaphysical Disputation on efficientcausality that such a form is required as an efficient cause of certain changes in nature.The causal role of the substantial form is particularly important in the case of sub-stantival generation. Suárez shared with Thomas the view that such generationinvolves the eduction of a substantial form that is contained in the potentiality of mat-ter (see §1.1.2). Suárez further insisted that the efficient causality of accidental formsis insufficient to account for this eduction, since a substantial form is “more noble”than an accidental form, and since the “principal cause”82 of an effect “must be eithermore noble than, or at least no less noble than, the effect” (MD XVIII.2, ¶2, 1:599).83

We will discover that Descartes accepts a version of the axiom from Suárez that acause must be at least as noble as the effect (see §2.1). Given his parsimonious ontol-ogy, however, Descartes could not accept the argument in Suárez that such an axiomrequires the postulation of substantial forms as efficient causes of substantival gener-ation. Indeed, Descartes rejects substantial forms on the basis of the fact that there canbe no natural generation of a substantial res. As he put the point in correspondencewith Regius, “[I]t is inconceivable that a substance should come into existence with-out being created de novo by God” (Jan. 1642, AT 3:505). Of course, Suárez wouldinsist that a secondary cause cannot produce a new substance without the help ofdivine concursus. Moreover, he could protest that the eduction of a substantial formdoes not amount to the creation of a substance insofar as a substance naturally sub-sists on its own, whereas a substantial form naturally composes a substance. Even so,Suárez’s metaphysical scheme requires that substantial forms are res distinct frommatter, and thus that in producing such a form, the secondary cause produces a beingthat can, at least miraculously, subsist on its own apart from matter (see §1.2.1). ForDescartes, this result is unacceptable, since any being that can subsist on its own, evenif only by God’s absolute power, is itself a substance.84 The dispute here is not simply


81. For a further consideration of Descartes’s various arguments against substantial formsand real qualities, see Rozemond 1998, ch. 4.

82. As opposed to an instrumental cause; see §1.2.3 (ii).83. Suárez also appealed to the efficient causality of the substantial form in explaining the

production of accidents that immediately derive from that form by means of a “natural ema-nation.” Thus, the substantial form of water is the efficient cause of the accident of coldnessthat naturally emanates from it. It is due to such an emanation that heated water will, whenremoved from the source of heat, reduce itself to its natural state of being cold (see MDXVIII.3, ¶4, 1:616). For more on the scholastic conception of substantial form, see the dis-cussion in Pasnau 2004. Pasnau documents the increasing emphasis in later scholastic thoughton the efficient cause role of substantial forms.

84. In his argument for mind–body distinctness in the Sixth Meditation, Descartes empha-sizes that “the question of what kind of power is required” to produce the separate existence

over the use of the term ‘substance’. Rather, the real question is whether somethingthat can be educed from the potentiality of matter is in fact a res distinct from matter.And when Descartes says in his letter to Regius that forms that merely “emerge fromthe potentiality of matter . . . should not be regarded as substances” (AT 3:505), hecan be seen as making the defensible point that something that is a res distinct frommatter cannot be contained in the potentiality of matter. For Descartes, what can beeduced from matter as res extensa is only local motion and, consequently upon that,different sizes and shapes.85 Since res extensa is itself a substance, it is something thatonly God can create.

In rejecting any res in matter distinct from divinely created res extensa,Descartes rejects as well the accidental forms that Suárez took to be res distinctfrom composite material substance that serve as efficient causes of natural acci-dental change. However, Suárez had a complex theory of the efficient causality ofaccidental forms that raises additional questions regarding Descartes’s conceptionof causation. Suárez’s theory starts from Aristotle’s list of predicamental acci-dents, which, as we saw in §1.2.1, distinguishes quantity, quality, relation, action,passion, time, place, position, and having.86 Of these categories, Suárez held thatonly qualities, and neither quantity nor relation nor the six minor accidents, can beprinciples of action. Among the qualities, principles of action include active (asopposed to merely passive) potentiae, habits and dispositions that yield specificactions (as opposed to general states), and sensible qualities. Among the sensiblequalities, some such as colors can produce “intentional species” of themselves butnot qualities similar to themselves, whereas others such as heat and light can pro-duce both intentional species of themselves and qualities similar to themselves.Suárez in fact explicitly denied that either shapes (in the category of quality)87 orlocal motions (as well as alteration in quality, augmentation in quantity, and sub-stantial generation) can serve as per se principles of action (see MD XVIII.4,1:624–27).

In Suárez’s view, then, Descartes’s claim in the Principles that his considerationof the material world “involves absolutely nothing apart from these divisions [inquantity], shapes and motions” (PP II.64, AT 8-1:79) requires the denial that any-thing in matter can serve as a principle of efficient causality. He therefore would takeDescartes’s radical alternative to the scholastic ontology of the material world to lead


of two objects does not affect the claim that they are really distinct (AT 7:78). For a discus-sion of the relation of this view to that of the scholastics, see Rozemond 1998, 130–33.

85. Descartes claims in the Principles that “any variation in matter or diversity of its manyforms depends on motion” (PP II.23, AT 8-1:52–53). For him, the “forms” intrinsic to theparts of matter can involve only modes of extension such as size and shape.

86. In his Disputations, Suárez devotes the following disputations to each of the cate-gories: XL–XLI to quantity, XLII–XLVI to quality, XLVII to relation, XLVIII to action, XLIXto passion, L to time, LI to place, LII to position, and LIII to habit.

87. On Suárez’s view that shape is a mode of quantity rather than a res distinct from it, see§1.2.1. As indicated in that section, this Suárezian view is reflected in Descartes’s characteri-zation of the scholastic position.

back to some form of occasionalism, at least with respect to the explanation ofpurely material change.

Whether Descartes would accept this implication of his ontology of the materialworld is a question we will address in due course. Even if Descartes were commit-ted to some form of occasionalism in the case of body–body interactions, however,it would be a mistake to see medieval Islamic occasionalism, rather than scholasticanti-occasionalism, as providing the proper context for a consideration of theaccount of causation in his physics. For one thing, Islamic occasionalism simply wasnot a live option during Descartes’s time in the way in which scholastic anti-occasionalist accounts of bodily causation were. Moreover, Descartes’s rejection ofthe scholastic ontology of the material world did not prevent him from adopting cer-tain general features of the account of causation that we find in Suárez. I havealready mentioned his endorsement of a version of the axiom in Suárez that a causemust be at least as noble as the effect. Given this endorsement, Descartes could nothave been sympathetic to the view in Islamic occasionalism, which Hume lateraccepts, that causal correlations can hold between any two distinct events.88 But aswill become evident in what follows, it is also the case that Descartes’s view ofGod’s causal activity draws on claims in Suárez concerning the relation betweendivine creation and conservation. I will be concerned to argue that this connection toSuárez provides a reason to reject the view of those who take Descartes’s theory ofcausation to include a form of temporal atomism that is similar to that of the Islamicoccasionalists (see §2.2). This connection to Suárez is significant for Descartes’stheory of causation given the fact, which I emphasize in chapter 3, that his accountof divine conservation is a central element of the metaphysical foundations that heprovides for his anti-scholastic physics.

The importance of the anti-occasionalist scholastic context is not restricted toDescartes’s account of causation in physics. In addition to the general metaphysicalprinciples in the work of the scholastics that I have emphasized in this chapter, thereare further specific claims concerning causation in Suárez and other scholastics thatwe must consider if we are to understand what Descartes has to say about forms ofcausation other than body–body interaction. In what follows, I note in particular therelevance of such claims for Descartes’s account of the action of body on mind (see§4.2.1) and of the action of mind on body (§4.3.1). The scholastic context will allowus to appreciate certain problems in Descartes for mind–body interaction that gobeyond the problem of the interaction of objects with differing natures that has tendedto dominate recent discussions of his theory of causation. Moreover, it will becomeclear in the final chapter that this context is essential for an adequate understandingof the sort of causation that Descartes takes to be involved in the free acts of our will.


88. For this view in Ghaza-lı-’s Incoherence, see §1.1.1. For a discussion of the relation ofGhaza-lı-’s position to Hume’s account of causation, see Nadler 1996. My view that Descartesdiffered from the Humean line on this point has been disputed in Della Rocca (forthcoming).According to Della Rocca, Descartes does not take the causal axiom he inherited fromthe scholastics to show that causes explain their effects. I defend my different reading ofDescartes’s axiom in §2.1.3.

Admittedly, as in the case of physics, so Descartes’s accounts of mind–body interac-tion and free human action presuppose a basic ontological framework that differs,sometimes radically, from a traditional scholastic framework. However, these unde-niably important differences should not blind us to the extent to which the problemsconcerning causation that Descartes confronts, and even aspects of his responses tothose problems, were bequeathed to him by his scholastic predecessors.


49

2

Two Causal Axioms

In contrast to Suárez, Descartes did not bequeath to posterity an extended treatise onthe nature of causality. Nevertheless, his remarks on causation in the Third Meditationprovide a natural starting point for a consideration of his theory of causation. For inthis text, Descartes emphasizes two conclusions regarding causation that he took tobe evident. The first, which is central to the main proof in the Third Meditation ofthe existence of God, is that “there must be as much in the efficient and total causeas in the effect of that cause” (AT 7:40). This is alternatively expressed by the claimthat the effect “cannot begin to exist unless it is produced by something in whichthere is formally or eminently all that is found” in the effect (AT 7:41). ElsewhereDescartes labels this as the “axiom or common notion” that “whatever there is ofreality or perfection in some thing, is formally or eminently in its first and adequatecause” (AT 7:165). Drawing on this label, as well as the claim in this passage thatthe reality or perfection is contained in the cause, I call this constraint on causationthe “containment axiom.”1 In addition to this axiom, there is Descartes’s argumenttoward the end of the Third Meditation that since “conservation differs solely in reasonfrom creation,” there must be “some cause that as it were creates me at this moment,that is, conserves me” (AT 7:49). Descartes also expresses this claim as the axiom that

1. In contrast to the English-language secondary literature on this topic (see note 7),Descartes typically speaks of causal axioms or notions rather than of causal principles. But hedoes indicate in correspondence that the term ‘principle’ can be used for “a common notionthat is so clear and so general that it can serve as a principle for proving the existence of allthe beings, or entities, to be discovered later,” as well as for “a Being, the existence of whichis better known to us than any other, so that it can serve as a principle for knowing them”(To Clerselier, June/July 1646, AT 4:444).

“no less a cause is required to conserve a thing than to produce it at first” (AT 7:165).I call this additional constraint on causation the “conservation axiom.”

Both of these axioms have a clear scholastic precedent in Suárez. Indeed, there isa suspicion among some commentators that the containment axiom, in particular, ismerely a scholastic holdover that has no real justification in Descartes’s system. Forinstance, Jonathan Bennett has concluded that “after decades of intermittently brood-ing” over this axiom, the axiom itself is “without value” and “seems not to reflect anydeeply considered views about the nature of causation” (Bennett 2001, 1:89).2 As wewill discover, however, other commentators have insisted that this axiom is significantfor Descartes insofar as it precludes the causal interaction of objects with natures thathe takes to be heterogeneous, most notably the interaction of mind as res cogitans andbody as res extensa.3

There is less disagreement in the literature over the value of Descartes’s conserva-tion axiom. However, there is an interpretation of this axiom that distances it from itsscholastic counterpart. Here again Bennett illustrates the point, claiming that the con-servation axiom leads Descartes to the position that “the continual preservation ofthings through time . . . is really the continual creation of successors to them” (Bennett2001, 1:98). This claim of course reflects the earlier interpretation of Descartes’s viewof divine conservation, mentioned in the introduction, that Norman Smith offered in1902.4 But this interpretation is perhaps developed most completely in the later workof Martial Gueroult.5 What neither Gueroult nor Smith nor Bennett emphasizes, how-ever, is that a re-creationalist account of the conservation axiom conflicts with the viewin Suárez and other scholastics that conservation requires not distinct acts of re-creation, but merely the continuation of the very same act by which God created in thefirst place.

A different view of the metaphysics of Descartes’s two causal axioms emerges, how-ever, once we take seriously their source in scholastic thought. The scholastic contextnot only allows us to understand the import the containment axiom had for Descartes,but also reveals that this axiom does not create the sort of difficulties for mind–bodyinteraction that critics have tended to emphasize. Moreover, Suárez’s version of theconservation axiom in fact provides a basis for rejecting the claim that Descartes iden-tified the conservation of the world with its continual re-creation. I noted in §1.3 thatDescartes offers a radical alternative to the sort of scholastic ontology that underliesSuárez’s account of causality. But this departure from scholasticism turns out to be


2. Bennett calls the containment axiom the “causal resources principle.” For further dis-cussion of Bennett’s treatment of the issue of causation in philosophers from Descartes toHume, see my review of Bennett 2001 in Schmaltz 2002b.

3. See the views of Radner discussed below. See also the comments in the introductionconcerning the so-called scandal of Cartesian interaction.

4. However, Bennett himself cites in defense of his re-creationalist interpretation ofDescartes a passage from Smith 1952, 218. For more on this interpretation, see note 92.

5. Where Bennett goes beyond Gueroult is in attributing to Descartes the position that Goddoes not conserve the very same object over time, but rather creates a series of nonidenticalsuccessors. I think that Bennett is correct in holding that this is an implication of the re-creationist reading of conservation, at least on one account of identity, but I argue in §2.2.2that such an implication reveals that this reading cannot reflect Descartes’s own views.

compatible in the end with the dependence of Descartes’s understanding of the meta-physics of causation on the views of his scholastic predecessors.

In §2.1, I begin my consideration of this account by focusing on Descartes’s con-tainment axiom. My statement above that this axiom “expresses” the claim that thecause contains at least as much reality as its effect actually begs the question againstthe view in the literature that there are two distinct constraints on causation here. SoI need to start by arguing that there is in fact only one axiom. Then I consider thesignificance of the fact that Descartes restricted his containment axiom to the “effi-cient and total cause” of an effect, as well as the precise meaning of the claim in thisaxiom that the effect is contained in the cause “formally or eminently.” Throughoutit proves useful to take into account remarks in Suárez, who anticipated Descartes’sstatement of the containment axiom and the technical terminology used therein.

In §2.2, I turn to the conservation axiom as explicated in the Third Meditation.Descartes indicates there that this axiom follows from the “nature of time,” and thatit yields the result that conservation is distinct only “in reason,” and not in reality,from creation. This result seems to be drawn straight from Suárez, though I have men-tioned the claim in Bennett, anticipated in Gueroult, that for Descartes divine conser-vation consists in a series of discrete creative acts rather than, as Suárez would haveit, in a continuation of God’s original creation of the world. But though there are somedifferences in the arguments for divine conservation in Suárez and Descartes, I under-stand both to agree that God conserves creatures by means of the continuation of thesame act by which he created them ex nihilo.

Even though the Suárezian context is essential for understanding Descartes’s con-tainment and conservation axioms, I claim in §2.3 that these axioms do not take himthe full way to Suárez’s own concurrentist position. The containment axiom leavesunresolved some basic issues concerning how an effect is actually produced. The con-servation axiom goes further in revealing that divine conservation plays an essentialbackground role in causal interactions. But there remains the metaphysical question—central to scholastic discussions of causality—of the precise nature of the creaturelycontribution to causality in nature. To address Descartes’s stance on this issue, wemust shift from a consideration of his abstract causal axioms to an exploration of thedetails of his accounts of various forms of causal interaction.

2.1. THE CONTAINMENT AXIOM

The main topic of the Third Meditation is “the existence of God,” and in the courseof offering his main proof there of God’s existence, Descartes appeals to the follow-ing as “manifest by the light of nature,” which I divide into two parts:

[1] There must be at least as much in the efficient and total cause as in the effectof that cause. For I ask, where could the effect receive [assumere] its reality,unless from the cause? And how could the cause give this to it, unless it also has[this]. For thus it follows that something cannot come from nothing, nor that whatis less perfect, that is, what contains more reality in itself, from what has less. . . .That is [Hoc est], [2] in no way can some stone, for example, which was notbefore, now begin to be, unless produced by another thing in which there is alleither formally or eminently that is found in the stone; nor can heat that was not

Two Causal Axioms 51

previously in a subject be induced, unless from a thing that is of at least the sameorder of perfection as heat, and so for the rest. (AT 7:40–41)

Both (1) and (2), as well as the view that they are intimately connected, are drawnstraight from the scholastic tradition. For instance, in his Disputations Suárez pro-poses that especially in the case of efficient causality, the following principle holds,which I divide into corresponding parts:

[1'] [A]n effect cannot exceed in perfection all of its causes taken together. It isproved that nothing of perfection is in the effect that it does not have from itscause; therefore [ergo] [2'] the effect can have nothing of perfection that does notpre-exist in any of its causes, either formally or eminently, because causes cannotgive what they in no way contain. (MD XXVI.1, ¶2, 1:916*)

Suárez’s (1') requires that “all causes taken together” contain “everything of perfec-tion” in their effect on the grounds that the effect can “have” its perfection only fromthese causes. Similarly, Descartes’s (1) requires that the “efficient and total cause”of an effect contain at least as much “reality” as its effect contains on the groundsthat the effect must “receive” its reality from its cause. And just as Suárez’s (2')requires that the perfection of the effect “preexist” in all of its causes “formally oreminently,” so Descartes’s (2) requires that the total cause contain in this way “all”that is in its effect.6

One important difference derives from the indication in Descartes that his causalconstraints apply not only to the “actual or formal” reality that an effect has from itscause, but also to the “objective reality” that his idea of that effect has. In the caseof (2), in particular, the causal constraint is said to require that “the idea of heat, orof the stone, could not be in me unless it is placed there by some cause in which thereis at minimum as much of reality as I conceive to be in heat or the stone” (AT 7:41).This extension of the causal constraint to the case of objective reality is of coursecentral to the Third Meditation argument that God must exist as the cause ofthe objective reality of our idea of God. I will have more to say presently aboutDescartes’s views on objective reality in relation to the very different views on thistype of reality in Suárez. But my main concern will be to address the following ques-tions concerning the passage above from the Third Meditation. First, there is thequestion of whether the constraints introduced in (1) and (2) amount to the sameor are distinct constraints. A second question concerns the import of Descartes’srestriction of the constraint in (1) to the “efficient and total cause.” Finally, there isthe question of what precisely Descartes meant by the claim in (2), anticipated inSuárez, that a cause must contain its effect “formally or eminently.”

2.1.1. How Many Causal Constraints?

Suárez links his two causal constraints by the term ergo, thus indicating that the factthat all perfections of an effect are contained formally or eminently in the total set


6. Whereas Descartes followed Suárez in holding that a cause need contain at least as muchreality or perfection as its effect, the version of the containment principle in the work of Proclusand other Neoplatonists requires the stronger condition that the cause contain more reality or per-fection than its effect. For a discussion of the Neoplatonic version of the principle, see Lloyd 1976.

of its causes (2′) follows from the fact that such causes together contain at least asmuch perfection as is present in this effect (1′). Indeed, the suggestion in Suárez isthat the two constraints come to the same thing. For a cause to contain at least asmuch perfection as its effect just is for it to contain formally or eminently everythingin its effect. Suárez took formal and eminent containment to exhaust the waysin which perfection can be contained. This same view seems to be reflected inDescartes’s remarks in the Third Meditation. For by introducing (2) by the term Hocest, he suggested that this constraint comes to the same as (1).

Nevertheless, there is the view in the literature that Descartes’s (1) and (2) are dis-tinct constraints insofar as (1) requires much less of the cause than does (2). Forinstance, Daisie Radner argues that whereas (1) explicates a relatively weak “realityprinciple,” which requires the containment in the cause of only at least as much real-ity as is found in the effect, (2) introduces a stronger “containment principle,” whichrequires further the containment in the cause formally or eminently of the specificfeatures of the effect (Radner 1985a, 41).7

When pressed to explain the sort of “reality” that he had in mind in asserting (1),Descartes explains that “substance is a greater thing than mode,” and that “if there isan infinite and independent substance, it is a greater thing than finite and dependent[substance]” (AT 7:185).8 What is suggested here is the following simple ontologicalhierarchy:

God infinite substance

minds finite substancesbodies9

thoughts modes of finite substances10

shapes/sizes/motions


7. I take the labels from the discussion of Radner’s position in O’Neill 1987, 231–32.Radner calls the reality principle the “at least as much principle,” and the containment princi-ple the “pre-existence principle.” O’Neill is inclined to Radner’s view that Descartes offeredtwo distinct causal constraints; see O’Neill 1987, 232. Radner also takes Descartes to offer adistinct “communication” principle on which the cause literally transfers to the effect what itcontains in itself. In §3.2.1 (iii), I consider the claim in Broughton 1986 that Descartes wasled by his views on causation to accept such a principle in the case of body–body interaction.

8. Descartes also includes in this hierarchy “real accidents, or incomplete substances” that“are greater things than modes, but less than complete substances” (AT 7:185). But hefamously rejects the existence of scholastic bodily accidents that can (at least miraculously)subsist apart from corporeal substance.

9. There is some dispute in the literature over whether Descartes allowed that particular bodiesare substances at all. On a view that Martial Gueroult defends, there is only one material substance,with particular bodies serving as modes; see Gueroult 1953, 1:107–18, and Gueroult 1968, 540–55.Cf. the recent development of this interpretation in Lennon 2007. However, Descartes himselfspeaks of the parts of corporeal substance as distinct substances (e.g., in PP I.60, AT 8-1:28–29), and he distinguishes between parts of a body and its modes (Sixth Replies, AT 7:433–34), thus sug-gesting that particular bodies are substantial parts of matter rather than modes of it. For an appealto these considerations in response to Gueroult’s interpretation, see Hoffman 1986, 347–49.

10. In a letter to Arnauld, Descartes emphasizes that one must distinguish betweenthought or extension insofar as it constitutes the nature of a substance and the variable

In terms of this hierarchy, the claim that a cause must have as much reality as itseffect requires only that the cause be on at least the same level of the ontologicalhierarchy as its effect. This seems to fall short of the requirement that the causecontain everything in the effect formally or eminently. According to Radner, therequirement here is that the cause possess not merely the general type of reality inthe effect, but the specific nature of the effect itself.11

In response to Radner, however, Louis Loeb denies the distinction between thetwo principles Radner claims to find in Descartes on the grounds that what is saidto be contained formally or eminently in the cause is simply the perfection or real-ity that the reality principle concerns. In Loeb’s view, to say that the cause mustcontain formally or eminently everything in the effect is just to say that the causemust contain something on either the same ontological level as its effect (in the caseof formal containment) or a higher ontological level than its effect (in the case ofeminent containment). Thus, the containment principle requires not that the cause“contain modes of the same kind” as it produces in the effect but merely that it con-tain the reality of the effect “qua degree of perfection” (Loeb 1985, 228).According to Loeb, then, the containment principle requires that in the case of theproduction of a mode, say, bodily motion, the cause that formally contains thiseffect possess not motion itself, but only something on the same ontological levelas this mode.

Loeb’s claim that the two causal principles are not ultimately distinct may seemto be supported by the fact that when attempting to formalize his system in theSecond Replies, Descartes offers only the one causal axiom, and explicates thataxiom in terms of his simple ontological hierarchy. The causal axiom, which I citedat the outset, is that the “first and adequate” cause contains formally or eminently“whatever there is of reality or perfection in the effect.”12 But this axiom is followedby a further axiom that explains the notion of reality or perfection by appealing tothe fact that “substance has more reality than accidents or modes, and infinite sub-stance, than finite” (AT 7:165). So the suggestion here is that the reality that thecause formally or eminently contains is simply the reality of the effect as infinitesubstance, finite substance or mode.


modes of that attribute, such as particular acts of thinking or particular shapes, sizes, ormotions (29 July 1648, AT 5:221). On his official view, the thought or extension that con-stitutes the nature of a substance is an invariable attribute that is only “distinct by reason”from that substance (see PP I.62, AT 8-1:30). This kind of attribute thus belongs on thesame level of reality as the substances to which they are attributed. In §2.2.2, I note the viewin Descartes that there is only a distinction of reason between a substance and its invariableattribute of duration.

11. As Radner puts the point, the further constraint on causation requires that in the casewhere the effect is a mode, the cause “communicates” this something that “pre-exists” in itselfand that what gets communicated is not merely “just modality or modeness” but rather a par-ticular kind of mode (Radner 1985a, 41).

12. I address presently the restriction of the axiom to the “adequate” or, what is the samefor Descartes, “total” cause.

Loeb’s deflationary version of the containment axiom suffices for the purposes ofthe main argument for the existence of God in the Third Meditation. As I have indi-cated, the central premise of this argument is that the cause must contain formally oreminently the reality that is present objectively in our idea of infinitely perfect sub-stance.13 But to contain something at the same level of reality as infinitely perfectsubstance just is to contain formally infinite perfection itself.

Nonetheless, a more robust sort of formal containment seems to be required forthe proof of the existence of the material world in the Sixth Meditation. After rul-ing out the possibility that his mind has an “active faculty” ( facultas activa) thatproduces the objective reality of his sensory ideas, Descartes notes that there mustbe “another substance distinct from me, in which all the reality must inhere[inesse] either formally or eminently, which is objectively in the ideas produced bythis faculty.” Either the substance is body, in which case the reality inheres for-mally, or it is God or “some creature more noble than [nobilier] body,” in whichcase the reality inheres eminently (AT 7:79). In terms of the simple ontologicalhierarchy, the claim that one created substance is more noble than another wouldseem to amount to the claim that the former is on a higher level in the hierarchythan the latter. But this claim is problematic given the implication of the simpleontological hierarchy that all substances other than God are on the same ontolog-ical level. As I indicate in my discussion below of Descartes’s view of eminentcontainment, this consideration reveals the need for a revised version of his onto-logical hierarchy. However, the relevant point here is that the mere containment ofsomething with the same amount of reality does not suffice for formal containmentin the Sixth Meditation proof. For other finite minds do contain something with thesame amount of reality as the bodily modes present objectively in our sensoryideas, namely, its own modes. But the proof makes clear that finite substancesmore noble than bodies contain the objective reality of the sensory features of bod-ies eminently rather than formally (AT 7:79). More needs to be said about theexact nature of the formal containment that Descartes has in mind here; we willreturn to this point presently. Yet even an initial consideration of the SixthMeditation proof of the material world indicates that formal containment requiresnot merely that what is contained be on the same ontological level as the effect,but also that it have the same nature as the effect. So at a minimum, that which for-mally contains the objective reality of our sensory ideas of bodies must have thesame nature as body.

My proposal is that Descartes offers a single causal axiom that requires that thecause contain the reality of the effect formally or eminently. Any apparent distinc-tion of causal constraints derives from the fact that he sometimes needed to considerthe reality or perfection of the effect only abstractly in terms of his simple ontolog-ical hierarchy, as in the case of the Third Meditation proof of the existence of God,


13. I have more to say in §2.1.3 (i) about Descartes’s account of objective reality and ofits distinction from formal reality.

whereas at other times he needed to consider the reality or perfection as reflected inthe particular nature of the effect, as in the case of the Sixth Meditation proof of theexistence of the material world. In the end, there seems to be no difference betweenDescartes and Suárez on the relation between the two causal constraints. For both,the requirement that the (total or adequate efficient) cause contain at least as muchperfection as its effect is to be understood in terms of the requirement that the (totalor adequate efficient) cause contain everything it produces in the effect formally oreminently.

To this point I have spoken only in general terms about the requirement in thecontainment axiom that the cause contain “formally or eminently” what is found inthe effect. We will discover that the notions of formal and eminent containment arenot entirely straightforward for Descartes. Before puzzling over the complications,however, we need to consider briefly the import of Descartes’s claim to Mersennethat when he said in the Third Meditation that there is nothing in the effect “not con-tained formally or eminently in its EFFICIENT and TOTAL cause,” “I added thesetwo words on purpose” (AT 3:274). At least initially, total causes are most usefullycontrasted with partial causes, and efficient causes with formal and final causes. Thescholastic context, particularly as provided in Suárez’s work, turns out to be crucialfor Descartes’s own understanding of these contrasts.

2.1.2. “EFFICIENT and TOTAL Cause”

(i) Total/Adequate versus Partial Causes

The “theologians and philosophers” gathered by Mersenne who wrote the SecondObjections argue that since living things are produced by the sun, rain, and earth,which lack life and therefore are “less noble” than what they produce, it is the case,contrary to what Descartes claimed in the Third Meditation, that “an effect mayderive from its cause some reality that is nevertheless not present in the cause”(AT 7:123). Descartes initially responds by insisting that life is a perfection that canbe explained in terms of the operations of inanimate bodies. Here he appeals to hisargument that it is only reason, particularly as manifested in language use, that can-not be so explained.14 Yet in his Second Replies, as well as in a related letter toMersenne, Descartes also allows for the possibility that living organisms includeperfections not present in the sun, rain, and earth, but concludes that if this is so, thenit shows only that these elements are not the total or adequate causes of what theygenerate.15

Descartes nowhere provided an analysis of total or adequate causes, or indicatedthe sense in which objects such as the sun, rain, and earth could be causes withoutbeing total or adequate causes. Yet at one point in the Third Meditation he does referto the possibility that several “partial causes” (causes partiales) contribute to his cre-


14. This is the argument in DM V, AT 6:55–59.15. Cf. AT 7:134, in which Descartes denies that they are adequate causes, and AT 3:274,

in which he denies that they are total causes.

ation (AT 7:50). Moreover, there is in Suárez an analysis of the distinction betweentotal and partial causes. In the Disputations, he defines the total cause as that “whichprovides the whole concursus necessary for the effect in its order,” and the partialcause as that “which per se alone does not contribute a sufficient and wholly neces-sary concursus” (MD XXVI.3, ¶1, 1:925–96*).16 In failing to contribute a sufficientconcursus, partial causes may seem to be similar to instrumental causes that, in hisview, must be subordinated to and act with other causes of the same order to produceeffects more noble than themselves (see §1.2.3 (ii)). Suárez cautions that thoughpartial causes also must act with other causes of their same order to produce theireffects, they are not subordinated to those other causes, and so are principal ratherthan instrumental causes (MD XVII.2, ¶18, 1:591). However, his claim that partialcauses require assistance from other causes of the same order allows him to hold thatsecondary causes can be total causes of their effects even though they can producethese effects only with the help of the concursus of the primary cause.

In terms of this analysis, Descartes could say that the sun, rain, and earth are nottotal or adequate efficient causes of living organisms because they do not provideeverything needed in the order of secondary efficient causes to produce their effect.The concursus of other organisms or, in the case of the original production of theorganism, of other kinds of bodies are required for this production. Since they areonly partial causes, the sun, rain, and earth need not contain formally or eminentlyeverything present in the organisms they produce.17 But given his containmentaxiom, Descartes must hold that the total efficient cause of the organisms, consist-ing of these partial causes together with the other organisms or bodies that contributeto their production, must so contain the effect. And on this point Descartes agreeswith Suárez, who asserts as certain that “the effect cannot exceed in perfection all ofits causes taken together.” For Suárez, as for Descartes, such a certainty reveals that“the effect can have nothing of perfection that does not pre-exist in some of itscauses, either formally or eminently” (MD XXVI.1, ¶2, 1:916*).

There is, however, one interesting complication for the view that Descartes canaccept the conclusion in Suárez that creatures as well as God can be the total causeof an effect. This complication derives from the so-called Conversation withBurman, a record of a 1648 interview that Descartes had in his country retreat inEgmont with the Dutch theological student Frans Burman. One portion of this con-versation concerned Descartes’s claim in the Third Meditation that given the fact thatGod has created him “there is a strong reason to believe that I have been made in


16. Suárez further distinguishes partial causes, which are principal causes that bring abouteffects with other causes of the same kind and order, from instrumental causes, which are notprincipal causes, since they bring about effects with other secondary causes of a higher orderto which they are subordinated; see MD XVII.2, ¶¶16–19, 1:590–91. For more on his view ofinstrumental causes, see §1.2.3 (ii). Here I focus on his account of principal causes.

17. Elsewhere, Descartes refers to the sun as a “universal” cause of its effects that requiresthe contribution of other “particular” causes; see To Elisabeth, 6 Oct. 1645, AT 4:314. In§2.2.1, I discuss Descartes’s view in this letter that the action of the sun as a universal causemust be distinguished from God’s action as “universal and total cause” of all effects.

some way in his image and likeness [imaginem et similtudinem], and that I perceivethat likeness, which includes the idea of God, by the same faculty that enables me toperceive myself” (AT 7:51).18 Burman objects to this claim that “surely God couldcreate you, and yet not create you in his own image.” Descartes is reported torespond, after citing the principle that “the effect is similar to the cause,” that since“God is my cause, I am His effect,” it follows directly that “I am similar to him.”When Burman rejoins that the builder who produces a house is not similar to it,Descartes notes that the fact that the builder “only applies activity to the passive”shows that “the work as a work is not itself similar.” He then claims that in the con-trasting case of “the total cause and [the cause] of being itself,” which “producessomething else ex nihilo (which is the mode of production that pertains to Godalone),” the effect must be similar to the cause. Thus since the total cause of beingis itself “being and substance,” it follows that what it produces “must at a minimumbe being and substance, and so in any case be similar to God and bear His image[imaginem]” (AT 5:156).

If ‘the total cause and the cause of being itself’ means “the total cause, that is, thecause of being itself,” then only God could be a total cause given the remark toBurman that the mode of producing being itself ex nihilo belongs to God alone.19 Onthis reading, the containment axiom could apply only to God. However, one couldread ‘the total cause and the cause of being itself’ as referring to something that isthe total cause and in addition the cause of being itself. The response to Burman maybe that a thoroughgoing similarity of effect to cause can be derived only in the caseof a total cause of that effect that is also the cause of the being of that effect.20 In thecases of total causes that bring about their effects by applying their activity to pas-sivity, one cannot argue to a similarity in being, since such causes do not produce thebeing of the patient, but merely alter a patient that already has its own being. To besure, Descartes must take the alteration to be contained in its total cause formally oreminently. Yet one cannot assume that the being of what is altered must be similarto the being of what alters it. The builder must (eminently) contain the plan of thehouse he will build, but the passive materials to which he applies his activity neednot be similar to himself. We will return in §2.2.1 to the question of whether God’stotal causality of the world precludes any other sort of causal input. But at least theargument in the Burman report that the similarity between cause and effect isrequired only in the case of the cause of being itself does not require the restrictionof total causality to God alone.21


18. Also at issue is Descartes’s claim in the Fifth Replies that “divine [creation] is closerto natural production than to artificial [production]” (AT 7:373).

19. Cf. Descartes’s remarks at AT 7:111. But see also the discussion in §1.2.3 (i) of reser-vations in Suárez of demonstrating on the basis of natural reason alone the conclusion thatGod alone can create ex nihilo.

20. I overlooked the possibility of this alternative reading in my analysis of the Burmanpassage in Schmaltz 2000.

21. There is a similar reading of the Burman passage in Pessin 2003, 43. Cf. the discus-sion of this passage in §2.2.1.

(ii) Efficient versus Formal and Final Causes

Previously we have considered the view in Suárez that since form and matter areintrinsic causes, they differ in kind from two kinds of extrinsic causes, namely, “finalcauses” that cause “by means of a metaphorical motion” insofar as they merelyincline other causes, and efficient causes that are the true source of effects through anaction (see §1.2.2). Though Suárez emphasized that efficient causality is the primarycase of causation, he was also willing to appeal to material, formal, and final causal-ity in his explanations of natural change. Descartes’s restriction of his containmentaxiom to efficient causality indicates this unwillingness to extend the notion ofcausality in a similar sort of way. Even so, he allows at times for something akin toformal causes, and he admits not only a rational teleology in the case of the actionsof created minds, but also a kind of natural teleology in the case of the soul–bodyunion.22 What we need to understand is how Descartes’s concessions are compatiblewith his emphasis on the exclusivity of efficient causality. Let us consider intrinsicformal causality first, then extrinsic final causality.

(ii.a) Descartes admits a kind of formal causality analogous to though distinct fromefficient causality in the course of commenting on his suggestion in the ThirdMeditation that God derives his existence from himself. The Dutch critic Johan deKater, or Caterus, protested in the First Objections that God can derive his existencefrom himself only in a negative sense, or not from another, and not in a positivesense, or from a cause (AT 7:95). In response, Descartes insists that it is legitimateto assume that everything requires a cause of its existence, and to inquire into itsefficient cause. He adds that even though the fact that God has “great and inex-haustible power” reveals that he does not require an external cause for his existence,still since “it is he himself who conserves himself, it does not seem too improperfor him to be called sui causa” (AT 7:109). Since God can be called a sui causa,“we are permitted to think that he stands in the same relation to himself as an effi-cient cause does to its effect, and hence to be from himself positively” (AT 7:111).

Dissatisfied with this explanation, Arnauld notes in the Fourth Objections that weare to understand the source of God’s existence not in terms of an efficient cause, butin terms of the fact that since his existence is identical to his essence, God requires noefficient cause. Arnauld adds that since nothing can stand in the same relation to itselfas an efficient cause does to its effect, God cannot stand in this relation to himself(AT 7:213–14).23

Though Descartes protests that Arnauld’s complaint “seems to me to be the leastof all his objections” (AT 7:235), he nonetheless responds to it at some length. Hebegins by insisting that he never said that God is an efficient cause of his own exis-tence, but only that he in a sense stands in the same relation to his existence as an


22. I take the terms ‘rational teleology’ and ‘natural teleology’ from Simmons 2001.23. Arnauld’s objection is relevant also to Descartes’s axiom in the Second Replies that

“no thing exists of which it cannot be asked what is the cause why it exists” (AT 7:164). I dis-cuss this axiom in §5.1.2.

efficient cause does to its effect.24 To explain more precisely the sense in which Godis the cause of his existence, Descartes appeals to the claim in Aristotle that theessence of a thing can be considered as a “formal cause” of certain features of thatthing (AT 7:242). He concedes to Arnauld that the fact that God’s existence is iden-tical to his essence reveals that it does not require an efficient cause, but he notes thatGod’s essence provides a formal cause of his existence that “has a great analogy tothe efficient [cause], and thus can be called an efficient cause as it were [quasi causaefficiens]” (AT 7:243).25

Even though he emphasizes the analogy to efficient causality, Descartes also sug-gests that there must be some room in his ontology for a species of causation distinctfrom efficient causation. After all, Descartes tells Arnauld that there is between anefficient cause and no cause “the positive essence of a thing” (AT 7:239). To be sure,he continues by allowing that the concept of an efficient cause “can be extended to”the concept of a formal cause, in the same way that the concept of a rectilinear poly-gon can be extended to the concept of a circle (AT 7:239). But just as a rectilinearpolygon remains something distinct in nature from a circle, so an efficient causeseems to remain something distinct in nature from a formal cause.26

In the exchange with Arnauld, the discussion of formal causality is limited for themost part to the special case of God’s existence. However, I have noted Descartes’sappeal to an understanding of formal causality in Aristotle that is not restricted in thismanner. Descartes cites in particular Aristotle’s claim in Posterior Analytics that thedefining form of a right angle is the cause of the fact that an angle in a semicircle is aright angle (II.11, 94a25–35, Aristotle 1984, 1:155). Given this citation, Descartescould extend the notion of formal causality to cover any case in which a feature of anobject derives from that object’s nature or essence. Though he himself does not speakin these terms, he could say that the extension that constitutes the essence of a body isthe formal cause of that body’s capacity to have certain kind of modes, in particular,modes of extension. Of course, this appeal could not explain why the body has certainmodes rather than others. In contrast to the case of God’s existence, such an explana-tion would need to invoke the efficient causes of the bodily modes. But also in contrastto the case of God’s existence, an explanation of these modes in terms of their efficientcauses seems to be perfectly compatible with an explanation of the ability of body topossess such modes in terms of the formal cause of the modes.


24. Descartes is not entirely innocent, though, since he does deny in the First Replies thathe said that it is impossible for something to be the efficient cause of itself, and he suggeststhat efficient causes need not be either prior to or distinct from their effects (AT 7:108). It isunderstandable that Arnauld takes this text (to which he had access when composing hisFourth Objections) to indicate that Descartes wanted to apply the notion of efficient causalityto the derivation of God’s existence from himself.

25. For a helpful discussion of Descartes’s exchange with Arnauld on this point, seeCarraud 2002, 266–88. Carraud draws on the discussion of Descartes’s conception of God ascausa sui in Marion 1996, 143–82.

26. Thus, there seems to me to be some reason to qualify Carraud’s conclusion that forDescartes “the expression ‘cause efficiente’ is henceforth redundant” (Carraud 2002, 179).Carraud cites the similar conclusion in Marion 1991, 286–87.

Thus, there may be no reason for Descartes to dispute the consequence in Suárez thatthe fact that an effect has a total efficient cause does not preclude the fact that it also hasa formal cause. Moreover, Descartes’s claim in the Fourth Replies that we conceive offormal causality in the case of God “by analogy with the notion of efficient causation”(AT 7:241) recalls the view in Suárez that formal causes can be called causes only byanalogy to efficient causes (see §1.2.2 (i)). Nevertheless, it is clear that the account offormal causality that I derive from the remarks in the Fourth Replies differs fundamen-tally from the account of such causality in Suárez. Descartes’s official doctrine in thePrinciples is that there is only a distinctio rationis, and not any distinction in reality,between the “principal attribute” of extension and the corporeal substance whose natureit constitutes (see PP I.62, AT 8-1:30).27 According to Descartes, then, anything thatinheres in matter can be only a mode of extension.28 Here he is of course concerned toreject the substantial and accidental forms that schoolmen such as Suárez took to be thesource of formal causality in the case of material substances.29 But Descartes iscommitted to rejecting as well the view in Suárez that formal causality involves an“intrinsic and formal union” of a form that is distinct in re from that with which it unites.Descartes therefore could not take formal causality to enter into an account of thecomposition of corporeal substance; at most, he could appeal to this kind of causalitymerely to anchor bodily modes in the extension that constitutes the essence of body.

But though a Suárezian account of the causal role of the forms of materialcomposites cannot provide a model for Descartes’s conception of formal causality,such a model is provided by something in Suárez that we have not yet considered,namely, the “metaphysical form” that he identified with “the form of the whole, noth-ing other than the whole essence of the substantial thing” (MD XV.11, ¶3, 1:558). Forif anything is a formal cause in a body, according to Descartes, it is the extension thatconstitutes the whole nature of that body. Yet Suárez himself denied that metaphysicalforms are formal causes in the case of material objects insofar as they already includeboth the matter and form of such objects and thus do not issue in “actualizing someother subject” (MD XV.11, ¶7, 1:559). Given this scholastic context, it is understand-able that Descartes felt no need to leave room in his physics for a kind of formalcausality that differs from the efficient causality governed by the containment axiom.

(ii.b) Descartes is famous for his rejection of appeals to God’s final causality. In theFourth Meditation, he argues for such a rejection by claiming that

since I now know that my own nature is weak and limited, whereas the nature ofGod is immense, incomprehensible and infinite, I also know without more adothat he is capable of countless things whose causes are beyond my knowledge.


27. Here Descartes is drawing on the theory of distinctions in Suárez. For a discussion ofthis theory, see §1.2.1.

28. In §3.2.2, I consider whether this implication of the doctrine is consistent withDescartes’s claim that bodies possess “forces” to persist in or to resist motion.

29. In a 1638 letter, for instance, Descartes asks his correspondent to “compare the sup-positions of others with mine that is to say all of their real qualities, their substantial forms,their elements and similar things, the number of which is nearly infinite, with this alone, thatall bodies are composed of some parts . . .” (To Morin, 13 July 1638, AT 2:200).

And for this reason alone I consider the customary search for final causes tobe totally useless in physics; there is considerable rashness in thinking myselfcapable of investigating the purposes of God. (AT 7:55)

This argument seems to allow for the possibility that God in fact has purposes, andindeed in the Fifth Replies Descartes granted his critic Gassendi that one mayconjecture about God’s purposes “in ethics” (AT 7:375). But in other places he wasconcerned to deny that God has at least a certain sort of purpose. Thus, in connectionwith his doctrine of the creation of the eternal truths, Descartes insists that God is com-pletely indifferent with respect to the question of what to create, to such an extent that

no good, or truth, no believing, or acting, or omitting can be feigned, the idea ofwhich was in the divine intellect before his will determines itself to produce suchan effect. And I do not speak here of temporal priority, but whatever is of order, ornature, or ratione ratiocinate, as they call it, such that this idea of good impelledGod to choose one rather than another. (AT 7: 432)

Given this view of divine indifference, it cannot be said that God had any purposethat led him to create as he did.30

It may be possible to reconcile this consequence with the suggestion in Descartesthat God can have hidden purposes by distinguishing between antecedent and con-sequent purposes. God has no purposes antecedent to the act of creation that leadhim to create in a certain way, but the act of creation itself could produce an idea ofthe good that conditions creatures. Divine purposes could perhaps be understood interms of this created idea of the good.

In any event, it is clear that for Descartes, we have no access by natural reason toany idea that would render intelligible the specific purposes deriving from God’sact of creation.31 It may seem, however, that this consideration does not rule outAristotelian final causes. For as we saw in §1.2.2 (iii), the orthodox Aristotelian viewis that that the forms even of beings that lack cognition and appetite are internalsources of final causality in nature. Given such a view, it might appear thatDescartes’s argument that we have no access to divine purposes is simply irrelevantto the issue of whether we are entitled to appeal to final causes. However, I alsonoted in this earlier section the clear position in Suárez that “natural agents” lackingcognition and appetite can be said to be final causes only insofar as their actionderives from God. This aspect of Suárez’s account of final causality reveals the depthof the confusion involved in Descartes’s persistent objection that in taking variousreal qualities and substantial forms to be responsible for various effects in nature, theschoolmen illicitly suppose that bodies have “tiny souls” that cognize the effects


30. Cf. the comment attributed to Descartes in the Conversation with Burman that we goastray when “we think of God as some great human being [magnum hominem], who proposesto himself such and such, and strives by such and such means, which certainly is most unwor-thy of God” (AT 5:158). See §5.1.2 for further discussion of Descartes’s doctrine of the cre-ation of the eternal truths.

31. I say ‘specific purposes’ to allow for Descartes’s claim, in the passage from his corre-spondence quoted toward the end of §1.2.2 (iii), that God created the world for his own sake.

they bring about.32 Far from holding that qualities and forms are quasi-mental causesthat cognize their ends, it is a consequence of the view of scholastics such as Suárezthat the notion of final causality has no application when nature is considered inabstraction from the ends that direct divine concurrence. So such a scholastic wouldin fact grant Descartes that were we not entitled to appeal to divine ends in physics,we could not speak of final causality in that realm.33

In fact, it seems that there is one respect in which Descartes is closer to the origi-nal Aristotelian stance than was Suárez. Whereas Descartes holds that divine ends areinscrutable to the philosopher of nature, he nonetheless insists that we do have accessto a kind of finality in the special case of the soul–body union. In the SixthMeditation, for instance, he takes experience to reveal that the sensations that derivefrom motions in the brain are “most especially and most frequently conducive to theconservation of the health of the human being” (AT 7:87). Here, it seems, the sensorysystem has the function of conserving the health of the soul–body composite.Descartes could not, consistent with his prohibition of the appeal to divine ends, con-clude that this function reflects God’s own purpose in creating the composite as hehas.34 But the function also cannot be referred to any other mind that cognizes the endof conservation. Thus we appear to have—what scholastics such as Suárez could notallow—an appeal to a kind of finality that is not grounded in a cognition of ends.35

But though Descartes seems to have allowed for a kind of finality in the case of thesoul–body composite, it is not clear that he allowed for the activity of final causes inthat case. After all, he took brain motions to be the source of the various sensationsthat serve the purpose of conservation of health, and he indicated repeatedly that thesemotions are efficient causes of the sensations.36 For Suárez, final causes could beinvolved in this case only by means of God’s concursus with the action of second-ary efficient causes. But Descartes eliminated this route to final causality when he


32. See To Mersenne, 26 Apr. 1643, AT 3:648; Sixth Replies, AT 7:441–42; PP III.56, AT 8-1:108.

33. In §2.1.3 (ii), however, I suggest that Descartes’s charge that the scholastics posit tinysouls may derive in part from his distinctive conception of eminent containment.

34. Admittedly, in the Sixth Meditation passage Descartes may seem to attribute the pur-pose of the sensory system to God. After all, he is concerned there to counter the objectionthat the fact that we are subject to “true errors of nature” in sensation conflicts with God’sgoodness (AT 7:85). But though this point requires further consideration than I can providehere, I would simply suggest that Descartes can be read as arguing not that God had goodintentions in creating the sensory system, but merely that the worthiness of this system showsthat true sensory error is not obviously incompatible with God’s goodness. In terms thatLaporte has introduced, the vindication of divine goodness requires an appeal only to the“internal finality” of the operation of the sensory system, and not to an “external finality”involving the ends that move God to create in a particular manner (Laporte 1928, 388).

35. For a further defense of the claim that this passage commits Descartes to a kind of nat-ural teleology, see Simmons 2001; cf. Laporte 1928, 385–96. There is a further discussion in§4.1 of the nature of the union in Descartes.

36. As indicated in §4.2, however, there are some important complications for his accountof the efficient causality of the motions in this case.

eliminated the appeal to divine ends. From Suárez’s perspective, then, he left us withefficient causes that exhibit a natural teleology ungrounded in final causes.

There is still rational teleology, which covers rational agents that act in accord withends they cognize. Though Descartes denies that we can explain divine action in thisway, he explicitly allows for this sort of explanation in the case of our own action. In theSecond Replies, for instance, he cites as an axiom that “the will of a thinking thing is car-ried [ fertur] voluntarily and freely (for this is the essence of the will), but neverthelessinevitably, toward a clearly known good” (AT 7:166). This “carrying” would seem tocorrespond to the sort of final causality that Suárez took to be present in cases where thewill of a created intelligent agent is inclined to act in a particular way by a cognizedend.37 However, it is important to recall the view in Suárez that the cognized object pro-duces in the will only a kind of “metaphorical motion,” and that strictly speaking it isonly the will itself that produces the actual volitional act as an efficient cause (see, again,§1.2.2 (iii)). For this reason, Gilles Olivo concludes that in the view of Suárez, “thecausality of the final cause is absorbed ultimately, that is to say, in its efficacy [effectivité],into that of efficient causality” (Olivo 1997, 99).38 Once more, Suárez provides the jus-tification for excusing Descartes from providing room in his system for causes in hisnatural philosophy other than the efficient causes governed by the containment axiom.

2.1.3. Formal and Eminent Containment

We have considered the requirement of the containment axiom that the reality of theeffect be contained in the total and efficient cause. Now we are in a position to con-sider the requirement of that axiom that such a cause contain this reality formally oreminently. Descartes’s language in the Third Meditation can suggest that he was ledto this requirement merely by the “light of nature,” with no dependence on previousteaching. But setting aside complications concerning objective reality (on whichmore presently), the requirement is straight from the scholastic tradition. As we havealready seen, Suárez affirmed prior to Descartes that “all causes taken together” mustformally or eminently contain the perfections they produce in their effect. We havealso seen Bennett’s claim that Descartes had no deep understanding of the notion ofcausal containment. In contrast, it is a central thesis here that Descartes offered thematerial for a conception of formal and eminent containment on which they differ inimportant respects from the corresponding kinds of containment that Suárez posited.

(i) Formal Containment

In the Third Meditation, Descartes illustrates his containment axiom by noting that heatcannot be induced in a subject “unless from a cause of at least the same order of per-fection as heat” (AT 7:41). Similarly, Suárez earlier used the case of “fire when gener-


37. However, the case emphasized in the Second Replies passage seems to involve whatfor Suárez is merely voluntary rather than free action (see chapter 1, note 66). In §5.2, Iconsider further the relation of the accounts of free human action in Descartes and Suárez.

38. Cf. Carraud 2002, 159.

ating fire” as an example of a “univocal cause,” that is, one that “effects an effect of thesame kind” (efficit effectum ejusdem rationis) (MD XVII.2, ¶21, 1:591). Yet the specificaccounts that Suárez and Descartes offer of the sort of containment present in this par-ticular case are significantly different. Whereas Suárez held that the heat of both thegenerating and generated fire is a real accident that is a res distinct from the fire itself,Descartes rejects the containment of any such res in a purely material being. InDescartes’s view, the physical heat (as opposed to the sensation of heat) that the bodycontains and produces can be only a certain kind of local motion of parts of matter.39

Descartes’s official explication of formal containment reveals an even deeper dis-agreement with Suárez. In the list of definitions that he provides in his “synthetic”presentation of his system in the Second Replies,40 Descartes includes the stipulationthat objects contain formally all that is “such as [talia . . . qualia] we perceive them”(AT 7:161). This follows his definition of the objective reality of an idea as “the entityof the thing [entitatem rei] represented by an idea, insofar as it is in the idea; . . . Forwhatever we perceive as in the objects of ideas, they are in the ideas themselves objec-tively” (AT 7:161). For Descartes, then, the paradigmatic case of formal containmentis one in which the object as it exists outside of our idea of that object conforms tothe objective reality of that idea.

Descartes’s understanding of this case of course relies on his account of the dis-tinction between formal and objective reality. According to the Second Replies, anobject formally contains what is present objectively in our idea of that object just incase it is “such as we perceive” it. What is odd, from a certain scholastic perspective,is the reference here to the correspondence of what is in the object to a distinct sort ofreality in the idea. Caterus protested in the First Objections that “objective being” ismerely “the act of intellect itself terminating through a mode of the object,” and thusis merely “an extrinsic denomination, and nothing real” (AT 7:92).41 This understand-ing admittedly reflects a Thomistic view, and Scotists were more inclined to posit an“objective concept” as a tertium quid between the act of intellect and the cognizedobject.42 But on this particular point Suárez sided with the Thomists, holding that thereis only a distinctio rationis between an act of intellect and its objective concept.43


39. See, for instance, Descartes’s account of heat in W II, AT 11:7–10.40. Descartes distinguishes a synthetic presentation that involves demonstrations with def-

initions, postulates, and axioms from an analytic presentation, illustrated in the Meditations,in which a method for discovering the truths is employed (AT 7:155–56).

41. Cf. the discussion of Caterus’s position in Armogathe 1995.42. On the difference between Thomists and Scotists on this point, and the relevance of

this disagreement to Descartes’s understanding of objective reality, see Dalbriez 1929. Thiswork is a critique of Gilson’s claim that “in scholastic thought, objective being is not a realbeing, but a rational being” (Gilson 1925, 321). For a reconsideration of this debate that issympathetic to Dalbriez’s position, see Ariew 1999, ch. 2. Cf. the Scotistic interpretation ofDescartes’s account of objective reality in Normore 1986.

43. Suárez was responding to the position of Durandus, which was defended by Suárez’scontemporary Vasquez. For discussion of this debate, with references, see, again, Dalbriez1929. But cf. Renault 2000, which takes Ockham to be the source of the anti-Cartesian under-standing of objective reality.

For Suárez, then, the reality that exists in an idea (or, as he put it, in an objectiveconcept) is just the reality as it exists in the object. This precision might not seem tobe so important; after all, it appears that Suárez could agree with Descartes on thebasic point that an object formally contains all that which is “such as we perceive it.”But the differences are significant in one case where Descartes’s explication of therelation between objective reality and formal containment is most problematic,namely, the case of sense perception.

As we have seen, Descartes argues in the Sixth Meditation that bodies must existas causes that formally contain what is present objectively in our sensory ideas. Butthere is scholarly disagreement over whether Descartes even allowed that bodily fea-tures are present objectively in sensory ideas.44 I myself take the argument in theSixth Meditation to indicate clearly enough that he did intend to allow for such con-tainment. Without the assumption that sensory ideas have an objective reality thatrequires a cause, this argument could not even get off the ground.45 Nonetheless, itmust be admitted that Descartes’s claim in the Second Replies that features that existformally in objects are “such as we perceive them” seems to fail in the case of sen-sory ideas. For Descartes himself warns after presenting the Sixth Meditation proofof the material world that bodies may not exist “in a way that is entirely such as[talia omnino . . . qualia] the senses comprehend them, insofar as the comprehen-sion of the senses is in many cases very obscure and confused” (AT 7:80). It wouldseem that bodies cannot formally contain the qualities that we sense in a confusedand obscure manner, and thus that there is no need for an external cause in the caseof such sensations.46

I think we can go some ways toward reconciling the proof in the Sixth Meditationwith the subsequent comment concerning the confused and obscure comprehensionof the senses by emphasizing the following claim elsewhere in this text:

[F]rom the fact that I sense diverse colors, sounds, odors, tastes, heat, hardnessand the like, I correctly conclude that there are other things in bodies from whichthese various sensory perceptions come [adveniunt], variations correspondingto them [i.e., to the variations among the sensations], though perhaps notsimilar to them. (AT 7:81)


44. The disagreement is most evident in the massive literature on Descartes’s account inthe Third Meditation of “material falsity.” For a representative discussion, see Kaufman 2000.

45. That is, the argument as presented in the Sixth Meditation. Interestingly, Descartesoffers a version of this argument in the 1644 Principles that does not appeal to the objectivecontainment in sensory ideas of what is formally contained in bodies; see PP II.1, AT 8-1:40–41. Even so, there is the point in this latter text that we know by means of sensory stim-ulation that matter “has variously different shaped and variously moving parts that give rise toour various sensations of colors, smells, pain and so on.” This point is connected to theaccount of objective containment in sensory ideas that I offer on Descartes’s behalf presently.Thanks to Marleen Rozemond for discussion of the significance of the differences betweenthe two versions of Descartes’s proof of the existence of the material world.

46. Here I draw on and further develop the position I proposed in my discussion of thisproblem in Schmaltz 1992b.

This passage indicates that sensory ideas that do not resemble bodily qualitiesnonetheless are systematically correlated with them. Because of these correlations,particular ideas can direct the mind to certain bodily qualities rather than others. Ofcourse, we cannot know, simply by introspection, which qualities these ideas repre-sent; that is why Descartes calls the ideas confused and obscure. Nonetheless, theideas can represent the qualities in the broad sense just indicated. In virtue of the factthat the ideas so represent, they possess some sort of objective reality. Bodies for-mally contain what is in the sensory ideas objectively, then, in the sense that theypossess the qualities to which these ideas direct the mind.47

Admittedly, this reading stretches thin the claim in the Second Replies that featurescontained objectively in the mind are contained formally in bodies only when they existoutside of the mind in a way that is “such as we perceive them.” But I take Descartes’sown remarks concerning confused and obscure sensory ideas to suggest a thin notion ofbeing “such as” these ideas reveal. Moreover, this thin notion allows for the passage fromthe Second Replies to be reconciled with the suggestion in the Sixth Meditation that eventhough the objective reality of sensory ideas corresponds to the formal reality of bodilyqualities, these qualities are often “not entirely such as” they are comprehended by sense.48

(ii) Eminent Containment

I have mentioned Descartes’s stipulation in the Second Replies that objects containformally all that is “such as we perceive them.” He continues by noting in that samepassage that objects contain eminently what “indeed is not such [as we perceive], butgreater, so that it is able to take the place of such a thing [that is as we perceive]” (AT7:161). This explication is less than transparent, to say the least. Indeed, critics suchas Radner have objected that Descartes offered no clear account of eminent contain-ment, and thus had no clear explanation of a case in which a cause produces an effectthat differs in nature from it.49 This is behind the charge in Radner and others thatDescartes’s containment principle rules out the causal interaction of objects with dif-ferent natures. To evaluate this charge, we need to determine whether we can makesome sense of Descartes’s claim that objects eminently contain what is not such as weperceive but is “greater” than and “able to take the place” of what we do perceive.


47. I take the account of the objective reality of sensory ideas that I attribute to Descartesto be similar to Locke’s view in Essay II.xxxi.2 that whether our simple sensory ideas “be onlyconstant Effects, or else exact Resemblances of something in things themselves,” still they“are all real and true, because they answer and agree to those Powers of Things, which pro-duce them in our Minds, that being all that is requisite to make them real, and not fictions atPleasure” (Locke 1975, 373). Locke’s claim that nonresembling sensory ideas “agree to” thebodily powers that produce them seems to me to be functionally equivalent to the view, whichI attribute to Descartes, that such ideas objectively contain the bodily qualities to which theydirect the mind.

48. See §4.2 for further discussion of Descartes’s account of the action of body on mind.In §3.2.1 (iii), I consider complications for formal containment connected to Descartes’saccount of body–body interaction.

49. Radner 1985b, 232, 233–34.

On one understanding, what is greater and able to take the place is simply thepower to produce the existence of the object we perceive. This understandinginforms the analysis of eminent containment that Eileen O’Neill has offered. On thisanalysis, Descartes held that

a property ø is eminently contained in X if and only if: ø is not formally containedin X [i.e., X does not contain at least n degrees of ø]; X is an entity displaying agreater degree of relative independence than any possible Y which could contain øformally (i.e., higher up in the ontological hierarchy than any such Y); and X hasthe power to bring about the existence of ø. (O’Neill 1987, 235)50

There is a weaker reading of the second clause, on which X is an entity displaying agreater degree of relative independence than ø, that is, is higher up in the ontologi-cal hierarchy than ø, as opposed to any possible Y that could contain ø formally. Thisweaker reading may seem to be supported by Descartes’s comment in the ThirdMeditation that since “extension, shape, position, and motion” are “merely modes ofa substance,” they can be contained in him eminently given that he is thinking sub-stance (AT 7:45).51 As we have seen, however, Descartes indicates in the SixthMeditation that certain finite creatures can contain bodily effects eminently in virtueof the fact that they are “more noble than” corporeal substance (AT 7:79). Here it isnot just the fact that the effects are mere modes that allows for eminent containmentin these other substances; in addition, there is the fact that these substances are morenoble than the corporeal substances that contain the effects formally.52

An initial problem for this analysis of eminent containment derives from the impli-cation of Descartes’s simple ontological hierarchy that mental and bodily substanceshave the same reality as finite substances. Given this hierarchy, it would seem that bod-ily effects cannot be contained eminently in a finite mind, contra the remarks in the SixthMeditation.53 However, we could get around this problem by appealing to Descartes’sown comment in correspondence that our soul “is much more noble [beaucoup plusnoble] than body” (To Elisabeth, 15 Sept. 1645, AT 4:292).54 One way in which mind is“more noble” is indicated in the Sixth Meditation, which includes the claim that “thereis a great difference between mind and body, inasmuch as body is by its very naturedivisible, whereas mind is utterly indivisible” (AT 7:85–86). This difference indicates the


50. To my mind, O’Neill’s analysis marks an advance over Clatterbaugh’s view that øis eminently contained in X if and only if X contains greater than n degrees of ø; seeClatterbaugh 1980, 391.

51. Thanks to David Ring, who pressed me to consider this point a number of years ago.52. Descartes could not have made this point that his mind is more noble than body in the

Third Meditation because he had not yet provided an account of the nature of body and of itsdistinction from mind.

53. On Gueroult’s interpretation of Descartes (see note 9), there would be no problem herefor eminent containment in mind insofar as particular bodies, as modes, are lower on the onto-logical hierarchy than mental substances. But in the Sixth Meditation, the stress is on the factthat certain substances, presumably mental, are more noble than the corporeal substances thatformally contain what is present objectively in our sensory ideas.

54. Thanks to Michael Della Rocca for drawing this passage to my attention.

greater nobility of mind given Descartes’s claim in the Second Replies that “it is knownper se that it is a greater perfection to be undivided than to be divided” (AT 7:138).

The implication in Descartes, then, is that though created minds are below Godinsofar as they are finite, still they are above bodies insofar as they are indivisible.This implication yields the following “enhanced” ontological hierarchy:

God infinite indivisible substance

minds finite indivisible substances

bodies finite divisible substances

thoughts modes of finite substancesshapes/sizes/motions

Given this enhanced hierarchy, O’Neill’s analysis is consistent with the claim thatparticular bodies and their modes can be contained eminently in finite minds.55

Even so, there remains a problem with the consequence of the last clause ofO’Neill’s definition that something can eminently contain ø only if it has the powerto bring about the existence of ø. In defense of this clause, O’Neill appeals to Suárez,and in particular to his claim that “what is said to contain eminently has a perfectionof such a superior nature that it contains by means of power [virtute] whatever is inthe inferior perfection,” where this power is said to be the power that “can produce[ potest . . . efficere]” the effects of inferior perfection (MD XXX.1, ¶10, 2:63*; citedin O’Neill 1987, 239).56 But though Descartes was obviously influenced by thescholastic view that the effect must be contained formally or eminently in its totalefficient cause, there are reasons to think that he did not adopt Suárez’s particularaccount of eminent containment. When he claims in the Third Meditation that hismind contains material things eminently, for instance, Descartes does not suggestthat he has the power to create the material world. Indeed, in a 1641 exchange withhis critic “Hyperaspistes,” Descartes makes clear his rejection of the claim that ourmind has such a power. This critic objected that in Descartes’s view, “since a corpo-real thing is not more noble than the idea that the mind has of it, and mind containsbodies eminently, it follows that all bodies, and thus the whole of this visible world,can be produced by the human mind” (AT 3:404). Such an implication is said to be


55. For complications concerning the eminent containment of matter in finite minds, seenote 57. Descartes also speaks of material things as being eminently contained in God’s mind(as in the Sixth Meditation proof of the existence of the material world, at AT 7:79). As I notein §5.1.2, however, Descartes’s doctrine of the creation of the eternal truths renders this sortof containment problematic.

56. As O’Neill notes, Suárez went on to claim that “formally speaking” the power to bringabout an effect cannot define eminent containment, since a cause is said to be able to producean effect in virtue of the fact that it eminently contains it (MD XXX.1, ¶10, 2:63*; cited inO’Neill 1987, 239). But Suárez also indicated that we cannot understand eminent containmentother than by its causal relation to the effect. In any event, it seems that having the power tocause an effect could be a necessary condition for eminent containment without beingdefinitionally equivalent to it.

problematic insofar as it undermines our confidence that God alone created the vis-ible world. In response, Descartes protests that we can produce “not, as objected, thewhole of this visible world, but the idea of the whole of things that are in this visi-ble world” (AT 3:428). The suggestion here is that even though the whole visibleworld is contained in our mind eminently, we do not have the power to produce itsextra-mental existence.57 Eminent containment would seem to be a necessary but nota sufficient condition for something to be able to produce features it does not for-mally contain.58

Nevertheless, it seems that we could substitute for O’Neill’s last clause the claimthat X has at least the power to bring about the reality of ø as present objectively inX’s idea of ø. It is this power that, in the terms of the Second Replies, is “not such”as we perceive “but greater, so that it is able to take the place” of what we perceive.This power is able, in particular, to produce bodily qualities insofar as they are pres-ent objectively in the mind.59 We therefore have the following alternative to O’Neill’sanalysis of eminent containment:

A property ø is eminently contained in X if and only if: ø is not formallycontained in X; X is an entity displaying a greater degree of relative independencethan any possible Y which could contain ø formally (i.e., higher up in theontological hierarchy than any such Y); and X has a power that suffices toproduce the objective reality that is present in X’s idea of ø.

This alteration of O’Neill’s account may seem to be minor, but in fact it serves tohighlight an important difference between the accounts of eminent containment inSuárez and Descartes. Suárez had no difficulty applying the notion of eminent con-tainment to the case of bodily causes, as for instance when he held that a heavy body(unum grave) that moves another heavy body to a particular place contains that place


57. The eminent containment of the whole of matter in finite minds is more problematicgiven Descartes’s view that this matter is an “indefinite” rather than a finite substance. Indeed,he tells Regius in correspondence that he could not think of indefinite extension “unless themagnitude of the world also was or at least could be indefinite” (24 May 1640, AT 3:64). Butthe “could be” perhaps suggests that there need not be anything indefinitely extended in orderfor our mind to think it. Moreover, there is the claim in the Fourth Meditation that our will isin some sense unrestricted, and that it is in fact in virtue of our possessing such a will that weunderstand ourselves to bear in some way the image and likeness of God (AT 7:56–57). Giventhis feature of mind, we could perhaps be said to contain even indefinite extension eminently.For discussion of the complications here, see Wilson 1999a.

58. Cf. the critique of O’Neill’s account of eminent containment in Gorham 2003, 11–13.Beyond objecting to the explication of eminent containment of an effect in terms of a causalpower to produce that effect, however, Gorham rejects in general any account on which theeminently contained effect is to be reduced to other features that the cause formally contains.The alternative to O’Neill’s account that I offer presently is reductionist in this sense.

59. It is important to hold that it is the power that “takes the place” rather than the objec-tive reality itself, given that Descartes emphasizes the difference between eminent andobjective containment when he notes that an effect must be contained in its cause “not merelyobjectively or representatively [objective sive repraesentative], but formally or eminently”(PP I.23, AT 8-1:11).

“virtually or eminently, but not formally,” since that place is contained only in the“active principle” that brings about the downward motion (MD XVIII.9, ¶10, 1:671).I mentioned above Descartes’s ridicule of the scholastic attribution of tiny souls tobodies, and at one point he expresses his objection in terms of the very example ofthe action of gravitas that Suárez used here. Thus, in the Sixth Replies Descartesnotes the misguided thought he had in his youth that “heaviness [gravitas] carriedbodies toward the center of the earth, as if it contained in itself some cognition ofthis” (AT 7:442).60 We have seen that the accusation that the scholastics attributedcognition of ends to natural beings is misplaced. But what is interesting isDescartes’s apparent assumption that a future effect not actually present in the causecan be contained in that cause only by means of cognition. Implicit in this assump-tion is the rejection of the view in Suárez that there can be qualities or powers in bod-ies “more noble than” bodily effects that eminently contain those effects.61 Descartesheld that all alterable features of body have the same kind of reality as modes ofextension, and that only infinite or finite indivisible minds can be more noble thanbodily substances and their modifications. For Descartes, then, the bodily principlethat Suárez posited as eminently containing its effect could be conceived only on themodel of a mind that acts in accord with its cognition of an end. In the case of allother total bodily causes, only formal containment can be at issue.62

2.2. THE CONSERVATION AXIOM

After concluding in the Third Meditation that it is manifest by the light of nature thatGod must exist as the cause of the objective reality of his idea of God as infinitely per-fect substance, Descartes notes that once his concentration on the argument for thisconclusion relaxes he no longer can remember why his idea of a being more perfectthan himself must be caused by such a being (AT 7:47). To remedy the uncertainty,


60. For further discussion of Descartes’s use of the heaviness analogy, see §4.3.2.61. But see also the discussion toward the end of §4.2.1 of one way in which Suárez’s

notion of eminent containment is more restrictive than Descartes’s.62. This implication of the view of eminent containment that I attribute to Descartes can

be contrasted with the position, which the English malebranchiste John Norris attributed to“the Modern Reformers of Philosophy,” that all Bodies that we call Hot, [are] so onlyEminently and Potentially, as they are productive of Heat in us.” This is said to be a replace-ment for the “Old Distinction” between heat as a quality formally contained in certain objectsand heat as eminently contained in bodily causes of this quality that eliminates the first partof this distinction (Norris 1693, 3:21–22). It is likely that Norris counted Descartes among themodern reformers, since he went on to say that Malebranche alone rejected this position in themoderns. But we have seen that Descartes himself took bodies to formally contain what ispresent objectively in our sensory ideas. And I know of no passage where he referred to theeminent containment of effects in bodily causes. I suspect that Norris simply assumed thatDescartes accepted the scholastic view that a quality that is present in a bodily power to pro-duce that quality is eminently contained in the body that has that power. Thanks to EileenO’Neill for drawing my attention to the passage from Norris.

he proposes another argument for the conclusion that he could not exist if God didnot. The new argument is not entirely distinct from the first argument, since it toorelies on an application of the containment axiom to the case of the objective realityof his idea of God. Descartes recognizes this point when he admits that the secondargument is not so much a new argument as “a more thorough examination” of theoriginal argument (First Replies, AT 7:106).63 But there is a distinctive element ofthe second argument that is crucial for our purposes. This element is introduced afterthe first portion of the proof, in which Descartes claims that he cannot have derivedhis existence from himself, since in that case he would have produced in himself allof the perfections he desires but lacks. He then considers the objection that he maynot need any cause of his existence now given the assumption that he has alwaysexisted. Descartes responds:

[S]ince the whole time of life can be divided into innumerable parts, each singleone of which depends in no way on the remaining, from the fact that I was shortlybefore, it does not follow that I must be now, unless some cause as it were createsme anew at this moment [me quasi rursus creet ad hoc momentum], that isconserves me. For it is perspicuous to those attending to the nature of time thatentirely the same force and action [eadem . . . vi et actione] plainly is needed toconserve a thing at each single moment during which it endures, as would beneeded to create it anew, if it did not yet exist; to the extent that conservationdiffering solely by reason from creation is also one of those things that is manifestby the natural light. (AT 7:49)

The claim here that “the same force and action is needed to conserve a thing . . . aswould be needed to create it anew” is reflected in the axiom in the Second Replies—which I have called the conservation axiom—that “no less a cause is required to con-serve a thing than to produce it at first” (AT 7:165). In the Third Meditation, theconservation axiom is said to be perspicuous to those who consider “the nature oftime,” and is said to yield the result that conservation differs “solely by reason” fromcreation.

In what follows I consider three aspects of the position indicated in the ThirdMeditation passage. The first is the nature of the conservation that Descartes takes tobe required for the continued existence of any creature. His ultimate view is that Godalone can conserve the being of created substances. The second aspect is Descartes’sunderstanding of “the nature of time.” It turns out that he offers an account of tempo-ral parts of duration that is incompatible with the temporal atomism that some com-mentators have attributed to him. Finally, there is Descartes’s argument that given thenature of time, conservation can differ from creation solely “by reason.” Though thisargument seems to conflict with an argument in Suárez for the lack of any real differ-ence between creation and conservation, in the end Descartes embraces the basic


63. Cf. Descartes’s letter of 2 May 1644, at AT 4:112. In the synthetic presentation of hissystem in the Second Replies, however, Descartes presents the two arguments as distinct; seepropositions II and III at AT 7:167–68.

Suárezian conclusion that the created world depends continually on the original divineact that resulted in the creation of that world.

2.2.1. Divine Conservation

In the Fifth Objections, Gassendi challenged the use of the conservation axiom in theThird Meditation by claiming that Descartes could have a power to conserve himselfthat is not a power to create himself anew, but rather a power “that suffices to guar-antee that you are preserved unless a corrupting cause intervenes” (AT 7:302). Thesuggestion here is that the presence of such a power is revealed by the fact that con-tinuation in existence is the default condition. What requires an external cause is notthis continuation, but only the initiation or cessation of existence.

I argue in the next chapter that Descartes invokes something very much like thetendency to persist in existence in his explanation in the Principles of his first law ofmotion (see §3.2.1 (i)). In his response to Gassendi, however, he insists that the claimthat conservation requires the “continual action of the original cause” is “somethingthat all Metaphysicians affirm as manifest.” Drawing on remarks from Thomas’sSumma Theologiae noted in §1.1.2, Descartes appeals in the Fifth Replies to the dis-tinction between causae secundum fieri, or causes of becoming, and causae secun-dum esse, or causes of being.64 Indeed, Descartes follows Thomas in illustrating thedistinction between these two kinds of causes by noting the difference between thebuilder as the causa secundum fieri of the house and the sun as the causa secundumesse of the light.65 His claim in the Fifth Replies is that just as the continuing actionof the sun is required for the light to remain in existence, so the continuing action ofGod is required for creatures to remain in existence (AT 7:369).66

The example of the sun landed Descartes in some trouble, since in correspondencewith him the pseudonymous Hyperaspistes attempted to defend Gassendi by notingthat the Bologna spur, a phosphorescent rock, can retain light in a closed room (July1641, AT 3:405). But Descartes responds that the light of the rock is perhaps not thesame as the light that constantly depends on the sun, and in any case that even if thesun example fails, it is “more certain that nothing can exist without God’s concursus,


64. Descartes indicates in a 1639 letter to Mersenne that he brought with him from Francea Bible and “une Somme de S. Thomas” (AT 2:630).

65. See also the example of the builder in the passage from the Conversation with Burmancited in §2.1.2 (i), at note 18.

66. Descartes’s conclusion that light requires the continuing action of the sun is no doubtconnected to his own view that light is not a motion but rather the instantaneous effect of pres-sure deriving from the source of illumination (see, for instance, PP III.64, AT 8-1:115).Whereas a motion could perhaps endure apart from the action that initially produces it, aninstantaneous effect could not. Interestingly, Thomas agreed that light cannot involve motion,since its diffusion is instantaneous (ST I.67.2). However, his insistence that light requires theaction of the sun derives rather from his view that the quality of light depends essentially onthe action of the substantial form of a self-luminous body (see chapter 1, note 21). This lineof argument obviously would not have been attractive to Descartes. Thanks to Andrew Janiakfor discussion of this point.

than no light of the sun without the sun” (Aug. 1641, AT 3:429). Here Descartes wasmoving toward the version of the Thomistic secundum fieri/secundum esse distinctionin Suárez, according to which God alone can be a cause secundum esse.67 In aSuárezian context, it is admittedly odd that Descartes refers to God’s concursus ratherthan his act of conservation. We will discover in the next chapter that there is in factsome question whether Descartes follows scholastics such as Suárez in thinking thatdivine conservation differs from God’s concurrence with the actions of secondarycauses. Even so, his use of the language of concursus in the passage above is supposedto indicate primarily that God must make a causal contribution for any dependentbeing to exist.

However, why not conclude with Gassendi that a being, once created, can persiston its own? To put the point in terms of Hyperaspistes’s example, why not hold thatobjects can continue to subsist without God’s concursus in the same way that theBologna spur can continue to glow without the influence of the sun? The answer inDescartes is connected to his version of the principle of sufficient reason, reflected inthe axiom in the Second Replies that every existing thing, including God, requires a“cause or reason” (causa sive ratio) of its existence (AT 7:164–65).68 I have men-tioned his conclusion that in the case of God, the reason is provided by God’s essence,which serves as the formal cause of his existence (see §2.1.2 (ii.a)). But since essenceis distinct from existence in the case of every other existing object, the cause or rea-son of beings other than God must be provided by some efficient cause. Because thiscause provides the reason for the existence of the object at any time, it is required notonly for the initial creation of the object but also for its subsequent conservation.

In addition, Descartes holds that the cause of an object secundum esse must beactive at each moment that object exists. Aquinas had concluded that just as somethingcannot be in the process of becoming without the action of its cause secundum fieri, sothat thing cannot subsist in being without the action of its cause secundum esse. If anobject needs the activity of a cause secundum esse at all, it needs the activity of thiskind of cause throughout its existence (ST I.104.1; see §1.1.2). As we know, Suárezalso accepted this Thomistic conclusion, arguing that a creature depends on God for itsexistence at each moment, since God is “its cause directly and per se with respect toits esse.” He held that there would be such a dependence even if—as Descartessupposes in the Third Meditation—the object were not created in time but existed frometernity. For even in this case, since the object is not God, it must have its being fromanother, and ultimately from God (see §1.2.3 (i)).

This line of argument in Suárez is reinforced by his analysis of efficient causality.In opposition to the view of some Thomists—but, he insisted, not of Thomas


67. See the discussion in §1.2.3 (i) of Suárez’s revision of the Thomistic distinction.68. For the link between this axiom and the principle of sufficient reason, see Carraud

2002, especially ch. 2. As Carraud emphasizes, however, Descartes’s version differs from themore familiar version in Leibniz insofar as Descartes insisted on the unintelligibility ofthe divine creation of eternal truths (Carraud 2002, 288–93). In §5.1.2, I consider this doctrineand its relation to Descartes’s axiomatic requirement that there be a “cause or reason” foreverything that exists.

himself—that a cause can remain efficacious even after its action has ceased, Suárezheld that an effect can depend on an efficient cause only insofar as that cause isactive (MD XVIII.10, ¶8, 1:682). Applied to the case of conservation, the conse-quence is that created objects that depend essentially on God for their being can con-tinue to so depend only insofar as God continues to produce that being through anaction.69 Descartes in effect adopts Suárez’s anti-Thomistic (though, for Suárez, notanti-Thomas) premise when he appeals in the First Replies to the fact, revealed bythe light of nature, that something “does not properly have the reason of a cause,unless for as long as it produces the effect, and thus is not prior to it” (AT 7:108).For Descartes, no less than for Suárez, an object cannot depend for its being on Godunless God is active as an efficient cause of that being at each moment it exists.70

In his letter to Hyperaspistes, Descartes offers a further Suárezian argument forthe necessity of divine conservation. Descartes insists there that “if God ceased hisconcursus, at once all that he has created would go into nothingness, because, beforethey were created, and he offered his own concursus, they were nothing.” He con-tinues by noting that “it is not possible that God destroy other than by ceasing hisconcursus, because otherwise he would tend to non-being through a positive action”(AT 3:429). These remarks are reminiscent of Suárez’s argument that God cannotconserve the being of an object merely permissively, through not depriving them oftheir being, rather than positively, by means of his efficient causation of that beingthat derives from an action. The argument is that since God is omnipotent, he canannihilate any creature, and since every action by its nature tends toward some pos-itive being, he can annihilate only by omitting some action (MD XXI.1, ¶14, 1:789).Thus, divine conservation must consist in the continuation of that action by whichGod gave being to creatures.71

Admittedly, Descartes’s claim in the passage from the Third Meditation concern-ing “the nature of time” is only that there must be some cause that “quasi createsanew” at each moment, not that God must be the cause. In the First Replies, how-ever, Descartes notes, “[W]hat I have not written before, that [conservation] can inno way come from any secondary cause, but altogether from that in which there issuch great power that it conserves a thing external to itself, so much the more con-serves itself by its own power, and thus is a se” (AT 7:111). The need for divine con-servation in the case of substances is clear from a passage from a 1642 letter toRegius, cited in §1.3, in which Descartes insists that God alone can create a sub-stance de novo (AT 3:505). According to Descartes, then, God alone can be the causesecundum esse of the initial existence of a substance, and thus he is the only beingwho can conserve this substance in existence through the continuation of the act ofcreating it.


69. I think this line of argument is implicit in Suárez’s discussion in MD XXI.1, ¶¶6–15,1:787–79; see §1.2.3 (i). In this earlier section, Suárez presents himself as correcting anddeveloping Thomas’s argument in ST I.104.1.

70. For this point, see also Secada 1990, 49–51.71. Cf. Thomas’s invocation of the claim in Augustine that nature is annihilated once God

withdraws his ruling power (ST I.104.1).

But a central question, given the scholastic rejection of occasionalism, is whetherDescartes allows for any causes other than God. In the passage from theConversation with Burman that I considered in §2.1.2 (i), he is said to hold that God,in producing the being of something ex nihilo, acts as a total cause, and thus pro-duces something similar to himself. I noted in that section that if the containmentaxiom requires that total causality involve creation ex nihilo, then given the remarksin this passage the axiom would be restricted to divine action. Indeed, Jean-LucMarion has suggested recently that Descartes restricts true efficient causality to Godalone.72 In addition to the remarks to Burman,73 Marion cites Descartes’s claim in a1645 letter, in defense of the conclusion that God is the cause of all effects of humanfree will, that

the distinction of the Schools between universal and particular causes is out ofplace here: because what makes the sun, for example, the cause of flowers is notthe cause of the fact that tulips differ from roses, [since] their production dependsalso on other particular causes to which they are not subordinated; but God is sucha universal cause of all that he is in the same way the total cause. (To Elisabeth, 6Oct. 1645, AT 4:314)74

However, the claim here is not that God is the universal and total cause in such a wayas to exclude all other causes. Rather, it is simply that the universality of God’scausality differs from the universal causality of the sun insofar as the former doesnot involve the contribution of particular causes not subordinated to it. All derivativecauses, whether universal or particular, are subordinated to God’s distinctive sort ofuniversal and total causality. Some combination of universal and particular deriva-tive causes can still be sufficient in their order to bring about their effect, and thusare still subject to the containment axiom. But this sort of total cause nonetheless isdependent on God’s total causality, which unlike derivative causes includes an act ofconservation not distinct from the act of creation ex nihilo.

We have seen the implication in Descartes that only God’s universal causality cancreate and conserve substantial being. However, there still seems to be room in hissystem for a derivative sort of causation of the modifications of substance. This cau-sation of modes must be subordinated to God’s causation of the substances themodes modify, since the existence of the modes themselves depends on the existenceof these substances. In this sense, God can be said to be the total causes of the effectsproduced by derivative causes. Even so, it seems that these causes could produceeffects that do not derive immediately from God. In terms of the Thomistic distinc-tion that Descartes employs, derivative causes could be causes secundum fieriof modes that God does not directly produce as the cause secundum esse of the


72. Thus Marion speaks of “the reduction of all kinds of causalities to the efficient causal-ity as only divine” (Marion 1991, 288).

73. Cited in Marion 1991, 289 n.24, in support of the conclusion that “God alone can exer-cise total causality.”

74. Cited in Marion 1991, 287. I will return in §5.3.1 to a consideration of the ramifica-tions of this passage for Descartes’s account of human freedom.

substances the modes modify.75 Descartes of course rejected the scholastic view thatderivative causes can educe from matter forms that are res distinct from matter. Butsince he follows Suárez in thinking that modes are not res distinct from the sub-stances they modify, Descartes’s rejection of this view does not prevent him fromholding that such causes educe from material substance modes that are contained init merely potentially. Whether Descartes can allow for this sort of derivative causa-tion of modes depends on whether his metaphysical system allows for the attributionof causal power to beings other than God; this is an issue that remains to be resolved.Even so, it seems clear enough that his view of God’s universal and total causalityof the created world does not straightforwardly commit him to occasionalism.76

2.2.2. “The Nature of Time”

In the Third Meditation passage that introduces the conservation axiom, Descartesclaims that careful attention to “the nature of time,” and in particular to the fact thattime is divisible into “innumerable parts,” each independent of the others, revealsthat conservation differs solely by reason from creation. §2.2.3 will be devoted to theargument that this feature of temporality provides support for the conservationaxiom. Here our concern will be to consider Descartes’s view of the nature of tem-poral parts and of their mutual independence.

In the Principles, Descartes holds that when time is considered in general, apart fromthe duration of particular objects, it is a mere “mode of thinking,” a mental abstraction(PP I.57, AT 8-1:26–27). What is mind-dependent here is a particular measure of dura-tion, such as when we measure our life in days and years by comparing it with otherregular motions. In line with a view that Suárez offered previously, however, Descartesholds that the duration that is measured is itself distinct only in reason, and not in real-ity, from the enduring object. In particular, he claims that we cannot distinctly conceiveof a substance apart from its duration, and also cannot so conceive the duration, as itexists in the substance, apart from the substance itself (PP I.62, AT 8-1:30).77


75. Cf. the discussion in §3.1.3 of the distinction between modal and substantial causesthat Garber attributes to Descartes.

76. On the basis of these considerations, I would dispute Gorham’s conclusion that theremarks in the letter to Elisabeth reveal that “God does not leave it to other causes to produceour diverse volitions and bodily movements, nor rely on them for assistance. Rather, he bringsabout all of the particular volitions and movements directly and by himself” (Gorham 2004,412–13). I think Descartes would agree that God produces the being of everything “directlyand by himself.” But in the case of becoming, his claim in the letter to Elisabeth seems to bemerely that his universal causality does not involve the causal contribution of particular causesthat are not subordinated to him. Gorham stresses the claim in Descartes’s letter that Godwould not be perfect “if there were something in the world that did not come entirely fromhim” (AT 4:314, cited in Gorham 2004, 412, n.104), but in light of the remarks that followI would read the claim that an effect “comes entirely from God” as saying that there is nocause of the effect that is not subordinated to a divine causality that is both universal and total.

77. See Suárez’s claim that there is merely a distinctio rationis between duration and theexisting object (MD L.1, ¶5, 1:914*).

So much for time and duration. What about their parts? Descartes notes in thePrinciples that the parts of an extended substance are themselves really distinct sub-stances, since each can exist on its own apart from the others (PP I.60, AT 8-1:28).78

If temporal parts are to be conceived in the same manner, however, we have thestrange result that all substances are composed of distinct substantial time-slices. Forwe have seen that when it is considered in objects, time is nothing other than anattribute that is distinct only ratione from the enduring substance. And if the dura-tion is composed of distinct substantial parts, the substance itself must be composedof distinct substantial parts over time, just as extended substance is composed of dis-tinct substantial parts at any one time.

There is reason to think that this result would be unacceptable for Descartes. Inthe conservation passage from the Third Meditation, he considers his own temporalduration as a thinking thing. Yet Descartes is clear in the Meditations that “it is oneand the same mind that wills, that senses, that has intellectual perceptions”(AT 7:86). Here it seems that the mind not only is “utterly indivisible” at a particulartime, but remains one and the same unified substance over time.79

There is an exchange relevant to this point in the Conversation with Burman.When confronted with the objection that it follows from the temporality of thoughtthat thought itself is extended and divisible, Descartes is reported to claim thatthough thought is “extended and divisible as far as duration, which can be dividedinto parts,” nonetheless “it is not extended and divisible as far as its nature, insofaras it remains unextended” (AT 5:148). The view here is that even though the mindhas an extended and divisible duration, it always remains unextended and indivisi-ble by its very nature, insofar as it is an immaterial res cogitans.80

I have suggested that given that the mind is indivisible by nature, the parts intowhich its duration can be divided cannot be substantial parts. What sort of parts,then? Of the three kinds of distinction that Descartes borrowed from Suárez—namely, the distinctiones realis, modalis, and rationis (see §1.2.1)—it would seemto be the modal distinction that is applicable to the case of temporal parts. Descartes


78. As Descartes makes clear elsewhere, the three-dimensional parts of body are distinctfrom its two-dimensional modes; see Fourth Replies, AT 7:250–51; Sixth Replies, AT7:433–34.

79. Descartes’s talk of extended substance taking on different modes over time suggeststhat he thinks that even though such a substance is divisible, it too can remain the same sub-stance over time.

80. This passage is admittedly suspect. Descartes is reported to continue by claiming thatGod similarly is an unextended being who has a duration that is divisible into parts, and thatsince we can now divide God’s duration, we can divide that duration as it was prior to his cre-ation of the world (AT 5:148–49). In his own correspondence around the same time of hismeeting with Burman, however, Descartes denies both that God has a successive duration (ToArnauld, 4 June 1648, AT 5:193), and that there was any such duration prior to the creationof the world (To More, 15 Apr. 1649, AT 5:343). But even though there may well be a cor-ruption in the report of Descartes’s remarks to Burman, I claim presently that Descartes canaccept a sense in which the duration of his indivisible mind is divisible into parts (though, incontrast to the case of the extension of a body, not into substantial parts).

indicates that a modal distinction holds between two modes of the same substance,since “we can know one mode without the other, and vice versa, but neither how-ever without the same substance in which they inhere” (PP I.61, AT 8-1:29). Myproposal is that in contrast to the case of the parts of extended substance, he takesthe different parts of the duration of a substance to be modally distinct from eachother.

A clear counterexample to this proposal may seem to be provided by Descartes’sclaim in the Principles that “in created things, that which never has in itself diversemodes, such as existence and duration in the thing existing and enduring, must becalled not qualities or modes but attributes” (PP I.56, AT 8-1:26). Isn’t the indica-tion here that duration is not subject to modification?81 However, we need to remem-ber Descartes’s position that thought and extension are also attributes that are notsubject to modification considered as such. As he notes in correspondence withArnauld,

[A]s extension, which constitutes the nature of body, differs greatly from variousshapes or modes of extension that it assumes, so thought, or thinking nature, inwhich I take the essence of the human mind to consist, is much other than this orthat act of thinking. (29 July 1648, AT 5:221)

In the same way, he could say that invariable duration differs greatly from the vari-ous modes that it has at different moments. Just as Descartes can distinguish the con-tinuing attributes of thought and extension from the varying modes that it assumes,then, so he can distinguish the duration of thinking and extended substances from thevarying modes that constitute its distinguishable parts.82

Thus far I have emphasized the importance of distinguishing Descartes’s view ofthe parts of time or duration from his view of the parts of extended substance. But thereis a reading of Descartes—more popular in the past, perhaps, than currently—onwhich the difference seems to be even greater than I have indicated. Descartes arguesexplicitly that there can be no indivisible atoms on the grounds that any portion ofextension can be divided into smaller parts.83 Yet Gueroult, most prominently, insiststhat he takes temporal duration to be a discontinuous collection of indivisible parts.84


81. In §3.2.2, I consider an objection along these lines that Alan Gabbey has offered.82. I will return to this interpretation of Descartes’s account of duration and its parts in the

course of my discussion in §3.2.2 of his view of bodily force.83. See, e.g., PP II.20, AT 8-1:51–51; To Gibieuf, 10 Jan. 1642, AT 3:475–77; To More, 5

Feb. 1649, AT 5:273–74.84. The prominence of Gueroult’s interpretation in the earlier literature is reflected in

Yvon Belaval’s remark that it is common knowledge that for Descartes time is discontinuous(Belaval 1960, 149). There is documentation of the relevant literature in Arthur 1988. For amore recent defense of an “atomist” interpretation of Descartes’s account of time, see Levy2005. Cf. Leibniz’s claim in a 1699 letter to De Volder that since “the Cartesians” hold that“God creates all things continually” and that “moving a body is nothing but reproducing it insuccessively different places,” they are committed to the conclusion that “motion in its essenceis nothing but a succession of leaps through intervening intervals, which flow from the actionof God” (Leibniz 1978, 2:193/Leibniz 1969, 521).

He argues that even though Descartes grants that time can be viewed as continuousfrom “the point of view of created things, or in the abstract” (Gueroult 1953,1:280–81), still he holds that there is “the point of view of creation and of the con-crete” from which time is conceived as a “repetition of indivisible and discontinuouscreative instants” (1:275). In Gueroult’s view, then, Cartesian conservation consists inthe successive creation of independent atemporal instants, the atomic parts of duration.

I have mentioned the emphasis in the work of Maimonides on the temporal atom-ism of the occasionalistic Mutakallimun (see §1.1.1). Even so, scholastic opponentsof occasionalism, represented most notably by Thomas, rejected this account oftime.85 Gueroult’s critics insist that Descartes also rejected the view of conservationas continual re-creation, arguing that he took durationless instants to be boundariesof temporally extended temporal parts rather than distinct temporal parts of time.86

However, the debate on this issue has tended to bog down on the interpretation oftechnical terminology in particular passages in Descartes, and there is the view in therecent literature that the texts do not decisively favor the thesis that instants are ulti-mate parts over the thesis that they are mere boundaries of parts.87

As will become clear in §2.2.3, my sympathies are with Gueroult’s critics. But Ithink that there are considerations against his interpretation of Descartes’s accountof temporality that do not rest solely on the issue of how he understood his techni-cal terminology. These considerations are related to the debate in contemporarymetaphysics between “endurantists,” who hold that persisting objects endure bybeing wholly present at different times, and “perdurantists,” who hold that persist-ing objects perdure by being composed of distinct temporal parts.88 The main issuehere concerns not the status of instants, but rather the nature of persisting objects.And on this point, Descartes seems to me to be clearly in the endurantist camp.89

For on the view that I have proposed above, he holds that there is an important dif-ference between spatial and temporal extension. The indication in Descartes is thatspatial extension is composed of parts that can exist on their own as substances. Buton my proposal, he was committed to denying that temporal extension is such acomposite. The sort of distinction that applies to temporal parts is not a realdistinction, as in the case of spatial parts, but rather a modal distinction. Insofar asthey are only modally distinct, however, temporal parts are modifications of the


85. Thomas accepted the Aristotelian view that time is the measure of the continuousmotion of the heavens, and so is itself continuous; see, e.g., ST I.10.6. Cf. Suárez’s endorse-ment of the view in Aristotle that “time has its extension from motion” and that “there is noother extension of time than the continuity of its succession” (MD L.8, ¶4, 1:949*).

86. See Beyssade 1979, ch. 3 and conclusion, and Arthur 1988, 373–75. This position isanticipated in Laporte 1950, 158–60.

87. See, for instance, Secada 1990 and Garber 1992, 266–73.88. For a helpful discussion of the various issues involved in the debate over endurantism

and perdurantism, see Haslanger 2003.89. Gorham 2006 defends a reading on which Descartes understood the different parts of

time to be substantially distinct, and thus was committed to perdurantism. As Gorham admits,however, this sort of perdurantism is incompatible with Descartes’s insistence on the simplic-ity of the soul. I take this conflict to provide reason to reject Gorham’s reading.

same underlying attribute of duration, that is, for Descartes, the same enduring sub-stance. Gueroult’s interpretation requires that there is for Descartes an importantdifference between spatial and temporal extension insofar as only the latter can becomposed of independent atomic parts. I agree that Descartes took these two kindsof extension to be fundamentally distinct, but I understand the relevant differenceto be between a spatial extension divisible into distinct substantial parts, on the onehand, and a temporal extension divisible only into distinct modes of a single attrib-ute of duration, on the other. As I argue presently, the persistence of this attribute iscoupled with a single act of divine conservation identical to the act by which Godoriginally created the substance that is only distinct by reason from that attribute.

2.2.3. Conservation and Creation

We can now turn to Descartes’s argument in the Third Meditation that it is evidentfrom the fact that the different parts of his life are independent of each other that “con-servation differs solely by reason from creation” (AT 7:49). In the Second Replies, thisconnection is reflected in the axiom that “the present time does not depend on the prox-imate preceding [time], and thus no less a cause is required to conserve a thing than toproduce it at first” (AT 7:165). We need to consider first how the fact that temporalparts are independent shows that the same sort of cause is required to conserve as tocreate, and then how this latter result is connected to the conclusion that conservationdiffers solely by reason from creation.

With respect to the first point, recall Descartes’s Suárezian conclusion that theactivity of the cause must be simultaneous with the production of its effect. Giventhat distinct parts of time are not simultaneous, it cannot be the case that causal activ-ity at a previous portion of time suffices for the existence of an effect at a subsequentportion of time.90 The conclusion that the simultaneous cause that suffices for theexistence of the effect during that portion of time must be of the same sort as a cre-ative cause depends on Descartes’s claim that any object that has an existence dis-tinct from its essence requires an efficient cause secundum esse that produces theexistence of this object at each moment it exists (see §2.2.1). In the case of a causesecundum fieri, the cause acts on preexisting material, and what is produced can con-tinue to exist after the cause has acted. But in the case of the cause secundum esse,there is no preexisting material, since the very being of the object is produced. Thisobject must therefore be produced ex nihilo, and in that respect this production doesnot differ from the initial creation of the object ex nihilo.


90. Here I follow the summary of Descartes’s line of argument in Gorham 2004, 391–400.Gorham considers the claim in Secada 1990 that Descartes cannot take the causal independ-ence of different parts of time to derive simply from the condition of causal simultaneity, sincecausal activity at an earlier portion of time could produce an effect at a later portion of time byproducing something during an overlapping interval that produces that effect. As Gorham indi-cates, the counterexample does not succeed, since the overlapping interval can itself be dividedat just the point where the initial points of time are distinguished, and that nothing prior to thatpoint can suffice for the production of something after that point (Gorham 2004, 398).

One problem, however, is that this conclusion seems to fall short of the doctrine thatconservation differs solely by reason from creation. For Descartes himself indicates thatwhen applied to God, this doctrine requires not only that conservation is the same typeof act as creation, or is merely type-identical to it, but also that it is the very same act,and so is token-identical. Thus, in the Discourse he notes that it is “an opinion com-monly received among the Theologians” that “the action by which [God] now con-serves [the world] is entirely the same as [toute la mesme que] that by which he hascreated it” (DM V, AT 6:45). This same theological opinion backs his later claim in thePrinciples that “the world now continues to be conserved by the same action [eademactione] as created it then” (PP II.42, AT 8-1:66).

In fact, this theological opinion can be found in Suárez. Recall his conclusion inthe Disputations that the creation of an object does not differ in reality from thedirect and immediate per se conservation of that object (see §1.2.3 (i)). Suárez didnote the objection that conservation seems to differ from creation insofar as theyoccur at different times. Far from making manifest that conservation does not differfrom creation, as Descartes claims, the distinction of the parts of time is here pre-sented as an obstacle to recognizing this conclusion. Suárez’s response to this objec-tion is that since the time of creation is joined continuously with the subsequent timeof conservation, we can hold that there is a single continuous effect deriving from asingle action. It is due to this sort of continuity, he concluded, that “Saint Thomassaid that conservation is as it were continued creation [quasi continuatam cre-ationem]” (MD XXI.2, ¶4, 1:791).

This last point is linked to Descartes’s claim in the Third Meditation that theremust be some cause that “as it were creates me anew at this moment” (me quasi rur-sus creet ad hoc momentum) (AT 7:49). The reference here to creation rursus mayseem to suggest that the creation that conserves is not merely a continuation of theact of creation, as in the case of Thomas and Suárez, but rather an entirely new act.Indeed, I noted at the outset of this chapter that Bennett takes Descartes to embracethis suggestion. But if we follow Bennett’s view on this issue, then the premise inDescartes that the parts of time are distinct not only cannot lead to the common the-ological opinion that initial creation and subsequent conservation are the same act,but actually conflicts with that opinion insofar as it requires that at each moment anobject must be created anew by a distinct act.

As in the case of Thomas, however, I think we must take seriously Descartes’squalification that he is only quasi created anew.91 Descartes’s main point inspeaking of new creation is that God does not conserve by acting on preexistingmaterial. Rather, he conserves by producing the being of an object ex nihilo. Inthis respect Descartes is simply following the Suárezian line, albeit by using lan-guage that may suggest an anti-Suárezian conclusion. Moreover, Descartes’sargument in the letter to Hyperaspistes that God annihilates only by ceasing hisconcursus indicates that he accepted the Suárezian view that God conserves thisbeing by continuing the initial act by which he created it. Whatever separation


91. Or, as Descartes puts it in other texts, veluti reproduced; see Fourth Replies, AT 7:110;PP I.21, AT 8-1:13.

there may be of distinct temporal parts, it is not a separation that requires con-servation by a series of distinct creative acts.92

Though Bennett does not address this point about distinct acts, Gueroult acceptsthat there is for Descartes only a single indivisible creative act in God.93 But it isunclear how this single act could result in the series of distinct creations thatGueroult (anticipating Bennett) takes Descartes to posit. Since there is only one cre-ative act, it would seem that there should be only one effect. I submit that in the caseof Descartes, this single effect is simply the attribute of duration in the persistingsubstance. Since this attribute is not distinct in reality from the substance itself, Godproduces this attribute in initially producing the substance as a cause secundum esse.Conservation is just the continued production of the substance that yields the con-tinuing presence of its attribute of duration.

I have mentioned the possibility that Descartes could allow for derivative causessecundum fieri of the modes of the substances that God alone can create and con-serve. Perhaps then there could be such causes of the modal features of the durationthat constitute the different parts of time.94 But even if Descartes grants that there aresuch causes, he still must hold that they depend essentially on God’s activity as causesecundum esse. For the production of the relevant modes of duration presupposesthe existence of the attribute that these modes modify. But God alone can be thecause that produces this attribute given Descartes’s claim both that the attribute is notdistinct in reality from the substance to which it is attributed, and that God alone canproduce the existence of the substance as a cause secundum esse.

According to Descartes, then, the cause of the duration of a substance is just thesame as the cause that gives that substance its existence in the first place. Here wehave an endorsement of the received scholastic position in Suárez that God con-serves the world by means of the very same act by which he created it. This simi-larity to Suárez is admittedly obscured somewhat by Descartes’s references toindependent parts of time and to conservation as renewed creation. But Descartes’sview of the nature of time militates against the view that temporal parts are separatedin such a way that their production requires a separate divine act. And his talk ofrenewed creation merely reflects his position, which Suárez had anticipated, thatconservation as well as creation involves a production of being from nothing. Thisis why his conservation axiom says that the cause that conserves is as great as the


92. Thus I stand in opposition to Gabbey, who endorses the view in Gilson that “Descartesand Aquinas differ in their respective interpretations of God’s conservation of the world: forAquinas the conservation is a simple continuation of the initial creative act, whereas forDescartes it is a re-creation at each (independent) instant” (Gabbey 1980, 302, n.40, citingGilson 1925, 340–42). As indicated in the introduction, this “re-creationalist” interpretationof Descartes is found also in Smith 1902. Cf. the more recent claim in Secada 2000 that forDescartes “God initially created and then recreates at every instant all that there is” (105).

93. “And certainly, the various creations are really only one, since the creative act of Godis in itself one, and since it would be inconceivable for them to be separated by intervals oftime” (Gueroult 1953, 1:280).

94. Indeed, my suggestion in chapter 3 is that we can conceive of the forces that Descartesattributes to bodies in terms of such causes.

cause that creates. And it is clear that Descartes himself takes this axiom to show thatthe power by which God conserves is not merely the same type as, but also token-identical to, the power by which he creates. This further result turns out to be cen-tral to Descartes’s argument—to be considered in the next chapter—that Godconserves the total quantity not only of the substance of matter but also of the motionof its parts.

2.3. FROM AXIOMS TO CAUSATION

I hope it is clear from what I have said in this chapter why I would resist saying ofDescartes’s account of the causal axioms, what Bennett has said about his accountof the containment axiom, that it “seems not to reflect any deeply considered viewsabout the nature of causation.” Indeed, I think that what Descartes has to say showsthat his account is backed by a rich view of causation that is profoundly conditionedby, though also departs on important matters of detail from, the scholastic accountof efficient causality that Suárez articulated with great sophistication. Now, however,I want to make the more deflationary point—more congenial, perhaps, to Bennett—that Descartes’s axioms do not provide an adequate basis for attributing to him thesort of concurrentist position that Suárez offered against his competitors. This isbecause the axioms themselves are easily rendered compatible not only with acertain form of occasionalism, but also with a “mere conservationist” position inDurandus that was an important scholastic alternative to concurrentism.

To show the conceptual flexibility of the axioms, I start with the containmentaxiom. I have mentioned the objection to this axiom in the literature that there is inDescartes no clear account of how causes that do not formally contain the reality oftheir effects can contain them eminently. In contrast, the main problem for this axiomthat I have stressed concerns formal containment. Recall that this problem derivesfrom the conclusion in the Sixth Meditation that bodies formally contain what is pres-ent objectively in our sensory ideas. Given Descartes’s emphasis in this text on thefact that bodies are often “not entirely such as” we sense them, it seems that they donot satisfy his official definition of formal containment, according to which an objectis “such as we perceive.” My solution was to offer on Descartes’s behalf the view thatbodies are “such as” we sense them insofar as they possess those features with whichour sensory ideas are systematically correlated. Descartes assumes that causal rela-tions between the bodily features and the ideas make this correlation possible. But itis not entirely clear that the solution requires a causal relation. For it seems that thecorrelations could be present even if bodies were causally inefficacious; an occasion-alist could explain them, for instance, by appealing to the nature of divine action. Themain point here is not that Descartes should be tempted by this occasionalist alterna-tive. Rather, it is that showing that bodies satisfy the requirement of formal contain-ment in the problem case of the objective reality of sensory ideas does not suffice toreveal that bodies have the power to cause this objective reality. To draw this conclu-sion, we must consider issues concerning the nature of body and of its connections tothe mind that go beyond the containment axiom. Even if we assume that bodies havesuch a power, moreover, the containment axiom itself does not reveal whether the


effect derives from this power alone, or rather, as on Suárez’s concurrentist view,requires a special divine concursus (see §1.2.3 (ii)). What needs to be sorted out hereis how precisely Descartes’s suggestion that bodies are total causes of the objectivereality of sensory ideas is related to his conclusion that everything derives from Godas a “total and universal cause.”

In contrast, the conservation axiom takes a clear stand on the nature of divine causal-ity. In particular, this axiom requires that God’s conservation of the world involve an actthat does not differ in kind from his act of creating that world. Descartes’s applicationof that axiom reveals further his commitment to the stronger conclusion, previouslyaccepted by Suárez and other scholastics, that there is only a distinction in reason, andno distinction in reality, between God’s conservation of an object and his creation of thatobject. Whatever Descartes thought about secondary causation, then, he clearly indi-cated that the continued existence of the world requires that God act in a particular wayas the primary cause.

But though it says more about God’s causal role than the containment axiom saysabout the role of secondary causes, the conservation axiom is nonetheless neutral onissues regarding divine causality that were central to earlier medieval and laterscholastic debates over causality. All of the main parties to these debates agreed thatdivine conservation is necessary for the continued existence of the created world.The differences concern the precise nature of God’s continuing causal contribution.Suárez characterized the occasionalist as arguing that since God brings a creatureinto existence by determining all of its features, God alone can be a real efficientcause (MD XVIII.1, ¶2, 1:593). We have seen that to avoid this occasionalist con-clusion, Durandus sharply distinguished the conservation of the being of the createdworld from the production of changes in that world. He held that whereas God alonecan bring about the former, the created powers that God conserves suffice to bringabout the latter (see §1.1.3).95 However, we also know that in response to Durandus,Suárez offered the compromise position, previously proposed by Thomas, thatthough created powers cannot produce effects without divine assistance, God can actwith those powers in a manner that allows them to make a genuine causal contribu-tion to the effect (see §1.2.3 (ii)).

In this chapter I have argued that Descartes’s claim that God is the “universal andtotal cause” does not preclude the view that creatures can be total causes, and thusdoes not rule out Suárez’s concurrentist position. But though this claim may seem torule out Durandus’s mere conservationism, it is not evident to me that this is the case.In the passage from the letter to Elisabeth cited in §2.2.1, Descartes emphasizes thatGod’s universal causality differs from the universal causality of the sun, sincewhereas there are particular causes of the effects of the sun that are not subordinatedto the sun, there are no particular causes of effects in the created world that are notsubordinated to God. The notion of “subordination” could be understood in a con-currentist manner, as indicating a dependence on a divine concursus that acts withthe created cause. But the notion also could be given a weaker sense that is more inline with Durandus’s mere conservationism. That is, subordination could be taken to


95. For a contemporary statement of the mere conservationist position, see Quinn 1988.

indicate only a dependence on God’s creation/conservation of substantial being.God’s total causality extends to this kind of being as the causality of creatures doesnot, and since there could be no secondary causality if there were no substances,such causality depends essentially on God’s action as a cause secundum esse. Butthis sort of subordination does not require a concursus with the action of secondarycauses that goes beyond God’s act of conservation. It thus allows for the view inDurandus that divine conservation exhausts God’s causal contribution to the pro-duction of effects by secondary causes. In terms of the Thomistic distinction we haveconsidered at several points, the view here is that though God alone is the causesecundum esse of such effects, only the secondary causes cause them secundum fieri.

The previously noted comment in Descartes’s letter to Hyperaspistes that “nothingcan exist without God’s concursus” may seem to put him on the side of the concur-rentists. However, it is not evident that he sharply distinguished this concursus fromthe divine act of conservation. Moreover, there is even the possibility of understand-ing the concursus in an occasionalist manner, as the exclusive locus of causal activ-ity. On this understanding, the contributions of creatures with which God “concurs”are restricted to merely passive aspects of the causal situation. The conservationaxiom alone cannot determine which of these readings is correct, since the axiomcould be acceptable to occasionalists as well as to concurrentists and mere conserva-tionists. As in the case of the containment axiom, so here we must descend from theabstract heights of the conservation axiom to Descartes’s treatment on the ground ofparticular kinds of causal interaction to determine whether he intended to allow forreal secondary causes and, if so, how he thought such causes produce their effects.

The account of the causation of motion in Descartes’s physics would seem to bea good place to start. After all, we will discover that the account that Descartes pro-vides in the Principles relies explicitly on the consideration, connected to the con-servation axiom, that God’s act of conserving the material world is identical to hisinitial act of creating it. Moreover, Descartes indicates the importance of the con-tainment axiom for his account in this same text of the communication of motionwhen he insists there that such a account be governed by the rule “that we neverattribute to a cause any effect that exceeds its capacity [potentiam]” (PP II.60,AT 8 1:76). It turns out that Descartes’s discussion in the Principles of body–bodyinteraction provides the material for a response to the argument in the literature thathe could not have taken the communication of motion to involve real bodily causesgiven that his spare ontology requires the reduction of matter to mere extension. Buteven if it is granted that Descartes rejects an occasionalist view of body–body inter-action, there remains the question—which a consideration of the scholastic contextof his theory of causation serves to highlight—of whether he intends to side withconcurrentism against mere conservationism.


87

3

Causation in Physics

Suárez offered a fairly luxuriant ontology of the material world, with corporeal sub-stances composed of really distinct matter and substantial form, in which inhere thereally distinct forms of quantity and various qualities along with their modifications(see §§1.2.1 and 1.3). I have noted the attempt on Descartes’s part to impose onto-logical austerity by identifying corporeal substance with extension and by reducingall additional features of the material world to modes of extension. Since the scholas-tics appealed to the various qualities and forms to provide causal explanations ofchange in the material world, however, Descartes’s eliminativism leaves him withthe burden of providing an alternative explanation of such change. One possibility,which Daniel Garber has proposed, is that Descartes turns to God, using him “to dowhat substantial forms did for his teachers.” In this respect, according to Garber,Descartes is in fact “the last of the schoolmen” (Garber 1992, 305).

When we turn to Descartes’s Principles, we see that he does indeed claim that Godis the “universal and primary cause of motion.” However, he emphasizes in this textthat there are in addition “particular and secondary causes” of motion, which he iden-tifies with “rules or laws of nature” (PP II.36–37, AT 8-1:61–62). The immutabilityof God as universal and primary cause is supposed to explain the fact that the totalquantity not only of matter but also of the motion and rest of its parts is conserved. Incontrast, the laws as particular and secondary causes are supposed to explain why par-ticular changes in the distribution of motion follow on collisions among these parts.

In §3.1, I begin my consideration of Descartes’s account of causation in physicswith his treatment in the Principles of God’s universal and primary causality. I havementioned his view in other texts that all effects depend on God as “universal andtotal cause.” But in the Principles Descartes holds that the conservation of specificfeatures of the material world follows from the immutability of God’s “ordinary con-cursus.” Descartes adopts the position, found also in Suárez, that the quantity of

matter is constant through all natural change. However, he further defends the posi-tion, rejected by Suárez and other scholastics, that such change does not alter the totalquantity of motion and rest. The discussion in the Principles focuses most on the con-servation of motion, and in later correspondence with the Cambridge Platonist HenryMore, Descartes develops the position that what is conserved in this case is the“force” that the quantity of motion measures.1 Though Descartes has less to say aboutthe conservation of rest, his suggestion is that it is only the quantity of matter at rest,and not the force involved in rest, that is conserved. Thus, he cannot treat the conser-vation of rest in precisely the same way that he treats the conservation of motion.Even so, Descartes insists that in both cases the conservation of the total quantity fol-lows from the premise that God’s act of conservation does not differ in reality fromhis initial act of creation. Drawing on my discussion of this premise in the previouschapter, I argue that Descartes ultimately takes the conservation of the various quan-tities to involve a continuation of God’s creation of matter with the motive force heinitially infused into it.

There remains the tangled problem of the ontological status of the various bodilyforces that Descartes posits in his physics. In §3.2, I attempt to unravel this problemby considering the meaning of Descartes’s claim in the Principles that his three lawsof nature serve as particular and secondary causes of motion. I offer a reading of thisclaim on which these laws are grounded in bodily forces that are themselves thecausal source of changes in motion. If this reading is correct, Descartes’s consideredview is that the nature of bodies is not exhausted by the purely geometric and kine-matic aspects of extension. Rather, this nature consists in an extension that has aduration that serves to ground the various forces and inclinations that causally deter-mine the particular motions that bodies possess. An understanding of force in termsof these durational tendencies reinforces the account of divine conservation consid-ered in §3.1. For on that account God conserves the total quantity of motive forcesimply by creating various parts of matter with various tendencies to continue toexist that together have a constant measurable quantity. There are admittedly diffi-culties for an explanation of bodily force in terms of features of bodily duration, butthe difficulties derive not from Descartes’s doctrine that the nature of body consistsin extension, but from his account of the specific kind of force associated with thebodily state of rest.

In defending this reading of Descartes’s view of body–body interaction, I sidewith commentators who reject an occasionalist interpretation of his physics of thesort that Garber offers. Given Descartes’s own reference to the conservation ofmotion and rest by means of God’s “ordinary concursus,” it is tempting to think thathe models his account of the relation of divine activity to bodily causes on the mainscholastic alternative to occasionalism, which appeals to God’s concurrence with theoperations of secondary causes. However, I argue in §3.3 that we need to resist thistemptation. A scholastic account of concurrence of the sort that we find in Suárezsimply does not fit the view in the Principles of the relation between the universal


1. For a discussion devoted to Descartes’s exchange with More on this issue, see Cottingham1997.

and particular causes of motion. In Descartes’s text, God’s universal causation ofmotion is to be understood in terms not of a divine concursus that varies with theparticular actions of bodily causes, but rather of God’s continuous conservation ofthe same total quantity of the motive forces responsible for particular changes inmotion. Ultimately, Durandus’s mere conservationist position provides a more suit-able model than Suárez’s concurrentism for the causal realism Descartes’s physicsrequires.

3.1. GOD AS UNIVERSAL AND PRIMARY CAUSE

In the second part of the Principles, Descartes offers an account of motion that relieson the distinction between motion in the strict sense (motus) and the “force oraction” (vis vel actio) that is the cause of motion. Whereas in earlier writings heridiculed the attempt to define motion,2 he here takes motion, as opposed to the forcethat brings it about, as the “transference [translationem] of one part of matter, or onebody, from the vicinity of bodies immediately contiguous to it and considered asbeing at rest in the vicinity of others” (PP II.25, AT 8-1:53).3 Though it has beenclaimed that this definition of motion is purely relativistic insofar as it does not allowfor a genuine distinction between motion and rest,4 it seems fairly clear thatDescartes himself recognizes such a distinction, and indeed that his account of col-lision in the Principles depends on it. For on the rules that this text offers, the out-comes of collisions differ in particular cases depending on whether one of thecolliding bodies is in motion or at rest.5 The need for a nonrelativistic understandingof motion is even more evident given the interpretation in §3.2 of Descartes’s accountof motive force.

Causation in Physics 89

2. See, for instance, RM XII, AT 11:426–27; To Mersenne, 16 Oct. 1639, AT 2:597. In thesetexts, Descartes ridicules the scholastic definition of motus as an actus entis in potentia, proutin potentia est (act of being that is in potency, insofar as it is in potency). Cf. W VII, AT 11:39.

3. Cf. his earlier view in The World that motion is that “by which bodies pass from oneplace [lieu] into another, successively occupying all the spaces [espaces] in between” (W VII,AT 11:40). In the Principles, Descartes intentionally avoids defining motion in terms of place,which he there distinguishes into the “internal place” (locus internus) that is identical to theextension of a part of matter and the “external place” (locus externus) that is an abstractlyconsidered relation among parts of matter; see PP II.10 and 13, AT 8-1:45 and 47.

4. See, for instance, Blackwell 1966 and Prendergast 1972.5. In particular, in the case where a larger body collides with a smaller body, the outcomes

differ depending on whether the larger or smaller body is at rest. According to the fourth rule,in the former case the smaller body is reflected with its original speed while the larger bodyremains at rest, whereas according to the fifth rule, in the latter case both bodies move togetherin the direction of the motion of the larger body (cf. note 28). For the point that these rules areincompatible with the view that the distinction between motion and rest is arbitrary, see Garber1992, 240–41. See also the conclusion in Blackwell 1966, 226–27, that the rules are inconsis-tent with Descartes’s relativistic account of motion. Whereas Blackwell takes this inconsistencyto tell against the rules, I take it to tell against the attribution of such an account to Descartes.

After the discussion in the Principles of the nature of motion as mere translatio(though, I would claim, a real bodily mode), Descartes turns to a consideration ofthe cause of the separation of parts of matter. He begins by noting that whereasGod is the “universal and primary cause” that is responsible for “all motions in theworld,” there are “particular causes” that are responsible for the fact that “singularparts of matter acquire motion they did not have previously” (PP II.36, AT 8-1:61).We know that the distinction between primary and secondary causes is a tradi-tional one, reflected in Suárez’s distinction between a “first” principal cause that“operates altogether independently,” namely, God, and a secondary principal causethat “is dependent, even if it operates by a principal and proportionate power [vir-tute]” (MD XVII.2, ¶20, 1:591; see §1.2.3 (ii)). As we will discover, however,Descartes’s use of this distinction is anything but traditional. In §3.2, I consider hisdistinctive account of the “secondary and particular causes” of specific changes inmotion. At present, though, the focus is on his understanding of the manner inwhich God serves as the universal and primary cause of matter as well as of itsmotion (and rest).

3.1.1. Universal and Total Causes

In the previous chapter, I noted the claim in Descartes that God is the “universal andtotal cause” of everything that occurs in nature (see §2.2.1). There I defended a read-ing on which such a claim endorses not the occasionalist conclusion that God is theonly cause of natural effects, but rather the more modest conclusion that all othercauses of natural effects are essentially subordinated to God’s universal causality.This modest conclusion would have been acceptable even to someone as critical ofoccasionalism as Suárez, who allowed that all secondary causes depend on God’sactivity as primary cause.

It is perhaps tempting to think that when he refers in the Principles to the fact thatGod is the universal cause responsible for all motions, Descartes has in mind the factthat as in the case of all other natural effects, motions depend on God’s universal andtotal causality. There is indeed one significant point of contact. I indicated previouslythat for Descartes, God’s universal and total causality involves his exclusive produc-tion of substantial being ex nihilo. One sense in which all other causes are subordi-nated to God is that the activity of all such causes depends on God’s creation andconservation of this kind of being. Likewise, Descartes stresses in the Principles thatby means of God’s universal causality “the world is now continuously conserved bythe same action that created it then” (PP II.42, AT 8-1:66). This universal causation,like the universal and total causality that Descartes elsewhere ascribed to God,involves the divine creation/conservation of the world.

Even so, it is clear from the discussion in the Principles that Descartes has inmind a thicker notion of universal causality than the notion employed in other texts.For elsewhere his claim is merely that all other effects are subordinated to God. Inthe Principles, however, he needs to derive the stronger conclusion that specific fea-tures of the material world are conserved. In particular, Descartes’s claim is that theuniversal cause of motion is


nothing other than God himself, who in the beginning created matter at the sametime with motion and rest [cum motu et quiete], and now, by his ordinary concur-sus [concursum ordinarium] alone, conserves the total quantity of motion and rest[motus et quietis . . . tota quantum] as he placed in it then. (PP II.36, AT 8-1:61)

The result here that there is a constant “total quantity of motion and rest” could notfollow simply from the fact that these quantities derive from God’s “universal andtotal” causality. For all that follows from this fact is that whatever quantities thereare depend essentially on God’s independent action. Nor does the fact that God iswholly immutable by itself show that the quantities are conserved. For Descartesholds that “there is always a single identical and perfectly simple act by means ofwhich [God] simultaneously understands, wills and accomplishes everything”(PP I.23, AT 8-1:14). Since this one immutable act produces a changing temporalworld, why could it not also produce changing quantities of motion and rest?

What Descartes actually argues is that the conservation of the quantity of motion andrest follows not simply from the fact that God acts immutably, but rather from the fur-ther assumption that God always acts on the material world “by his ordinary concursusalone.” We will need to puzzle about the precise nature of this ordinary concursus. Butfirst let us consider precisely what sorts of quantities this divine action is supposed toconserve.

3.1.2. Conserved Quantities

In the Principles, Descartes claims that “God moved parts of matter in various wayswhen he first created it, and now conserves the whole of this matter in the same wayand with the same plan [eademque ratione] by which he first created it” (II.36, AT8-1:62). There are three sorts of quantities that Descartes takes to be conserved inthis text: first, the total quantity of matter; second, the total quantity of motion in thisworld, which involves also the total quantity of the force associated with motion; andthird, though not as obviously from the passage just cited, the total quantity of rest,which ultimately cannot include the force associated with rest. Let us consider thesequantities in turn.

(i) Quantity of Matter

Prior to his discussion in the Principles of God’s universal causation of motion,Descartes argued that there is only one kind of matter that occupies all imaginablespace, and that this matter is simply extended substance itself (PP II.22, AT 8 1:52).Earlier in this text he also had noted that there is no distinction in reality betweenthis extended substance and its quantity (PP II.9, AT 8-1:45).6 The fact that thisquantity is naturally conserved follows further from Descartes’s claim that any sortof natural change in this quantity can occur only in virtue of “the divisibility and


6. Cf. the view in the Fifth Meditation that matter can be clearly and distinctly conceivedonly when conceived as identical to “quantity, which the common philosophers call continu-ous” (AT 7:63).

mobility in respect of its parts, and its resulting capacity to be affected by all theways that we perceive to be possible following the motion of its parts” (II.23, AT8-1:52). So whereas quantity can be divided into different parts that can be trans-ferred from each other in different ways, such changes in division and motion donot involve any absolute creation or annihilation of quantity itself.

For reasons indicated below, I take Descartes’s claim that the quantity of mat-ter is conserved when God acts by his ordinary concursus to indicate that it is con-served in all natural (i.e., nonmiraculous) interactions. The issue of whether sucha conclusion is acceptable in a scholastic context is complicated by the fact thatamong the scholastics, all but the nominalists departed from Descartes’s officialposition that these two are distinct only by reason and not in reality.7 For scholas-tic realists, including Suárez, quantity is an accident that can exist apart both fromthe material substance composed of prime matter and substantial form, and fromprime matter itself. However, most scholastics could agree that prime matter can-not be produced or destroyed by natural means. For they took this matter to be thesubstrate presupposed by all material processes, including the generation and cor-ruption of material substances, and so not something that can be produced ordestroyed by these processes. Among the Thomists, this argument for the conser-vation of prime matter could not be extended to the case of quantity. Given theirview that quantity is an accident that naturally inheres in, and thus is bound to, aparticular material substance, the replacement of one such substance by another(for instance, in the case where a living body becomes a corpse) involves thereplacement of one quantity by a qualitatively identical but numerically distinctquantity.8 However, Suárez departs from Thomistic orthodoxy in holding thatquantity inheres directly in prime matter rather than in the composite material sub-stance.9 Moreover, he insists that quantity cannot be “per se produced de novo, orconcomitantly, so that what was not now begins to be through the actions of natu-ral agents” insofar as “quantity is coeval with matter” (MD XVIII.4, ¶3, 1:624).This is so because quantity is a “fundamental and radical property” that resultsdirectly from the being of the prime matter itself (MD XIII.14, ¶¶15–16,1:459–60*). Despite disagreeing with Descartes’s view of the relation of quantityto material substance, then, Suárez, at least, could accept his conclusion thatquantity can be neither produced nor destroyed by natural means.


7. As Suarez describes it, the nominalist position is that “bulk quantity” (quantitas molis)is “not a thing distinct from substance and material qualities. Rather, the being [entitas] ofeach of them by itself has that bulk and extension of parts that is in bodies; that being is called‘matter’ insofar as it is a substantial subject, and ‘quantity’ insofar as it has extension and adistinction of parts” (MD XL.2, ¶2, 2:533*). Suárez lists as advocates of this position PeterAureoli, Ockham, Gabriel Biel, Adam of Wodenham, John Major, and Albert of Saxony.

8. For discussion of the Thomistic position on this point, see Des Chene 1996, 146–47.9. The main point at issue was whether prime matter has some sort of reality apart from

form, as Suárez claimed, or whether such matter is purely potential and has no such reality, asthe Thomists insisted; see MD XIII.4, 1:409–14*. Cf. the metaphysical differences betweenSuárez and the Thomists considered in §1.2.1.

(ii) Quantity of Motion

Though the result in Descartes that quantity is conserved was not entirely foreign toSuárez, there is no analogue in his thought of Descartes’s position in the Principlesthat the total quantity of motion remains constant in all natural change.10 EvenDescartes recognizes that the case of the conservation of the quantity of motionseems to differ from the case of the conservation of the quantity of material sub-stance insofar as motion is a mere mode and not an ultimate subject of change. Henonetheless insists that each token of the modal type of motion has a certain quan-tity that combines with the quantities of all the other tokens of that type to constitutea total quantity of motion “in the universe as a whole” (PP II.36, AT 8-1:61). Forhim, there is as good an argument for the conservation of this modal quantity as thereis for the conservation of quantity as such.

The Jesuit order in effect rejected the equivalence of these two cases in a formalprohibition of various Cartesian propositions that its general, MichelangeloTamburini, issued in 1706 (Rochemonteix 1899, 4:89–93). Among the thirty con-demned propositions, the tenth asserted that “in order to admit that some quantity ofmotion that God originally impressed on matter is lost, one would have to assumethat God is changeable and inconstant,” whereas the sixteenth claimed that “there is,in the world, a precise and limited quantity of motion, which has never been aug-mented or diminished.” Thus it became official Jesuit doctrine that a natural changein the quantity of motion is possible, and that such a change is perfectly compatiblewith divine immutability.

As is often the case with formal condemnations, there is room to argue that thepropositions condemned do not match precisely the views of the condemnedauthor.11 Particularly in the case of the sixteenth proposition, there are certain claimsthat go beyond what Descartes actually says in the Principles. For instance,Descartes never claims in that text that the total quantity of motion is “limited.” Moreimportant, he argues there not that there has been no change in the quantity ofmotion, but merely that there can be no such change given the assumption that Godacts only by his ordinary concursus. Indeed, Descartes explicitly allows in this sec-tion of the Principles for changes in the quantity of motion “that evident experienceor divine revelation renders certain” (PP II.36, AT 8-1:61). Presumably, Descartes


10. For a discussion of the novelty of Descartes’s views concerning the quantity ofmotion, see Dubarle 1964. But see note 25.

11. Admittedly, some of the propositions listed in the Jesuit condemnation most likely weredrawn more from certain followers of Descartes than from Descartes himself. This is so, forinstance, in the case of the nineteenth proposition that “only God can move bodies; angels,rational souls, and bodies themselves are not the efficient causes, but the occasional causes ofmotion,” and in the case of the twentieth proposition that “creatures do not produce anything asefficient causes, but God alone produces all effects, ad illarum praesentiam.” There is reason tothink that Descartes was not the main target here, but rather later Cartesian occasionalists, andmost especially Malebranche, who made claims very close to those found in these propositions.Even so, the seventh and sixteenth propositions seem clearly to be drawn from Descartes’s dis-cussion in the second part of the Principles of God’s universal causation of motion.

does not want to claim that either evident experience or divine revelation compro-mises divine immutability.12

Nevertheless, from a Jesuit perspective even the more modest implication inDescartes that divine immutability is incompatible with natural change in the quan-tity of motion is bad enough. To be sure, most Jesuit scholastics accepted the viewthat the circular motions of the celestial spheres are constant.13 However, suchscholastics, including Suárez, insisted that sublunar motion can be corrupted by nat-ural means.14 Indeed, on the traditional Aristotelian view this sort of corruption isinevitable given that the natural state of body is to be at rest in its proper place. Thus,all bodies naturally resist motion from such a place, and this resistance is the sourceof the corruption of motion. Nor, for Suárez and other scholastics, does this sort ofcorruption bespeak some inconstancy in God, for the change derives from thenatures of the bodies that God immutably creates and conserves.

There is the influential claim in the work of Pierre Duhem that the modern empha-sis on the conservation of motion has its roots in scholastic thought. In particular,Duhem takes scholastic discussions of projectile motion to provide the basis for theconclusion in Descartes and other seventeenth-century mechanists that motion is con-served in all natural change (Duhem 1955, vii, 49). Aristotle claimed that the moverof the projectile imparts a portion of its motive force to the medium, with the impartedforce keeping the projectile in motion until it is depleted. However, fourteenth-century scholastics developed the view of earlier Aristotelian commentators that themover impresses an “impetus” on the projectile itself. What Duhem emphasizes asthe decisive moment on the road to the modern thesis of the conservation of motionis the proposal in the work of the scholastic master John Buridan that impetus in bod-ies is “something permanent by nature” (see Duhem 1955, 34–53).15

So do we not have here in the scholastics something akin to Descartes’s permanentquantity of motion? Anneliese Maier shows that we do not. In response to Duhem,she notes that Buridan proposed the permanence of impetus only in the case of celes-tial motion, where there is no resistance to the impetus. Buridan allowed that in thecase of terrestrial motion, impetus is opposed by contrary tendencies in bodies, whichlater scholastics identified with natural resistance to any violent motion (Maier 1982,89–91).16 Suárez was among those who accepted the theory of impetus, but also


12. It is true, however, that in the earlier World Descartes suggested that no change innature can be attributed to God given the immutability of his action; see W VII, AT 11:37. Forfurther discussion of the clause in the text from the Principles that exempts changes revealedby evident experience or divine revelation, see §4.3.3.

13. For a discussion of scholastic views on this matter, which derive from the work ofAristotle and the Neoplatonists, see Menn 1990, 218–26.

14. Thus, at one point Suárez argued that the motion of a projectile can be corrupted bythe weakening of its quality of impetus due to the resisting action of the heaviness of thatobject (MD XXI.3, ¶27, 1:801). I discuss presently the nature of this quality.

15. The proposal, from the twelfth question of Buridan’s Questiones octavi libri physico-rum, is cited in Duhem 1955, 44.

16. For the Aristotelian, violent motion is simply motion that does not have as its termi-nus rest in the natural place of the body that is moving.

stressed that when an impetus is impressed on a body, “there is no requirement that itbe conserved there permanently” and that “because in other respects the subject of thequality is always resisting it and its action, the nature of such a quality requires thatit should stop being conserved little by little” (MD XXI.3, ¶27, 1:801).17

Descartes’s view in the Principles that the total quantity of motion cannot bealtered by natural means thus involves a marked departure from the view of ascholastic such as Suárez. Descartes introduced such a view already in The Worldwhen he referred to God’s conservation of the “quantity of motions” (quantité desmouvemens) (W VII, AT 11:43), though he did not indicate there what sort of quan-tity he had in mind.18 By the time of the Principles, however, he settled on the viewthat motion is a state that has a quantity measured by the product of the size (i.e.,volume) of a moving body and the scalar speed of its motion.19 The argument in thistext is that not only the total quantity of matter but also the total quantity of all of themotions of its parts are conserved insofar as God acts by his ordinary concursus.

Nevertheless, there is a complication for Descartes’s account of the quantity ofmotion that is relevant to scholastic impetus theory. For Suárez and other scholasticproponents of this theory, the impetus of a projectile is a quality that is distinct fromthe motion of this object insofar as it is the cause of that motion. But Descartes alsoseems to allow that there is more to conserved motion than its quantity. Thus hewrites in 1641 that one must consider a rock that collides with the earth as having“motion, or the force to move itself [la force a se mouvoir] as a quantity that is nei-ther augmented nor diminished” (To Mersenne, 17 Nov. 1641, AT 3:451).20 There isalso the view in the 1644 Principles that there is a certain “quantum of force [quan-tum . . . virium], either to move, or to resist motion” (PP II.45, AT 8-1:67). Finally,Descartes claims in his 1649 correspondence with More that strictly speaking it isnot the various changing modes of matter that remain constant, but only “the forceimpelling its parts” (vis eius partes impellente), a force that “applies itself now toone part of matter, now to another” (To More, Aug. 1649, AT 5:405).21

There is the question whether Descartes’s conserved force, like scholastic impe-tus, is a cause of motion that is distinct from the motion it produces. I need to


17. It is worth noting that in this passage Suárez spoke of the reduction of impetus asinvolving a reduction not in its quantity, but rather in its intensity (intensio). This serves todistinguish impetus from quantity strictly speaking given his official position that quantity assuch cannot have intensity (MD XLI.5, ¶6, 2:601*).

18. For the suggestion that the use of the plural indicates that Descartes is thinking herethat God conserves the number of motions or the number of bodies in motion, see Costabel1967, 250–51.

19. See his claim that “if one part of matter moves twice as fast as another that is twice aslarge, we consider that there is the same quantity of motion in each part” (PP II.36, AT8 1:61). Prior to this time Descartes corrected his reference in The World to the quantity ofmotions by speaking of a certain total quantity of all motions; see To Debeaune, 30 Apr. 1639,AT 2:543; To Mersenne, 17 Nov. 1641, AT 3:451.

20. Even earlier, he wrote in The World that “the ability [vertu] or power [puissance] tomove itself that is found in body cannot entirely cease to be in the world” (W III, AT 11:11).

21. See the discussion in §3.2.1 (iii) of this passage from the More correspondence.

postpone consideration of this question until the discussion below of his account of“secondary and particular causes.” But at least it can be said at this point that inDescartes, the measure of the total quantity of motion, namely, the sum of the prod-ucts of the sizes and speeds of each moving part of matter, also serves to measurethe total quantity of the force impelling these parts. Moreover, his assumption seemsto be that if the total quantity of the force is conserved, then so is the total quantityof the motion associated with the force.22

Though the result that the total quantity of motion and its force are conserved isradical enough given the scholastic context, it is important to recognize how limitedthis result is. For instance, such a result does not by itself preclude action at a dis-tance. As long as this kind of action does not produce an increase or decrease in thetotal quantity of motion, the principle of conservation is satisfied.23 Moreover, theresult that quantity of force/motion is conserved says nothing about the directionalityof the bodily motions. Several incompatible scenarios concerning changes in direc-tions of bodies subsequent to collision, or even to action at a distance, seem to beconsistent with this result. I do not think that Descartes himself has any illusions con-cerning the robustness of his conclusions regarding the conservation of motion/force.Indeed, I argue in §3.2 that he invokes particular and secondary causes precisely tounderwrite claims concerning the necessity of contact action and the constraints onthe direction of motion that his conservation principle alone does not yield.

(iii) Quantity of Rest

In his discussion of God’s universal causality in Principles II.36, Descartes emphasizesthe conservation of motion. He begins this article by referring to “the general cause ofall motions that are in the world,” and he ends it by insisting that God “always con-serves just as much motion in [matter]” as he created in it initially (AT 8-1:61–62).Even so, he makes a passing reference in this same article to the fact that God con-serves the total quantity of rest (tota quantum quietas). Unfortunately, in contrast to thecase of motion, he does not specify how the relevant quantity is to be measured. It isperhaps tempting to think that he takes the same formula that applies to the case ofmotion, namely, size times speed, to apply to the case of rest as well. But if we applythe formula in this way, the result of course is that the quantity of rest is nill.24


22. As Garber observes, there is the suggestion in a note from Descartes, dating perhapsfrom the 1630s, that the uniform application of force (vis) results in uniform acceleration (AT11:629–30*, cited in Garber 1992, 354–55, n.17). I concur in Garber’s judgment that this sug-gestion is not in line with Descartes’s later account of force (though in §3.2, I provide an inter-pretation of this later account that differs from the one that Garber defends).

23. As Des Chene notes, Descartes’s conservation principle seems to allow for God toproduce changes “that to us would look like action at a distance,” providing no reason “whybodies in motion and about to collide should not be brought to a standstill, and their motiondistributed to others” (Des Chene 2000b, 149).

24. In his Search after Truth, Malebranche offers an account of rest on which it is this sortof limiting case of motion, though he does so in direct opposition to Descartes’s account of the“force of rest”; see bk. VI-2, ch. 9, Malebranche 1958–84, 2:428–29/Malebranche 1997, 515.

A more promising proposal is that the total quantity of rest is the same as the totalquantity of the parts of matter at rest. To conserve the total quantity of these parts istherefore to conserve the total quantity of their rest. On this understanding, the claimthat the total quantity of rest is conserved would be simply a corollary of the doc-trine of the conservation of the total quantity of matter. The clearest evidence thatDescartes himself understood conservation of rest in this manner is found not in thePrinciples but rather in his correspondence with More. More initially asked if“matter, whether we imagine it to be eternal or created yesterday, left to itself andreceiving no impulse from anything else, would move or be at rest?” (More toDescartes, 5 Mar. 1649, AT 5:316). Descartes at first ignored the question, but whenMore repeated it (More to Descartes, 23 July 1649, AT 5:381), he is forced torespond. Descartes writes, with respect to More’s reformulation of his query, that“I consider matter left to itself and receiving no impulse from anything else, asplainly at rest. But it is impelled by God, conserving as much motion or transference[translationem] in it as he put there from the first” (To More, Aug. 1649, AT 5:404).So the suggestion is that God produces a particular quantity of rest simply by pro-ducing a particular quantity of the matter and then “leaving it to itself.” This is incontrast to his production of a particular quantity of motion, which involves not onlyhis producing a particular quantity of matter but also his adding to that quantity aparticular amount of impulse, that is to say, force.25

So far, then, the quantity of the rest of a body seems to reduce to the quantity ofthe matter of that body. But there is a joker in the deck. Just as Descartes posits aforce or impulse in the case of motion, so he holds that there is a force associatedwith rest. Thus, in a 1640 letter to Mersenne, he notes that

from the fact alone that a body begins to move, it has in itself the force tocontinue to move [il a en soy la force de continuer à se mouvoir]; just as, from thefact alone that it is stopped in a certain place, it has the force to continue toremain there [la force de continuer à y demeurer]. (To Mersenne, 28 Oct. 1640,AT 3:213)

Later, in the Principles, he holds that the effects of bodily collisions are determinednot only by a “force for acting” (vis ad agendum) in parts of matter that are inmotion, but also a “force for resisting” (vis ad resistendum) that is found in anymaterial parts in such collisions that are at rest (PP II.43, AT 8-1:66–67).26 So justas motion is not simply a mode with a certain quantity but also involves force, so, itseems, rest must involve a force in addition to its quantity. Even in the letter to Morein which he suggests that God produces motion by simply producing matter without


25. The view here that motion requires a continually conserved impulse from God showsthe need to qualify Dubarle’s claim that Descartes’s originality lies in his conception ofmotion “as a fact of the nature of the universe, no longer as an action deriving causally froma mover external to the thing moved” (Dubarle 1964, 121). Even so, I argue in §3.2 that thereis a sense in which motion is grounded in something internal to the moving body (even if orig-inally impressed and subsequently conserved there by God). I owe the point about the needfor qualification, though not the more concessive point, to Garber 1992, 362, n.28.

26. In §3.2.2, I propose an account of the nature of these bodily forces.

impulse, Descartes notes that “the resting thing, from the fact alone that it rests, hasthis effort [renixus], [but] this effort is not therefore rest” (To More, Aug. 1649, AT5:403). The “effort” that Descartes mentions here, which More identified with resist-ance to motion,27 therefore is something that he took to be distinct from rest itself.If, as seems likely, Descartes identifies this effort with the force for resisting that heposited in the Principles, then he has good reason to distinguish it from the state ofrest. For whereas he claims in this text that the total quantity of rest is conserved, itturns out that his account there of the force for resisting precludes him from holdingthat the quantity of this force is conserved as well.

To see that Descartes cannot hold that this force is conserved, we need to considerhis discussion in the Principles of the seven rules governing the effects of bodily col-lisions.28 The force for resisting comes into play only in the case of the fourth andsixth rules, which concern the collision of a moving body with a body at rest that iseither larger than or equal in size to it.29 The suggestion in the Latin edition of thistext, at least, is that this force is to be measured by the product of the resting bodyand the speed of the moving body that collides with it.30 Such a force is present onlyin the case of collision, and it ceases to exist after it has acted. In this sense, it is akinto the vis impressa that Newton later introduces in his Principia, which consistssolely “in the action exerted on a body to change its states” and that “does not remainin the body after the action has ceased” (Newton 1999, 405). In contrast, the forcefor acting in the moving body not only acts in collision, but also maintains the bodyin its motion. This force therefore has aspects of Newton’s vis inertia, which explains


27. More understood Descartes to hold that “rest is an action, truly a certain effort orresistance [renixum sive resistemtiam]” (23 July 1649, AT 5:380*).

28. The original version of the seven rules is set out in the Latin edition of the Principles(PP II.46–52, AT 8-1:68–70). The rules there are as follows: the first covers the collision ofbodies moving in opposite directions with equal sizes and speeds, the second covers the col-lision of a body with a smaller body moving in the opposite direction with the same speed, thethird covers the collision of bodies moving in opposite directions with equal sizes but differ-ent speeds, the fourth covers the collision of a moving body with a larger body at rest, the fifthcovers the collision of a moving body with a smaller body at rest, the sixth covers the colli-sion of a moving body with a body at rest that is equal in size to it, and the seventh covers thecollision of bodies with varying sizes or speeds that are moving in the same direction.

29. There is a much-expanded version of the rules that cover collisions involving bodiesat rest (viz., the fourth, fifth, and sixth rules) in the 1647 French edition of the Principles, atAT 9-2:90–92, which was preceded by a reconsideration of such collisions that Descartesoffered in a 1645 letter to Clerselier, at AT 4:183–87. For more on the evolution of Descartes’sviews concerning such collisions, see Garber 1992, ch. 8. See also note 30.

30. Cf. Gabbey’s view, drawing on the version of the collision rules in the 1647 Frenchedition of the Principles, that the force for resisting is proportional not to the speed of themoving body but rather to the quantity of motion that the resting body would receive from themoving body were the later able to impose it on the former (Gabbey 1980, 269–70). I concurin Garber’s judgment that this account of resisting force cannot be found in the original Latinedition of this text (Garber 1992, 358, n.16). Even so, I think that the points I want to makego through even on the account of the force for resisting that Gabbey attributes to Descartes(as Gabbey himself seems to recognize; see note 32).

the fact that a body perseveres in its state (Newton 1999, 404).31 Whereas Descartescan hold that the force for acting has a quantity that is continuously conserved, hecannot say the same about the force for resisting. As I argue below, this differencereveals that the metaphysical foundations for his physics are much less amenable toforces for resisting than they are to forces for acting.32

3.1.3. Conservation and Ordinary Concursus

We can now turn to the claim in the Principles that the conservation of the totalquantity of matter, along with the total quantity of the motion and rest of its parts,follows from the assumption that God acts subsequent to creation “by his ordinaryconcursus alone” (PP II.36, AT 8-1:61). I claimed previously that the reference hereto ordinary concursus indicates that there is conservation in all natural change. Thisimplication pertains less to the fact that God acts by concursus than to the fact thatthis concursus is ordinary. There was a common scholastic distinction betweenGod’s “absolute” and “ordinary” (or “ordained”) power. Drawing on the account ofthis distinction in Thomas, Suárez held that God’s absolute power (potentia abso-luta) is his power to affect anything “apart from any respect toward the nature ofthings or toward other causes,” whereas his ordinary power ( potentia ordinaria) isinvolved in his action “according to the common laws and causes that he has estab-lished universally” (MD XXX.17, ¶32, 2:216*, citing ST I.25.5, ad 1). It is becauseGod’s power is absolute in this way that he can miraculously produce logically pos-sible effects that do not follow from the natures of objects in the created world. Incontrast, God produces all effects that follow from these natures by means of hisordinary power. By excluding from consideration any changes in the material worldguaranteed “by divine revelation” (PP II.36, AT 8-1:61), Descartes indicates as wellthat God’s ordinary concursus excludes miraculous action.33


31. Newton adds that inertial force explains perseverance of the body in its state quantumin se est. We will discover presently that this qualification is important for Descartes’s ownconception of bodily force.

32. Cf. Gabbey’s claim that whereas “the force maintaining a moving body’s motion is the sameas the force with which it resists change of direction or acts to change the state of other bodies,” theforce maintaining rest must differ from resisting force, since the former “appears to have no mean-ingful measure analogous to that of the force maintaining motion,” but resisting force “does have ameasure in terms of the motion that the striking body ‘tries’ to impart to it” (Gabbey 1980, 267–68).Though I emphasize the difficulties of accommodating resisting force within Descartes’s account ofGod’s conservation of motion and rest, Gabbey himself takes the collision rule that introduces thisforce to be “the most seminally valuable of the seven” and the force itself to take us “half-way to thefully Newtonian conception [of resisting force]” (Gabbey 1980, 269–70).

33. The further exclusion of changes guaranteed “by our own experience” most likelyindicates a bracketing of changes in motions produced by finite minds. See, for instance,Descartes’s claim to Arnauld that the fact that an incorporeal mind can set body in motion isrevealed “by the most certain and most evident experience [certissima & evidentissima expe-rientia]” (29 July 1648, AT 5:222). In §4.3.3, however, I indicate that he has problemsallowing for the production of a new quantity of motion by finite minds.

The fact that a consideration of the scholastic context sheds some light on the ordi-nary nature of the concursus that Descartes takes to underwrite divine conservationmay lead us to believe that this same context can help in the explication of his conceptof concursus. After all, we know that ‘concursus’ was a scholastic term of art. I arguein §3.3 that there are in fact good reasons to resist an interpretation of Descartes’saccount of God’s causation of motion in terms of a standard sort of scholastic concur-rentism. However, it is worth noting here that from the perspective of a traditionalscholastic concurrence theory, there are initial difficulties with aspects of his discus-sion of concursus.34 The structure of Suárez’s Metaphysical Disputations makes clearthat divine conservation and concursus involve different aspects of the creature.Disputation XXI is devoted to the essential dependence of the being of creatures onGod’s conservatio, whereas disputation XXII is devoted to the essential dependence ofthe operation of creatures on God’s concursus (see §1.2.3). Even though creaturescould not act without God’s concursus, in Suárez’s view, they could continue to exist.He indicated that in the biblical miracle of the three men in the fiery furnace, God con-tinued to conserve the fire but prevented it from burning the men by withholding hisconcursus with the action of the fire (MD XXII.1, ¶11, 1:804). To annihilate the fire,God would need to withhold not only this concursus but also the conservatio by meansof which the fire remains in existence.

In the article of the Principles that concerns God’s universal causation of motion,however, there is no corresponding distinction between conservatio and concursus.Descartes indicates in this article that God’s ordinary concursus is to be explicatedin terms of his conservation of matter “in the same manner and according to the samereason that he first created it” (PP II.36, AT 8-1:62). Moreover, Descartes empha-sizes in a later article that a portion of his third law of nature is “demonstrated by theimmutability of the operation of God, now continually conserving the world bythe same action with which he created it then” (PP II.42, AT 8-1:66). As indicatedpreviously, Descartes adopts the Suárezian position that the divine conservatio isidentical to God’s initial creation of an object (see §2.2.3). In suggesting that ordi-nary concursus is identical to God’s initial act of creation, then, he seems to indicatethat it is identical as well to continued divine conservation.

The identity of conservation and concursus is further reinforced by Descartes’sremarks on concurrence outside of the Principles.35 Thus, in a 1641 letter to


34. Cf. the claim of J. A. van Ruler that “the fact of Descartes’ commitment to the scholas-tic concept [of concurrence] cannot conceal the ambiguous way in which he uses it” (Ruler1995, 271).

35. Here I draw on the argument in Ruler 1995, 271–78. Cf. the response to this argumentin Pessin 2003, 36–39. But though Pessin is correct that Descartes sometimes used the term‘concursus’ when speaking of causal contributions of partial causes that do not involve con-servation (e.g., in the Fourth Meditation, at AT 7:50 and 56), his point that Descartes linkeddivine concursus to the laws of nature does not seem to me to establish a concurrentist read-ing of his account of God’s causation of motion (see note 88). I take this same point to applyto the different version of concurrentism attributed to Descartes in Hattab 2000 and 2007 (seenote 73). In §3.3, I argue more explicitly against the view that Descartes’s physics can beexplicated in terms of scholastic concurrentism.

Mersenne, he protests that he never held that “God does not concur immediately[concourt immediatement] in all things,” and notes that he even “expressly assertedthe contrary” in his response to the theologian Caterus (To Mersenne, 21 Apr. 1641,AT 3:360). But this response, from the First Replies, explicitly concerns the con-servation axiom discussed in the previous chapter. In the Replies Descartes repeatsthe conclusion in the Third Meditation that he could not continue to exist “unlesssome cause as it were causes anew [quasi rursus efficiat] at each moment” and addsthat “I do not hesitate in calling this cause, which conserves me, efficient” (AT7:109). Given this background information, we are to take the “immediate con-course” that Descartes mentions in his letter to Mersenne to be a continued conser-vation of the world that is identical to the efficient causal act by which God initiallycreated it.36

Descartes suggests the identification of concursus with conservation as well in pas-sages that concern the nature of substance. In the Synopsis to the Meditations, henotes that all created substances “are by their nature incorruptible, and cannot evercease to be unless they are reduced to nothing by God denying his concursus to them”(AT 7:14). The implication here is that the divine concursus is not only necessary butalso sufficient for the continued existence of created substances. The claim concern-ing the necessity of the concurrence is picked up in the Latin edition of the Principles,where Descartes claims that whereas God is substance in the primary sense since he“depends on no other thing to exist,” created substances “cannot exist unless by thework of the concursus of God” (PP I.51, AT 8-1:24). However, the claim concerningsufficiency is reflected in the comment added in the later French edition of this textthat created substances “need only the ordinary concourse [concours ordinaire] ofGod” to exist (AT 9-2:47).

We do not have to speculate that the use of the notion of concursus in these pas-sages would have been objectionable to those with a scholastic sensibility. One ofDescartes’s Dutch critics, Jacobus Revius, the author of Suarez repurgatus,37 tookissue with the remarks concerning substance in the Principles when he announcedin a 1650 text that “I am ashamed for the love of God, I am ashamed about yourignorance, Descartes. That such a great philosopher as yourself has not learned todistinguish between conservation and concursus!”38 Revius made what forscholastics would be the crucial point that an object that is nothing apart from thedivine act of conservation cannot be said to concur in that act.39 In the scholasticview, creatures concur not in the conservation of their substantial existence, which


36. See also Descartes’s repeated reference to God’s concursus in his discussion of divineconservation in To Hyperaspistes, Aug. 1641, AT 3:429–30.

37. See Revius 1643. In this text, Revius attempted to present a Suárezian system purgedof what to his Calvinist sensibility was Suárez’s own “pelagian” emphasis on our freedom toreject divine grace.

38. From Revius’s Statera philosophiae cartesianae (Revius 1650), cited in Ruler 1995,276. Here I follow Ruler’s translation. For an overview of Revius’s critique of Descartes’s sys-tem, see Verbeek 1992, 40–46, 78–81.

39. See the passage from Revius 1650, 73, cited in Ruler 1995, 277, n.51.

God alone brings about,40 but only in the production of their effects, to which theiroperations contribute.

Revius also raised the question of how Descartes could speak of concursus at all inthe context of his physics given that there can be no vis agendi in his world of mereextension, and so no “acting with” in the case of bodies.41 Such a question of coursebroaches the issue of Descartes’s views on the efficacy of secondary and particularcauses, an issue that I will address in §3.2. At this point, though, it is worth consider-ing why Descartes would even use the language of concursus when discussing matterspertaining to conservation. One answer suggested by the text of the Principles is thatwhat God contributes to body–body interaction is the conservation of the total quan-tity both of the matter of the parts that compose the material world and of the motionof these parts. This contribution therefore goes beyond the mere conservation of thequantity of matter, and so in this sense can be said to involve an additional concursus.However, this concursus is nothing beyond the act of conserving the total quantity ofmotion, an act that is identical to the act by which God created this quantity ex nihilo.

It is important to distinguish this suggestion of the identity of concursus with cre-ation of motion from what Garber has called the “cinematic view” of God’s causa-tion of motion.42 As Garber describes it, such a view holds that “motion is simplythe divine recreation of bodies in different places with respect to one another in dif-ferent moments; God, on this view, moves bodies by his recreation alone” (Garber1992, 275). Though others, such as Gueroult, attribute the cinematic view toDescartes, Garber is careful to note that such a view is something to which Descartesnever explicitly commits himself (Garber 1992, 275–76).43 On the basis of my ear-lier argument against Gueroult (see §2.2), I think we can go further by claiming thatDescartes in fact rejects the view of divine conservation as continual re-creation andoffers in its place the more traditional position that such conservation consists in thecontinuation of the initial act of creation. But an additional point is that the identifi-cation of divine concursus with conservation of matter in motion need not entail thelack of a distinction between the creation of matter simpliciter and the causation ofits motion. In the Principles itself, Descartes emphasizes that God not only origi-nally created a certain quantity of matter but also “moved the parts of matterdiversely, when he first created them” (PP II.36, AT 8-1:62). The divine concursusto changes in motion consists in a continuation of God’s act not only of creating the


40. Suárez allowed that certain accidents can depend on secondary causes for their con-servation (as light depends on the sun, or motion on a mover), but concluded that God alonecan immediately conserve a sort of being that can be created only ex nihilo; see MD XXI.3,1:794–801. As indicated in §1.2.3 (i), however, Suárez acknowledged the difficulty in pro-viding a demonstration that God cannot communicate to creatures a derivative sort of powerto create ex nihilo.

41. Again from Revius 1650, 73, cited in Ruler 1995, 277.42. Following Bergson, who labels this the conception cinématographique of motion; see

Bergson 1907, 295, 356, 373, cited in Gueroult 1953, 1:274.43. Though Garber does say that “it is quite possible that the cinematic view is a correct

representation of [Descartes’s] thought in The World or the Principles” (Garber 1992, 276).

matter, but also of initially moving its parts.44 In contrast, on the cinematic view thedifference between these two acts can be revealed not in the initial moment of cre-ation but only at later moments, when material parts are re-created either with thesame relations of distance or different relations.

Though Garber himself is skeptical that we can find any clear repudiation of the cin-ematic view in the Principles, he takes such a repudiation to emerge in Descartes’s latecorrespondence with More. On the account that Garber finds in this correspondence,which he calls the “impulse view,” God’s causation of motion consists not in his con-servation of matter, but rather in his producing a “divine shove” that serves to impelmatter (Garber 1992, 278). Garber finds such a view to be indicated in the remark inan August 1649 letter to More, cited previously, that though matter is at rest when it is“left to itself ” and receives “no impulse from anything else,” in fact it has been“impelled by God, conserving as much motion or transference in it as he put there fromthe first” (AT 5:404). The divine shove is simply that by which God introduced motionin the first place, and to conserve this motion he merely needs to keep shoving.

One reason to take the impulse view to depart from the Principles is that it seems todistinguish God’s causation of motion more radically than this text does from his cre-ation/conservation of the material world. Garber explains this difference in terms of thedistinction between causes that produce the modes of a thing, or “modal causes,” andcauses that produce the substantial being of a thing, or “substantial causes” (Garber1992, 277). The suggestion in the Principles is that God’s activity as a substantial causein creating and conserving the material world also explains his conservation of the totalquantity of motion. As Garber sees it, however, the view in the More correspondenceis that God acts merely as a modal cause in initially impelling matter and in conservingthis impulse. As evidence that Descartes thought in his later writings that God is merelya modal cause of motion, Garber cites his claim in an April 1649 letter to More that

although I believe that no mode of acting belongs univocally to God andcreatures, I confess, however, that I can find no idea in my mind that representsthe mode by which God or an angel can move matter different from that whichexhibits to me the mode by which I am conscious that I can move my own bodythrough my thought. (AT 5:347)

In moving our body, it seems evident that we act as a modal and not a substantialcause. But if we do not conceive of the mode of God’s action in moving matter asdiffering from the mode of our action in moving our body, then it would seem thatwe must conceive of God as moving matter as a modal rather than a substantialcause. Garber conclusion is that on Descartes’s view in the More correspondence,“there would appear to be a distinction between God as sustainer of the world, as asubstantial cause, keeping things in existence, and God as cause of motion, a modalcause, causing bodies to have the particular motion they have, determining, at leastin part, their modes” (Garber 1992, 277).


44. Of course, Descartes’s official view is that there is a single act of creation here.However, the point is that in the cases of creating matter without motion and creating matterwith motion, the acts must differ given the differences in the objects.

It can be noted initially, in response to Garber’s claim here, that Descartes pref-aces his April 1649 remark to More by saying that “it is not a defect for a philoso-pher to hold that God can move body, even though he does not hold that God iscorporeal; so also it is not a defect to judge the same of other incorporeal sub-stances” (AT 5:347). It seems that Descartes’s main concern in comparing the con-ception of the way we move bodies to the way God does is to counter the positionthat God could move bodies only if he were corporeal. But we also need to takeseriously his warning in this passage that “no mode of acting belongs univocally toGod and creatures.” Even if we as well as God move by an impulse of some kind,it must be recognized that both the effect of the impulse and its causal source arevery different in the two cases. Descartes suggests in his correspondence withElisabeth that we have a primitive notion of the union of mind and body “on whichdepends [our idea] of the force [force] that our soul has to move the body” (21 May1643, AT 3:665).45 But as Descartes admits to More, this force must differ from thesort of force involved in God’s causation of motion. Our force is a mere mode ofour mind, whereas there can be no modes in God (Aug. 1649, AT 5:404).46 In addi-tion, Descartes notes in this same letter, in a passage cited earlier, that it followsfrom “the laws proposed in articles 45 and following of the Second Part” of thePrinciples that God moves body by conserving “the force impelling its parts, whichforce applies itself now to one part of matter, now to another” (AT 5:405). Howeverit is that we move our body, it surely is not the case that we act by continuing toconserve the force we originally placed in it.

We can understand the distinction here between the cases of God and finiteminds in terms of the familiar Thomistic distinction between causes secundumfieri and secundum esse. I have noted the claim in Descartes that God alone can bethe cause secundum esse of substances. But whereas finite minds could perhaps becauses secundum fieri of certain changes in motion, God himself cannot be char-acterized as such a cause, principally because he is not a cause of change. Rather,he is the cause of a constant quantity of motive force in the world. So thoughmotive quantity is a mode, God can be considered as the cause secundum esse ofthis quantity insofar as he creates/conserves a quantity that is the subject of vari-ous modifications.47

In §4.3, I consider Descartes’s account of the forces in finite minds that accountfor changes in the material world. What is relevant here, however, is his conceptionof the various bodily forces that he posits in his physics. With respect to such forces,there would seem to be only three options: either these forces exist in bodies, or they


45. For further discussion of Descartes’s account of the union and its primitive notion, see §4.1.46. This view follows from the claim in the Principles that since “no variation is intelligible

in him,” God can possess only invariable attributes and no variable modes (PP I.56, AT 8-1:26).47. In his 1704 Use of Reason (see chapter 1, note 77), the Cartesian Regis claims that God

“produces the substance of formal motion,” whereas “the modes of this motion . . . dependimmediately on creatures” (I-2.15, Regis 1996, 296). Though Regis’s account of the productionof motion is idiosyncratic in certain respects (see note 87), he allows for the basic view, whichI find in Descartes, that God does not produce motion in the way in which other causes do.

exist in God, or they do not exist at all.48 The claim that they exist in God is pre-sumably ruled out by the fact that the various forces for acting and for resistingposited in the Principles vary, whereas again, as Descartes emphasizes to More,there can be no variation in God. But the claim that the varying forces exist in bod-ies seems to involve an appeal to something akin to the scholastic forms and quali-ties that Descartes wanted to banish from his physics. This would seem to leave uswith the last option, which Garber attributes to Descartes, that “force is nowhere,strictly speaking, not in God, who is the real cause of motion in the inanimate world,and not in bodies, which are the recipients of the motion that God causes” (Garber1992, 298).49

To determine Descartes’s own intentions with respect to this issue, we need toreflect on his account in the Principles of the ways in which “secondary and partic-ular causes” contribute to God’s universal causation of motion. I take this account tosupport a version of the first option, namely, that variable forces exist in bodies. HereI draw on the view in the seminal work of Gueroult and Gabbey that the bodilyforces that Descartes posits in this text are to be understood in terms of the existenceor duration of the bodies that God continuously conserves.50 In contrast to thesecommentators, however, I explicate the differences among bodily forces in terms ofthe differences among the modes of the duration of the bodies that possess theforces. Moreover, my position is distinctive in emphasizing that an account of bod-ily force in terms of duration better accommodates the forces for acting in movingbodies than it does the forces for resisting in resting bodies. I do not take this prob-lem with forces for resisting to constitute an objection to my interpretation ofDescartes since—as will become clear—I think that such forces are not easily incor-porated into his own view in the Principles of the ordinary concursus involved inGod’s continued creation of the material world.

3.2. LAWS AS PARTICULAR AND SECONDARY CAUSES

Descartes’s claim in the Principles that “rules or laws of nature” (regulae sive leges nat-urae) are “particular and secondary causes” no doubt strikes the contemporary reader asodd. From our post-Humean perspective such rules or laws would seem to be mere


48. As I indicate presently, Gueroult and Gabbey are the main representatives of the positionthat Descartes takes the first option, whereas Garber is the main representative of the position thathe accepts the third option. The second option is often attributed to Hatfield on the basis of hisdiscussion in Hatfield 1979, but he has indicated to me in conversation that when he said thatforce is in God, he meant only that God is the sole efficacious cause in body–body interactions,and not that varying forces are literally in God. Hatfield’s account of his original intent isreinforced by the discussion of his position in Garber 1992, 636–37, n. 41.

49. See also the view in Des Chene 1996, which he presents as a reconciliation of the firstand third options, but which I take to be a variant of the third, on the basis of his claim that“if by ‘force’ one means a mode whose intensity would be measured by the quantity Mv (in amoving body), then there is no such thing” (Des Chene 1996, 340).

50. Cf. Gueroult 1980, 198, and Gabbey 1980, 253–58.

empirical generalizations, hardly the sort of thing that could serve as a cause. One ques-tion that we need to address, then, is what Descartes could mean when he speaks of rulesor laws as causes. Yet there is a further puzzle concerning the relation of the laws toGod’s action as the universal and primary cause of motion. As we have seen, this divineaction consists in the conservation of the total quantity of motive force by means of thecontinuation of God’s initial act of creating the material world. But in the Principles, onlythe third law of nature, concerning the distribution of motion subsequent to collision, isderived explicitly from God’s action as primary cause of motion. The first law, whichrequires the conservation of the state of an object quantum in se est, is not restricted tothe case of motion, and the second law, which concerns the rectilinear determination ofmotion, does not concern the quantity of motion at all.51 It turns out that the heterogene-ity of the three laws cannot be eliminated entirely. It also is the case that the third law ofmotion reveals most directly Descartes’s view of the nature of the particular causes ofmotion; indeed, we will discover that this law introduces some interesting complicationsfor his containment axiom in the Third Meditation. Even so, a closer consideration ofeach of the three laws will put us in a better position to understand why Descartes takesall of them to be grounded in the divine conservation of the material world.

3.2.1. Three Laws of Nature

(i) Perseveret quantum in se est

In the article of the Principles that follows the one in which he discussed God’sactivity as primary cause of motion, Descartes offers the following as the first of thethree laws of nature that serve as particular and secondary causes of motion:

[E]ach and every thing, insofar as it is simple and undivided, remains, insofar as ithas it in itself, always in the same state, and never changes unless by externalcauses [unamquamque rem, quantus est simplex et indivisa, manere, quantum inse est, in eodem semper statu, nec unquam mutari nisi à causeis externis].(PP II.37, AT 8-1:62)

This law has an earlier counterpart in The World. In this text, Descartes introducesthe law that “each part of matter, in particular [chaque partie de la materiere, en par-ticulier], continues always to be in the same state, so long as its encounter [le ren-contre] with others does not force it to change” (W VII, AT 11:38). This law itselfderives from Descartes’s early work on free fall. Thus, he notes in a 1629 letter toMersenne that he will try to demonstrate the principle that “what is once in motionwill, in a vacuum, always remain in motion” in his “treatise,” presumably, The World(18 Dec. 1629, AT 1:90).52 But what is found in The World, as well as the Principles,is not a law restricted to motion, but one that applies to any state of a “part of mat-ter in particular” or to a “thing, insofar as it is simple and undivided.”


51. I am indebted here to the clear statement of this problem in Des Chene 2000b, 153, n.5.52. The principle itself dates from a 1618 note on free fall that Descartes wrote for Beeckman,

in which he claims that “in a vacuum, what is once moved always moves” (AT 10:78*).

It is difficult to know whether it was significant for Descartes that the later for-mulation of his law applies to any res, and not just to material parts. Though the sug-gestion in the Principles is that the law applies to minds as well as to matter, to myknowledge Descartes never discussed its applicability to the case of mind, whichsurely for him is a simple and undivided object.53 We will discover in chapter 5,moreover, that he allowed that the mind can be a cause of changes in its internalstates through acts of will. It is not clear how such immanent action could be con-sistent with the requirement that even minds will persist in their states unless alteredby external causes.54 For our purposes here, we can ignore the broader scope of theversion of the first law in the Principles and apply it only to the case of bodies.

But what can it mean to say that bodies are “simple and undivided” things? Herethe simplicity cannot be the same as the simplicity of mind given Descartes’s viewthat any body is divisible, as no mind is, into really distinct parts. However, it is sig-nificant that he says in the Principles only that the thing is undivided, and not that itis indivisible. In this text he distinguished between the division of a body into partsthat “occurs simply in our thought” and its division into parts by means of actualmotions (PP II.23, AT 8-1:52–53). The simplicity of the bodies that the first law con-cerns would not be the metaphysical simplicity that the mind has in virtue of lack-ing any really distinct parts, but only the physical simplicity of a body that lacks anyinternal parts distinguished by motions. Such a body has no actual internal com-plexity to compromise its tendency to persist in its state quantum in se est.

Descartes refers in Principles II.36 to the fact that particular causes bring it aboutthat “singular parts of matter acquire motion they did not have previously” (AT8 1:61). With respect to the first law, however, the focus is not on the external causeof a bodily state, but rather on the internal source of the persistence in that state.55

This is indicated by the appeal in the first law to persistence quantum in se est. AsI. B. Cohen has shown, this phrase, which has a long history dating back to


53. On the indivisibility and simplicity of mind, see the Sixth Meditation, at AT 7:85–86,and PS I.47, AT 11:364–66.

54. For more on this issue, see chapter 5, note 42. Interestingly, later Cartesians held thateven in the cases of minds, changes in state require an external cause. A case in point is pro-vided by an exchange between Regis and his critic Jean Du Hamel. In response to the axiomin Regis’s 1691 System of Philosophy (Système de philosophie) that “each thing persists initself to remain in the state it is in,” the Paris academic Jean Du Hamel objected that we knowby faith that angels “turn themselves, some toward the good, others toward evil” (Du Hamel1692, 56). Regis rejoined that the axiom holds even in the case of mind, since all changes inwill derive from changes in intellect, and since there can be changes in intellect only if themind is acted on by a body to which it is united (Regis 1692, 28–29). Regis was following theview here of the French Cartesian Robert Desgabets that any “pure mind” not united to anybody must think “indivisibly and irrevocably.” For further discussion of this position inDesgabets and of its relation to Regis’s views, see Schmaltz 2002a, ch. 4.

55. See Spinoza’s helpful gloss on this passage in Descartes’s Principles of Philosophy(Renati Descartes Principiorum Philosophiae Pars I et II ): “[I]f we attend to no external, i.e.,particular causes, but consider the thing by itself, we shall have to affirm that insofar as it canit always perseveres in the state in which it is” (IIp14dem, Spinoza 1985, 277).

Lucretius, was often understood as being equivalent to sua sponte, ex natura sua,and sui vi, that is, from the internal power or nature of a thing (Cohen 1964, 148).So the main claim in the first law is that there is something in an object, in est, thatexplains the fact that that object persists in its state. There is the derivative claim thatsuch persistence holds except in the case of changes due to “external causes.”However, nothing is said about how such causes operate. It is consistent with the firstlaw, as expressed in the Latin edition of the Principles, that the causes bring abouttheir changes through action at a distance. The counterpart of this law in The Worldmay seem to preclude this option when it refers to changes forced by an “encounter”(le rencontre) with other parts of matter. But it is not clear how from the result thatan unimpeded body continues in its same state that only an impediment that involvesactual contact can produce a change in that state.56 As in the case of God’s universalcausality, so the particular internal cause of persistence does not require by itself thatbody–body interaction take place only by means of contact action. This requirementin fact depends on further considerations introduced by the other two laws of nature.

(ii) Motus ex se ipso rectus

In Principles II.39, Descartes summarizes his second law of nature as follows:

[E]ach and every part of matter, taken separately, never tends to move so that itperseveres according to oblique lines, but only according to straight lines[unamquamque partem materiae, seorsim spectatem, non tendere unquam ut secundùmullas lineas oliquas pergat moveri, sed tantummodo secundum rectas]. (AT 8-1:63)

This law seems to be connected to Descartes’s investigations during the late 1620s ofthe nature of the propagation of light.57 But whereas one can find formulations of thefirst law that predate The World, the second law is introduced in that text, though asthe third of the three laws (for reasons I indicate below). In The World, it is said that

when a body moves even though its motion is made most often along a curved line,and even though it can never make any that is not in some manner circular, as wassaid earlier, still each of its parts in particular tends always to continue its [motion]in a straight line. And thus their action, that is to say the inclination [l’inclination]that they have to move, is different from their motion. (W VII, AT 11:43–44.)


56. In the French edition of the Principles, the reference in the Latin edition to being changed“by external causes “by external causes” (a causis externis) is replaced by the reference to beingchanged “by encounter with others” (par la rencontre des autres) (PP II.37, AT 9-2, 84). Butthough this change brings the position in this text into line with what Descartes said in the World,there is no more of a justification here than in that earlier text for the limitation to collision.

57. As proposed in Garber 1992, 210. In the early unfinished Rules for the Direction of theMind (abandoned in 1628), there is an appeal to “the nature of light” with respect to the problemof how light passes through a transparent body (RM VIII, AT 11:394–95). Later, in the Dioptrics(1637), Descartes emphasizes that rays of light must be held to be propagated in a straight line whenthey pass through a uniform and transparent body (AT 6:88–89). Finally, Descartes applies the sec-ond law explicitly to the case of the propagation of light in the Principles (PP III.55, AT 8-1:108).

In both The World and the Principles, there is an emphasis on the fact that a movingbody has a tendency to move in a straight line even in the case where it is alwaysconstrained to move along a curved line. In the earlier work, Descartes takes this factto show that the “action or inclination” that bodies have to move (l’action, c’est àdire l’inclination qu’elles ont à se mouvoir) differs from the motion itself. Even inthe case of circular motion, the body has an inclination at each instant (instant) tohave the simplest motion, namely, motion in a straight line (AT 11:44–45). The dis-tinction between motion and inclination at an instant is also present in the Principles,where Descartes notes that although no motion takes place in a single instant (in sin-gularis instantibus), still at each instant the moving body “is determined to continueits motion [determinatum esse ad motum suum continandum] toward another partaccording to a straight line” (PP II.39, AT 8-1:64).

The language in the Principles is a bit confusing, since the sort of determinationreferred to here differs from the determination of motion in a certain direction thatis important to the third law of nature in that text (see PP II.41, AT 8-1:65–66).Descartes holds that this latter sort of determination is a modification of an actualmotion (see To Mersenne for Hobbes, 21 Apr. 1641, AT 3:355–56). In contrast, theinstantaneous determination to motion, or what he calls in The World the inclinationto motion, is not a mode of motion but only a “first preparation for motion” that canbe present even in something, such as the pressure of light, that does not involveactual motion (PP III.63, AT 8-1:14).

In The World, Descartes argues that since at each instant God can conserve onlythe inclination to motion that is present at that instant, and since it is only the incli-nation to move in a straight line “that is entirely simple, and the whole nature ofwhich is comprised in an instant,” God must conserve at each instant the inclinationto move in a straight line (W VII, AT 11:44–45). His later claim in the Principles isthat God’s immutable and simple action can conserve nothing “unless in the preciseform it is at the moment of time that he conserves it” (PP II.39, AT 8-1:64). DennisDes Chene takes this claim to deviate from the position in The World insofar as itattributes the simplicity to the divine operation rather than to the motion itself (DesChene 1996, 283–84). However, the view in the Principles that God conservesmotion “in the precise form it is at the moment he conserves it” indicates that theform intrinsic to the motion at an instant involves the inclination to rectilinearmotion. The argument here, consistent with that in The World, is that since at eachinstant motion can have only the inclination “the whole nature of which is comprisedin an instant,” and since only the inclination to rectilinear motion has such a nature,a simple and immutable divine concursus can result only in the conservation at eachinstant of the inclination to rectilinear motion.58


58. It might be thought that Descartes’s remarks here favor the view of Gueroult, criticizedin §2.2, that God conserves the world by re-creating it at each instant. But Descartes himselfindicates that the fact that moving bodies have durations composed of instants “does not inter-rupt the continuity of their motions” (PP III.63, AT 8-1:115). For God to conserve the rectilin-ear inclination of a motion at an instant is just for him continuously to conserve the motion thathas such an instantaneous inclination. Thanks to Andy Pessin for pressing me on this point.

I leave for §3.2.3 a consideration of the assumptions in this argument regarding thesimplicity and immutability of the divine concursus. For now, I want to consider theclaim that rectilinear motion itself is the simplest in form. Such a claim went againstthe traditional Aristotelian position—which influenced even critics of Aristotelianismsuch as Galileo—that circular motion is the most basic found in nature.59 In thePrinciples, Descartes’s argument against this position relies on the assumption thatonly the inclination to rectilinear motion can be comprised in an instant, since any otherinclination requires a comparison to motion “that was a little while earlier” (PP II.39,AT 8-1:64). If a point lies on a curved path, one needs information about previous (andsubsequent) points on the path to determine its nature. But in the case of a straight pathfrom a point, any direction of the path from that point will yield such a line.60

Just as the first law emphasizes what follows from an object quantum in se est, sothe second law focuses on the results of an instantaneous inclination internal tomotion itself. As I have noted, however, the first law had little to say about the exter-nal causes that disrupt persistence. We are now in a position to see the implicationof the second law that in the case of bodies, at least, only something that blocks thepath of a moving body could frustrate the inclination to rectilinear motion. The rejec-tion here of any change in motion produced by action at a distance is even clearer inthe case of Descartes’s third law.

(iii) Quantum motus transfert

In Principles II.40, Descartes offers the following as the third law of nature:

[W]here a body that moves encounters another, if it has less force for perseveringaccording to a straight line than this other has to resist it, then it is deflected inanother direction, and, retaining its motion, gives up only its determination [tocontinue its motion in a particular direction]; but if it has more, then it moves theother body with it, and whatever it gives of its motion to [the other body] it losesjust as much [ubi corpus quod movetur alteri occurrit, si minorem habeat vim adpergendum secundum lineam rectam, quàm hoc alterum ad ei resistendum, tuncdeflectitur in aliam partem, & motum suum retinendo solam motûs


59. Aristotle’s argument in De Caelo is that since the simplest bodies have the most per-fect motion, and since circular motion is the most perfect insofar as it has a starting point thatis identical to its ending point, and so is complete in itself, circular motion is natural for thesimplest bodies (I.2, a17–30, Aristotle 1984, 1:448–49). In the Two Chief World Systems(Dialogo Sopra i Due Massimi Sistemi del Mondo) Galileo’s spokesman Salviati attacks theAristotelian view of the rectilinear motion of the elements by insisting that uniform circularmotion is most natural to terrestrial objects, since they share in the Earth’s rotation (Galileo1967, 28, 148). Even Descartes’s Dutch mentor Beeckman, who was a critic of Aristotelianphysics, argued that motion along a curved path continues along this path if unimpeded, justas unimpeded motion along a straight path continues along that path (Beeckman 1939, 1:253).

60. My summary of Descartes’s argument here has been influenced by the discussion inDes Chene 1996, 283–86. But see also the reservations concerning Des Chene’s reading indi-cated in the previous paragraph.

determinationem amitti; si verò habeat majorem, tunc alterum corpus secummovet, ac quantum ei dat de suo motu, tantundem perdit]. (AT 8-1:65)

In contrast to the first two laws, this law directly concerns the case where a bodybrings it about that “singular parts of matter acquire motion they did not have previ-ously,” and so serves as a particular cause in the sense defined in Principles II.36.What is central here to the causality of colliding bodies is either the force for resist-ing that produces a change in the determination of motion, or the force for acting thatproduces a redistribution of motion.

The reference to the determination of the motion in a certain direction explainswhy this law is presented in the Principles after the second law. For though I haveindicated that the instantaneous inclination to motion to which the latter refers mustbe distinguished from the determination that modifies motion, this determinationnonetheless derives from that inclination. It is in virtue of that inclination, after all,that an unimpeded moving body is determined to move along a straight path. Thecounterpart of the third law in The World, however, makes no mention of determi-nation. There the law is simply that “when a body pushes another it can give it nomotion without losing as much of its own; nor take away any without augmenting itsown by the same amount” (W VII, AT 11:41). This is the second law in that text, pre-sented before the law concerning the inclination to rectilinear motion. And it couldbe so presented, since, in contrast to the case of the third law in the Principles, itrefers only to the quantity of motion and not to its directionality.

As I indicated previously, the Principles also differs from The World insofar as itrelies on a specific measure of the quantity of motion (see §3.1.2 (ii)). It follows fromthe account in the former text of God’s action as universal cause that there is a globalconservation of the quantity of motion. What the third law adds is that this quantity islocally conserved in cases of collision. Since this law is supposed to govern “all theparticular causes of changes that happen to bodies, . . . at least those which themselvesare corporeal,”61 it follows from conservation of the quantity of motion in cases of col-lision that particular causes do not bring about a change in the total quantity of motion.

Even so, both The World and the Principles include an appeal to bodily forces inthe explanation of transfers of motion in collision. The suggestion in the earlier workthat bodies “retain or transfer [motions] one to another, as they have the force [ force]to do so” (W VII, AT 11:43) becomes the more specific position in the later text thata resting body has a “force for resisting” (vis ad resistendum) that inclines it toward“persevering in its state of rest,” whereas a moving body has a “force for acting” (visad agendum) that is responsible for its tendency “for persevering in its motion,that is, in motion with the same speed and toward the same part” (PP II.43, AT


61. The Cottingham translation has this passage say that the third law “covers all changeswhich are themselves corporeal” (Descartes 1984–85, 1:242), thereby taking the changesrather than the causes to be corporeal. Though the Latin is ambiguous on this point, it wouldnot seem to be necessary to emphasize that a law concerning changes in motion governs onlycorporeal changes. For a criticism of the Cottingham translation along these lines, see DellaRocca 1999, 52–54.

8 1:66–67). According to the third law in the Principles, these forces are supposedto explain the redistributions of motion that occur in collision.

The claim in the Principles that the third law governs all particular causes that arecorporeal, coupled with the claim there that such causes produce changes only in thecase of collision, yields the further result that all body–body interaction occurs bymeans of contact action. Thus the third law makes explicit, what was merely implicitin the second law, that no corporeal causes can produce changes in motion at a dis-tance.62 We have discussed the requirement in Descartes that an efficient cause mustbe simultaneous with its effect, since it cannot act at a time distant from that of itsaction (see §2.2.1). Now we have the requirement that in the case of particular bod-ily causes, at least, an efficient cause must be contiguous with body it acts on, sinceit cannot act at a place distant from that action.63

Both the third law in the Principles and its counterpart in The World suggest thefurther requirement that a particular corporeal cause acts by “giving” to its effect aportion of its quantity of motion that it “loses.” We seem to have here an endorse-ment of what Janet Broughton has called “the migration theory of motion,”according to which motion is transferred in collision by means of the migration of aparticular mode of motion from one body to another. Broughton’s proposal is thatDescartes was led to this theory by the fact that the containment axiom from theThird Meditation requires that in cases of formal containment, the cause not onlycontains something similar to what is found in the effect, but also imparts the verysame thing it formally contains to its effect (Broughton 1986, 120–21).

If Descartes’s containment axiom required the migration theory of motion, therewould be good reason to challenge his claim that such an axiom is revealed to beevident by the “natural light” (AT 7:40).64 But though there is no doubt thatDescartes’s language in certain texts clearly suggests such a theory,65 it is not evident


62. Cf. the discussion of Descartes’s rejection of action at a distance in Suppes 1954. HereI agree with Suppes’ conclusion that Descartes has a priori grounds for restricting body–bodyinteraction to contact action, though my account of Descartes’s derivation of the restrictiongoes beyond what is found in his article. Cf. Clarke’s argument that Descartes did not derivehis laws of nature from metaphysical principles (Clarke 1982, 100–104). In contrast to Clarke,however, I take Descartes’s appeals to experience in his discussion of the laws to be an attemptto illustrate rather than to justify the laws.

63. In §4.3.2, we will consider whether Descartes held that some version of the require-ment of contiguity applies to the case of mind-to-body action.

64. For the suggestion that Descartes’s containment principle requires a “transmission”view of causation, see Lloyd 1976. Bennett objects to Descartes’s treatment of the contain-ment axiom (or what he calls the “causal resources principle”) on the grounds that it equivo-cates between the unacceptably strong view of “causing as giving” and the trivial view of acause as having the power to produce its effect (Bennett 2001, 1:88–89). I argued in the pre-vious chapter that there is more to Descartes’s axiom than this trivial view. Presently I arguethat this axiom does not require an implausible version of the strong view.

65. Besides the passages already cited, see the claim in The World that “the virtue orpower to move itself, which is found in a body, can well pass wholly or in part into another[passer toute ou partie dans un autre], and thus no longer be in the first, but it cannot nolonger be at all in the world” (W III, AT 11:11).

that his containment axiom requires it. With respect to this last point, it is worth con-sidering the case of Suárez. As I noted in the previous chapter, Suárez accepted a ver-sion of Descartes’s containment axiom. Yet we also saw in the first chapter hisemphasis on the fact that an efficient cause is “extrinsic” insofar as it produces in theeffect a being distinct from its own (see §1.2.2). His basic point is that the efficientcause that formally contains its effect does not pass on the very same feature that itcontains; rather, it only produces a feature of the same type.

For Suárez, the type here is the kind of form or quality that the cause induces inits effect. A paradigmatic case would be one in which fire generates the same kindof “real quality” of heat in something else that it itself possesses.66 Suárez alsoallowed that there can be cases of bodily interaction where the “principal”67 effi-cient cause does not possess the same kind of quality, but is an “equivocal” causethat contains the quality it induces only eminently. He held, for instance, that heatedwater contains the quality of coldness “virtually” insofar as it has by its nature theactive power to cool.68 Descartes of course rejected the whole category of real qual-ities and active powers that informs Suárez’s account of efficient causality on thegrounds that all bodily accidents are mere modes of extended substance.69 As I indi-cated earlier, however, he in effect rejected as well the claim that there can be qual-ities or power in bodies “more noble than” bodily effects that also eminentlycontain these effects (see §2.1.3 (ii)). Descartes therefore could not accept Suárez’sposition that there can be adequate bodily causes that eminently contain theireffects.

Even so, Suárez’s analysis of efficient causation serves to counter Broughton’ssuggestion that Descartes’s containment axiom, when applied to the case ofbody–body interaction, requires a migration theory of motion. It seems thatDescartes could hold that a body that is an efficient cause of motion itself containsa motion that is similar to, but nonetheless numerically distinct from, the motion itinduces in another body. Indeed, in a passage from his correspondence with Morethat Broughton herself cites, Descartes explicitly rejects the migration theory ofmotion. No doubt prompted by the passage from the Principles that is suggestive ofthe migration theory, More protested that a mode of motion cannot “pass over”


66. As I indicate toward the start of §3.2.2, Suárez had a complex view both about whichaccidental features of bodies could be causes and what sort of effects the causally efficaciousfeatures could produce.

67. Recall here the distinction in Suárez, considered in §1.2.3 (ii), between instrumentaland principal causes.

68. MD XVIII.3, ¶¶4and 7, ¶19, 1:616 and 635–36. For Suárez’s account of equivocalcausality, see MD XVII.2, ¶21, 1:591–92.

69. See, for instance, Descartes’s claim in a 1643 letter to Mersenne, cited in §1.2.1, that“I do not suppose any real qualities in nature, which are joined to substance, as little souls totheir bodies, and which can be separated from it by divine power, and thus I attribute no morereality to motion or to all the other varieties of substance that one calls qualities no more real-ity than the philosophers commonly attribute to shape, which they call not real quality butonly mode” (AT 3:648–49).

(transet) or “migrate” (migret) from one corporeal subject to another (23 July 1649,AT 5:382*). Descartes’s response, a portion of which I cited previously, is as follows:

[T]his is not what I have written; indeed I think that motion as a mode continuallychanges. . . . But when I said that the same motion always remains in matter,I understood this to concern the force impelling its parts [de vi eius partesimpellente], which force applies itself now to one part of matter, now to another,in accord with the laws proposed in articles 45 and following of the Second Part[of the Principles]. (30 Aug. 1649, AT 5:405)

In light of his talk in the discussion in the Principles of the third law of a body as“giving” to another the motion that it “loses,” it seems somewhat disingenuous forDescartes to claim that he never wrote that motion can be passed from one body toanother. But his main point is perhaps that when he wrote that, he meant only thatforce is applied to a body in a manner that produces an alteration in its motion. Andin the Principles itself, Descartes emphasizes that changes in the speed or directionof motion derive from “the quantum of force [quantum . . . virium] in [bodies], eitherto move, or to resist motion” (PP II.45, AT 8-1:67/CSM 1:244). The indication hereis that the communication of motion is to be understood in terms not of the migra-tion of the very same mode of motion, but rather of the production of a numericallydistinct mode of motion by means of the application of a force.

Nevertheless, there remains the requirement of Descartes’s containment axiom thatthe total or adequate bodily cause of changes in motion contain formally or eminentlyall of its effects in other bodies. If my analysis above of Descartes’s relation to Suárezis correct, eminent containment cannot be at issue here. So the question is whether theformal containment requirement is satisfied in cases of body–body interaction. In mostof the basic cases of such interaction that he considers, Descartes in fact can allow thatthe efficient cause actually contains something similar to what it produces in its effect.In his discussion in the Principles of the seven collision rules that supplement the thirdlaw, he indicates that in all cases where a moving body collides either with a movingbody smaller than or equal to itself or with a smaller body at rest, the bodily cause con-tains modes of the same type as those it produces.70 Matters are less straightforward,however, in cases where a moving body collides with a body at rest that is either largerthan or equal in size to it. Descartes’s fourth and sixth rules of collision require that thebody at rest produce an instantaneous reversal of the direction of the motion of themoving body in such cases, even though the resting body does not actually possess anymotion, and thus does not possess the particular directional determination it producesin the moving body. So in these cases, the resting body that serves as the efficient andapparently total or adequate cause of the reversal of the moving body71 seems not tocontain formally or eminently everything that it produces in the effect.


70. For a summary of the rules, see note 28. My comment applies to the first, second,third, fifth, and seventh rules.

71. With respect to bodies, at least; here I bracket God’s contribution as “universal andprimary cause” to changes in motion.

As I have indicated, however, the third law requires that both the resting body thatcauses a change and the moving body that is changed possess “force”—force forresisting, in the former case, and force for acting, in the latter. Even though the rest-ing body does not formally contain motion with a certain speed and directionaldetermination, it does formally contain a force that is at least as great as the forceresponsible for such motion. The resting body could therefore be understood to for-mally contain that motion in this attenuated sense.72 This sense would be all thatDescartes required to adhere to his rule in the Principles—cited in the course of hisdiscussion there of the implications of the third law—that one must “never attributeto a cause any effect that exceeds its capacity [potentiam]” (PP II.60, AT 8-1:76).

The notion of bodily force that is so prominent in the version of the third law in thePrinciples serves to connect it directly to the first law in that text. For Descartesemphasizes that “the force of any given body to act on another, or to resist the actionof another,” consists “in this one thing, that each and every thing tends, quantum in seest, to persist in the same state it is in, as posited in the first law” (II.43, AT 8-1:66). Itis in virtue of the fact that a body has a tendency to persist in the same state quantumin se est that it has the force either to remain at rest or to persist in its motion.According to the second law, moreover, the persistence of the moving body involves atendency toward rectilinear motion that derives from the instantaneous inclinations tomotion in that body. The indication here seems to be that the efficacy of the laws asparticular causes is to be explained in terms of the internal bodily forces and inclina-tions that are themselves responsible for changes in the distributions of motion.73 Tobe sure, Descartes claims in the Principles that both the second and the third laws fol-low from the divine attributes.74 It is certainly possible to read Descartes’s talk of inher-ent forces and inclinations—as Garber has, for instance—as a mere façon de parler, away of describing the changes that God alone produces by means of his continuous


72. In §4.2.2, I propose that the innate mental faculty that Descartes posited in the 1647Comments as a cause of certain features of sensory ideas must also be understood to formallycontain those features in an attenuated sense.

73. In contrast, Hattab offers a reading on which Descartes holds that the laws themselves,and not any features of bodies, are secondary and particular causes (see Hattab 2000 and2007). She claims that this “seems to be the only reading that does not contradict Descartes’sontology,” and in particular his purported view that the nature of body consists in extensionalone (Hattab 2000, 116; cf. 98–100). It is difficult to see how on this reading the causal effi-cacy of the laws could have a source distinct from God’s action. Such a reading thereforeseems to be simply a version of an occasionalist interpretation of Descartes. In §3.2.2,I address the objection in Hattab and others that the attribution of causal efficacy to the bod-ies themselves conflicts with Descartes’s ontology of the material world.

74. In PP II.43 Descartes claims that the demonstration of the part of the third law con-cerning transfer of motion is provided by “the immutability of the operations of God, nowcontinuously conserving the world by the same action with which he created it then” (AT8 1:66), whereas in PP II.39 he holds that the “cause” (causa) of the second law is “theimmutability and simplicity of the operations by which God conserves motion in matter” (AT8-1:63). In The World all three laws are said to “depend only on the fact that God conserveseach thing by a continuous action” (W VII, AT 11:44).

conservation of the material world. However, Gueroult and Gabbey also suggest onDescartes’s behalf that the bodily forces that ground the laws of nature are explicablein terms of the existence or duration of bodies and their motions. There is of course thequestion of whether such a view is compatible with the reduction of matter to exten-sion that Descartes’s ontology requires. But there is also the question whether heoffered a notion of force that applies equally to the cases of rest and motion.

3.2.2. The Ontology of Bodily Force

Toward the end of §3.1.3, I mentioned that the various forces that Descartes positsin his physics can be attributed either to bodies, or to God, or to nothing at all.I also noted that the second option is ruled out by the implication in Descartes thatthe various forces for acting and for resisting can change through time, given hisfirm position that nothing in God can be subject to change. But the first optionwould seem to be ruled out as well. For on Descartes’s official position in thePrinciples, “extension in length, breadth and depth constitutes the nature of cor-poreal substance,” and thus “everything else that can be attributed to body presup-poses extension, and is merely a mode of an extended thing” (PP I.53, AT 8-1:25).Whatever forces for acting and for resisting are, it seems that they cannot be meremodes of extension akin to various shapes, sizes, and motions. But then it appearsthat such forces would be akin to the real qualities that the scholastics posited assecondary causes in nature and that Descartes himself wanted to banish fromphysics. We apparently are forced to the final option that the forces Descartesposited in his physics are mere fictions.

Before acquiescing in this fictionalist interpretation of Descartes’s view of bodilyforce, however, it is worth pausing to consider the comparison to scholastic real quali-ties. I indicated in §1.3 the view in Suárez that among the predicamental accidents, onlycertain qualities, and not quantity, shape (in the category of quality) or local motion, canserve as principles of efficient causality. Clearly, Descartes rejected the assumption inSuárez that there are principles of action in bodies other than quantity, shape, and localmotion. But did he accept the Suárezian position that these bodily features could not beprinciples of action? If so, Descartes would be forced to conclude that there are no suchprinciples in bodies. However, it seems to me that the claim that Descartes in fact drewthis conclusion is often based less on the textual evidence than on intuitions about whathis identification of body with extension requires. In particular, the guiding intuition isthat bare extension is something that is purely passive, the mere instantiation of a purelygeometrical essence, and not something that can ground causal activity.

Some of Descartes’s statements may seem to indicate that he conceived the exten-sion that constitutes the nature of bodies that actually exist in the material world injust this manner. There is, for instance, his claim in the Principles that “I admit noother principles in physics but those in geometry or abstract mathematics” (PP II.64,AT 8-1:78), as well as his earlier remark in a letter to Mersenne that “my physics isnothing but geometry” (27 July 1638, AT 2:268). But Descartes’s own account of thelaws of nature belies this simple identification of physics with geometry. For theobjects of geometry, as present objectively in the mind that considers them, do nothave either the tendency to persist in a particular state quantum in se est (as required


by the first law), the instantaneous inclinations to rectilinear motion (as required bythe second law), or the particular forces that require changes in the speeds and deter-minations of motions (as required by the third law). The reason that purely geomet-rical objects lack these features seems to be that they lack any sort of existenceexternal to mind. But could Descartes hold that they possess the features simply invirtue of possessing this additional extra-mental existence?

The answer to this question is yes, according to the most sophisticated argumentin the literature for the view that Descartes attributes real causal efficacy to bodies.75

Thus, Alan Gabbey urges that he distinguishes “between a body’s essence as an idea,that is as existing objectively in the intellect, and the body’s existence outside themind,” and that “force depends on extension in the sense that extension is presup-posed in saying that something corporeal exists or endures” (Gabbey 1980, 238).Here Gabbey is drawing on the claim in Gueroult that according to Descartes,

in reality, force, duration, and existence are one and the same thing (conatus) underthree different aspects, and the three notions are identified in the instantaneousaction in virtue of which corporeal substance exists and endures, that is, possessesthe force which puts it into existence and duration. (Gueroult 1980, 197)

The difference between a purely geometrical object and an actually existing body isthat only the latter possesses an existence or duration that, on Descartes’s officialview in the Principles, is distinct merely in reason and not in reality from the sub-stance that exists and endures (see PP I.62, AT 8-1:30). If force is simply identifiedwith the attribute of existence or duration, then it seems to be something that con-cretely existing bodies, but not abstract geometrical objects, possess.

There is the question, however, whether the forces that Gueroult identifies with thedurational existence of bodies can themselves be identified with the various forces foracting and resisting that Descartes posits in the Principles. It may seem that they infact cannot be so identified for the simple reason that whereas Descartes provides theexample of existence or duration as something “that always remains unmodified,” andso is an attribute in the strict sense (PP I.56, AT 8-1:26), his view that the forces ofparticular parts of matter constantly change due to collision indicates that they arevariable, and so must be considered as modes rather than attributes.

Gabbey attempts to supplement Gueroult’s interpretation in a manner that addressesthis problem. He retains the view in Gueroult that there is a sense in which force canbe identified with the invariable bodily attribute of duration. However, Gabbey intro-duces an additional sort of force that is present in bodies as a variable mode rather thanan attribute. He explains this distinction between the two kinds of bodily force in terms


75. Cf. the causal realist interpretations of Descartes’s physics in Westfall 1971, ch. 2, andCottingham 1997. However, Westfall emphasizes the problematic nature of dynamical con-cepts in Descartes, and Cottingham concludes that “when talking about impact, impulse andtransfer of motion between bodies, [Descartes] seems not to have given any serious attentionto the precise meaning of the concepts he used” (Cottingham 1997, 164). But see the “con-currentist” interpretations of Descartes in the literature cited in the introduction, note 11,which tend to emphasize the coherence of Descartes’s account of causation.

of the Thomistic distinction in Descartes between causes secundum fieri and secundumesse (see §§1.1.2 and 2.2.1). On Gabbey’s reading of Descartes, “force as causa secun-dum esse is . . . an attribute of body, in the sense that qua cause it is necessarily entailedby a body’s duration, viewed simpliciter and irrespective of mode,” whereas when theforces are viewed “as quantifiable causes of change in the corporeal world, or as rea-sons . . . explaining absence of change of a certain kind in particular instances, they arecausae secundum fieri” that are “clearly in body diverso modo, and so are modes ofbody, rather than attributes” (Gabbey 1980, 236, 237).

But when forces are present in bodies diverso modo, what sort of modes are they?Gabbey does not say, but there is an answer that is connected to the interpretationI offered earlier of Descartes’s account of temporal parts (see §2.2.2). I tookDescartes’s considered position to be that these parts are modally distinct features ofthe attribute of duration. It might seem natural, on Gabbey’s view, to identify thevariable forces that serve as the quantifiable causes secundum fieri with the variousmodal parts of the duration of interacting bodies. Gabbey himself indicates that hewould reject this sort of move on the grounds that “a body cannot have ‘more or less’duration: either it exists or it does not, and if it does exist, in whatever modal dispo-sition, it necessarily endures without ‘existential variation’ ” (Gabbey 1980, 237).But I previously anticipated this line of objection when I noted that whereas dura-tion is not subject to variation when considered just as an attribute, Descartes indi-cates clearly enough that this attribute nonetheless has various distinguishable parts.

I have already proposed that we take Descartes to hold that God is the cause secun-dum esse of the constant quantity of motion (see §3.1.3). My suggestion now is that weread him as saying that the various modes of bodily duration are causes secundum fieriof changes in the distribution of this quantity among the parts of matter. But this way ofputting the suggestion is perhaps too abstract and disconnected from Descartes’s discus-sion in the Principles of forces for acting and resisting. To make this suggestion moreconcrete, let us start with the point in this text at which the third law of nature makes con-tact with the first. I have noted the claim in Principles II.43 that the forces that the thirdlaw posits as responsible for changes in motion due to collision consist simply “in thisone thing, that each and every thing tends, quantum in se est, to persist in the same stateit is in, as posited in the first law” (AT 8-1:66). What is in est, according to the first law,is simply the tendency to continue in the same state. What the later reference to this lawmakes clear is that this tendency is not constant but varies depending on the nature of themode involved. The measure of the tendencies to persist in motion and rest is just thesame as the measure of the forces involved in those states. Thus, in the case where onebody is double the size of another body moving at the same speed, the first body has dou-ble the tendency to persist in its state of motion that the other body has to persist in itsstate. In the case where a moving body collides with a body at rest, the strength of thetendency of the latter to persist in its state of rest is measured by the product of the sizeof that body and the speed of the body that collides with it.76 These tendencies can be


76. At least this is so with respect to the 1644 Latin edition of the Principles. As indicatedin note 30 above, the 1647 French edition of this text provides a more complicated measurefor the force for resisting.

said to be something in est insofar as they are simply varying modal features of the dura-tions of moving and resting bodies.

As we know, Descartes claims that duration, considered as an attribute, is distinct onlyratione, and not in re, from the substance that endures (PP I.62, AT 8-1:30). On the viewI propose, he also holds that the strength of duration, or force, is only rationally distinctfrom the features of motion and rest that possess that sort of duration. One reason that itseems that force is not intrinsic to these modes is that the modes themselves can be con-sidered abstractly as purely mathematical objects, the nature of which is exhausted bytheir geometric and kinematic aspects. In this respect, these modes are similar to the tri-angle that, as Descartes writes in a letter to an unknown correspondent, can be consid-ered merely with respect to its essence, in abstraction from its existence. In this case,according to Descartes, the thought of the essence of a triangle differs modally from thethought of the existence of that triangle. But he continues in this letter by noting that out-side of thought the essence of a triangle and its existence are “in no way distinct” (1645or 1646, AT 4:350). Descartes could allow similarly that the thought of force, that is, ofstrength of duration, is modally distinct from the thought of the modal feature that hasthat force. However, it still seems open to him to say that the force or strength of dura-tion is in no way distinct from that duration as it exists external to mind.

In light of this distinction between the two ways of considering the modes of motionand rest, we can discern an important ambiguity in Descartes’s official position that allbodily modes must be conceived through extension. For as Gabbey emphasizes, exten-sion itself can be conceived merely abstractly, as present objectively in the intellect, orconcretely, as something not distinct from its durational existence in reality. If themodes are understood in terms of abstract extension, then force cannot be conceivedthrough the nature of body. This nature would be exhausted by purely mathematicalfeatures. But if the modes are understood in terms of concrete extension, force can beconceived through the nature of body insofar as this force is identified with the strengthof the duration that does not differ from the modes in reality.

I have mentioned in passing Descartes’s claim in correspondence that determina-tion is a mode that inheres in motion. There he is addressing Hobbes’s objection thatdetermination cannot so inhere given that motion is a mode and not a subject. ButDescartes insists that determination can inhere “as in a subject” even though that inwhich it inheres is a mode rather than a substance (To Mersenne for Hobbes, 21 Apr.1641, AT 3:355–36). As other commentators have emphasized, determination is forDescartes a composite mode involving two further modal features of motion, namely,direction, on the one hand, and quantity of motion (scalar speed × size), on the other.77


77. See Gabbey 1980, 248–50; Garber 1992, 188–93; McLaughlin 2000, 87–97.Descartes indicated that the nature of determination is fixed not only by the direction of themotion but also by the quantity of the motion determined. A component of the determinationof a particular quantity of motion can be shared by motions that have different quantities, butthe composite determination can belong only to a motion with a particular motive quantity.Thus, to take an example from Descartes’s Dioptrics (at AT 6:97), though a tennis ball canhave the same horizontal determination before and after a refraction that involves a change inspeed, it cannot have the same overall determination in these two cases. For documentation ofthis point in Descartes, see McLaughlin 2000, 94–97.

The force for acting in the moving body would be simply the strength of the durationof the motive quantity that modifies the motion in that body. As long as this quantitycontinues to endure, so does the force that is bound up with that duration.

Matters are less straightforward in the case of rest. As I noted in §3.1.2 (iii), thereis the implication of Descartes’s remarks in his correspondence with More thatforce for resisting is distinct from the state of rest itself. And there is in fact good rea-son for him to refrain from identifying this force with the duration of the rest, sincethe force is supposed to be proportional to the speed of a body that collides with it.Thus, the resting body seems to have this force only at the instant of impact. It is dif-ficult to see how the force could be identified with the duration of features of the modeof rest. I argue below that this difficulty in turn renders problematic an account ofGod’s conservation of the quantity of rest in terms of this force.

The case of the instantaneous inclinations posited in the second law raises differentproblems. As we have seen, Descartes sharply distinguishes the quantity that measuresthe force of motion from the directionality of that motion. Since the inclinations concernthe direction of motion exclusively, they cannot be folded into motive force. Moreover,Descartes holds that though motions can have varying forces, there is no difference inthe strengths of the rectilinear inclinations of those motions. Nevertheless, it is clear thatthe inclinations can change over time. This is shown by Descartes’s own example of thestone in a sling, which at different instants of its circular motion has inclinations to moveoff in straight paths at different tangents from the circle (PP II.39, AT 8-1:64–65).78

Inclinations are therefore as much variable modes of a motion as the forces derivingfrom its various quantities. The fact that the inclinations are instantaneous precludes anyexplication of them in terms of the duration of motion. Even so, the inclinations seemto be as much internal features of motion as the forces tied to its duration.

It might be thought that Descartes himself offered a different account of inclina-tion that is more in line with Garber’s fictionalist interpretation. There is for instanceDescartes’s point in the Principles, also found in The World, that the claim that bod-ies recede from the center of a circle indicates “only that they are so situated, and soincited to motion, that they will really go [away from the center] if they are impededby no other cause” (PP III.56, AT 8-1:108; cf. W XIII, AT 11:84). The suggestionhere seems to be that the inclination is not a real feature of the motion, but merely adisguised counterfactual conditional of the form that if there were no impediment,bodies in circular motion would recede from the center.79 However, notice that in thePrinciples the inclination is said to explain not only what would happen in certaincounterfactual circumstances, but also what does happen at the actual instant thebodies possess this inclination. Thus, Descartes observes in his discussion of the sec-ond law in this text that “our hand can experience this in the stone itself while


78. Admittedly, there is little in the way of an explicit argument in this section for the con-clusion that the motion must be at a tangent from the circle as opposed to some other straightpath. For a criticism of an argument for this conclusion in The World, see Garber 1992,221–23. See also the discussion in Des Chene 1996, 281–82, of different arguments for thisconclusion in the work of later Cartesians.

79. A similar view of this passage is suggested in Garber 1992, 219.

we turn it in the sling,” since we feel the stone pulling away from us (PP II.39, AT8-1:64).80 So just as forces are real modes in bodies that produce changes in motiondue to collision, so inclinations are real modes of motion that produce the pull of abody in circular motion away from the center.

If this account is correct, then for Descartes the bodies in motion that God con-tinuously conserves have as modes of their duration various forces that determine theoutcomes of collisions, just as the motion that he conserves has as modal features ofitself various inclinations that determine not only how it would proceed if unim-peded, but also certain effects that a moving body actually does have. These forcesand inclinations are therefore true causes secundum fieri that produce the particularchanges due to contact among bodies. But if the three laws of nature are groundeddirectly in such bodily causes, it might seem odd that Descartes consistently pre-sented them as deriving from God’s activity as universal cause of motion. On theview that there are no causally efficacious inclinations or forces in bodies, thisappeal to God would seem to be straightforward, since God would be required as thetrue cause of the effects attributed to these bodily features. However, if one takesseriously the persistent suggestion in Descartes that inclinations and forces arecausally efficacious features of bodies and their motions, there is a need to recon-sider his account of the manner in which the activity of such features as particularcauses is related to the universal causality of God’s ordinary concursus.

3.2.3. Divine and Bodily Causes

In 1678, the French Cartesian Nicolas Malebranche criticized those who attributedto Descartes the view that bodies can move each other, “against what he saysexpressly in articles 36 and 37 of the second part of these Principles of Philosophy”(Malebranche 1958–84, 3:238/Malebranche 1997, 677). More recently, Garberclaims that “it seems to me as clear as anything that, for Descartes, God is the onlycause in the inanimate world of bodies, that bodies cannot themselves be genuinecauses of change in the physical world of extended substance” (Garber 1993, 12).An initial consideration of the articles that Malebranche cited seems to confirm anoccasionalist reading of Descartes’s account of body–body interaction. For in article36 of the second part of the Principles Descartes holds that the conservation of thetotal quantity of motion (and rest) follows simply from the divine attribute ofimmutability. Though he also refers in article 37 to laws of nature as secondary andparticular causes of motions in particular bodies, it is understandable that some havetaken these laws as well to follow directly from divine immutability. After all,Descartes holds that the reason for the second law is the same as the reason for thefirst, namely, “the immutability and simplicity of the operation by which God con-serves motion in matter” (PP II.39, AT 8-1:63). And he notes that the proof of thesecond part of the third law, concerning the transfer in collision of a quantity ofmotion from a stronger body to a weaker one, consists in “the immutability of the


80. The last point about the pull of the stone is made explicitly in the version of this pas-sage in the French edition, at AT 9-2:86.

operations of God, now continually conserving the world by the same action withwhich he created it then” (PP II.42, AT 8-1:66). For Garber, the clear indication hereis that “God stands behind the world of bodies and is the direct cause of theirmotion” (Garber 1993, 14). As I indicated at the outset of this chapter, he takesDescartes to substitute God for the substantial forms that the scholastics posited toexplain natural change in the material world.

It is beyond doubt that Descartes rejects the appeal to material substantial formsand real qualities to explain natural change. What is less clear is that his alternativeaccount of body–body interaction in the Principles takes God alone to stand in forsuch forms and qualities. For as we have seen, there is an explicit reference in thediscussion of the three laws of nature to internal features of bodies and their motionsthat provide the basis both for certain kinds of persistence and for changes in motiondue to collision. The immutability of God’s ordinary concursus is of course alwaysin the background, but the official view in this text is that this action is only auniversal cause that requires supplementation by the particular causes of motion.

We can understand the division of labor suggested here by drawing on Descartes’sview in the Principles that God’s ordinary concursus consists simply in the fact thathe “diversely moved parts of matter when he first created it, and now conserves thiswhole matter with the same plan [eademque ratione] by which he first created it” (PPII.36, AT 8-1:62). Given my reading above of Descartes’s account of the ontology ofbodily force, we can take the claim here to be that God originally created movingparts of matter with durations that have degrees of strength measured by the quantityof their motions. Just as these parts are modified by a quantity of motion, so the wholeof matter is modified by a quantity that is the sum of the quantities of motion thatmodify its parts.81 The object of God’s act of creation/conservation is simply the totalquantity of matter as modified by the total quantity of the motion of its parts (seefigure 3.1). God conserves the total quantity of motive force—the superadded“impulse” mentioned in the correspondence with More—simply by conserving thetotal quantity of the durational strength of the moving parts that he infused into mat-ter at the start.82 But this act of determining the total quantity itself underdeterminesthe precise manner in which the quantities are distributed across the parts over time.The determination of this distribution requires further the action of the motive forcesof the individual parts, which produce changes only in the case of collision.

Interestingly, a similar sort of division of labor in the production of motion isfound in the critique of Malebranche’s occasionalist physics in Fontenelle’s Doubts


81. The quantities of the parts are therefore modes of the total quantity that is itself a modeof matter as a whole. In contrast, as we have seen, the parts themselves are substantial partsrather than modes of the whole matter (see chapter 2, note 9). The result here is that movingparts are substances that possess as modes quantities that are themselves modifications of thetotal quantity of motion that modifies the material substance that includes these parts.

82. Cf. Clarke’s view that according to Descartes, “God imparted a real quality called forceor power to physical bodies, and that the amount of this power is fixed by the immutability ofhis creative action” (Clarke 1996, 335). Clarke characterizes this quality as a mode of body, butdoes not indicate precisely what sort of mode it is. My identification of forces with modal fea-tures of the durations of bodily motions can be seen as a friendly amendment to his view.

concerning . . . Occasional Causes (1686), a work that I cited at the outset of chap-ter 1. In this text, Fontenelle proposes that “the moving force of God is that by whichhe produces a motion that was not,” whereas “the moving force of creatures is thatby which they pass a motion that is already there from one body to another”(Fontenelle 1989–2001, 1:562). We have seen that in his correspondence with More,Descartes attempts to distance himself from the view that motion is literally passedfrom one body to another in collision. However, I think he could accept Fontenelle’ssuggestion that God creates and conserves a certain total quantity of motion in mat-ter by means of his universal causality, whereas particular bodies change the distri-bution of this total quantity among themselves by means of collision.83

According to Descartes, however, God continuously conserves not only the totalquantity of motion, but also what is essential to particular motions. More specifi-cally, he so conserves the instantaneous inclinations to rectilinear motion. Theseinclinations in turn explain the presence of the centrifugal pressure involved incircular motion. In a striking passage from The World, Descartes attempts to illus-trate the differences between the contributions of the universal and particular causesof determination by drawing an analogy to God’s role in the production of sinfulaction. He notes there that “the theologians teach that God is also the author of ouractions, insofar as they exist and insofar as they have some goodness,” and that “it isthe various dispositions of our wills that can render them evil.” Just so, according toDescartes, “God is the author of all the motions in the world insofar as they exist and


Total Quantity of Matter( = sum of quantities of parts)

TotalQuantity of

Motion( = sum of quantitiesof motion in moving

parts)

FIGURE 3.1 Object of Divine Concursus

83. See also the claim in Regis’s Use of Reason, cited in note 47, that whereas God pro-duces the “substance” of motion, secondary causes produce various modifications of thismotion.

insofar as they are rectilinear,” whereas “it is the various dispositions of matter thatrender them irregular and curved” (W VII, AT 11:47–48).84

Toward the start of §3.2, I noted the puzzle that Descartes’s three laws of natureseem to be too heterogeneous to be grounded in the same way in God’s conservationof the quantity of motion by means of his ordinary concursus. We now have thematerial for a solution to this puzzle. God conserves motion or motive force simplyby continuing his act of creating matter divided into parts with motions that possesscollectively a certain total strength of duration, and that possess individually instan-taneous inclinations to motion along a straight path. The continuation of the initialact of creation results in the continuing presence of the forces and inclinations thatprovide the foundation for the three laws of nature. The laws therefore follow fromGod’s immutability in the sense that they follow from the matter in motion that heimmutably creates/conserves as universal and primary cause. The laws are particu-lar causes, moreover, in the sense that they reflect the nature of the inclinations andforces that are themselves the particular and secondary causes of changes inmotions. Contrary to what one might think initially, then, Descartes does not holdthat God directly creates the laws, which in turn condition matter in motion. Rather,the view that I find in him is that God directly creates matter in motion, and that thelaws merely reflect the natures of what God has created.

To this point I have emphasized the case of God’s conservation of the quantityof motion through his conservation of the total quantity of acting forces—the viresad agendum posited in the Principles. As I have indicated, however, there are seri-ous complications in the case of God’s conservation of the quantity of rest.Whereas motive forces have a quantifiable strength even prior to collision, theresisting forces in resting bodies have such a strength only at the instant those bod-ies collide with other bodies. Thus, the activity of these resisting forces—the viresad resistendum of the Principles—cannot depend solely on God’s conservation ofa quantifiable tendency in bodies to persist in a certain state. Rather, what isrequired is some additional impulse from God at the moment of impact that servesto resist the action of the moving body in cases where this body is equal to orsmaller than the body at rest. As we will discover, however, the suggestion thatthere is a special sort of divine concurrence with the action of the resting body isout of line with the strongly conservationist character of Descartes’s account in thePrinciples of the ordinary concursus that exhausts God’s natural contribution tobody–body interaction.


84. In the first volume of his Search after Truth (1674), Malebranche appeals to a similaranalogy in holding that just as “all motions make a straight line” and deviate from this onlydue to “some foreign and particular causes that determine them and that change them intocurved lines by their oppositions,” so “all inclinations that we have from God are straight”insofar as they are directed toward the “good in general,” but only “if they have no foreigncause that determines the impression of nature toward evil ends” (bk. 1, ch. 1, Malebranche1958–84, 1:45/Malebranche 1997, 4). For discussion of Malebranche’s use of this analogy toexplain human freedom, and some indication of the problematic nature of this use, seeSchmaltz 1996, 220–22.

3.3. DESCARTES’S CONSERVATIONIST PHYSICS

It is clear from several passages cited in §3.1.3 that Descartes helps himself to thescholastic notion of divine concursus in describing God’s activity as universal cause.I also noted in this section, however, that in a majority of these passages Descartestends to conflate God’s concursus with his act of conservation. The question I nowwant to address is whether his account of God’s universal causality of motion sys-tematically conflates the two, or whether there is room in his physics for somethinglike a traditional form of scholastic concurrence.

We know that Suárez distinguished sharply between God’s conservation of thebeing of creatures, which does not differ in reality from his act of creating that being,and his concursus with secondary causes, which varies with the actions of thosecauses (see §1.2.3 (ii)). Given this Suárezian position, our question now is whetherDescartes’s physics requires an act of concursus on God’s part that differs from hisact of conservation and that varies with the action of the secondary cause. But oncethe question is put in this manner, the answer would seem to be clearly in thenegative. For Descartes consistently indicates that there is only a single unvaryingaction that God contributes as universal cause of motion. Just as he insists in TheWorld that the laws of nature “depend only on the fact that God conserves each thingby a continued action” (W VII, AT 11:44), so he claims later in the Principles thatthe law governing transfers of motion “is demonstrated by the immutability of theaction of God, continually now conserving the world by the same action by whichhe created it then” (PP II.42, AT 8-1:66).

Of course, it was a theological commonplace that God himself is wholly immutable.Moreover, Suárez allowed that divine concursus can be said to be always the same ifone is speaking of “the internal concursus or volition by which God concurs” (MDXXII.4, ¶8, 1:831). What changes is only the external concursus, the action that residesin the patient. However, Descartes insists not only that the principle of action in God isimmutable, but also that what he produces is always the same. Thus, he claims in TheWorld that it follows from divine immutability that God “always produces the sameeffect,” and in particular always produces “a certain quantity of motions in matterin general” (W VII, AT 11:43).85 And though Descartes seems to allow in the Principlesthat God could produce changes by means of acts that go beyond his ordinaryconcursus, he continues to hold that this concursus itself yields a constant effect.

This argument for the constancy of the effect of divine activity appears to beproblematic. After all, Descartes allows, with Suárez, that an eternal principle inGod can yield temporal effects. Why not grant, with Suárez again, that an immutabledivine principle can yield an inconstant concursus? Here I think it is best to under-stand Descartes’s line of argument in terms of Suárez’s account not of the divineconcursus, but rather of divine conservation. Suárez argued that God’s creation of anobject cannot be distinguished from his subsequent conservation of it, since in bothcases the agent is producing in the same way, namely, ex nihilo, the very same effect,


85. On the possible significance of the fact that this passage refers to motions rather thansimply to motion, see note 18.

namely, the esse of the object. We can derive the sameness of the act from the factthat the mode of production and the terminus remain the same (MD XXI.2, ¶3,1:791). Descartes assumes that God produces motion as a universal and primarycause by means of a single concursus identical to his creation of matter in motion.Drawing on Suárez’s view of conservation, however, we can infer that since this actremains the same, it must have some common terminus, which for Descartes, as I seeit, would be the existence of a constant total quantity both of matter and of the dura-tional strength (i.e., force or impulse) of its moving parts.

If my reading of Descartes is correct, then there can be in his physics no differencebetween God’s ordinary concursus and his continuous conservation of matter inmotion. Since this continuing action must have a constant effect, the changes in motionproduced by bodily collisions must be due not to that action, but rather—as Descarteshimself indicates—to the particular and secondary causes of motion. We are far herefrom the concurrentist position in Suárez that the diverse actions that produce suchchanges are identical to God’s action. Instead, Descartes seems to me to be closer tothe mere conservationism of Suárez’s opponent Durandus. Recall that in Durandus’sview, though God is the immediate cause of the being of secondary causes, his onlycontribution to the action of secondary causes is his conservation of such causes (see§1.1.3). In Descartes’s case, the view is that God’s ordinary concursus is exhausted byhis continuous conservation of matter with the forces of its parts and inclinations of itsmotions. These forces and inclinations, rather than the divine concursus itself, are theimmediate causes of changes due to collisions among these parts.

Thus, far from placing Descartes with the occasionalists, I see him as advocatingan alternative to this position that departed from occasionalism more radically eventhan Suárez’s concurrentism. There is admittedly one passage in which Descartesseems to favor the view, directly opposed to Durandus’s mere conservationism, thatGod is in fact the immediate cause of our actions. This passage is from a 1641 letterto Regius in which Descartes offers his Dutch correspondent a response to the objec-tion of their common critic Gisbertius Voetius that mental faculties rather than ourmind itself must be the immediate principle of volitional and intellectual acts.

What Voetius marks down here of you in no way opposes you. For sincetheologians indeed say that no created substance is the immediate principle of itsoperation, they understand this as follows: that no creature can operate without theconcursus of God; not, however, that it ought to have a certain created faculty,distinct from itself, through which to operate. For it would be absurd to say thatsuch a created faculty could be the immediate principle of a certain operation, butthe substance itself could not. (May 1641, AT 3:372)86

But notice that his main conclusion here is merely that it does not follow from thefact that no created substance is the immediate principle of its operation that everysubstance acts through a faculty that is distinct from it. There is no requirement thatthe divine concursus on which all effects in the created world depend is in fact


86. This is one of the passages that Pessin cites in support of his view that Descartesaccepted a kind of “nomic concurrentism” (in Pessin 2003, 37). See note 88.

identical to the operations through which creatures produce those effects. We there-fore cannot assume that Descartes’s talk of God’s immediate concursus indicates apreference for Suárez’s concurrentism over Durandus’s mere conservationism. Andgiven his emphasis on the constant nature of the product of God’s ordinary concur-sus, there is even reason to think that Descartes could not distinguish concursus fromconservation in the manner Suárez required.87

But though Descartes himself did not sharply distinguish God’s conservation ofthe world from his concursus with secondary causes, we can reconstruct such a dis-tinction in terms of the contrast in his correspondence with More between God’s cre-ating matter at rest and “leaving it to itself ” and his creating matter with an“impulse” that involves the motion of its parts. In the former case there would be nomotive forces or inclinations with which God could concur, whereas in the latter caseGod’s initial shove of matter would yield forces and inclinations with which Godmust concur. Even so, on Descartes’s official view this concursus involves nothingmore than God’s conservation of matter with a total quantity of durational strengthand of motions with their instantaneous inclinations.

Des Chene has proposed an alternative manner of distinguishing between con-servation and concurrence in Descartes’s physics.88 In his view, God’s concurrenceconsists in the “moving force” by which he causes a moving body to change thestates of other bodies with which it collides, whereas his conservation consists in the“resisting force” by which he resists changes in collision to the state of rest in a rest-ing body (Des Chene 1996, 334–36). Since Des Chene takes Descartes to deny that


87. In his Use of Reason, Regis follows Descartes in speaking of God’s concours with sec-ondary causes (see chapter 1, note 77), but also indicates that in the case of body–body inter-action, this concourse involves only God’s creation of motion simpliciter, and not the particulardeterminations of motion that secondary causes produce (I-2.32, Regis 1996, 384–85; cf.I 2.11, 271–74). Regis’s position admittedly requires the conclusion, not to be found inDescartes, that the motion that God produces is an atemporal nature that particular motionsexpress in a temporal manner (see the discussion of this position in Schmaltz 2003). Despitethis important difference from Descartes, the case of Regis serves to reinforce the lesson thatwe cannot simply assume on the basis of the fact that Descartes used concurrentist terminol-ogy that he is committed to a traditional scholastic understanding of divine concursus.

88. See also Pessin’s attribution to Descartes of a “nomic concurrentism” on which “Godconcurs with bodies in their effects on each other via willing the laws of motion; but theselaws produce their effects only in conjunction with the relevant ‘initial’ conditions, such asstates of matter” (Pessin 2003, 40). In contrast to conservation, concurrence involves not onlyGod’s willing the laws, but also a contribution on the part of the material conditions. GivenPessin’s insistence on the intrinsic passivity of the matter Descartes posited (40), however, thematerial conditions would seem to serve merely as occasions for divine action. Pessin’s viewtherefore seems to be a version of an occasionalist reading of Descartes. Moreover, there is noreference in Descartes to a separate willing of the laws of nature on the part of God. The indi-cation in the Principles is rather that God’s contribution to the laws is exhausted by his con-tinued conservation of the matter and motion that he first created in the world. The fact thatthese particular laws hold is simply a consequence of this divine action, and not something tobe explained by divine volitions that have these laws as their content.

bodies can be real efficient causes, he admits that in concurring God is not genuinelyacting with secondary causes (336–37). Nevertheless, he insists that Descartes canpreserve “some sort of distinction between conservation and concurrence” that has“been refashioned to suit a physical world without active powers” (341).

I have already argued that Descartes’s intention is to allow for a physical worldthat has an internal source of activity. But even apart from this point, the suggestionthat moving force in Descartes pertains to concurrence rather than to conservationseems to me to be problematic. For Descartes indicates in the Principles that boththe total quantity of matter and of the motion (and rest) of its parts are conserved bya single divine act, namely, the ordinary concursus that is merely a continuation ofthe original act by which God created the material world. Indeed, if any additionalact of concurrence be involved at all, it would seem to pertain to the case of resist-ing force. In the case of moving force, it can be said that there is a modal quantitythat is continuously conserved in all natural interactions. As I have noted, however,the resisting force that Descartes posits in his fourth and sixth collision rules is nota conserved quantity, but rather something that enters the scene only at the momentof collision and that is not conserved after that moment. Thus this force cannot bereferred solely to the ordinary concursus by which God produces a constant effect.But the clear view in the Principles is that this concursus provides the sole divinefoundation for the laws of nature. It therefore seems to me that far from providing abasis for a conception of divine conservation, as Des Chene suggests, resisting forcescannot be explained in terms of the sort of conservation that Descartes took to con-stitute God’s activity as universal cause of the material world.

Even if Descartes can conceive of God’s concursus with the motive forces ofmaterial parts in a conservationist manner, however, special problems seem to arisewhen one considers his views on mind–body interaction. For one thing, there is theobjection, from Descartes’s time to our own, that he is not entitled to the claim thatthere is such interaction given the implication of his dualism that minds and bodieshave radically heterogeneous natures. I argued in the previous chapter thatDescartes’s containment axiom does not preclude such interaction. However, I alsoconceded toward the end of this chapter that this axiom does not provide much helpin explaining precisely how unextended minds and extended bodies produce changesin each other. In the case of body–body interactions, Descartes at least can appeal tobodily forces that have comparable degrees of strength. But what sort of forces couldbring objects as dissimilar as mind and body into causal relation with each other?And what precisely is God’s role in bringing about this sort of interaction? It turnsout that Descartes has even less to say about mind–body interaction than he doesabout body–body interaction. Nevertheless, we will discover that he says enough toindicate that it is his considered position that though mind-to-body action is in somerespects less problematic than body-to-mind action, even the latter is possible withthe help of divinely instituted natures. And as was the case with his accounts ofbody–body interaction and the causal axioms, we will discover that we cannot fullycomprehend the details of his views on mind–body interaction in abstraction fromtheir scholastic context.


129

4

Causation in Psychology

Scholastic psychology, as the science of the soul (in Greek, psuche-), was a study of lifein general. Guided by the procedure in Aristotle’s De Anima, the scholastics positedsouls to explain not only the rationality of human beings or the sentience of animals, butalso the vegetative functions of plants.1 Descartes rejects this sort of science of the soul,as indicated by his protest in a 1641 letter to Regius (then a friend; see §4.2.2) that thevegetative and sensory souls of the scholastics are nothing more than “locomotive pow-ers” that consist in certain arrangements of bodily parts, and that are thus distinct in kindfrom the rational soul (May 1641, AT 3:371). It is true that Descartes at times helps him-self to the scholastic view that the rational soul is the “substantial form” of the humancomposite.2 However, he could not accept, in the manner scholastic traditionalists did,the project of showing how the rational soul provides the foundation for all the func-tions of the living human body.3 As he insists to Regius, “no actions can be reckonedhuman unless they depend on reason” (3:371). More accurately, Descartes’s view is thatonly those actions that involve some kind of thought can be the subject of humanpsychology. What is new to his psychology is the exclusive focus on the relation ofthought to the motions in the machine that for him constitutes the human body.

1. The best recent discussion of the scholastic science of the soul is found in Des Chene2000a. Des Chene 2001 is a companion volume that considers the alternative in Descartes tothe portions of this scholastic science that concern living bodies.

2. See, for instance, To Regius, Jan. 1642, AT 3:503, 505.3. I say ‘traditionalist’ because there were scholastics who deviated from the standard

Thomistic line, to which Suárez adhered, that there is only one substantial form of the humancomposite. In the letter I have quoted, Descartes tells Regius that his claim that human beingshave a threefold soul “is a heretical thing to say” (May 1641, AT 3:371).

We will discover that Descartes often speaks as if thoughts and correspondingmotions casually interact. It would appear that given the restriction of his containmentaxiom to efficient causes and their effects (see §2.1.2 (ii)), this interaction involvesefficient causality. Notoriously, however, it has seemed to many that the claim thatmind and body are real efficient causes of changes in each other is particularly prob-lematic for Descartes. I mentioned at the outset of the first chapter the claim inFontenelle’s Doubts that the system of occasional causes arose from the “extreme dis-proportion” that Descartes introduced between mind and body. Robert Richardsonhas dubbed the view that such a disproportion renders mind–body interaction prob-lematic “the problem of heterogeneity” (Richardson 1982), or what I call, for short,the heterogeneity problem. The assumption central to this problem is that causal inter-action requires a kind of likeness between cause and effect that is in fact missing inthe case of objects as heterogeneous in nature as Descartes took mind and body to be.

Sometimes Descartes’s way with the allegedly scandalous heterogeneity problemis short. In response to Gassendist questions concerning the intelligibility of the unionof mind and body, for instance, Descartes notes curtly, in a section added to the 1647French edition of the Meditations, that “the whole difficulty that [such questions] con-tain proceeds solely from a supposition that is false and cannot in any way be proved,namely that, if the soul and the body are two different substances with diverse natures,this prevents them from being able to act on each other” (AT 9-1:213).4 This line ofresponse is supported to some extent by the result of our discussion in chapter 2 thatDescartes’s containment axiom is consistent with at least some forms of mind–bodyinteraction. Even so, when Princess Elisabeth raises the question in 1643 of how animmaterial soul can determine the bodily animal spirits so as to perform voluntaryactions, Descartes replies that this question “seems to me to be that which one can askme most properly in view of the writings I have published” (21 May 1643, AT 3:664).He then attempts to address this question by appealing to the “primitive notion” of thesoul-body union. But though the suggestion here is that this primitive notion isrequired to explain any sort of mind–body interaction, Descartes’s considered posi-tion is more complicated than this. For reasons that are intelligible in light of thescholastic context of his thought, Descartes never admits the possibility of bodilyaction on a mind not united to it, but explicitly allows that disembodied minds notonly can but actually do act on body.

To explain Descartes’s complex views on mind–body interaction, I begin in §4.1with his account of the soul–body union, starting with the one he provides in his1643 correspondence with Elisabeth. It turns out that both Elisabeth’s objections andDescartes’s response in this correspondence address difficulties concerningmind–body interaction and the union that differ from those deriving from theheterogeneity problem.

In §4.2, I consider Descartes’s account of that portion of the union involvingbody-to-mind action. The difficulties with this account can be understood in termsof the scholastic problem of the relation of the bodily senses to intellectual cogni-tion. Such difficulties are reflected most clearly in a famous passage in which


4. See also To Hyperaspistes, Aug. 1641, AT 3:425–26.

Descartes refers to motions in the brain as “occasions” for the formation of sensoryideas by an innate faculty of mind. But though it is tempting to take the occasioninghere to be noncausal, there is evidence that Descartes wants to make room for theposition that brain motions serve as a special sort of efficient cause.

In §4.3, I turn to Descartes’s account of mind-to-body action. Here again, there isa link to earlier scholastic discussions—in this case, concerning the action of incor-poreal spirits on corporeal objects. What is particularly relevant to Descartes’s vari-ous comments regarding mind-to-body action is the insistence in scholastics such asSuárez that spirits must be “present” to bodies to move them. This scholastic back-ground helps to make sense of Descartes’s attempts to explicate the action of mindon body in terms both of the action of the quality of heaviness and of an “extensionof power” found in minds. What is not anticipated in the scholastic discussions, how-ever, is the problem in Descartes—to which Leibniz most notably drew attention—of the compatibility of mind-to-body action with the basic principle in Cartesianphysics of the conservation of the total quantity of motion. In fact, Descartes him-self never explicitly endorsed the position—which Leibniz attributed to him andwhich later Cartesians clearly embraced—that finite minds can only change thedirection of moving bodies, and not create new motion. Even so, there are reasonsderiving from his conservationist physics for him to adopt such a position.

4.1. MIND–BODY INTERACTION AND UNION

When addressing the issue of mind–body interaction, Descartes often emphasizeshis position that our soul is united in a special way to the human body. This is under-standable given the implication of this position that the union involves a special sortof interaction. However, in considering his views on this topic, it is important not toconflate the issues of union and interaction. After all, we will discover that Descartesinsists on the possibility that certain immaterial entities, such as God and angels, canact on bodies even though they are not united to any of them in the way in which thehuman soul is united to a body. I also find in his writings the implication that dis-embodied minds can have no sensory states. On my reading, then, the problem in thecase of the human soul of how the body to which it is united can cause its sensationssimply does not arise for such minds.

In contrast to the heterogeneity problem, which yields a single problem withmind–body interaction, there are in Descartes several different problems with suchinteraction, only some of which concern the union. In what follows, I attempt to dis-tinguish various issues concerning mind–body interaction and union and to show howsuch issues are or are not related to Descartes’s account of the union. I begin with hisimportant exchange with Elisabeth, which broaches both issues. Then, following thelead of Descartes’s remarks to Elisabeth, I focus on the issue of the union. Though hiswritings may seem at times to suggest that the union merely consists in a certain set ofconnections between mental and bodily states, I take his more considered position tobe that the union also involves something underlying those states that serves to explaintheir connections. Finally, I consider Descartes’s view that our sensory and volitionalstates reflect the special sort of interaction that occurs in the case of the union.

Causation in Psychology 131

4.1.1. Elisabeth’s Objections and Descartes’s Response

Perhaps Descartes’s most famous comments on the issue of mind–body interactionoccur in his 1643 correspondence with Princess Elisabeth of Bohemia. Elisabeth wasthe daughter of Elizabeth Stuart, electress Palatine and queen of Bohemia (and sis-ter of Charles I), and Frederick V of Bohemia. Elisabeth’s family went into exile inThe Hague after Frederick was deposed, and there is a report that Descartes spentsome time at the house of Elizabeth Stuart during 1634/35 (Frederick had died in1632). He may have met the daughter Elisabeth at that time, but his first mention ofher is in a 1642 letter to their mutual friend Alphonse Pollot (also Pallotti), in whichhe noted her interest in his work on metaphysics (6 Oct. 1642, AT 3:577–58). In May1643 Descartes made an unsuccessful attempt to meet with Elisabeth in The Hague.Touched by the effort, and heartened by good wishes that Pollot had passed fromDescartes to her, Elisabeth sent Descartes a letter of thanks a few days later.5 In thatletter, she also broached a particular problem concerning the relation of the soul tothe body. She expressed this problem as follows:

I ask you to tell me how the human soul can determine [determiner] the bodilyspirits, to make voluntary actions (it being only a thinking thing). For it seems thatall determination of motion takes place by the moved object being pushed, by theway in which it is pushed by what moves it, or by the qualification and shape ofthe surface of the latter. Touch is required for the first two conditions, extensionfor the third. You entirely exclude the latter from your notion of the soul, and theformer seems to be incompatible with an immaterial thing. (6/16 May 1643,3:661*)

It is sometimes claimed that the heterogeneity problem lies behind the difficultythat Elisabeth raised here concerning interaction.6 Notice, however, that her diffi-culty concerns not the general question of whether substances with distinct naturescan causally interact, but the more specific question of whether the soul is able to acton the body. In particular, the question is whether something that lacks extension isable to “determine” the motion of a body by pushing it. What seems to render thissort of determination problematic is the assumption that only something that can bein contact with a body can move it.7


5. For more on Elisabeth and her interactions with Descartes, see Nye 1999; Shapiro1999b; and Hutton 2005.

6. See, for instance, Margaret Wilson’s identification of “the difficulty of rationalizingcausal relations between distinct sorts of substances” with the difficulty on which Elisabethinsisted (Wilson 1978, 215). Cf. Radner 1971; Mattern 1978; and Broughton and Mattern1978. But cf. the claim in Shapiro 1999b that “Elisabeth’s questions of Descartes go beyondthis problem” of the causal interaction of really distinct substances (506).

7. There is a question here whether Elisabeth had in mind here the technical understand-ing of determination in Descartes, according to which it is a modal feature of motion that canbe changed apart from any change in the speed of that motion (see §3.2.1 (ii)). For the claimthat she did have such an understanding in mind, see Tollefsen 1999. However, as I indicatepresently, Elisabeth restated her original objection without appealing to the notion of deter-mination, asking instead about how the soul “can move [peut mouvoir] the body” (June 10/20,

In a subsequent letter to Descartes, Elisabeth again mentions the problem of“how the soul (nonextended and immaterial) can move the body.” However, shealso introduces a new problem concerning the action of body on the soul, notingthat it is “very difficult to comprehend that the soul, as you have described it, afterhaving had the faculty and habit of reasoning well, can lose all that by some vapors,and that, being able to subsist without the body and having nothing in common withit, it is so ruled by it” (10/20 June 1643, AT 3:684–85*). Elisabeth’s claim here thatthe soul has “nothing in common with” the body that acts on it may seem to sug-gest that the heterogeneity problem is behind her new concern with interaction.Nonetheless, her worry is not merely that soul and body have distinct natures, butrather that the soul has a power of reasoning that is superior to bodily vapors, andthus should be unaffected by them. This superiority is indicated by the fact thatthe soul is “able to subsist without the body” in virtue of possessing this power.Elisabeth’s remarks recall Descartes’s argument in the Discourse that an immate-rial soul must be invoked to explain language use. The argument appeals to the factthat we can provide the appropriate verbal responses in innumerable different cir-cumstances in support of the claim that our language use derives from reason “asa universal instrument” rather than from the limited dispositions of our bodilyorgans.8 Descartes concludes that it is in virtue of possessing reason that “our soulis of a nature entirely independent of the body, and consequently is not bound to diewith it” (DM V, AT 6:56–61). Elisabeth’s challenge to Descartes is to show how thereasoning of the soul can be affected by bodily vapors given that the rational soulis so superior to anything found in body. What seems to create difficulty here is theunderlying assumption that an object cannot act on, and thus “rule over,” anotherobject more perfect than it.

In Elisabeth’s two letters, then, we have two different problems regarding thecausal interaction of soul and body. The first is that the soul, as an unextended thing,does not have the sort of contact with body required to act on it, whereas the secondis that the body, as more imperfect than a soul capable of reason, does not have thesort of perfection required to act on the soul. As we will discover, both of these dif-ficulties with interaction have analogues in scholastic discussions, particularly in thework of Aquinas and Suárez, concerning the causal interaction of immaterial andmaterial beings. The first difficulty is linked to the issue in scholasticism of whethera “spiritual presence” is required for spirits to move bodies (see §4.3.1), and the sec-ond to a problem in scholastic thought concerning the relation of bodily phantasmsto intelligible species in the incorporeal intellect (see §4.2.1).

Descartes’s response to Elisabeth focuses not so much on these particular difficul-ties, however, but more on the nature of the union of our soul with body. This empha-sis on the union is somewhat surprising. Though Elisabeth mentioned the human soul


AT 3:684). Moreover, none of Descartes’s responses to Elisabeth appeal to this notion. Asindicated in §4.3.3, however, Descartes’s views on determination are important for laterCartesian discussions of the causation of motion by finite minds.

8. Chomsky has emphasized this argument as an early version of the “poverty of stimu-lus” argument for innatism; see Chomsky 1966.

in particular when expressing the problem in her initial letter concerning the action ofthe soul on body, the problem itself does not appear to be restricted to this particularcase. For the question of whether an unextended thing can push a body would seemto apply also to God or, if an appeal to divine omnipotence solves the problem inthis case, to finite spirits distinct from human souls, such as angels. Nevertheless, inhis reply to Elisabeth’s initial letter, Descartes focuses on the fact that “there are twothings concerning the human soul on which depend all the knowledge that we canhave of its nature, the first that it thinks, the other, that being united to a body, itcan act and be acted on by it [peut agir et patir avec lui] (21 May 1643, AT 3:664).He acknowledges that he said “almost nothing” (quasi rien) in his writings aboutthe second feature of the human soul,9 but adds that we have a “primitive notion” ofthe union that is distinct from the primitive notion of the body alone and of the soulalone. In the case of body, we have “the notion of extension, from which followsthose of shape and motion,” and in the case of the soul, we have the notion “ofthought, in which are included the perceptions of the understanding and the inclina-tions of the will.” The notion of the union is distinctive, however, since it is that “onwhich depends that of the force [force] that the soul has to move [mouvoir] the body,and the body to act [d’agir] on the soul, in causing [causant] its sensations and pas-sions” (AT 3:665).

Though this passage mentions both the action of body on soul and the action ofsoul on body, in this letter Descartes follows the lead of Elisabeth’s initial query infocusing on the latter form of action. Thus, he emphasizes the need to refrain fromconceiving “the manner in which the soul moves the body by that in which a body ismoved by another body” (AT 3:666). In particular, he holds that we must refrain fromthinking that the soul moves a body by means of “a real contact of surfaces” (AT3:667). Even so, when Elisabeth subsequently protested that “it seems to me easier toconcede matter and extension to the soul, than the capacity to move a body . . . to animmaterial being” (10/20 June 1643, AT 3:685*), Descartes invites her “to attributematter and extension to the soul,” since to conceive of the soul as having this sort ofextension is “nothing other than to conceive it as united to body.” Yet he also warnsthat the “extension of thought” that is attributed to the soul must be distinguishedfrom the extension of matter, which unlike the extension of thought “is determined toa particular place, from which it excludes all other extension of body” (28 June 1643,AT 3:694).

I will return in §4.3.2 to the positive account that Descartes offers in the corre-spondence with Elisabeth and elsewhere of the extension of thought, or what he else-where calls the extension of power. For the moment I restrict myself to a couple ofobservations concerning his response to Elisabeth. The first is that in attributingextension to the soul, he directly addresses Elisabeth’s initial objection that an imma-terial thing does not have the sort of contact with a body required to move it. Such anattribution allows him to hold that there is in fact some sort of contact between thetwo, albeit not a real contact of surfaces. Though some of Descartes’s language in his


9. One puzzle, which I address in §4.1.3, is how Descartes could say that he said almostnothing about the union even though he previously argued in the Sixth Meditation that natureteaches him that he is “very closely joined and as if intermixed” with a body (see AT 7:81).

correspondence may seem to indicate that he was merely patronizing the princess,10

on this issue, at least, he seems to me to take her objection to heart.My second observation is related to the initial impression of a confusion in

Descartes of the distinct issues of the union and interaction. It seems that Descartescannot simply claim, as he does in his letter to Elisabeth, that the conception of thesoul as having a special sort of extension simply amounts to the conception of it asunited to a body. For again, Elisabeth’s original objection concerns the intelligibilityof the conception of the action of any immaterial thing on a body, whether or not thatthing is akin to a human soul in being united to a body. If an extension of thoughtmust be attributed to an immaterial substance to conceive its action on body, thensuch an extension must be attributed to all immaterial substances that can so act,including God and angels as well as human souls.

Once the notion of the “extension of thought” has been distinguished from thenotion of the union, though, it is natural to wonder what real work the latter is doingfor Descartes. To see what work it is doing, I think it is important to deviate fromDescartes’s procedure in his correspondence with Elisabeth by addressing issues thatgo beyond the intelligibility of interaction per se. A good place to start is with thequestion that Margaret Wilson has raised concerning the nature of the union thatDescartes posited. Wilson’s question is whether he accepted a “natural institutiontheory,” on which the union simply consists in a particular kind of interaction, orrather a different sort of “co-extension theory,” on which the union involves an“intermixture” that makes possible certain correlations of mental and bodily states(Wilson 1978, 204–20). Whereas Wilson’s judgment is that the natural institutiontheory is most plausible for Descartes, I hope to show that despite its problems, thereis at least one important respect in which the co-extension theory is superior.

4.1.2. Two Accounts of the Union

As Wilson defines it, the natural institution theory holds that “what we call the closeunion or intermingling of this mind with this body is nothing but the arbitrarilyestablished disposition of this mind to experience certain types of sensations on theoccasion of certain changes in this body, and to refer these sensations to (parts of)this body” (Wilson 1978, 211).11 One piece of evidence for the attribution of thistheory to Descartes is drawn from his remarks concerning the union in the SixthMeditation. There Descartes claims that even though “human nature could have beenso constituted by God that the same motion in the brain would exhibit somethingelse to the mind,” still God so constituted this nature that


10. See especially his suggestion that it would be best for her to refrain from too muchmetaphysical meditation and to give herself over to use of the imagination and the senses (28June 1643, AT 3:692–94).

11. Cf. the view of Baker and Morris that for Descartes our nature as a soul-body union“consists in a strict correlation between thoughts and movements of the pineal gland” (Bakerand Morris 1996, 172). In opposition to the emphasis in Wilson, discussed below, of the “arbi-trary” or contingent nature of the correlation, however, Baker and Morris insist that the cor-relations are necessary (167).

when the nerves in the foot are moved violently and more than is usual, thatmotion of them, passing through the medulla of the spine to the innermost parts ofthe brain, there gives a sign to the mind to sense something, namely pain as ifexisting in the foot, by which the mind is excited to remove the cause of pain, asharming the foot, so far as it can. (AT 7:88)

The view that the union is an “arbitrarily established disposition” is supposed toreflect Descartes’s position that God could have constituted human nature differ-ently than he has. Moreover, Wilson finds nothing here to indicate that the union issomething more than the set of correlations between mental and bodily states thatGod has arbitrarily established. For a mind to be united to a body is nothing morethan for it to have certain sensations follow from certain bodily “signs,” as in thecase of pain, and also, presumably, to have certain motions in the brain follow fromcertain volitions.

There is an initial question whether the evidence supports Wilson’s claim thatDescartes took the correlations that constitute the union to be “arbitrary.”12 To besure, his admission that it was within God’s power to change the connections sug-gests that they are contingent in some sense. However, Descartes’s emphasis in theSixth Meditation is on the fact that the connections that God has established are“conducive to the well-being of the body” (AT 7:88). Descartes’s rejection of divineteleology commits him to the conclusion that (apart from revelation, at least) we can-not know God’s purposes for establishing the union.13 However, he seems to allowthat experience can reveal the purposes that the connections involved in the unionserve. In some significant sense, then, these connections are not arbitrary but are ren-dered intelligible by the end of the preservation of the union.14

In Wilson’s view, Descartes’s claim that motions in the brain are mere “signs” tothe mind to have certain sensations indicates further the arbitrariness of the connec-tions involved in the union.15 Below I offer a reading of this claim on which issuesconcerning the arbitrariness of the causal connections, or even concerning the pos-sibility of any causal relation, do not play a central role (see §4.2.2). For the moment,though, I want to turn from the issue of the nature of the causal connections involvedin the union to the implication of the natural institution theory that the union consistsin nothing over and above these connections. As Wilson notes, this implicationseems to conflict with Descartes’s view in the Sixth Meditation that various internalsensations “are nothing other than certain confused modes of thought arising from[ab . . . exorti] the union and the intermixture as it were [quasi permixtione] of themind with the body” (AT 7:81). The suggestion here is that confused thought is not


12. In addition to Wilson 1978, 209 and 211, see Wilson 1999b, 43 and 57; and Wilson1999c, 480.

13. See my discussion of this point in §2.1.2 (ii.b); cf. chapter 2, note 34.14. For a similar criticism of Wilson, see Loeb 2005, 67–69. In §4.2.3, however, I criti-

cize Loeb’s concession to Wilson that the connections involved in the union are “brute.” SeeSimmons 2001 for an emphasis on the teleological nature of Descartes’s explanation ofsensation.

15. Cf. Wilson 1978, 207, and Wilson 1999b, 43.

merely a constituent feature of the union, but also something that, as arising from theunion, is an effect of it.16

Wilson takes this suggestion to lead us to what she has called the co-extension the-ory, on which the mind has a special sort of presence in the body that is a prerequi-site for the production of states of the union. With respect to this theory, Wilsonhighlights Descartes’s claim, cited previously, that his correspondent Elisabeth shouldfeel free to attribute extension to the soul, since to do so is “nothing other than to con-ceive it as united to body” (AT 3:694). So also, in a passage from the Sixth Repliesrelated to this claim, Descartes draws an analogy between the human soul and thescholastic real quality of heaviness that “while it remained coextensive with the heavybody, it could exercise its power at any part of it” (AT 7:442; see §4.3.2). Accordingto the co-extension theory, then, the experience of the union is not restricted to theexperience of certain states associated with it. Rather, as Wilson notes, “we experi-ence the co-extensiveness of mind throughout the body . . . and (perhaps by this veryfact) experience something called the mind–body union” (Wilson 1978, 216).

I have some sympathy for Wilson’s conclusion that the co-extension theory is prob-lematic for Descartes. Whereas the problem that she emphasizes is that co-extensiondoes not seem to render mind–body interaction intelligible (Wilson 1978, 215),17 theproblem I find is that the sort of co-extension he posits in the correspondence withElisabeth is not clearly relevant to the special intermixture that he takes to be presentin the union. As indicated in §4.1.1, Descartes there invokes the extension of thoughtto address Elisabeth’s worries about how any immaterial thing could move a body.Neither these worries nor the purported solution seems to concern the distinctive sortof connection present in the case of the relation of the human soul to its body.18

Moreover, the implication of the “heaviness analogy” that the soul is presentwherever it can act on the body does not support the view that it is co-extended withthe whole of the body to which it is united. For Descartes makes clear in the Passionsof the Soul that “there is a certain part of the body where [the human soul] exercisesits functions more particularly than in others,” namely, the pineal gland (PS I.31, AT11:352). It seems to follow from the fact that the soul is extended only where it canact that it is co-extended merely with this specific part of the brain, and not with thehuman body as a whole.19


16. See also the claim in the Sixth Replies that sensations “arise from” (oriri ex) the union(AT 7:437). But cf. Loeb 1981, 131–32, n.9, for an attempt to argue that there is no serioussuggestion in Descartes that mind and body interact in a certain manner in virtue of the union.Here I follow Wilson in thinking that there is evidence of such a suggestion in Descartes. Theimplications of my interpretive differences with Loeb on this point are revealed in §4.2.3.

17. Cf. Rozemond 2005, 360–62.18. For a similar objection, see Rozemond 2005, 350, 356.19. It is possible that Descartes simply confuses the co-extension required for the soul’s

direct action on the body with the co-extension that derives from the fact that the body withwhich the soul interacts has parts that must be interconnected to sustain the union. In fact, thearticle containing the passage from the Passions just quoted is immediately preceded by anarticle in which Descartes emphasizes the sort of co-extension that involves the unified natureof the human body (PS I.30, AT 11:351).

Finally, the heaviness analogy appears to be irrelevant to the portion of the union,to which Descartes himself draws Elisabeth’s attention, that concerns the “force” ofthe human body “to act [d’agir] on the soul, in causing [causant] its sensations andpassions” (AT 3:665). But this sort of force seems to be essential for the kind of inter-mixture posited in the Sixth Meditation insofar as Descartes is concerned in this pas-sage with the fact that the human soul has certain confused sensations of its body. Tobe sure, he does note, in comments on this passage in the Sixth Replies, that such sen-sations arise “from the fact that the mind is so intimately conjoined with the brain[cerebro]” (AT 7:437). As in the case of the action of the soul on body, so too, itseems, the action of body on soul requires that the agent be in some sort of contactwith the patient. Nevertheless, Descartes emphasizes that the conception of the actionof the quality of heaviness is drawn from the conception of the action of the soul onbody (see §4.3.2). Given this emphasis, the former conception cannot contribute to anexplication of the force of the body to act on the soul in sensation.

In spite of its very real drawbacks, the co-extension theory does reflect what I taketo be Descartes’s considered view that the union consists in more than a certain set ofcorrelations between mental and bodily states. The extension of the human soul issupposed to allow it to have the sort of presence required for it to move the body. Soalso, it seems, the union of the soul with the brain is needed for certain motions thereto be able to cause certain sensations in the soul. Even if the extension of thought thatDescartes mentions in his letter to Elisabeth cannot be identified with the union, then,at least it is similar to the union in being a prerequisite for interaction.

Admittedly, Descartes is not entirely clear on what precisely the union is sup-posed to be when it is distinguished from the particular states it produces. The claimin the correspondence with Elisabeth that we have a primitive notion of the unionakin to primitive notions of thought and extension may seem to indicate that theunion is similar to thought and extension in being a principal attribute of a substance.This impression is strengthened by Descartes’s comment in a 1642 letter to Regiusthat “the mind is united in a real and substantial manner to the body” so that itforms with the body a true ens per se (Jan. 1642, AT 3:493).20 But though it is tempt-ing to think that Descartes at least entertained a “trialism” on which the union is acreated substance along with mind and body,21 ultimately it seems that he cannotaccept such a position. Perhaps the most serious problem for trialism derives fromDescartes’s claim in the Synopsis to the Meditations that all created substances are


20. In an earlier letter, Descartes advises Regius to hold in his dispute with Voetius thatbody and soul are “incomplete substances” in relation to the human being and thus constitutean ens per se when joined (Dec. 1641, AT 3:460). See also Descartes’s remarks in the FourthReplies at AT 7:222–23.

21. Indeed, I succumbed to this temptation in Schmaltz 1992a. For a trenchant critique ofmy view in this article, see Rozemond 1998, 191–203. But cf. the defense of the view thatDescartes’s union involves a unified substance in Hoffman 1986 and 1999. I borrow the term‘trialism’ from Cottingham 1985, though Cottingham’s claim is that Descartes added the thirdcategory of the union “without proceeding to reify it as a separate substance” (Cottingham1985, 229).

“incorruptible, and cannot cease to be unless they are reduced to nothing by God’swithholding his concursus from them” (AT 7:14).22 The problem is that he indicatedin the same passage that the human body, and thus the union, “can very easily per-ish” by natural means. But since Descartes claims that everything produced naturallyderives from God’s ordinary concursus, the implication here is that the union can bedestroyed even when God continues to offer this concursus, and so cannot properlybe said to constitute a substance.

If Descartes’s union does not constitute a substance, however, in what sense is itsomething that is distinct from and explains the correlations between mental and bod-ily states of the human being? A clue to the answer is provided in the Passions, in asection entitled “how the soul and the body act on each other.” There Descartes claimsthat whereas the union involves a soul “whose nature is such that it receives . . . asmany different perceptions as different motions occur in [the pineal] gland,” it alsoinvolves a body the mechanism of which “is so constructed that simply by the gland’sbeing moved in any way by the soul . . . it drives the surrounding spirits toward thepores of the brain, which direct them through the nerves to the muscles; and in thisway the gland makes the spirits move the limbs” (PS I.34, AT 11:355). In the case ofthe union, the correlations between mental and bodily states are to be explained interms of the coordinated natures of the human soul and the bodily mechanism towhich that soul is united. Given that the soul and body possess these natures, certainmotions in the brain can produce certain sensations in the soul, and certain volitionsin the soul can produce certain motions in the brain.

This explanation of the correlations involved in the union may seem to conflictwith Descartes’s occasional suggestion that there is not much that can be said aboutthe union itself. There is, for instance, his claim to Elisabeth that “what belongs tothe union of the soul and the body is known only obscurely by the intellect alone oreven by the intellect aided by the imagination, but is known very clearly by thesenses.” Thus the conception of the union derives from “the ordinary course of lifeand conversation, and abstention from meditation and from the study of things thatexercise the imagination” (28 June 1643, AT 3:692). Moreover, he writes later toArnauld that the fact that a mind united to a body can act on it “is shown us not byreasoning or comparison with other things, but by the most certain and most evidenteveryday experience” (29 July 1648, AT 5:222). So isn’t Descartes’s view that thereis no need for an explanation of the union apart from an appeal to the fact that wejust feel it?

The problem here, I think, is that Descartes is not always careful to distinguishthe fact of the union from the explanation of this fact. Certainly his view is thatimmediate sense experience is the best source of our knowledge of the fact of theunion. But though he presented this experience as yielding a full primitive notion ofthe union, in the correspondence with Elisabeth, or a complete understanding of theaction of our mind on our body, in the letter to Arnauld, his remarks elsewhere in


22. As I indicated in §3.1.3, this claim is related to the position, added to the French edi-tion of the Principles, that created substances are such that they “need only the ordinary con-course [concourse ordinaire] of God” to exist (PP I.51, AT 9-2:47).

these same texts indicate that more is required. For in the correspondence withElisabeth, he appeals to the fact that the notion of the union include the notions of“the force that the soul has to move the body” and the force of “the body to act onthe soul, in causing its sensations and passions” (AT 3:665), and it would seem thatthese forces are theoretical posits that go beyond what it revealed in immediate senseexperience. Likewise, in the letter to Arnauld, Descartes follows up on his claim thatthe action of mind on body is immediately evident by attempting to explain thisaction in terms of the heaviness analogy that he had used previously for this samepurpose (AT 5:222–23). Such an explanation seems to go beyond our immediateexperience of our action insofar as it concerns the underlying source of that action.23

Even so, some of Descartes’s remarks draw attention to the distinctive nature ofthe states of the union that are immediately evident in our sense experience. Thus, inhis “proof ” of the union in the Sixth Meditation there is an emphasis on the fact thatthe “intermixture” of mind and body yields “confused” sensations such as hungerand thirst that differ phenomenologically from the clear perceptions that a pure intel-lect would have of the body (AT 7:81). Though to my knowledge Descartes nevercharacterized the volitions by which the human soul moves its body as confused ina similar manner, the passage from the Passions quoted previously suggests thatthere is something distinctive about them insofar as they depend for their efficacyon the construction of the bodily mechanism. Before turning to the details ofDescartes’s account of the kinds of interaction that the union involves, I want to con-sider his view of how the union is reflected in the sensory and volitional states towhich we have access in basic sense experience.

4.1.3. Sensation and Human Volition

In response to Arnauld’s objection that his dualism suggests the “Platonic opinion”that a human being is “a soul using a body” (Fourth Objections, AT 7:203), Descartesclaims that the argument that he used in the Sixth Meditation to show that the humanmind is “substantially united” with the body is “as strong as any I have read” (FourthReplies, AT 7:228). In this argument he appeals to the “teaching of nature” that he has“sensations of pain, hunger, thirst, etc.,” and so is “not only present in my body as asailor in a ship, but most closely conjoined and as it were intermixed with it, so thatI form a unit with it” (AT 7:81). The image of the sailor in the ship derives fromAristotle, who used it to characterize a Platonic view of the relation between the souland body.24 Descartes therefore indeed attempted to distance himself from a Platonicconception of the human being. What is supposed to show that he is not merely using


23. Cf. Wilson’s observation that “the senses could hardly be supposed to tell us—‘in theordinary course of life,’ as Descartes adds—that brain states give rise to mind states accord-ing to correlations instituted by nature” (Wilson 1999b, 58). Wilson is here criticizingRichardson’s view that Descartes’s remarks in the Elisabeth correspondence indicate thatmind–body interaction is an irreducible theoretical primitive (see Richardson 1982, 25–26;see also Richardson 1985).

24. See De Anima II.1, 413a8, Aristotle 1984, 1:657.

a body is that when his body is damaged or needs food or drink, he does not merely“expressly understand” that this is so, but rather perceives it by means of the “con-fused sensations” of pain, hunger, and thirst. As he indicates in a letter to Regius, theinference here is that these sensations are not “pure thoughts of a mind distinct froma body,” but rather “confused perceptions of a mind really united to a body” (Jan.1642, AT 3:493).

In this same letter, Descartes illustrates the difference between a purely intellec-tual mind and the embodied soul by noting that “if an angel were in a human body,it would not have sensations as we do, but would simply perceive the motions thatare caused by external objects, and in this way would differ from a real man” (AT3:493). Though this passage may seem to leave open the possibility that the angelicperception of the motions, as well as the motions themselves, have bodily causes, infact Descartes is committed to the conclusion that bodies cannot act on disembodiedminds. His official position in the Principles is that everything in the mind that doesnot pertain to it solely as an intellectual thinking thing must be “referred to” (ad . . .referri) the “close and intimate union of our mind with the body” (PP I.48, AT8-1:23).25 Moreover, he makes clear in the Fifth Replies that “the brain cannot in anyway be employed in pure understanding [pure intelligendum], but only in imaginingand sensing” (AT 7:358).26 For him, a mind not united to a body could have onlypure intellect, and thus would have no perceptions that derive from the force of thebody to act on the mind.

Despite Descartes’s claim in the Principles that all of his confused sensationsmust be “referred to” the union, his Sixth Meditation proof of the union emphasizesonly a certain class of such sensations, namely, his “internal sensations” of states ofhis own body. One natural question is why this argument focuses on these sensationsto the exclusion of external sensations of sensible qualities such as colors, sounds,and tastes.27 As Alison Simmons has shown, the answer is that internal sensationssuch as those of pain, hunger, and thirst lead the soul to regard a particular body asits own.28 Descartes’s emphasis in the Sixth Meditation passage is on the fact that apure intellect that had merely an explicit understanding of the body could not asso-ciate itself with the body in the same way. But it seems that even a soul that hadin addition to such an understanding confused external sensations of the sensible


25. The French edition says that such mental elements must be “attributed to” (atribuéesà) the union (AT 9-2:45).

26. As Wilson notes, this feature of Descartes’s dualism distinguishes it from versions ofCartesian dualism in contemporary discussions that require only that mental states are notidentical to bodily states, and thus allow for the thoroughgoing correlation of such states(Wilson 1978, 180).

27. For the distinction between internal and external sensation, see PP IV.190–98, AT8-1:316–23. In this text internal sensations are lumped together with the passions and distin-guished from external sensations, whereas in the later Passions there is a threefold distinctionof perceptions referred to external bodies (external sensations), perceptions referred to ourown body (internal sensations), and perceptions referred to our own soul (passions) (PSI.22–25, AT 11:345–48).

28. See Simmons forthcoming.

qualities of its body could no more become attached to this body than it could to anyother body it perceives by means of such sensations. Thus it is not only the fact thatthe internal sensations are confused that reveals the union with a body, but also thefact that they are confused in a way that produces in the soul a sense that it “formsa unit” with that body.

It is important that the teaching of nature is here restricted to our sense that weform a unit with a body. There is no claim that nature teaches us how it is so united.The absence of such a claim helps to explain Descartes’s admission to Elisabeth thathe said “almost nothing” in his previous writings, including his Meditations, aboutour knowledge that our soul is such by nature “that being united to a body, it can actand be acted on by it” (AT 3:665). Though we have seen that he was not always clearon this point, his suggestion in this passage is that he has provided no explanation ofhow the mind is able to be united with the body so as to have the states revealed byhis immediate sense experience.29 Such an explanation requires a further analysis ofthe kind of “forces” that underlie the union.

Though the Sixth Meditation proof of the union emphasizes the importance of ourinternal sensations, Descartes indicates in his letter to Elisabeth—in line with his offi-cial doctrine in the Principles—that all sensations and passions derive from the union.Moreover, he claims in this letter that the primitive notion of the union includesthe notion of the force of the soul to move the body. Yet there is the suggestion in theDiscourse that the volitions that produce bodily motions are not as clearly connected tothe union as the sensory states that the body produces. Descartes writes in this text that

it does not suffice that the [rational] soul is lodged in the human body as a pilot ina ship, except perhaps in order to move its limbs, but it must be joined and unitedto it more closely in order to have, in addition, sensations and appetites similar toours, and thus compose a true man. (DM V, AT 6:59; emphasis mine)

As noted below, Descartes insists in his late correspondence with More that Godand angels, as well as human souls, can move bodies (see §4.3.2). Since he extendsthe power to move bodies to disembodied minds, Descartes has reason to concludethat the mere fact that the soul can move its limbs does not clearly reveal that it isclosely joined to the body it moves. And given that he denies that disembodiedminds have the confused sensations that derive from bodily action on the mind, hehas reason to take our experience of such states to reveal the close union of oursoul with our body.

Nevertheless, the volition to move bodies that Descartes posited in the case ofhuman souls seems to have a feature that distinguishes it from the correspondingkind of volition in disembodied minds. To see the difference, consider again his viewin the Passions that though it is united to the whole body, the human soul “exercisesits functions more particularly” in the pineal gland (PS I.31, AT 11:352). To move


29. See also Descartes’s claim, added to the 1647 French edition of the Meditations, thathis response to Gassendist objections to the possibility of the interaction of human souls withbodies “presuppose an explication of the union of the human soul with the body, with whichI have not yet dealt” (AT 9-1:213).

any part of our body, then, we must act on this gland in a particular manner.30 But itis no part of Descartes’s view that this action is brought about by a volition that hasthe change in the gland as part of its intentional content. He indicates that we moveour arm by willing this motion, not by willing a change in our pineal gland. In con-trast, presumably, a disembodied mind would need to will that a change occur in thisgland to produce that same effect.

The implication in Descartes that human volition need not include its immediateeffect in its intentional content tends to undermine Bernard Williams’ puzzling argu-ment that Cartesian interactionism is unsatisfactory given that “one’s control overone’s body could not be understood as internal, localized psychokinesis” (Williams1978, 289). Williams’ assumption is that Descartes’s view that we can move our armonly by acting on our pineal gland commits him to the conclusion that we can pro-duce this internal action directly by a kind of psychokinesis.

However, Williams himself admits that Descartes allowed that there are certainchanges that I cannot produce by directly willing, such as the dilation of my pupils(290). Why couldn’t Descartes hold that changes in the pineal gland are among thosethat I cannot produce by directly willing? Indeed, it seems that this is just whatDescartes does hold, and so he is not committed by his account of human volition tothe existence in the case of our voluntary motion of a kind of psychokinesis.31

We have considered the view in Descartes that the union gives rise to sensoryrepresentations of states of our own body—such as those present in the cases of oursensations of pain, hunger, and thirst—that are confused in the sense that they fail toyield an “explicit understanding” of the nature of those states. Though he did notspeak in these terms, it seems that he also could say that the union includes volitionsto move our body that are confused in a related sense, since their content does notreveal the nature of their immediate effects in the brain. These “confused” volitionscan nonetheless produce their intended effects in virtue of the special nature of thehuman body. As Descartes tells Arnauld, although we are conscious of the action bywhich our soul brings about a certain change in the brain, the fact that it producesthis change is due to “the appropriate way in which the body is constructed, of whichthe mind may not be aware” (29 July 1648, AT 5:222). And even if we were awareof this construction through the study of anatomy, we still could not produce thechange in the brain by willing it directly (thus the impossibility in our case of inter-nal psychokinesis). Given that our soul is united to a body, according to Descartes,the only way in which we can bring about the right sort of change in our brain is byproducing a volition for some effect, such as the motion of our arm, which volitionitself brings about that change due to the nature of the union.


30. As we will discover in §4.3.3, there is some dispute over whether Descartes held thatwe act on the gland by moving it or merely by “determining” its motion.

31. Williams’ claim that for psychokinesis to be possible it must be possible “for mind toinfluence matter separate from it” (Williams 1978, 289) suggests that he takes psychokineticaction to involve action at a distance, or at least action without any sort of immediate presenceof the mind to that on which it acts. As indicated in §4.3.2, however, there is reason to thinkthat Descartes would deny that any sort of mind-to-body action is psychokinetic in this sense.

Later occasionalists cite the fact that we often do not know how we produce thebodily effects we will in support of the conclusion that we do not have the power tomove our body. Thus, Malebranche argues as follows in his 1674/75 Search afterTruth:

For how could we move our arms? To move them, it is necessary to have animalspirits, to send them through certain nerves toward certain muscles in order toinflate and contract them, for it is thus that the arm attached to them is moved.And we see that men who do not know that they have spirits, nerves, and musclesmove their arms, and even move them with more skill and ease than those whoknow anatomy best. Therefore, men will to move their arms, and only God is ableand knows how to move them. (Bk. VI-2, ch. 3, Malebranche 1958–84,2:315/Malebranche 1997, 449–50)

Here the insistence that causal power requires knowledge of its exercise is limited tothe case of mind-to-body action. But in True Metaphysics (Metaphysica vera), pub-lished posthumously in 1691, the Flemish Cartesian Arnold Geulincx cites as ageneral causal principle that “it is impossible that one make what one does not knowhow to make” (impossible ese, ut is faciat, qui nescit quomodo fiat) (Geulincx1891–93, 2:150). It also seems to be in line with Geulincx’s position that there mustbe, in addition to knowledge of how to make the effect, an efficacious volition to pro-duce what is necessary to bring about that effect. In the case of our action on thepineal gland, though, we have no such volition. And according to Geulincx’s princi-ple, if we cannot have this sort of volition (either because we do not know the meansor because we could not directly will them if we did know them), we cannot producethe effect.

However, Descartes takes it to be a distinctive feature of human volition that it canimmediately produce an effect in a body that it does not know how to make, or at leastthat is not included in its intentional content, due to the fact that the human soul isunited to that body. This union is supposed to confer a special sort of force on thisvolition that it would not have apart from the union. It is this sort of force, presum-ably, that the volition of a disembodied mind would lack. Of course, one could takeGeulincx’s principle to show that the human soul cannot possess a force of this kind.But Descartes’s comment to Arnauld that our “most certain and most evident experi-ence” proves that such volitions in fact move our body indicates that he would takethis experience to reveal as well the falsity of Geulincx’s principle.

This is not the end of the matter, however, since we still have Elisabeth’s worrythat an unextended thing does not have what is required to move a body. As I haveemphasized, this worry concerns not just the special sort of volitions involved in theunion, but mind-to-body action as such. If I am correct in thinking that Descartes iscommitted to denying that body can produce perceptions in a disembodied mind,there is no problem for body-to-mind action that extends beyond the union. But thereis still the worry, again from Elisabeth, that something as imperfect as body couldnot have an effect on any mind capable of rational thought. To reinforce the pur-ported dictate of experience that the human soul interacts with its body in a specialmanner, Descartes needs to address these different problems concerning body-to-mind and mind-to-body action.


4.2. BODY-TO-MIND ACTION

I have noted the claim of Descartes’s critics that his troubles with mind–body inter-action derive from a heterogeneity problem that applies equally to the cases of body-to-mind and mind-to-body action. In contrast, I start here with a problem withinteraction in the work of the scholastics, and more specifically the writings ofAquinas and Suárez, that pertains exclusively to the case of body-to-mind action.Then I turn to the version of this scholastic problem in Descartes’s account of thecausation of sensation. Here I emphasize his startling claim in a late text that motionsin the brain merely “give occasion” to the mind to form its own sensory ideas bymeans of a faculty innate to it. In contrast to commentators who see this passage asderiving from a recognition in Descartes of the impossibility of the action of bodyon mind, however, I argue that Descartes attempts to retain the view that body issome (albeit special) sort of efficient cause of sensation. In so doing, I set myselfapart not only from occasionalist readings of Descartes’s account of body-to-mindaction, but also from Steven Nadler’s recent claim that he posited an “occasionalcausation” distinct in kind from efficient causation. Whereas Nadler sees Descartesas denying that occasional causation derives from any real power in the occasionalcause, I find in the Cartesian texts the suggestion that the occasional connectionbetween motions and sensations is grounded in divinely instituted natures. The factthat I find such a suggestion in Descartes explains my resistance to the view in theliterature that he took mind–body interaction to involve “brute” psychophysical lawsthat lie outside the scope of natural philosophy. On my reading, then, Descartes doesnot in the end entirely reject the scholastic understanding of psychology as a scienceof the soul.

4.2.1. The Scholastic Problem

In a discussion of Descartes’s views on the action of body on mind in sensation,Marleen Rozemond has drawn attention to the importance of a scholastic problemwith interaction that I wish to emphasize here.32 As Rozemond indicates, this prob-lem pertains not to the case of sensation as such, but rather to the relation of sen-sation to intellectual acts. To be sure, in his Summa Theologiae, Aquinas broacheda possible problem for the conclusion that the body acts on the mind in sensa-tion when he noted the view, which he attributed to Plato, that “the soul is some-how excited so as to form sensible species in itself.” He further noted, cautiously,that Augustine “seems to touch on” the opinion of Plato, citing his claim that “thebody does not feel, but the soul through the body, which it uses as a kind of mes-senger to form [ad formandum] in itself what is announced from without” (STI.84.6).33 However, Thomas countered with the Aristotelian position that sensationis a passive feature of the soul-body composite, rather than an activity of the humansoul. Since the sensible object actually possesses the sensory form that the bodily


32. Rozemond 1999; cf. Rozemond 1998, 178–79.33. This passage, cited in ST I.84.6, is from Augustine’s De Genesi ad litteram, pt. XII, ch. 24.

senses possess only potentially, this object is able to actualize this form when it actson the senses.34

This is not the end of the story, however, insofar as Aquinas also endorsed theargument, which he took to be common to Plato and Aristotle, that since the intel-lect has an operation more perfect than anything found in a body, and thus is morenoble than body, nothing bodily can act on it (ST I.84.6).35 Bodily operations canproduce sensible species in the sense organs, and these species can lead to the pro-duction of phantasms in the imagination. But on their own, those phantasms cannotproduce corresponding “intelligible species” in the intellect. The problem here, asAquinas expressed it, is that “in order to cause an intellectual operation, accordingto Aristotle, an impression of sensible bodies is not enough, but something morenoble is required, because what acts is more noble than what is passive, as he him-self says” (ST I.84.6). In opposition to the view in Plato that these “more noble” ele-ments are the separate intelligible Forms that illumine the intellect, Aquinas positedan “agent intellect,” a power of the intellect to impress intelligible species in itself(ST I.79.3). Because the activity of the agent intellect is required to produce the intel-ligible species, “it cannot be said that sensible knowledge is the total and perfectcause of intellectual knowledge” (ST I.84.6).

Even so, Aquinas insisted that intellectual knowledge must be “caused by” thesenses, and thus that the phantasms must be at least “in a way the matter of the cause”(ST I.84.6). This requirement derives from Aquinas’s commitment to the position thathuman intellectual knowledge, as opposed to the knowledge of separate intellectssuch an angels, depends essentially on sensory operations. In particular, this depend-ence is reflected in the fact that the agent intellect can produce species only by aprocess of abstracting intelligible content from phantasms. Phantasms are thus “thematter of the cause” in the sense that they provide the material for abstraction. As heexpressed the point in Questions on Truth (Quæstiones de Veritate), a phantasm is an“instrumental and secondary agent” (agens instrumentale et secundarium) that pro-vides a model for the production of intelligible species by the agent intellect as “prin-cipal and primary agent” (agens principale et primum) (QT X.6, ad 7, TA 14:554).

In effect, then, Aquinas was trying to find a middle way between the position thatphantasms are total causes of intelligible species and the view that they play nocausal role in the formation of these species. This was only one of a number ofaccounts of intellectual cognition offered in the later medieval period, which rangedfrom the Platonist view of James of Viterbo, on which phantasms merely trigger the


34. There is a debate in a series of articles in Philosophical Review over whether Aquinastook the reception of the sensible species to be merely a “physical event,” or whether itinvolves in addition some “spiritual” change; see Cohen 1982, which defends the former posi-tion; Haldane 1983, which defends the latter position; and Hoffman 1990, which attempts tosplit the difference.

35. In ST I.75.2, Thomas argued that intellectual cognition cannot be an operation of abodily organ, since such cognition has universal scope whereas the bodily operations have amore determinate nature. This line of argument is strikingly similar to Descartes’s argumentin the Discourse, mentioned in §4.1.1, that since the reason involved in language use is a “uni-versal instrument,” it cannot be explained in terms of the limited dispositions of bodily organs.

production of inborn intelligible species, to the nominalist view of Durandus, onwhich phantasms produce intellectual cognition directly, without the mediation ofagent intellect or intelligible species.36 However, Thomas’s influence seems to bereflected in the position of our main scholastic, Suárez, that a phantasm “instrumen-tally attains to the production of an intelligible species” (MD XVII.2, ¶11, 1:586).Since this text defines an instrumental cause as an efficient cause that “concurs in oris elevated to the effecting of something more noble than itself” (¶17, 1:590), thesuggestion here, as in Aquinas, is that the phantasm is a real though subordinate effi-cient cause of the species.

However, Rozemond draws attention to the much less causally robust role thatSuárez attributed to the phantasm in his commentary on De Anima.37 In this text, hedid allow that the phantasm “concurs in some mode in the production of the [intelli-gible] species” (DA IV.2, ¶4, Opera 3:716*). Yet Suárez also emphasized there—indirect opposition to the views of Thomas and the Thomists38—that since the phan-tasm is corporeal, and thus cannot act on something spiritual, it “concurs not effec-tively [effective] but materially to the production of the species, and the agent intellectalone effects it” (¶10, Opera 3:719*). Earlier in this text Suárez had described thephantasm as “only a prerequisite, or as the exciting occasion [occasio excitans], or asexemplar, or at most, as instrument elevated by the spiritual light of the same soul”(DA I.11, ¶21, Opera 3:550*). This passage does leave open the possibility of describ-ing the phantasm as an instrument for the production of the species. Yet the moredominant suggestion there is that it is an “exciting occasion” that serves merely as aprerequisite for the efficient causal activity of the agent intellect.39

Interestingly, Suárez distanced himself from Thomas even when purporting to behewing to the Thomistic line. Thus, after citing with approval Thomas’s denial thatthe phantasm is a “total and perfect cause” of intellectual cognition, Suárez noted hisown view that the phantasm is related to the intelligible species as the inner sensesare related to the appetitive power. Yet earlier he had claimed that “apprehension is


36. On Viterbo, see Spruit 1994–95, 1:238–40; on Durandus, see Spruit 1994–95, 1:281–83.37. See the discussion in Rozemond 1999, 440–44; cf. Spruit 1994–95, 2:301–305.

Neither Rozemond nor Spruit notes the apparent tension between the views in De Anima andthe Metaphysical Disputations. Though De Anima was first published in 1622, twenty-fiveyears after the Disputations, it is based on work that predates the later text. It may be, then,that Suárez’s thought shifted toward a more Thomistic position. Or it might be that the viewin the Disputations is less in line with such a position than it initially appears to be, and thusthat Suárez’s thinking on this issue was more consistently anti-Thomistic than I have allowed.Thanks to James South for discussion of this issue.

38. See, for instance, his critique of the view of Cajetan and the Thomistae that “phan-tasms are united with the agent intellect as an instrument with its principal agent” (DA IV.2,¶¶6-7, Opera, 717–18*), and his critique of the view in Thomas that the phantasm must serveas some sort of efficient cause of the species (DA IV.2, ¶13, Opera 3:720*). For discussion ofvarious later Thomistic developments of Thomas’s account of the relation of the phantasm tointelligible species, see Spruit 1994–95, 1:360–85, and 2:111–28, 274–93.

39. For a comparison of the uses of occasionalist terminology in the work of Descartesand various scholastic authors, see Specht 1966, chs. 1–3, as well as Specht 1971 and 1972.

required, not as an efficient cause, but as an application to the object” (DA V.3, ¶8,Opera 3:760*).40 Moreover, just prior to the citation of the passage from Thomas,Suárez insisted that the contribution of the phantasm to intellectual cognition “doesnot occur by way of any influx of the phantasm itself, but by providing the matterand as it were an exemplar to the agent intellect in virtue of the union that they bothhave in the same soul” (DA IV.2, ¶12, Opera 3:719*). But Suárez’s official view isthat causation is “nothing other than that influx, or concursus by which each causein its kind actually flows being into the effect” (MD XII.2, ¶13, 1:387*). Given hisdenial of any influx of the phantasm in the production of the intelligible species,Suárez is committed to the conclusion, not to be found in Thomas, that the phantasmplays no efficient causal role in this production.41

Though Aquinas and Suárez agreed that the phantasm cannot be the total efficientcause of intelligible species, then, they offered different accounts of the manner inwhich it is related to the intellect. For Thomas, the phantasm is still a true efficientcause, albeit a “secondary and instrumental” one, that determines the content of theintelligible species. At least in De Anima, however, Suárez gave the impression thatthe phantasm is a noncausal occasion for the production of the species and that theagent intellect is their sole secondary efficient cause.

The scholastic problem of the relation of the phantasm of the soul-body compos-ite to intellectual cognition does not arise for Descartes. He replaces the phantasmwith two distinct “grades” of sensation, the first of which consists in “nothing but themotion of the particles of the [sense] organs, and any change of shape and positionresulting from this motion,” and the second of which consists in “all the immediateeffects produced in the mind as a result of its being united with a bodily organ thatis affected in this way” (Sixth Replies, AT 7:437). Nonetheless, the scholastic prob-lem is similar to a problem that arises for Descartes’s account of the “immediateeffects” of the union. For just as Aquinas and Suárez had to explain how changes inan incorporeal intellect are related to a corporeal phantasm, so Descartes has toexplain how sensory modes of an immaterial mind can result from bodily motions.Moreover, we will discover presently that Descartes speaks at times of motions as“giving occasion” to the mind to form sensory ideas, thus recalling Suárez’s talk ofphantasms as the “exciting occasion” for the production of intelligible species by theagent intellect.

The link to the scholastic problem is reinforced by the fact that Descartes’s con-tainment axiom seems to create difficulties for body-to-mind action that do not applyto the case of mind-to-body action. The Platonic axiom in Thomas and Suárez thatthe less perfect cannot act on the more perfect precludes only the action of body onintellect, not the action of intellect on body. Similarly, Descartes’s requirement thatthe “total” or “adequate” cause must contain its effects “formally or eminently”applies asymmetrically to the cases of mind and body. As I noted in chapter 2,


40. Suárez cites Aquinas’s account in ST I.82.4 of how the intellect moves the will, thoughthere is no indication in the cited text that the apprehension of an object cannot serve as anefficient cause of an appetite.

41. For this point about Suárez, I am indebted to the discussion in Rozemond 1999, 440–44.

Descartes’s account of eminent containment allows for minds to so contain bodilyeffects, but not for bodies to so contain mental effects (see §2.1.3 (ii)). Thus, totalbodily causes must formally contain their mental effects. I also suggested in thischapter that there is a sense in which bodily causes could formally contain the“objective reality” of sensory ideas. However, it seems that they cannot so containeverything in sensations, and in particular not those features of the sensations that donot resemble what is in bodies (see §2.1.3 (i)). So just as the scholastics had to con-front the question of how intellectual aspects of the intelligible species that gobeyond the corporeal can derive from the phantasm, so Descartes has to confront thequestion of how mental aspects of sensations that go beyond what can be found inbodies can nonetheless result from motions in the brain. But though Descartes’s useof occasionalist terminology may seem to indicate an affinity to Suárez’s more rad-ical views, I hope to show that he is in fact closer to Aquinas in attempting to makeroom for a genuine causal role for body in sensation.

4.2.2. Sensation in the Comments

Perhaps Descartes’s most provocative remarks on the causation of sensation arefound in his 1648 Comments on a Certain Broadsheet (Notae in Programma quod-dam). The “broadsheet” in question is that of Henricus Regius (1598–1679), oncethe primary exponent of the new Cartesian philosophy at the University of Utrecht.Though Descartes had previously befriended Regius, their relationship took a turnfor the worse with the publication in 1646 of Regius’s Fundamenta physices.42

Descartes complained in the preface to the French edition of the Principles that inthe Fundamenta his former disciple erred in denying “certain truths of metaphysicson which the whole of physics must be based” (AT 9-2, 20). Regius responded in1647 by publishing anonymously a pamphlet, entitled “An account of the humanmind, or rational soul, where it is explained what it is, and what it can be,” in whichhe outlined his position on the nature and knowledge of mind and body. TheComments is Descartes’s point-by-point response to the twenty-one theses offered inthis pamphlet.43

Regius declared in the twelfth thesis of his pamphlet that the mind is devoid of“ideas, or notions, or axioms that are innate,” and he concluded in the thirteenth that“all common notions that are engraved in the mind have their origin in observation


42. Regius held a chair in medicine at the University of Utrecht. Descartes had defendedRegius when Gisbertius Voetius, the rector of the university, attempted to remove him fromhis chair on the grounds that his views were heretical. In an open letter to Voetius, publishedin 1643, Descartes said that he was so confident of Regius’s intelligence that any view ofRegius could be attributed to him (AT 8-2:163). This contrasts sharply with the charge of pla-giarism, and inaccurate plagiarism at that, in the preface to the Principles (AT 9-2:19), in a1646 letter to Mersenne (AT 4:508–11), and in the Comments (AT 8-2:365). For more on theexchange between Regius and Descartes, see Verbeek 1993.

43. In this work, Descartes responded also to some of the objections to his system pro-vided in Consideratio theologica, a pamphlet published by a theologian at Utrecht, JacobusRevius, whose critique of Descartes is discussed in §3.1.3, at note 37.

of things or in verbal instruction” (AT 8-2:345). In response, Descartes takes excep-tion, in particular, to the purported suggestion in Regius that “the faculty of thoughtis not able to accomplish anything by itself, could never perceive or think anythingexcept what it receives [accipit] from observation or tradition, that is, from thesenses.” He adds that

whoever correctly considers how far our senses extend, and what it is precisely thatis able to come [pervenire] to our faculty of thought from them, must admit that inno case do they exhibit to us ideas of things, just as we form them in thought. Somuch so that nothing is in our ideas which is not innate to the mind, or to thefaculty of thinking [nihil sit in nostris ideis, quod menti, sive cogitandi facultati,non fuerit innatum], except only those circumstances that pertain to experience:such as that we judge that these or those ideas, which we now have present to ourthought, are referred to certain things placed outside us. (AT 8-2:359)

The material for judgments relating to experience does not consist of ideas that thesense organs “transmit to our mind through the sense organs” (nostrae menti perorgana sensuum immiserunt). Rather, these organs “transmit something that givesoccasion to itself [i.e., the mind] to form [the ideas], at this time rather than another,by means of a faculty innate to it” (aliquid immiserunt, quod dedit occasionem adipsas, per innatum sibi facultatem, hoc tempore potius quam alio, efformandas).Descartes explains that “nothing approaches [accedit] our mind from external objectsthrough the sense organs, except certain motions.” Citing the discussion in the 1637Dioptrics, he asserts that the figures arising from bodily motions “are not conceivedby us just as they are in the sense organs,” and concludes from this fact that

the ideas of motions and figures themselves are innate to us. And so much themore must the ideas of pain, colours, sounds and the like be innate, so that ourmind is able, on the occasion of certain corporeal motions [occasione quorundammotuum corporeorum], to exhibit them to itself; for they have no likeness[simultudinem] to corporeal motions. (AT 8-2:359)

Before considering the precise import of the claim that bodily motions merely“give occasion” for an innate mental faculty to form sensory ideas, we need to con-front the objection that such a claim is inconsistent not only with Descartes’s viewselsewhere, but also with other remarks in the Comments itself. Let us start with thecharge of an inconsistency internal to the Comments. In response to Regius’s twelftharticle, Descartes endorses the distinction in the Third Meditation of adventitious,factitious, and innate ideas (AT 8-2:358; cf. AT 7:57–58). The implication hereseems to be that adventitious sensory ideas are distinct from innate ideas. But how,then, could Descartes say in the very next paragraph that sensory ideas are innate?44

We can easily dispel this tension by appealing to the suggestion in Descartes thatsensory and intellectual ideas are innate in different ways. The production of innatesensory ideas follows immediately on the transmission of motions to the mind. In


44. This same problem applies to the claim, attributed to Descartes in the Conversationwith Burman, that the author of the Comments “says not that all ideas are innate in him, butthat some are also adventitious” (AT 5:165).

contrast, Descartes indicates in the Comments that the formation of innate intellec-tual ideas, such as the idea of God, is much less tightly linked to the motionsinvolved in sense experience. In particular, he claims there that the motion that givesrise to “verbal instruction or the observation of things” is a “remote and merely acci-dental cause” that “induces us to give some attention to the idea that we can have ofGod.” In this respect the motion is similar to the orders for certain work, whichmerely “give occasion to” the workers to produce the work as its “proximate andefficient cause” (AT 8-2:360). Not only are the orders remote in the sense of beingtemporally prior to the actual production of the work; they also are accidental in thesense that the workers might have produced the work without the orders. Butmotions are neither remote from nor accidental to the production of sensory ideas inthis way. They are not remote, since the ideas are formed at the exact moment thatthe motions “approach” the mind, and they are not accidental, since, it seems, thepresence of the motions is required to prompt the activity of the innate sensory fac-ulty. The Comments thus includes the suggestion that though sensory ideas areinnate in the sense that they derive from an innate mental faculty, still they are adven-titious insofar as they possess a tight connection to bodily motions that is missing inthe case of innate intellectual ideas.45

More recalcitrant, however, is the impression that the account of sensation in theComments is in tension with the Sixth Meditation argument for the existence of thematerial world. Recall from a previous discussion the conclusion in the latter textthat bodies possess an “active faculty” that is the cause of the objective reality of sen-sory ideas (see §2.1.1). Descartes also notes there that God would be a deceiver ifthese ideas “were emitted [emitterentur] by something other than corporeal things”(AT 7:80). The language here, as well as in other texts,46 seems to indicate that sen-sory ideas are not formed by an innate mental faculty, as in the Comments, but ratherare “emitted” from the bodies that formally contain what they produce objectively inthe ideas.

It would not do here to dismiss the view in the Comments as aberrant, sinceDescartes offers a similar position elsewhere, most notably in a 1641 letter toMersenne in which he wrote that all ideas “that involve no affirmation or negationare innate in us; for the sense organs do not bring [rapportent] us anything that is


45. Cf. the response to this problem of consistency in Schmaltz 1997, 40, which does notemphasize the point that occasioning causes are remote and accidental only in the case ofinnate intellectual ideas. Nadler takes Descartes to have held in the Comments that occasionalcauses in general are remote and accidental causes distinct from the proximate and efficientcause of their effects (Nadler 1994, 48). Although I suggested a similar position in Schmaltz1992b, I now read the Comments as indicating that the motions that occasion sensory ideasare not merely remote and accidental. See §4.2.3 for further reservations concerning theaccount of occasional causation that Nadler attributes to Descartes.

46. See, for instance, the claim in the Principles that “whatever we sense undoubtedlycomes [advenit] to us from something that is distinct from our mind” (PP II.1, AT 8-1:40),and the claim in the Passions that various perceptions, including sensations, are passionsbecause “it is often not our soul that makes them such as they are, and the soul always receives[reçoit] them from things that are represented by them” (PP I.17, AT 11:342).

such as the idea that arises in us [se reveille en nous] on their occasion, and thus thisidea must have been in us previously” (22 July 1641, AT 3:418).47 So we do need toconfront the apparent conflict between the accounts of the causation of sensation inthe Sixth Meditation and the Comments. One way to lessen the conflict would be toemphasize that the discussion in the Sixth Meditation concerns the manner in whichbodies external to us cause sensory ideas in us, whereas the Comments is concernedwith the manner in which motions internal to our brain give rise to these ideas. Theactive faculty in bodies could produce sensory ideas in us by producing thosemotions in our brain that trigger the operation of our innate sensory faculty. In thissense, sensory ideas could still derive from the active faculty of an external bodywithout arising solely from motions in the brain. Indeed, in the version of the proofof the existence of the material world in the French edition of the Principles,Descartes speaks of “the idea formed in us on the occasion of bodies from without”(l’idée se form en nous à l’occasion des corps de dehors) (PP II.1, AT 9-2:64). Yethe claimed in this same article that “everything we sense comes [vient] from some-thing other than our thought” and that our sensation “depends on that thing thataffects [touche] our senses” (AT 9-2:63).48 The indication here is that the claimthat motions in the brain give occasion for the production of sensory ideas is con-sistent with the claim that such ideas come from external objects insofar as theydepend on such objects for their production.49

Even so, the passage from the Comments does say that with the exception of judg-ments pertaining to experience, “there is nothing in our ideas that was not innate tothe mind or the faculty of thinking” (AT 8-2:358; my emphasis). Given this claim,there seems to be no room for anything in the sensations to come from externalbodies.50 I indicate presently that Descartes seems to have been most concerned hereto deny that the mind literally sees anything in the brain. Yet it is also significant thathe continued to claim in this text that brain motions “transmit something” to themind that triggers the formation of sensory ideas by the innate mental faculty (AT8-2:359). Even here, then, he wanted to allow that the body makes some sort of con-tribution to what is produced in the mind in sense experience.51


47. As I indicate presently, Descartes’s position in the Comments is related to the view inhis earlier writings that the relation between motions and sensations can be understood interms of the relation between signs and what they signify.

48. This additional claim supports Scott’s rejection of Garber’s tentative suggestion thatthe French edition of the Principles indicates a shift in Descartes’s thought away from theposition that bodies cause sensations; cf. Garber 1993, 22, and Scott 2000, 508–10.

49. Cf. the defense of the consistency of the accounts of sensation in the Sixth Meditationand the Comments in Schmaltz 1997, 41–44.

50. Thanks to Gonzalo Rodriguez-Pereyra for this objection.51. For the point that Descartes allowed in the Comments for some sort of causal transmis-

sion from bodily motions to mind in sensory experience, see Scott 2000, 516–20. Scott is partic-ularly concerned to counter Nadler’s account of “occasional causation” in Descartes. In contrastto the view I offer against Nadler in §4.2.3, however, Scott claims that transmission is to be con-trasted with “efficacious or productive agency” and that it occurs at a “pre-efficiency” stage

But what sort of contribution, exactly? My suggestion above, based on the remarksin the Sixth Meditation, was that bodies contribute the objective reality of the sensoryideas, that is, the bodily features the ideas represent insofar as these features are con-tained objectively in the mind (see §2.1.3 (i)). To be sure, Descartes does not refer inthe Comments passage to the objective reality of sensory ideas. However, I take hisstrange talk in this text of brain motions “approaching” or being “transmitted to” themind to reflect his earlier view that bodies send or emit the features that are containedobjectively in our sensory ideas. Even if the ideas do not resemble the motions in ourbrain that external bodies cause, and even if there is nothing that is literally transferredinto the mind from the motions, still these motions bring it about (albeit with the assis-tance of the activity of an innate mental faculty) that the mind has sensory ideas thatdirect it to certain bodily features rather than others. What the motions contribute, then,is the link between the ideas and the features of bodies that these ideas represent.

Admittedly, the claim in the Comments that sensory ideas are formed by an innatemental faculty seems to add something to the more spare account of the causation ofsensation in the Sixth Meditation. Indeed, I think that the Comments in fact reflectssome sort of development in Descartes’s view of how bodily motions bring aboutsensory ideas.52 So the question I need to address is why Descartes is concerned inthis text to distance himself from the view encouraged by some of his own remarksthat bodies serve as the total efficient cause of sensory ideas.

In addressing this question, Janet Broughton proposes that it was Descartes’scommitment to the containment axiom that led him away from the view that bodiesserve as such causes. More than this, her view is that this axiom led him to the con-clusion, which she finds in the Comments, that “the mind alone causes sensations.”Broughton reconstructs Descartes’s argument in this text as follows:

[T]he formal containment principle [i.e., the containment axiom as it concernscases where eminent containment is ruled out] applied to brain movements as thecause of sensation yields this requirement: Brain movements must contain exactlywhat is in the objects of sensation. Thus Descartes has shown in the [Comments]passage that a necessary condition for causation, as well as origination, has notbeen met. By showing that sensations do not originate from brain movements,Descartes has also shown that they are not caused by brain movements.

. . . The formal containment principle plus the [Comments] argument entail adenial that brain movements cause sensations. (Broughton 1986, 118–19)53


“immediately prior to the actual production of ideas by the mind’” (Scott 2000, 520). For yetanother interpretation of the Comments passage, on which Descartes held that the motionsserve as causal triggers but do not produce the content of sensory ideas, see Rozemond 1999,456–62.

52. See Wilson’s claim that her proposed reconciliations of Descartes’s various remarkson the causation of sensation “do not fully accommodate Descartes’s actual statements” and“do not at all resolve the problem about how the mind if truly ‘passive’ in sensation, can helpto bring about sensory ideas” (Wilson 1999b, 52). In moving closer to this claim, I am heremoving away from the more firmly “reconciliationist” position in Schmaltz 1997.

53. See the similar reading of the argument in the Comments in Secada 2000, 103–7, andGorham 2002.

There are two immediate questions that Broughton’s reconstruction of the argu-ment raises: (1) Does Descartes’s containment axiom plus his argument in theComments commit him to the denial that motions in the brain cause sensations?; and(2) is there even an implicit appeal in the Comments to the containment axiom?

First question first. We can understand this question in terms of our discussion in§4.2.1 of the scholastic problem of the connection between phantasms and intelligi-ble species. We have seen that though Thomas indicated that the phantasm can be a“secondary and instrumental” cause of the species, Suárez used occasionalist termi-nology in a manner that indicated that (apart from the divine concursus, of course)the agent intellect is the sole efficient cause of the species. What we need to ask iswhether Descartes’s claim in the Comments that brain motions merely serve as theoccasion for the formation of sensory ideas reveals that such motions are similar toSuárezian phantasms in lacking any sort of efficient causality, or whether such aclaim allows for something akin to the derivative causality of Thomistic phantasms.

I noted earlier the suggestion in Descartes that there can be efficient causes that arenot total or adequate causes, and so are not fully subject to the containment axiom (see§2.1.2). In light of this suggestion, one could argue on Descartes’s behalf that thoughbrain motions do not formally contain everything that comes about in the mind, theyare still partial efficient causes of sensory ideas. Perhaps the total and efficient causein this case is the combination of the motions and the innate mental faculty.

An initial worry, however, is that Descartes’s account of the activity of the innatemental faculty conflicts with his containment axiom. In particular, the concern is thatthe mental faculty responsible for forming sensory ideas can contain neither for-mally nor eminently those features of sensory ideas not contained formally in brainmotions. For Descartes indicates in the Comments that prior to their formation intel-lectual ideas exist merely potentially in the faculty that forms them (AT 8-2:361),and he presumably thinks that prior to their formation sensory ideas exist in the sameway in the innate sensory faculty in the mind. So if formal containment requires theactual presence of what is formally contained, sensory ideas cannot be formally con-tained in the faculty that forms them. But on the account of eminent containment thatI have attributed to Descartes (see §2.1.3 (ii)), they cannot be eminently contained inthat faculty either, since they are not features of objects lower on the ontologicalscale than minds.

One option would be for Descartes to allow for a weaker sort of formal contain-ment than actual possession. Even if those features of sensory ideas that derive fromthe innate mental faculty are not actually present prior to their formation by themind, the reality of those features could be contained in the reality of the innate men-tal faculty that forms them. It might seem that Descartes precludes this option whenhe insists in the Third Meditation that “the objective being of [an] idea can be pro-duced not only by potential being [esse potentiali], which strictly speaking is noth-ing, but only by actual or formal [actuali sive formali]” (AT 7:47). Given the claimhere, it may appear as if the reality of sensory ideas as present potentially in theinnate mental faculty could not play a role in the causation of the ideas themselves.54


54. Thanks to Dan Kaufman for this objection.

In this particular passage, however, what is “strictly speaking nothing” are divineperfections that cannot be contained in a finite mind. What worries Descartes is thatthere would be certain features of the reality contained in the idea of God that wouldhave no foundation in the reality of its cause. But it does seem that the reality of thesensory ideas that the mind has a faculty to form does have a foundation in an actualbeing, namely, in that very faculty. If so, then the mind could be said to formally con-tain the reality of those features of its sensory ideas that do not derive from motionsin the brain.55 The total and efficient cause of these ideas, consisting of motions inthe brain and the innate mental faculty, could be said to formally contain all of theeffect produced, and thus the containment axiom would be satisfied.

One might well protest at this point that the innate mental faculty all by itselfcould be the total cause of sensory ideas, in just the way that Suárez’s agent intellectis said (in his commentary on De Amina, at least) to be the total (secondary) efficientcause of intelligible species. What reason would Descartes have to resist the posi-tion—in line with one of Suárez’s views of the role of phantasms in the productionof intelligible species—that brain motions merely occasion rather than genuinelycause the formation of sensory ideas? One motivation for resistance emerges fromthe alternative in Suárez to a causal explanation of the link between the presence ofthe phantasm and the formation of the intelligible species. In De Anima, Suárezappealed to the fact that “the phantasm and the intellect of a human being are rootedin one and the same soul,” and thereby “have a wonderful ordering and harmony[mirum ordinem et consonantiam] in their operation” (DA IV.2, ¶12, Opera 3:719).Thus instead of invoking God as the direct cause of the harmony of operation, in amanner out of line with his anti-occasionalist sensibilities, Suárez held that the coor-dination of phantasm and intelligible species derives from the fact that the same soulis the efficient cause of both.56

An analogous explanation of the coordination of brain motions and sensory ideasis not open to Descartes. For unlike Suárez, he cannot say that it is the same soul thatunderlies the formation both of the motions and of the ideas. His dualism requiresthat he attribute the first two grades of sensation to radically different substances. Tobe sure, he does claim, in the passage from the Sixth Replies cited in §4.2.1, that sen-sations are a “result of ” the mind’s “being united with a bodily organ.” As indicatedin §4.1.2, though, Descartes cannot take this union to involve the formation of a sin-gle substance that encompasses both the motions in this organ and the sensationsin the mind. There remains the occasionalist option that God directly brings it aboutthat the motions are coupled with the sensations. Significantly enough, however, there


55. In contrast, Suárez claimed that in all cases of immanent action, the cause that pro-duces the effect is an equivocal one, and thus only eminently contains the effect; see MDXVIII.9, ¶10, 1:671.

56. Similarly, Suárez held that though a cognition of an object by means of external orinternal senses cannot be an efficient cause of the act of appetite that constitutes the desire forthat object, the “sympathy” between cognition and desire can be explained by the fact that therelevant faculties all are rooted in the same soul (DA III.9, ¶10, Opera 3:649*; V.3, ¶6, Opera3:759*). I am indebted here to the discussion of Suárez’s position in Rozemond 1999, 441.

is no appeal to God in the Comments.57 What Descartes claims there is rather that amental faculty forms the ideas at the same time that certain motions “approach” themind. One way of reading this claim, which is left open by his containment axiom, isthat the approaching motions and the mental faculty work together as partial efficientcauses of the same sensory effects.

But is there even an implicit appeal to this axiom in the Comments? This was thesecond question we asked of Broughton’s analysis of the Comments. To answer thisquestion, it is important to follow up on the reference in this text to the account ofsensation Descartes “explicated at length in Dioptrics.” In this earlier text, appendedto the 1637 Discourse, he argues that we do not perceive an object by viewing pic-tures of it in the brain, “as if there are other eyes in our brain with which we couldperceive it” (AT 6:130). The concern here therefore is to argue that the mind doesnot come to have the perception by viewing pictures composed of motions in thebrain.

The argument in the Dioptrics is in fact linked not to considerations involvingthe containment axiom, but rather to an account of sensation in even earlier writings,in which Descartes invokes an analogy to signs. In The World, which he aban-doned in 1633, Descartes responds to the claim that sensory ideas must be similar to“the objects from which they proceed” by noting that words need not be similarto what they signify.

Now if words, which signify nothing except by human institution, suffice to makeus conceive of things to which they bear no resemblance, then why could naturenot also have established a certain sign which would make us have the sensationof light, even if this sign has nothing in itself similar to the sensation? Is it notthus that nature has instituted laughter and tears to make us read joy and sadnessin the face of men? (W I, AT 11:4)

On the view presented here, bodily images that give rise to sensations are nonre-sembling signs, rather than perfect images, of external objects. Sensations are linkedto the world by virtue of being linked to bodily signs of objects.

In using the sign analogy, Descartes appears to suggest that the mind is aware ofthose bodily motions that are responsible for sensation. After all, the mind is awareof the laughter and tears by means of which it reads joy and sadness. More gener-ally, signs seem to be able to function as signs in virtue of the fact that they are reador comprehended. Yet as I have mentioned, Descartes indicates in the Dioptrics thathe does not wish to take the analogy this far when he ridicules the position that oneis aware of something in the brain, “as if there were yet other eyes within our brainwith which we could perceive [appercevoir] it.” To avoid this implausible position,


57. As admitted in Garber 1993, 23–24. However, Garber also sees Descartes as movingin his later writings away from the view in the Sixth Meditation that bodies are genuine causesof sensation and toward “something closer to what his occasionalist followers held, that Godis the true cause of sensations on the occasion of certain motions in bodies” (24). For adetailed argument against this hypothesis in Garber that I take to be decisive, see Scott 2000,504–15. See also note 48.

he offers the alternative view that motions in the brain “are instituted by nature tomake [the soul] to have such sensations” (AT 6:130).58

However, there is another aspect of the sign analogy that Descartes at least some-times explicitly embraces. This analogy can be seen as implying that the mind isactive in sensation. Just as the mind actively interprets written or spoken signs, so itresponds actively to corporeal motions by forming sensations. Descartes indicatesthat mental activity is a consequence of the sign analogy when he states in The Worldthat “it is our mind which represents to us the idea of light each time our eye isaffected by the action which signifies it.” The fact that the mind is active is linkedwith the fact that bodily motions serve as signs. He continues this remark by claim-ing that because physical processes merely signify sensory ideas, the latter are“formed in our mind on the occasion of our being touched by external bodies” (W I,AT 11:4-6). This view, recorded in a relatively early work, is retained in theComments, for in the latter the occasional connection between the presence of cer-tain motions and the formation of sensory ideas, on the one hand, and the possessionby mind of a productive faculty, on the other, are affirmed in the same sentence.

What we have in the Comments, then, is not the recognition of restrictions thatthe containment axiom places on body-to-mind action, but rather the development ofan account of sensation present in more primitive form in the Dioptrics and even ear-lier discussions of sensation. The concern throughout is to provide a replacement forthe view that the mind senses objects by viewing resembling images of them in thebrain. The alternative proposed is that what is present in the brain merely provides asign to the mind to form nonresembling ideas of the objects in itself. Moreover, thereis no indication that providing a sign is to be contrasted with causing the mind toform the ideas. In fact, Descartes speaks in The World of the mind as forming theidea of light when it is “affected by the action that signifies it” (AT 11:4), just as hespeaks in the Dioptrics of the brain motions being instituted to produce sensationsas “acting directly on the soul” (AT 6:130). This context of the argument in theComments reveals that it is removed not only from considerations involving the con-tainment axiom but also from the conclusion that bodily motions play no causal rolein the formation of sensory ideas.

4.2.3. Occasional Causation and Psychophysical Laws

On my interpretation, the final position in the Comments is that brain motions arereal efficient causes of sensory ideas, albeit causes supplemented by the activity ofthe innate mental faculty. There is the argument, however, that Descartes takesthese motions to be in a class of causes distinct from efficient causes. In particular,Steven Nadler finds in the Comments the position that by some means other thanefficient causality, the motions “induce” the innate mental faculty to be the efficientcause of sensory ideas. More generally, his proposal is that Descartes takes there to


58. Even so, in this same text Descartes suggests the position that the mind sees somethingin the body when he claims that we sense the size of an object by comparing our knowledgeof the distance to “the size of the images they imprint on the back of the eye” (AT 6:140).

be a class of “occasional causes” that provide the occasion for something else to bethe efficient cause of an effect. Nadler emphasizes that this sort of occasional cau-sation is to be distinguished from occasionalism. Occasionalism is a species ofoccasional causation, on which all causes other than God are occasional causes thatinduce God to be an efficient cause of their effects. However, occasional causationdoes not require occasionalism. Indeed, on Nadler’s view Descartes endorses occa-sional causation in the Comments but also holds in this text that the innate mentalfaculty rather than God is the proximate efficient cause of our sensory ideas (Nadler1994, 36–44).59

Nadler insists that though Descartes’s occasional causation is distinct from effi-cient causation, it “is still a real causal relation, albeit an inferior or secondaryvariety.”60 In contrast to the case of efficient causes, the relation of an occasionalcause to its effect is “not grounded in some ontically real power” in that cause.However, this relation is “not just a Humean relationship of succession,” but involvesgenuine nomological correlations. In Descartes’s case, according to Nadler, the lawsare grounded not in the nature of the occasional causes but in God’s will (Nadler1994, 42–43).

Nadler takes the suggestion in the Comments to be that motions in the brain areable to elicit the efficient causal activity of the innate mental faculty due to the factthat “God, in establishing the union of mind and body, has ordained that . . . partic-ular motions in the body should occasion the mind to produce particular ideas”(Nadler 1994, 50). He cites in this connection Descartes’s claim in his early work,the Treatise on Man, that

when God unites a rational soul to this machine . . . he will place its principal seatin the brain, and will make its nature such that the soul will have different sensationscorresponding to the different ways in which the entrances to the pores in theinternal surface of the brain are opened by means of the nerves. (AT 11:143–44)

Though God is invoked as an explanatory principle here, he is supposed to establishthe occasional causal relation between body and mind not by means of his continualactivity, as the occasionalist would have it, but “once and for all.” Nadler’s final judg-ment on Descartes’s account of occasional causation is: “Deus ex machina? Yes.Occasionalism? No” (Nadler 1994, 51).

I have noted the admission in Descartes of formal causes and (perhaps, withrespect to rational teleology) final causes distinct from the efficient causes governedby the containment axiom (§2.1.2 (ii)). Given Nadler’s view, we would need to add


59. Cf. the discussion of Descartes’s various uses of the term occasio in Specht 1966,41–56, and 1972, 12–27. Along with Nadler (see Nadler 1994, 40 n. 13), Specht is critical ofthe claim in Henri Gouhier that Descartes did not in fact use this term in a philosophically rig-orous manner (Gouhier 1926, 83–88). However, Specht differs from Nadler in takingDescartes to adopt the view in late scholasticism that occasional causes are efficient causesthat are accidental, indirect, or assisting causes (see, e.g., Specht 1972, 19).

60. As Nadler notes, however, this claim causes problems for his view that occasionalismis a species of occasional causation, given that occasionalists have typically been concernedto deny that occasional causes have any sort of causal efficacy (Nadler 1994, 42, n. 16).

occasional causes as another sort of nonefficient cause.61 However, I am not per-suaded that Descartes takes occasional causes to produce effects by some meansother than efficient causality. Nadler finds in Descartes’s account of body–bodyinteraction the suggestion that in the case of efficient causation, “something literallypasses from cause to effect, either because the cause gives up something to the effector because it multiplies something of its own to share with the effect” (Nadler 1994,38). But though Descartes does suggest at times what Broughton calls the “migra-tion theory of motion,” I have argued that his more considered view is that a bodyserves as an efficient cause of motion not by transferring its own mode of motion,but rather by applying its force in a manner that results in the production of a numer-ically distinct mode of motion (see §3.2.1 (iii)). Thus, the fact that motion cannot lit-erally transfer a sensory idea into the mind does not show that it cannot serve as anefficient cause of such an idea.

Moreover, there is reason to doubt Nadler’s claim that the psychophysical lawsthat Descartes posited to explain the formation of sensory ideas are not grounded inthe natures of the objects these laws relate. Indeed, in the passage from the Treatiseon Man that Nadler cites, Descartes appeals to the fact that God made the nature ofthe human soul such that it has sensations corresponding to different states of thebody to which it is united. To be sure, it is the nature of the soul that is mentionedhere rather than the nature of the body that acts on it. However, the indication is thatthe nature of the soul is such that it is affected in particular ways by the action ofbody on it. Likewise, in a 1647 letter Descartes claims that motions can be connectedto thoughts they do not resemble since

our soul is of such a nature that it can be united to a body, it also has this propertythat each of its thoughts can be associated with certain motions or otherdispositions of this body so that when the same dispositions recur in the body,they induce [induisent] the soul to have the same thought. (To Chanut, 1 Feb.1647, AT 4:604)

This passage has been read (though not by Nadler) as a “foreshadowing of a Humean,constant conjunction, or regularity analysis of causation.”62 However, the reference tothe fact that the mind has a “nature” that explains the connection of its thoughts tobodily motions and to the fact that these motions “induce” the thoughts in the mindindicates that Descartes was thinking of a causal relation grounded in the natures ofthe interacting entities. In particular, the human soul has a nature that explains howmotions in the human body can cause it to have various sensations. In light of the


61. Cf. the view in Baker and Morris 1996 that Descartes took occasional causation tolack the sort of intelligible connection between causes and effects that is present in the caseof efficient causation (see, for instance, 155–56). In suggesting that Descartes’s appeals tooccasional causes lack the sort of rigor present in the case of causal explanations in hisphysics, Baker and Morris offer a view that is similar to the interpretation of Descartes in Loeband Alanen that I consider presently.

62. This reading is found in Loeb 1981, 137. See the discussion below of Loeb’s morerecent view of the mind–body relation in Descartes.

remarks in the Comments, we can take Descartes’s position to be that the soul has aninnate faculty that forms certain sensory ideas when certain motions in the body towhich that soul is united act on it. Insofar as God connects motions and sensations bycreating human souls and their bodies with particular natures that themselves carrythe causal load, we have some reason to conclude, with respect to Descartes’s accountof occasional causation: Deus ex machina? No.

In fact, the account of body-to-mind action that I find in Descartes is similar inimportant respects to the “conservationist” account of body–body interaction thatI have attributed to him (see §3.3). In particular, I have argued that he takes God’s“ordinary concursus” with such interaction to consist in his continued conservationof the matter in motion that, given certain initial conditions, itself determines sub-sequent changes in the distribution of motions. Likewise, the suggestion in theComments is that God brings about body-to-mind action by creating the human mindwith a faculty that itself responds to the action of certain bodily motions by formingparticular sensory ideas. In the case neither of body–body interaction nor of body-to-mind action does God bring about the relevant connections by directly producingthe corresponding states or by instituting the governing laws by means of a separate“law volition.” Rather, in both cases he simply creates and conserves objects withnatures that themselves determine the laws that are followed.

It is true that Descartes does not emphasize that the laws connecting motions tosensations derive from God’s immutability, as he did in the case of the laws govern-ing body–body interaction. But there is one intriguing passage that appeals to divineconstancy in the case of body-to-mind action. In the Conversation with Burman,Descartes is reported to have responded to the question of why God allows for sen-sory deception by claiming that

God fabricated our body as a machine, and willed that it act as a universalinstrument, which always operates in the same manner according to its laws. Andthus when it is well disposed, it gives [dat] to the soul a correct thought; whenpoorly [disposed], it nonetheless still affects [afficit] the soul according to its laws,so that there must result a thought by which it is deceived; for if the body did notsupply [the deceptive thought], it would not act uniformly [aequaliter] andaccording to its universal laws, [and] there would be a defect in God’s constancy,since he would not be permitting [the body] to act uniformly, when there areuniform laws and modes of acting. (AT 5:163–64)

I am reluctant to stake too much on this passage, since it is not from Descartes’s ownhand. However, the claim here that God makes the human body such that it “affects”the soul in accord with its universal laws complements Descartes’s view elsewherethat God makes the nature of the soul such that motions in its body regularly induceit to have certain sensory thoughts. Divine constancy would seem to ensure thathuman souls and their bodies retain their same natures, and thus that they continueto follow the laws that those natures determine.

In contrast to Nadler’s claim that Descartes takes the occasional causationinvolved in mind-to-body action to derive from laws that are not grounded in onti-cally real powers in objects, I find in the texts the suggestion that these laws have afoundation in the natures that God creates and conserves. The depth of my interpre-


tive dispute with Nadler is indicated by his remark in a note that the ontologicalstatus of the power of an occasional cause

can be compared to the ontological status of force in Descartes’s physics: force isnothing really (i.e., ontically) in bodies . . . , but rather is simply the conformity ofbodies in motion and at rest to certain laws of nature due to the way they aremoved by God (whose will is, in a sense, the real locus of force). Likewise, A’s“power” to occasion e is simply the lawlike correlation between A and e, asestablished by God. (Nadler 1994, 43, n.18)

Now I agree with Nadler that the powers of occasional causes can be compared tothe ontological status of force in Descartes’s physics. But whereas Nadler attributesto Descartes the view that these powers, like the forces, are nothing real in objectsbut are due solely to divinely imposed laws, I take him to offer the position that thepowers, like the forces, derive from natures in objects that reflect the relevant laws.It is true that in contrast to the case of body–body interaction, Descartes holds thatmotions are merely occasional causes of sensory ideas. But this is not because bod-ies cannot be real efficient causes of ideas. Rather, whereas bodily forces can betotal causes of changes in motions, such forces can bring about the formation ofsensory ideas only due to the distinctive nature of the mind on which they act.Without the presence of an innate mental faculty that is sensitive to these forces, themotions would not be able to act on the mind at all. Insofar as this faculty pertainsparticularly to the union, this account of the manner in which motions serve asoccasional causes is in line with Descartes’s remark to Elisabeth that it is only bymeans of the primitive notion of the union that we can conceive “the force . . . ofthe body to act on the soul, in causing its sensations and passions” (21 May 1643,AT 3:665).

In taking the remarks to Elisabeth to indicate that there is an ontological groundfor psychophysical laws, I deviate not only from Nadler but also from commentatorswho find in Descartes the position that these laws are “brute” facts that, in contrastto laws governing motion, are incapable of yielding true scientific explanation. Forinstance, Louis Loeb claims that “in Descartes’s view, the mind–body union consistsin a set of connections between specific types of brain states and specific types ofthoughts . . . that are not themselves subsumed under more general connections fromwhich they can be derived” (Loeb 2005, 70). He notes the similarity of this claim tothe conclusion in the work of Lilli Alanen that there is in Descartes “no room . . . fora scientific . . . psychology accounting for the laws of our mental life as embodied,human persons” (Alanen 1996, 6).63

Is it so clear, however, that Descartes has no room for a scientific psychology?After all, in a famous passage from the preface to the French edition of the Principles,


63. Cited in Loeb 2005, 85, n.70. Alanen develops her position further in Alanen 2003.See also the conclusion in Della Rocca forthcoming that the causes involved in mind–bodyinteraction do not render their effects intelligible. But cf. Hatfield’s thesis, closer to my ownview, that “Descartes included (at least some functions and states of) mind as part of nature,[and] despite his dualism he continued an established tradition of treating the operations of thesenses as open to empirical investigation” (Hatfield 2000, 631).

he claims that the “tree of philosophy” that has metaphysics as its roots and physicsas its trunk yields as branches the three “principal sciences” of medicine, mechanics,and morals. Moral science in particular is presented as the most important of thesesciences, “which presupposes a complete knowledge [connoissance] of the other sci-ences and is the ultimate level of Wisdom” (AT 9-2:14). This highest science ofmorals seems to depend on the systematic classification Descartes later offers in thePassions of the perceptions of the human soul that derive from certain motions in thebody to which it is united. In this same text, he insists that the soul has a “nature” suchthat “it receives as many different impressions, that is, it has as many different impres-sions as there occur different motions” in the pineal gland (PS I.34, AT 11:354). Itseems that the psychophysical laws that Descartes invokes are not brute or unsub-sumed, but rather are to be explained in terms of the special nature of the human soul.We know from the Comments that this nature consists in part in the presence of aninnate mental faculty for forming sensory ideas.64

There is admittedly the troubling appeal to ends in Descartes’s account of theunion. Here we would seem to have an important difference from his nonteleologi-cal physics.65 Even so, an appeal to ends need not preclude an investigation of thevarious efficient causes responsible for the fact that human beings have the variousstates that serve the natural end of the preservation of the union. In particular, theends could be “hardwired” into the innate mental faculty responsible for the forma-tion of sensory ideas. The nature of this faculty could be such that it forms just thosesensory ideas that serve this end when the appropriate motions in the brain act on thehuman soul.66 Even given this view, both the motions and the faculty remain efficientcauses that jointly explain the presence of the sensory connections governed by psy-chophysical laws. For Descartes, of course, God must be in the background here, forit is he who must create and conserve human souls and their bodies with the naturesthat make such connections possible. Even so, it is important to emphasize that hetook God to be in the background, and that he related psychophysical laws mostimmediately to the conserved natures.


64. Loeb objects to the related proposal in Rozemond that Descartes can explain sensationin terms of innate causal capacities in the mind on the grounds that investing “the mind witha plethora of fine-grained causal capacities does not secure an explanation of the specific con-nections in terms of the mind’s essence” (Loeb 2005, 85, n.64, citing Rozemond 1999,455–56, 464–65). But even if there is no explanation here in terms of mind’s essence as apurely intellectual substance, there does seem to be an explanation in terms of the innate men-tal faculty responsible for the formation of sensory ideas.

65. See Loeb’s claim that Descartes’s appeals to teleology in his explanation of the unionreveal that he exempted “his account of the mind and its relationship to the body not only frommechanism, but also from the sorts of explanatory requirements he himself regards as properto physical science.” Just prior to making this claim, however, Loeb admits that this feature ofDescartes’s thought conflicts with the fact that he “aspired to a scientific treatment of a widerange of broadly ‘mental’ phenomena” (Loeb 2005, 79).

66. Cf. the argument in Simmons 2001 that Descartes’s appeal to the ends of the union isconsistent with his restriction of causal explanations of the states of the human being to effi-cient causes.

4.3. MIND-TO-BODY ACTION

Just as we have had to overcome a fixation on the heterogeneity problem to under-stand Descartes’s struggles with body-to-mind action, so we will need to look beyondthis problem to discern the issues he was concerned to address with respect to actionin the other direction, namely, mind-to-body action. Our consideration of Descartes’saccount of body-to-mind action started with a scholastic problem concerning interac-tion that pertains exclusively to this sort of action. The treatment here of his accountof mind-to-body action will start with a different scholastic problem concerning inter-action that derives from the Aristotelian prohibition of action at a distance. For manyscholastics, this prohibition precluded the possibility of any action of spirits on bod-ies that does not involve their spatial “presence” to those bodies.

This scholastic position provides the proper context for Descartes’s two mainattempts to defend his position that an immaterial substance can act on an extendedthing. The first involves his claim in the Sixth Replies and the 1643 correspondencewith Elisabeth, among other places, that we can understand this action in terms of theaction of the scholastic quality of weight or heaviness. The second involves the claimin the Elisabeth correspondence as well as in his later correspondence with More thatwe can understand an immaterial substance that acts on a body to have a special sortof extension, which he calls in the More correspondence an “extension of power.” Inboth cases, I take the claims to indicate—in line with a traditional scholastic view—that minds are indeed spatially present to the bodies on which they act.

However, Descartes’s discussion of mind-to-body action broaches issues that gobeyond not only the heterogeneity problem, but also scholastic accounts of the actionof spirits on bodies. One notable example is the issue—which Leibniz popularized—of the compatibility of the action of mind on body with the principles of Descartes’sphysics, and in particular with his principle of the conservation of total quantity ofmotion in the material world. Here I enter a debate in the literature over Leibniz’s viewthat Descartes was led by the constraints of his conservation principle to limit the actionof mind on body to changing the direction of motion. In the end I side with the con-clusion of critics of this view that Descartes in fact never endorsed the “change of direc-tion” account of such action, but also with the conclusion of Leibniz’s defenders thatthe constraints of Descartes’s physics provide good reason for him to endorse it.

4.3.1. Scholastic Spiritual Presence

Descartes of course was not the first to claim that the mind is immaterial, and so tobroach the problem of how the mind can act on something that is material. The doc-trine of the immateriality of the intellect goes back at least to Plato, and was stan-dardly accepted by the scholastics. Thus, Aquinas insisted that purely spiritualsubstances, and even the rational souls that serve as the form of human composites,are incorporeal entities that can or do exist apart from body.67 However, the principles


67. See, for instance, his argument in ST I.75.2 for the conclusion that the human soul issomething that can subsist by itself apart from body, and thus is an incorporeal substance.

underlying his account of the action of the immaterial on the material emerge mostclearly in his discussion not of the action of finite spirits on body, but rather of God’s“presence” in the material world. Aquinas argued that God must be present in everyplace in this world by appealing to the axiom in Aristotle that “no action of an agent,however powerful it may be, acts at a distance, except through a medium.”68 Givenhis conclusion that every being and every action depends immediately on God’saction (see §1.1.2), it follows from the “no distant action” axiom that nothing is dis-tant from him, “as if it could be without God in itself ” (ST I.8.1, ad 3).

It may seem that in attributing ubiquitous material presence to God, Thomas alsoattributed to him extension, and thus divisibility. Yet he attempted to avoid the het-erodox implication of divine divisibility by holding that God does not inhabit placesin the way in which bodies do. To illustrate the special sense of spiritual presenceinvolved in the case of God, Thomas turned to the case of the presence of the rationalsoul in the human body. As a “perfect form,” this soul is so present “whole in wholeand whole in each part” (tota in toto et tota in singulis partibus).69 Since it does nothave parts spread out in the parts of the body, the soul remains indivisible even whilebeing present throughout the body. Likewise, according to Thomas, God is “whollyin all things and in each one” (ST I.8.2, ad 3).

In this very same text, however, Aquinas explained the distinctive nature of God’sspiritual presence by appealing to something other than the manner in which the rationalsoul serves as the form of the human body. In particular, he appealed to the fact thatincorporeal beings are present where they act “not by contact of dimensive quantity, asbodies are, but by contact of power” (ST I.8.2, ad 1). This notion of presence by “con-tact of power” is prominent also in the question of the Summa Theologiae devoted tothe relation of angels as purely spiritual substances to place. There it is said that an angelis “in a corporeal place by application of the angelic power in any manner whatever toany place” (ST I.52.1).

Admittedly, Aquinas went on to compare this sort of presence to the presence ofthe rational soul in a human body, noting, for instance, that an incorporeal substancethat is in a place by contact of power contains but is not contained by a place, as “thesoul is in the body as containing it, not as contained by it” (ST I.52.1). It thus mayappear that the difference between the presence of the human soul in its body andthe presence of an angel by contact of power is not significant after all. However,some sort of distinction is required by his admission that both a human soul and anangel can be co-present in a body.

To see why this admission requires such a distinction, we need to considerAquinas’s argument that two angels cannot exist in the same place at the same time.This argument begins with the premise that an angel can be present in a place only


68. In ST I.8.1, Thomas cited Aristotle’s discussion in Physics VII as the source for thisaxiom.

69. The contrast here is with imperfect forms, such as those plants and lower animals, orbruta, which have parts that are distributed in the parts of the body to which it is united thatcan exist apart from the whole (as in cases of plant cutting or the division of a worm). On laterscholastic accounts of divisible souls, see Des Chene 2000a, ch. 9.

if it perfectly contains it, and it can so contain such a place only if it is a total causeof what occurs at that place. But Aquinas rejected the possibility of the sort of causaloverdetermination that occurs when an effect has two different total causes of thesame order.70 Thus his conclusion is that there can be only one angel present in aplace at a time. However, he noted as a possible counterexample to this conclusionthe—admittedly quaint—case of demonic possession. The problem in this case isthat we seem to have the co-presence of two spiritual entities—namely, the demonand the possessed soul—in the same body at the same time. However, Aquinasresponded that the demon and the soul do not have “the same relation of causality”to the body insofar as the soul serves as the substantial form of the body whereas thedemon does not (ST I.52.3, ad 3).71 Though the demon and the possessed soul areco-present in the body, then, they are present there by means of different sorts ofcausality, and thus there is no instance here of causal overdetermination.72

Despite this admission of different ways in which incorporeal entities can be causallyrelated to bodies, it continues to be the case for Thomas, given his understanding of theAristotelian no-distant-action axiom, that any sort of action of such entities on bodyrequires that they be spatially present where they act. However, this requirement on spir-itual action was not universally accepted. A generation after Thomas, for instance, JohnDuns Scotus (†1308) insisted that it is essential to divine omnipotence that God can pro-duce effects without being present where he acts.73 Moreover, Thomas’s critic Durandusargued around the same time, in his Sentences, that even finite angels act through theirintellect and will alone, and thus need not be present in a place in order to act there (SI.37.1, 1:101–2*).74 To be sure, Durandus’s claim that the angel is present “by opera-tion” (per operationem) or “the application of power to a place” (applicationem virtutead locum) (S I.37.1, ¶39, 1:102*) may not seem to be far from Thomas’s view of angelicpresence via a “contact of power.” Nevertheless, Durandus explicitly rejected the appli-cation of the Aristotelian no-distant-action axiom to the case of angelic action on body,arguing that in this case it is only the effect, and not the intellectual power responsiblefor the effect, that must have a spatial location.75


70. The qualification “of the same order” is needed, of course, to allow for Thomas’s “causalcompatibilist” position that both God and creatures can be total causes of the same effect giventhat creatures belong to a causal order that is subordinated to divine action (see §1.1.2).

71. This response draws on the position in ST I.51.3 that angels cannot exercise the vitalfunctions of a body, since they are not united to a body.

72. Perhaps the view here is that the demon is the total cause of the soul’s being the totalcause of certain bodily actions.

73. For discussion of Scotus’s position with references, see Sylwanowicz 1996, 174–81.74. This quaestio is a commentary on ST I.52.1. Durandus also insisted, in response to

Thomas’s views in ST I.8.2, that God is in all places “not in himself, but according to his effects”(S I.37.1, ¶6, 1:100*). In line with his “mere conservationist” position, Durandus held that theaction by which God is present in all places is his conservation of the existence of all materialbeings.

75. So also, Durandus argued contrary to Thomas’s view in ST I.53.1 that since angelscannot occupy a place, they cannot be said to move locally, that is, from one place to another(S I.37.2, 1:102–3*).

In his Disputations, Suárez defended the need for spiritual presence in the case ofangelic action on body against the objection in Durandus that in such a case what isrequired is not the proximity of position but an immediacy of intellectual power(cited in MD XVIII.8, ¶7, 1:652). Suárez insisted that an angel must have “a sub-stantival and real delimited presence in accord with which it must be conjoined to apatient when it acts on that patient,” and that such a presence is “perhaps whatSt. Thomas meant by ‘spiritual contact with a body’” (MD XVIII.8, ¶42, 1:666).76

In response to Durandus, Suárez noted that since the angel can will any object indif-ferently but cannot move just any object, its power to move must be distinct from itswill. The efficacy of this power requires that the angel be present to the body to moveit (¶43, 1:666–67; cf. MD XXXV.6, ¶¶22–23, 2:474–75*).

To explain the nature of the “substantival and real delimited presence” requiredhere, Suárez appealed to our experience of “the rational soul that is diffused throughthe whole body and necessarily is as whole is in whole and whole is in each part”(MD XXX.7, ¶44, 2:109*). As in the case of Aquinas, the point of the comparison tothe soul’s union with body is to show that what is present in a divisible body neednot itself be divisible. Even so, Suárez allowed that the substantival presence of anangel in a body differs in important ways from the presence of the rational soul inits body. Whereas the soul can be united only to a body specially suited to it, forinstance, the angel is not limited to a particular sort of body (MD XXXV.6, ¶23,2:475*).77 On the other side is the fact that the rational soul, as the substantial formof a body, can be the source of vital actions of that body. Since the angel is not a sub-stantial form, it cannot produce such actions, but can bring about in a body only localmotion (¶¶15–16, 2:472–73*).

This last difference indicates an interesting point of contact between Aquinas andSuárez. Aquinas had argued that though an angel is more perfect than a body, still itcannot cause those actions of the body that derive from its form, since it is not suf-ficiently similar to a substantial form (ST I.110.2).78 Suárez attempted to reinforcethis line of argument by appealing to the fact that “a given thing’s being more per-fect is not sufficient for its being able to effect a less perfect thing.” What is requiredfurther is that the cause be “proportioned to the type of action in question.” Sinceangels are not proportioned to the actions of substantial forms, however, they cannotproduce such actions (MD XVIII.1, ¶17, 1:598). Suárez linked this inability to thefact that angels “do not contain corporeal things eminently.” I have noted thatDescartes’s account of eminent containment is more restrictive than Suárez’s insofaras the former has the implication missing in the latter that only minds can eminentlycontain objects (see §2.1.3 (ii)). But now we can discern a sense in which Suárez’saccount is more restrictive than Descartes’s. For we discovered earlier the implica-


76. Suárez noted, however, that the claim that a spiritual entity must have such a presencedoes not itself commit one to a view on whether the operation constitutes the nature of thispresence or whether (as Suárez himself held) the presence is prior to and independent of thatoperation (MD XVIII.8, ¶42, 1:666).

77. For the similar position in Thomas, see ST I.110.3, ad 3.78. Cf. note 71.

tion in Descartes that a finite mind can eminently contain the entire visible worldeven though it cannot produce such a world. Missing in Descartes, then, is Suárez’srequirement that what eminently contains an object must be proportioned to the pro-duction of that object.

Given his rejection of the claim that angels can eminently contain bodies, how-ever, Suárez might seem to be committed to the conclusion that they cannot producelocal motion in bodies either. Yet he argued that since local motion is not bound upwith the form of a body, it is something to which an angel can be proportioned (MDXXXV.6, ¶16, 2:478*). Here he was drawing on the claim in Thomas that angels arecapable of producing local motion, since such motion is “perfect” in the Aristoteliansense of not requiring any intrinsic change in the moving object (ST I.110.3).79

Since Descartes holds that all change in body is due to the local motion of its parts(see, e.g., PP II.23, AT 8-1:52–53), he need not take the result in Aquinas and Suárezthat angels are limited to local motion to indicate any significant restriction onangelic action. Moreover, he is quite willing to admit that angels can act on body, aswhen he claims in his 1649 letter to More that “an angel can exercise power now ona greater and now on a lesser part of corporeal substance” (15 Apr. 1649, AT 5:342).However, his further claim in this letter that angels and other incorporeal substanceshave “no extension of substance [extensio substantiæ], but only an extension ofpower [extensio potentiæ]” (5:342) raises the question of whether he could acceptthe claim in Aquinas and Suárez that there can be no spiritual action on a body with-out the presence of that spirit to the body it moves. In other words, does Descartes’s“extension of power” involve something as thick as Suárez’s “substantival and realdelimited presence,” or rather only something as thin as Durandus’s “application ofpower to a place”? With respect to this question, it turns out to be relevant that whenElisabeth challenged Descartes to explain how an immaterial substance could affectan extended thing, he appeals to the scholastic conception of the manner in whichthe real quality of weight or heaviness acts on body.

4.3.2. The Heaviness Analogy and the Extension of Power

Elisabeth’s initial objection to Descartes regarding the conceivability of mind–bodyinteraction was that it is not conceivable that an immaterial mind could “determinethe motion” of a body given that it is not literally in contact with that body (see§4.1.1). To show how he conceived of the interaction in this case, Descartes respondsby appealing to the scholastic account of heaviness in terms of the real quality ofweight (pesanteur). On that account, this quality consists simply in “the force tomove the body in which it is toward the center of the earth,” and even though thequality is supposed to be a being distinct from body, “we have no difficulty in con-ceiving how it moves this body or how it is joined to it; and we never think that[motion] is made by a real contact between two surfaces, because we experience inourselves that we have a particular notion to conceive it.” This notion is simply that


79. In this text Thomas cited Aristotle’s claim in Physics VIII.7 (261a14, Aristotle 1984,1:436) that local motion requires no change in the quality or quantity of the object moved.

of the union, or more particularly of the force of the soul to move the body. Descartesemphasizes that we misuse this notion when we use it to conceive of the action ofthe scholastic quality of weight, since this quality “is nothing really distinct frombody, as I hope to show in my Physics” (21 May 1643, AT 3:667–8). Even so, thepoint of the analogy is to draw Elisabeth’s attention to the fact that the notion pro-vides a manner of conceiving of the action of the soul on body, despite the fact thatit is something that is really distinct from body.

In his letter to Elisabeth, Descartes refers back to his discussion in the SixthReplies of the action of heaviness (gravitatem) (AT 3:666).80 He observes in thattext that he had in his youth conceived of heaviness as a quality insofar as “I referredit to the bodies in which it inhered,” but also as real insofar as “I really took it tobe a substance: in the same way that clothing, considered in itself, is a substance,although when it is referred to a clothed man, it is a quality.” He had further con-ceived of heaviness as something which, though it is “spread out through the wholebody that is heavy,” nonetheless does not possess the same kind of extension thatbody possesses, since “the true extension of body is such that it excludes the pene-tration of all parts,” whereas heaviness is something “extended throughout the heavybody.” The co-extension of this quality is also tied to the fact that “it could exerciseall of its power [vim] in any part of [the heavy body],” as when that body pulls on arope attached to one of its parts “just as if this heaviness were only in the part touch-ing the rope instead of also being spread through the other parts.” This conceptionof the power of heaviness is supposed to be akin to the conception of the soul asrelated to the body in which it is united as “whole in whole, and whole in each of itsparts” (AT 7:441–42).

However, this comparison seems to be weakened by the fact that the co-extensionof the human soul does not include the ability of that soul to exercise its power in anypart of its body. As I have indicated, Descartes makes clear that the human soul canbring about voluntary motion only by acting on the pineal gland (see §4.1.3). To besure, he allows in the Passions that this soul is related to the whole body given that thisbody is “in a sense indivisible because of the arrangement of its organs, these being sorelated to one another that the removal of any one of them renders the whole bodydefective” (PS I.30, AT 11:351). Insofar as the soul is joined to a body that is indivisiblein this way, it can be said to be “whole in whole, and whole in each part.” Nevertheless,this relation of the soul to the whole extension of the body does not involve the powerof the soul to act directly on any part of that extension. In this respect, then, the co-extension of the soul must be distinguished from the co-extension of heaviness.81

I have drawn attention to the confusion in Descartes of the conception of mind asbeing able to act on body with the conception of the human soul as united with aparticular body (see §4.1.1). In his discussion of the heaviness analogy, we havethe related confusion of the extension that the mind must have to act on body with


80. See also the references to this discussion in To Hyperaspistes, Aug. 1641, AT3:424–25; and To Arnauld, 29 July 1648, AT 5:222–23.

81. Cf. the criticism of Descartes’s use of the heaviness analogy in Rozemond 2005,which has influenced my discussion here.

the extension that the soul has in virtue of its union with body. To act on a body, mindmust have a power that is somehow present to that body. But the union with a bodypresupposes the different condition that this body consist of parts that are systemat-ically interrelated in a particular manner.

Nevertheless, there is one point of contact between the cases of mind-to-bodyaction and soul-body union that may help to explain why Descartes confuses them.Both this action and the union require that the mind or soul have a location, and so notbe something that lacks any sort of extension. Even though Descartes held in his youththat the quality of heaviness is really distinct from body, still he was able to conceiveof it as being located in the parts of the body where it can act. And he was able to con-ceive of heaviness in this way because he already had the notion of his soul as animmaterial thing that is nonetheless present in the body to which it is united. ThoughDescartes is not careful to note that the sort of presence involved in heaviness differsfrom the sort of presence involved in the union, this need not undermine the basic pointthat the immateriality of the soul does not preclude its contact with a body. And thispoint is all that he needs to address Elisabeth’s original worry that the touch requiredto determine the motion of body is incompatible with the immateriality of mind.

Descartes does not simply drop the claim in his correspondence with Elisabeth thatthe soul can be conceived to have some kind of extension. Indeed, this claim plays acentral role in his late correspondence with Henry More.82 More took exception toDescartes’s doctrine that the nature of body consists in extension on the grounds that“God, and an angel, truly any other thing subsisting per se, is a res extensa” (11 Dec.1648, AT 5:238*). For More, anything that exists must be somewhere, and so musthave an extension. Whatever lacks extension can only be nowhere, and thus be noth-ing. This explains why More later charged that those who follow Descartes in takingminds to lack extension are “Nullibists,” literally, nowherists.83

To be sure, Descartes does deny in his initial response to More that “real exten-sion [veram extensionem], as commonly conceived by everyone, is found either inGod, or in angels, or in our mind, or in any substance that is not a body.” The factthat such substances do not have real extension is revealed by the fact that “if therewere no body, I would understand no space with which angels or God would becoextended.” Nevertheless, Descartes also insists in this response that immaterialsubstances can have “capacities or powers” (virtutes aut vires) that “apply them-selves to extended things” (5 Feb. 1649, AT 5:269–70). In a previously cited passagefrom a subsequent letter to More, he expresses this point by saying that even thoughminds do not have “an extension of substance,” nonetheless they do have an “exten-sion of power,” as shown by the fact that “an angel can exercise its power now in agreater, now in a lesser part of corporeal substance” (15 Apr. 1649, AT 5:342).

This response did not placate More. He protested that it implies “a contradictionthat a power of the mind is extended, when the mind itself is in no way extended.”


82. For a comprehensive survey of More’s attitude toward Cartesianism, see Gabbey1982. See also the more recent discussion in Jesseph 2005 of More’s contribution to the recep-tion of Descartes in England.

83. See the discussion in More’s 1671 Enchiridion Metaphysicum (More 1995, 27.2).

More was particularly concerned to argue that we could not distinguish between exten-sion of power and extension of substance in God’s case given the doctrine of absolutedivine simplicity. But he also made the point that even in the case of finite minds, therecannot be a distinction between the two kinds of extension given that the power that isextended is merely “an intrinsic mode of the mind, it is not outside of the mind itself”(23 July 1649, AT 5:379*). Since a mode is merely the manner in which a substanceexists, the mode cannot have any extension if the substance it modifies lacks extension.

Descartes ultimately concedes that any distinction of God’s substance from hispower is problematic. He is led thereby to the Thomistic view of God’s ubiquity,holding that “God’s essence must be present everywhere for his power to be ableto manifest itself everywhere.” But there is no indication that Descartes is willing togive up his earlier claim that extension is not essential to God insofar as God couldexist in the absence of any body. One has the sense here that he is simply having dif-ficulty expressing this point in a manner consistent with the doctrine of divine sim-plicity.84 Nevertheless, in the case of finite minds, there is no similar barrier to hisassertion that it is not essential to an immaterial thing that it have a relation to exten-sion. In this sense, Descartes is indeed a “nowherist.”

Even so, this sort of nowherism admits only the possibility that minds exist in aworld without bodies, and thus without any extension. There remains the question ofwhether Descartes is a nowherist with respect to minds that act on bodies in theactual world. To address this question, we need to consider further the sort of “exten-sion of power” that he attributed to minds. The discussion in §4.3.1 of the scholas-tic debate over spiritual presence is in fact germane to such a consideration. We haveseen that Thomas was led by his acceptance of the Aristotelian no-distant-actionaxiom to the conclusion that there must be a “contact of power” in the case of theactions of spirits on bodies. In Suárez, this contact of power required in turn a “sub-stantival and real delimited presence” of the spirit to the body. If Descartes’s exten-sion of power were equivalent to Thomas’s contact of power, particularly as Suárezinterpreted it, then he would be committed to the conclusion that this extensionrequires that minds that act on bodies are really located where those bodies are.

However, there is also the possibility that the Cartesian extension of power is justa version of Durandus’s “application of power to a place,” which requires the loca-tion only of the effects of the action of a spirit on body, and not of the spirit itself. IfDescartes embraced Durandus’s alternative conception of spiritual presence, hecould have responded to More’s main objection by simply denying that the mentalpower to act on body, as opposed to the effects of this power, has a location.

It is perhaps significant that Descartes does not respond to More in this manner, butsimply insists that minds lack any essential relation to extension. Even more significant,though, is his persistent suggestion that the scholastic conception of the relation of heav-iness or weight to the body in which it inheres is modeled on the conception of the rela-tion of our soul to the body with which it is united. For surely Descartes does not take


84. See Descartes’s earlier claim to More that he prefers to discuss the nature of extensionwith respect to finite minds, “which are more on the scale of our own perception, rather thanto argue about God” (15 Apr. 1649, AT 5:343).

the scholastics to hold that the quality of heaviness lacks any location. Indeed, the indi-cation in the Sixth Replies is that this quality is in fact located throughout the heavybody. Given that our idea of the soul provides the model for this conception of heavi-ness, it would seem that this idea presents the soul as something that also is locatedin the body with which it is united. Contrary to the view in Durandus, then, it is not justthe effects of our soul that are present in our body, but our soul itself.85

Of course, we must keep in mind Descartes’s warning in his correspondence withMore, as well as in his correspondence with Elisabeth, that immaterial things are notpresent in the material world with the same sort of extension that bodies possess.Whereas the extension of bodies entails impenetrability, and thus an exclusion of allother bodily extension, such is not the case with the extension of minds (cf. To More,Aug. 1649, AT 5:403, and To Elisabeth, 28 June 1643, AT 3:694–95).86 Yet whatmay seem strange, on a contemporary understanding of Descartes’s dualism, is thefact that he attributed any sort of extension to mind. In light of scholastic discussionsof spiritual presence to bodies, however, this attribution is not only not unusual, butalso understandable as a means of avoiding action at a distance, or at least actionwithout proximity, in the case of mind-to-body action.87

4.3.3. Voluntary Motion and the Conservation Principle

In chapter 3, I noted the doctrine in Descartes—central to his physics—that God’sordinary concursus does not bring about any change in the total quantity of motionor motive force that he originally created in matter. One natural question is how


85. As indicated in Reid 2008, later Cartesians such as Geraud de Cordemoy (1626–84)and Johann Clauberg (1622–65) insisted that finite spirits are present in a place merely “oper-ationally” and not substantially.

86. More would agree that the extension that the mind has differs from the divisible andimpenetrable extension that is found in bodies. However, after Descartes’s death he came toreject the scholastic view, which Descartes himself endorsed, that the soul is present in thebody as “whole in whole and whole in each part.” More came to hold that such a view, whichin his Enchiridion Metaphysicum he labels “Holenmerism” (see More 1995, 98), must bereplaced by the position that the extension of the soul, like the extension of space, has distinctbut “indiscerpible” (i.e., indivisible) parts. For further discussion of the evolution in More’sviews on these issues, see Reid 2007.

87. But cf. the view in Reid 2008 that though Descartes in the end was forced to concedethat God’s activity in the material world requires his substantial spatial presence, on balancethe evidence favors the view that he took the action of finite minds on bodies to require onlyan “operational” presence. For the reasons indicated in the text, I take Descartes’s use of theheaviness analogy and his appeal to an “extension of power” to indicate more than an opera-tional presence. Moreover, whereas Reid sees Descartes’s claim in the More correspondencethat finite minds are not present in a place “in virtue of essence” (15 Apr. 1649, AT 5:343) assupporting the conclusion that they are present merely operationally, I read it as saying onlythat having a location is not essential to such minds, since they could continue to exist eventhough there are no bodies. Nevertheless, I concede that Descartes himself was not careful toclose off an operationalist view of spiritual presence.

Descartes’s admission of mind-to-body action can be reconciled with his principleof the conservation of motion. In contrast to other aspects of his account of thisaction, his views on this issue have no clear scholastic precedent, since this princi-ple has no analogue in the scholastics and in fact was singled out for condemnationby the Jesuits (see §3.1.2 (ii)). In attempting to relate Descartes’s views of causa-tion in psychology to his views of causation in physics, then, we must strike out ona new path.

One place to start is with Leibniz’s various comments regarding Descartes’s viewof the manner in which the soul moves the body. In correspondence with Arnauld,Leibniz discusses the position, which “M. Descartes apparently intends,” that

the soul, or God on its occasion, changes only the direction or determinationof the motion and not the force that is in bodies, because it does not appearprobable that God is constantly transgressing, on the occasion of every act of willby minds, the general law of nature that the same amount of force must continueto exist. (30 Apr. 1687, Leibniz 1978, 2:94/Leibniz 1985a, 117)

In the New Essays concerning Human Understanding, which he finished by 1704,Leibniz attributes to “the Cartesians” the position that “immaterial substances atleast change the direction or determination, if not the force, of bodies” (I.1, Leibniz1978, 6:72/Leibniz 1981, 72). However, in other mature writings, he is not reluctantto attribute this position to Descartes himself. Thus, Leibniz writes in the 1710Theodicy that

Descartes wished . . . to make a part of the body’s action dependent on the soul.He believed in the existence of a rule of Nature to the effect, according to him,that the same quantity of motion is conserved in bodies. He deemed it notpossible that the influence of the soul should violate this law of bodies, but hebelieved that the soul notwithstanding might have the power to change thedirection of the motions that are made in the body; much as a rider, though givingno force to the horse he mounts, nevertheless controls it by guiding that force inany direction it pleases. (I.60, Leibniz 1978, 6:135–36/Leibniz 1985b, 156)88

On the view here, then, not only later Cartesians but also Descartes concluded on thebasis of constraints deriving from his physics that immaterial minds cannot producenew motion in the material world, but can only change the direction of motions thatalready exist.

It is beyond doubt that certain Cartesians accepted the “change of direction”account of voluntary action. There is, for instance, a 1660 letter from Descartes’sliterary executor, Claude Clerselier, appended to the third and last volume ofClerselier’s Lettres de Descartes, which concerns “l’action de l’Ame sur le Corps.”In this letter, Clerselier claims that God alone can “give the first motion to body, orto imprint in it a completely new motion that increases the quantity of that which isalready in the world.” But though no finite mind can add to the quantity of motion,still such a mind “could have the power to change the direction of motion that is inbodies” (Clerselier 1667, 641). Here is a possible source for Lebiniz’s claim in the


88. See also the 1714 Monadology (§80, in Leibniz 1978, 6:620–21/Leibniz 1989, 223).

Theodicy that the soul can have “the power to change the direction of the motionsthat are made in the body.” However, there is even clearer evidence of the influenceof the discussion of mind-to-body action in Of the Body and Soul Conjoined in Man(Corporis et Animæ in Homine Conjunctio), a work of the Dutch-trained CartesianJohann Clauberg.89 In his text, Clauberg claims that the commonsense view that “themind produces certain changes in the body” seems contrary to the fact that “God istruly the efficient, creative, and conserving cause of all motions,” whose action, asimmutable, accounts for the fact that the quantity of motion in the world can beneither increased nor diminished.90 He proposes to eliminate this tension by appeal-ing to the fact that “the human mind is not a physical cause of bodily motions in theman, but only a moral one, since it no more than guides and directs those other[motions], and forms those motions, which already are in the body, by which meansthat part [of the human body] is agitated.” Significantly, given Leibniz’s remarksin the Theodicy, Clauberg continues by comparing the action of the soul on body tothe case where “the driver joins the horse to the wagon, and turns to that place, andthus directs the motion of the wagon that the horse truly produces as the physicalcause” (Clauberg 1968, 1:221). There is thus some reason to think that when heattributes the change-of-direction account to Descartes, Leibniz is thinking not onlyof Clerselier’s appeal to our “power to change the direction of motion,” but also ofClauberg’s position that the human mind merely guides and directs the motion thatGod has already created and conserved.

Whatever the actual source of Leibniz’s understanding of the change-of-direc-tion account, though, there is still the question whether this account can be foundin Descartes. Over a century ago, Norman Smith argued for the negative on thegrounds that

though Descartes frequently speaks of the motion of the “animal spirits” as beingmerely directed (not originated) by the movements of the pineal gland, he never,so far as we know, suggests that those movements of the pineal gland, which areinvolved in voluntary action, can be explained in a similar manner as previouslyexisting and merely guided by the mind. (Smith 1902, 83, n. 2)

Though this judgment has its defenders in the recent literature,91 Peter McLaughlinhas insisted that Leibniz “was basically right both on the historical question of whatDescartes meant to say about conservation and change of direction and on the philo-sophical question of why he had to mean this” (McLaughlin 1993, 157).

On the historical question, the evidence seems to me to favor the Smith readingover McLaughlin’s alternative. In a series of articles in the Passions that concern the


89. This text was originally published in 1664, though Leibniz may have had access to theversion published in 1691 in Clauberg’s posthumous Opera Omnia Philosophica (of whichClauberg 1968 is a reprint).

90. Clauberg here cited theses concerning motion found in his Physica Contracta,§§146–47, in Clauberg 1968, 1:7. Cf. his more complete account of motion in DisputatioPhysica, ch. 13, in Clauberg 1968, 1:97–103.

91. See Remnant 1979 and Garber 1983.

precise nature of this action, Descartes speaks of the soul as “making the glandmove” ( fait que la petite gland . . . se meut) (PS I.41, AT 11:360), as “making thegland lean” ( fait que la gland se penchant) to one side or the other (PS I.42, AT11:360), and as “having the force to make the gland move” (a la force que la glandese meut) in a particular manner (PS I.43, AT 11:361). The final claim that the soulhas “the force” to move the gland is particularly significant given Descartes’s earliercontention to Elisabeth that the notion of the union involves the notion of “the forcethat the soul has to move the body” (AT 3:665). Here, it seems, the soul does notmerely guide motion, but has a force that produces such motion.

McLaughlin cites as evidence for the Leibnizian reading passages in whichDescartes speaks of the mind as merely “determining” the motion of the animal spir-its.92 However, this evidence is far from conclusive. For one thing, it is relevant,given Smith’s comment quoted above, that these passages concern the motion of theanimal spirits rather than the motion of the pineal gland. They therefore seem toleave open the possibility, to which Smith drew attention, that the soul indirectlydetermines this motion by directly moving the pineal gland. Moreover, the referencein Descartes to the force of the soul to move body is significant in light of his viewin a letter to Mersenne, which McLaughlin himself cites, that “force is neededonly to move bodies and not to determine the direction in which they are to move”(11 June 1640, AT 3:75). Whereas McLaughlin concludes that “the Cartesian mindacts on the body in a forceless manner,” since it merely determines its direction(McLaughlin 1993, 163), I think the conclusion must be rather that Descartes takesour mind to have the force needed to move bodies.

It must be admitted, however, that we do not have a clear and final word fromDescartes on the status of mind-to-body action. Though he had a long-standing goalof providing a treatment of such action that relates it to his mechanistic account ofbody, such a goal was never fully realized. In the early Treatise on Man, Descartespromised a treatment of “the body on its own,” “the soul, again on its own,” and“how these two natures would have to be joined and united in order to constitute themen who resemble us” (AT 11:119–20), but in fact considered mainly the humanbody, with some remarks concerning the manner in which “nature” has linked bod-ily motions to sensations. Later, in the Principles, he noted that his third law ofnature, governing changes in motion due to collision, covers only corporeal causesof such changes,93 and that he was reserving consideration of “the power that humanor angelic minds have of moving bodies” (PP II.40, AT 8-1:65) for another treatise.However, this treatise was never written, and so whatever opportunity there was toaddress directly the manner in which mind-to-body action relates to the three lawsof nature and the conservation principle was lost. We do have the Passions, but thereis no consideration there of how the power of the soul to act on the body relates tothe laws that govern body–body interaction. This text merely appeals to the fact that


92. The evidence is from the Meditations and the Description du corps humain; seeMcLaughlin 1993, 166–67.

93. For more on the view that this passage restricts the law to corporeal causes, see chap-ter 3, note 61.

in certain instances, at least, the soul affects the animal spirits by moving the pinealgland.

It might be thought that Descartes provided a hint of his final view when heallowed for the possibility in the passage from the Principles just cited that mind-to-body action is exempt from the third law. Daniel Garber has recently drawn atten-tion to this “escape clause,” as well as to Descartes’s claim in the Principles that hisconservation principle allows for changes in the quantity of motion that “evidentexperience or divine revelation render certain” (PP II.36, AT 8-1:61).94 The refer-ence here to “evident experience” is particularly telling given his comment in the1648 letter to Arnauld that such experience reveals that our soul can move body. Onthe basis of these texts, Garber finds reason to conclude that “Descartes might haveanswered Leibniz’s attack on interactionism by simply denying that the conservationlaws hold for animate bodies” (Garber 1983, 116).

My own sense is that if pressed, Descartes’s initial reaction would indeed havebeen to deny that the conservation principle and related laws governing motionapply to the case of mind-to-body action. Certainly, he gave no signs of a willing-ness to sacrifice the dictate of evident experience that our soul has a real force tomove our body. But though the evidence seems to me to be stacked againstMcLaughlin’s answer to the historical question of whether Descartes embraced thechange-of-direction account of voluntary motion, there remains the philosophicalquestion of whether there are reasons internal to his system for him to accept thisaccount. And with respect to this question, McLaughlin seems to me to have theupper hand.95 Garber has claimed that though Descartes’s conservation principlerequires that God cannot add to the total quantity of motion in the world, it providesno reason to think that finite minds cannot add to this quantity (Garber 1983,115–16).96 But it is far from clear that additions to motion by finite minds leave theconstancy of God’s action untouched. I have emphasized the conservationist viewin Descartes’s physics that God’s ordinary concursus consists simply in the contin-uation of his act of creating particular quantities of matter and motion (see §3.3).Insofar as there are no additions to these quantities, Descartes could say that thisconcursus alone suffices for the conservation of the material world. If finite mindscould add to the quantity of motion, however, then that additional quantity could be


94. McLaughlin takes Garber to task for his assumption that this passages concerns changesin the quantity of motion rather than changes in the distribution of motion (McLaughlin 1993,177, n. 52). However, the exceptions that PP II.36 allows are clearly to the constancy of thequantity of motion. The distribution of the motions is not at issue in this article.

95. Even so, McLaughlin’s insistence on the point that the material world is a closedcausal system (e.g., McLaughlin 1993, 158–59) does not seem to me to be clearly reflected inDescartes’s texts. I think we can do better by focusing on the implications of Descartes’s con-servationist physics.

96. Cf. Remanant’s view that the doctrine of primitive notions in Descartes’s 1643 corre-spondence with Elisabeth suggests that animate bodies are governed by a set of laws distinctfrom those governing inanimate bodies (see Remnant 1979, 382–84). As Garber notes, how-ever, it cannot be the case that animate bodies are exempt from all laws governing inanimatebodies, including those governing the geometrical properties (Garber 1983, 118–19).

retained only if God supplemented his initial act of creation with additional actsthat yield the conservation of this new quantity.97 But then it would seem thatDescartes could not say, as he apparently wants to say in the Principles, that God’sordinary concursus consists simply in the continuation of his initial act of creation.We have seen the suggestion in the Principles that there could be exceptions to theconservation principle that are revealed by “evident experience,” that is, experienceof mind-to-body action, or “divine revelation,” and in particular the revelation ofmiracles. But the exceptions that experience reveals would be a greater threat toDescartes than the miraculous exceptions that divine revelation requires, insofar asonly the latter are supposed to leave God’s ordinary concursus untouched.

Precisely this threat is reflected in the claim in the passage above from Leibniz’sletter to Arnauld that “it does not appear probable that God is constantly transgress-ing, on the occasion of every act of will by minds, the general law of nature that thesame amount of force must continue to exist” (Leibniz 1978, 2:94/Leibniz 1985a,118). A world in which finite minds could add to the quantity of motion over timewould be a world that God cannot conserve by his ordinary concursus alone, at leaston the account of that concursus that Descartes provided in the Principles. In such aworld there would indeed be a constant transgression of the immutability of God’sconservation of the world by means of the same act with which he created it.

In requiring such a transgression, the case of the addition of motion by finiteminds is similar to the problematic case of the “force for resisting” in Descartes’sphysics (see §§3.1.2 (iii) and 3.3). I have argued that this case is problematic becauseit requires not just the conservation of a constant quantity, but also the addition incollision of an instantaneous impulse to resist motion. It is difficult to see how theintroduction of these momentary resistive impulses could be said to derive simplyfrom the divine concursus that Descartes identifies with God’s continuation of hisoriginal act of creating a world with a particular quantity of matter in motion. But inthe same way, it is difficult to see how such a concursus alone could yield the con-servation of any new motion that finite minds produce.

In contrast, any changes in the direction of motions that finite minds introducewould not threaten the conservation principle. For as Descartes makes clear, thatprinciple concerns only the total quantity of motion and not the direction of the motions.So considerations drawn from the metaphysics of Descartes’s physics ultimately doseem to me to support the change-of-direction account of voluntary motion, just as


97. Garber notes that Descartes’s claim in a letter to More that God’s motive force con-sists in his conserving “the same amount of transference in matter as he put into it in the firstmoment of creation” (Aug. 1641, AT 5:404) seems to suggest the “strange position” that Godfails to conserve any additional motion that finite minds add (Garber 1983, 132, n. 70).Garber’s conclusion is that we should not read this remark so literally, but should rather takeDescartes to indicate only that God conserves just the amount of motion that existed duringthe previous moment. I find this reading of Descartes’s remarks to More to be strained, andwould suggest as an alternative that he simply carries over his view in the Principles of God’sordinary concursus without realizing the complications required for the admission of addi-tions to the quantity of motion provided by finite minds.

Leibniz suggested. McLaughlin concludes from this fact that we should attribute toDescartes “the strong philosophical position attributed to him by Leibniz, which isat least compatible with the texts” (McLaughlin 1993, 182). But given that the posi-tion Leibniz attributes to Descartes in fact is not compatible with the texts, a betterconclusion would seem to be that Descartes never quite got around to thinking hisaccount of mind-to-body action through to its foundations. Had he done so, he couldhave discovered the need to reconcile such action with the sufficiency of the divineconcursus that his conservationist physics requires.

Descartes’s conservation principle need not create similar problems for hisaccount of body-to-mind action. It is not necessary that bodies gain or lose motionin order to trigger the activity of the innate mental faculty that Descartes posited inthe Comments (see §4.2.2). Rather, the nature of this faculty is such that the merepresence of these motions affects it in such a way that it forms particular sensoryideas. The fact that Descartes feels compelled to posit a mental faculty and to speakof the motions as “giving occasion” for this faculty to produce the ideas draws atten-tion to complications for body-to-mind action that are not present in the case ofmind-to-body action. But such complications do not derive from Descartes’s con-servationist physics.

There is another case of casual action that the constraints of his physics leavesuntouched, namely, the mind’s causation of its own states. In the Passions, theseactions are said to result in volitions that we experience “as proceeding directly fromour soul and as seeming to depend on it alone” (PS I.17, AT 11:342). Though, as itstitle indicates, this text focuses on the passions of the soul rather than its actions,an account of volition is nonetheless crucial for its central teaching that we shouldrestrict our desires to what depends only on our free will (see, e.g., PS I.144, AT11:436–37). Since the production of volition does not itself involve any gain or lossof motion, there is no threat here to the divine concursus that yields a constant quan-tity of motion in the material world. However, Descartes’s insistence that we freelyproduce our own volitions may seem to threaten his acceptance of a strong formof divine providence. Indeed, he himself notes this tension when he claims in thePrinciples that it is beyond our power to reconcile our freedom with the fact that“the power of God, by which he not only foreknew from eternity all that is or canbe, but also willed and preordained it, is infinite” (PP I.40, AT 8-1:20). Of course,Descartes was not the first to worry about the compatibility of divine preordinationwith human freedom. As we will discover in the following chapter, scholastic dis-putes concerning this issue are germane to his own struggles with it. But the refer-ence in the passage from the Principles to the divine preordination of what can bebroaches what is perhaps the most idiosyncratic feature of Descartes’s theory of cau-sation, namely, his famous (and notorious) doctrine of the creation of the eternaltruths. As we will discover, this doctrine plays an important though subtle role in hisaccount of the relation of our freedom to divine providence.


178

5

Causation and Freedom

In an entry from a notebook that he composed as a young soldier traveling throughEurope, Descartes observes that “the Lord has made three miracles: something fromnothing, free decision, and God in man” (Tria mirabilia fecit Dominus: res ex nihilo,liberum arbitrium, et hominem deum) (AT 10:218). The first two of these miraclesare directly relevant to his theory of causation.1

The miracle of “something from nothing” of course recalls Descartes’s conclu-sion that the act by which God conserves the world is not distinct from the act bywhich he creates it ex nihilo (see §2.2). But this chapter focuses on the miracle offree decision, or more specifically the freedom of our will, which as I hope to showis an important element of Descartes’s theory of causation.

To understand the connection between the issues of causation and human free-dom in Descartes, it is helpful once again to consider the scholastic context. Towardthe end of §1.2.3 (ii), I noted the objection to Suárez, deriving from Durandus, thatthe concurrentist conclusion that God is the “proximate and immediate cause” of oursinful action conflicts with the claim that such action is free. As we know, theresponse in Suárez is that God’s concursus with our free action is compatible withour ability—essential for our freedom—to do otherwise even given that all the con-ditions for action have been posited. Though our will determines such action, forSuárez, there is nothing, including the divine concursus, that determines the will todo so.

1. Though the last mystery of “God in man” is less obviously relevant to Descartes’s the-ory of causation, it bears some relation to the union of mind with body, for in both cases it isthe unity of a composite of parts with diverse natures that is at issue (see §4.1).

In §5.1, I begin with a consideration of two problems that confront Suárez’saccount of human freedom. The first concerns the relation between the will andjudgment. Suárez’s view of judgment presupposes a scholastic account of the men-tal faculties that differs in fundamental respects from the account that Descartes laterdefended. Nevertheless, he considers an objection to his account of free will that isrelevant to Descartes’s treatment of freedom, namely, the objection that our will doesnot seem to have freedom of indifference given that it is determined by our intellect.Suárez’s response to this objection turns on the claim that in this life, at least, anysort of intellectual determination depends on a prior act of our will that is itself unde-termined. The second problem for Suárez’s account of human freedom concerns thecompatibility of his insistence on the indifference of free human action with the tra-ditional Catholic doctrine that God exercises complete providential control over cre-ation. This problem is connected to a heated theological dispute in the late sixteenthand early seventeenth centuries that pitted Jesuits against members of the Dominicanorder. Suárez attempts to defend the Jesuit line by developing the view in the workof his contemporary Luis de Molina that divine providence is made possible by aspecial “middle knowledge” of undetermined human action. I argue in this chapterthat ultimately we must place Descartes with Suárez on the Jesuit side of this debate.However, I also claim that Descartes must be distinguished from Suárez and otherMolinists insofar as his account of freedom is informed by his distinctive but diffi-cult doctrine of the creation of the eternal truths. I close this first section with a con-sideration of Descartes’s account of created truth, with particular attention to thecomplications it introduces for his theory of causation in general and for his treat-ment of human freedom in particular.

I then turn in §5.2 to the issue of Descartes’s somewhat conflicted attitude towardthe Jesuit position in Suárez that human freedom requires “indifference” in action. Inthe Fourth Meditation, Descartes not only offers an account of judgment very differ-ent from what we find in Suárez, but also indicates that the will is most free when itis determined in a manner that wholly excludes indifference. Nevertheless, there isreason to think that even in this case Descartes accepts the view in Suárez thatperceptions are not efficient causes of the action of the will. Moreover, in both thePrinciples (1644) and correspondence with Jesuits dating from the mid-1640s, there issome movement toward Suárez’s position that free human action precludes any sort ofthoroughgoing intellectual determination. These texts still have a non-scholasticemphasis on the role of free will in assent to truth. However, the role of free will in thepursuit of the good—which is more familiar from Suárez—is prominent in the accountof human freedom that Descartes provides in his last published work, the Passions ofthe Soul.

One reason that scholars have taken Descartes to be a “compatibilist” with regardto free human action is that some of his statements seem to indicate a commitment tothe position that God fully determines everything in nature, including our free action.If, as I argue in this chapter, his final account of human freedom is fundamentally“incompatibilist” in nature, we need to rethink his account of divine providence and itsrelation to free human action. In §5.3, I offer an interpretation of this account thatdraws on Suárez’s claim that God’s middle knowledge of our free action derives fromhis comprehensive grasp of the dispositions in us from which this action results in a

Causation and Freedom 179

nondeterministic manner. The corresponding position that I attribute to Descartes isthat God produces our free action by making it the case that this action derives in anondeterministic manner from the inclinations of our will. For Descartes, then, just asGod can make essences from which truths derive necessarily, so he can produce incli-nations from which our actions derive predictably but nondeterministically. The impli-cation here that God determines free human action only indirectly, by producing aspecial sort of volitional inclination, ultimately serves to reinforce the conservationistreading of Descartes’s theory of causation that I have been concerned to defend.

5.1. JESUIT FREEDOM AND CREATED TRUTH

5.1.1. Suárez on Human Freedom

Suárez’s views on human freedom belong to what can be called the “voluntarist” tra-dition in scholasticism. This sort of voluntarism must be distinguished from the “vol-untarist axiom” that I discussed earlier with respect to Suárez’s renovation ofscholastic metaphysics (see §1.2.1). Recall that that axiom affirmed the power ofGod to produce in separation really distinct creatures. But Bonnie Kent has notedthat in the context of later-medieval scholasticism, “ethical voluntarism” indicates“a strong emphasis on the active character of the will, the claim that the will is freeto act against reason’s dictates, and the conviction that moral responsibility dependson this conception of the will’s freedom.”2 In contrast to the case of the voluntaristaxiom, ethical voluntarism is concerned primarily with the nature of free humanaction, and not with the nature of divine power.

Nevertheless, the two kinds of voluntarism are rooted in the Paris Condemnationof 1277. As we have seen, the voluntarist axiom derives from the rejection in thisproclamation of certain Thomistic restrictions on divine power.3 However, theCondemnation includes a section of twenty articles on the topic of the human will.The overall concern in this section is to condemn the position that the will is subor-dinated to reason. For instance, this section prohibits the following article: “Thatthe will necessarily pursues what is firmly believed by reason and that it cannotabstain from that which reason dictates. This necessitation, however, is not compul-sion but the nature of the will” (Lerner and Mahdi 1995, 350). The “intellectualist”view here that the will is by its nature subservient to reason was popular among cer-tain Thomists. Just as Suárez’s acceptance of the voluntarist axiom set him against ahard-line Thomist ontology, then, so his emphasis on the importance of indifferencefor human freedom set him against a Thomistic form of intellectualism.4


2. Kent 1995, 94–96, citing the characterization of voluntarism in Bourke 1970, 1:138, 147.For more on discussions of human freedom in later medieval philosophy, see Eardley 2006.

3. As indicated in chapter 1, note 35.4. As indicated in chapter 1, notes 35 and 37, Cajetan is one of the main Thomistic oppo-

nents of the voluntarist axiom. Among the Thomists that Suárez cites as opposed to hisaccount of freedom is the Dominican Chrysostom Javellus (1470–1538) (MD XIX.6, ¶3, 720).

There is an issue here whether Thomas himself was as much of an intellectualistas some of his followers claimed. In fact, Suárez insisted that his voluntarist positionwas for the most part in line with Thomas’s own position.5 But the focus here is noton Suárez’s Thomas scholarship, but rather on the manner in which he responded tothe account of human freedom that some Thomists claimed—rightly or wrongly—to derive from Thomas’s writings. I indicated at the outset of this chapter two fea-tures of this account that Suárez, in line with other Jesuits, took to be problematic.The first concerns the intellectualist view that the will is determined by intellectualjudgment. On this point, Suárez held that our will has a sufficient amount of indif-ference to act against such judgment. The second concerns the claim that divineprovidence requires a knowledge of free human action that is made possible byGod’s predetermination of such action by means of “physical premotion.” Againstthis claim, Suárez insisted that divine providence involves a middle knowledgeof this action that does not compromise its undetermined nature. Let us consider inturn these elements of Suárez’s response to the Thomists.

(i) Judgment and Indifference

Suárez and his Thomist opponents shared an Aristotelian conception of the will asrational appetite. Qua rational, the will is a power of the intellectual soul rather than,as in the case of sense appetite, the soul-body composite. However, both the will andthe sense appetite, qua appetite, are directed toward the good as presented eitherby the senses or by the intellect. In contrast, the senses and intellect are apprehen-sive powers directed toward the true, with the senses being powers of the compositeand intellect a power of the soul alone. As opposed to the view that Descartes laterdefended, most scholastics, including Suárez, simply assumed that judgment con-cerns truth and, as such, is an operation of an apprehensive faculty. Whereas sensoryjudgment is an operation of the senses that involves the apprehension of sensible fea-tures of objects, rational judgment is an operation of the intellect that involves theapprehension of intelligible features of objects.

Beyond these basic points, however, there was considerable controversy withinscholasticism concerning the relation of the will to rational judgment and the rela-tion of both to “free decision” (liberum arbitrium). Whereas intellectualists under-stood the conception of the will as rational appetite to indicate that the will issubordinate to rational judgment and cannot act contrary to it, Suárez follows othervoluntarists in insisting that the will is more noble than and thus can control the


5. For Suárez’s invocation of Thomas, see, among many other places, MD XIX.8, ¶10,1:729. As Kent notes, however, late-thirteenth-century masters who defended a voluntaristaccount of the will, such as William de la Mare (†c. 1290) and Walter of Bruges (†1307), didso in explicit opposition to Thomas (Kent 1995, ch. 3). The question of whether Thomas wasmore of an intellectualist or a voluntarist is still a matter of dispute today; cf. Pasnau 2002,220–33, which defends a primarily intellectualist (and compatibilist) reading of Thomas’saccount of freedom; and Stump 1997, which defends a primarily voluntarist (and incompati-bilist) interpretation of this account.

intellect. In contrast to those who held that free decision is rooted in the intellect,moreover, he argues that the intellect is wholly determined to assent to the true anddissent from the false, and so cannot be the source of freedom. Since the will isdirected to the good of the object, however, and since some objects can be both goodand evil in different respects, there can be some indifference on the part of the willwith respect to whether to pursue or avoid such objects. The conclusion here is thatfree decision is rooted in the will rather than the intellect (MD XIX.5, ¶14, 1:716).6

Suárez notes that there are two sorts of necessity with respect to acts of will. First,there is the necessity that involves coercion, which is to be contrasted with uncoercedand voluntary acts of will. But there can be a sort of necessity even with respect tovoluntary acts in cases where the goodness of the object that the intellect presentsdetermines the will to act in a particular manner. Suárez holds, for instance, that thewill of the blessed is determined by the beatific vision to love God (MD XIX.8, ¶8,1:728–29).7 Moreover, he grants that even in this life, our will is not able to will con-trary to an end proposed under the concept of a universal good; in his terms, this enddetermines the will with respect to the specification of its act (¶15, 1:730). Yet Suárezinsists that though the will acts voluntarily in these cases, it lacks the freedom thatprecludes any determination to a particular act (MD XIX.2, ¶9, 1:695). This robustsort of freedom, which adds “indifference” in action to voluntariness, is required forthe justice of reward and punishment for our actions (¶16, 1:698).

It may seem that given his emphasis on the value of the freedom that involvesindifference, Suárez would claim that the will is less perfect when the intellect deter-mines its voluntary act of will than when the will itself determines its voluntary actin a manner that involves freedom of indifference. However, he in fact says that theperfection of the acts of will is determined by the rule that

an object be loved in proportion to its worthiness or capacity; so that if a givenobject is the highest and most necessary good, then it is loved with completenecessity, whereas if it is a lesser or non-necessary good, then, correspondingly,an act with respect to it is under the control of the one who loves it. (MD XIX.8,¶21, 1:732)

When the will is not determined to pursue nonnecessary goods, it has a perfection ofcontrol over action that allows for merit. But even though this perfection is lackingboth in the case of the blessed who love God necessarily and in our own case whenwe are unable to act contrary to the universal good, still there is in these cases theperfection of the will that consists in its determination to the good.

Nevertheless, Suárez holds that in this life, at least, the perfection of the will con-sists more in the perfection of control than in the perfection of determination to thegood. Even in the case of the end of the universal good, where our will lacks free-dom of specification, he insists that we are free to refrain from acting altogether, and


6. In taking free decision to be a power of the will, Suárez opposes not only those whotake free decision to reside in something external to the will, but also those who identify itwith an act or habit rather than with a faculty; see MD XIX.5, ¶4, 1:712.

7. As Suárez indicates in this text, he sides with Thomas against Scotus on this point.

so retain freedom of exercise. And in the case of all particular goods that our intel-lect naturally apprehends in this life, including God, our will has both freedom ofexercise and freedom of specification. All such objects can be considered as evil orinconvenient, and thus as something that the will can refrain from loving or can evenhate (¶16, 1:731). Though judgment concerning the goodness of particular objectscan move the will as a final cause, the fact that it does not determine either the exer-cise of the will or the specification of its act reveals that it is not the efficient causeof the decision of the will.8

Cases where judgments seem to be efficient causes of free decision are to beexplained by the volition to pursue particular goods as ends or means (¶8, 1:722).Suppose, for instance, that I resolve to satisfy my craving for sweets. Given this res-olution, the judgment that particular objects are conducive to this end compels meto pursue those objects. However, the efficacy of the judgment is conditional on myresolution. If I withdraw the resolution (due, for instance, to concerns about myweight), then the judgments will not necessarily lead me to pursue the objects theypresent. Since, according to Suárez, I as a free agent have some indifference to allresolutions concerning particular objects, and thus can withdraw such resolutions,there is no judgment concerning such objects that can, by itself, determine my will.

Suárez’s concern with the pursuit of particular goods distinguishes his discussionof human freedom from the remarks on this topic in Descartes, which, as we will see,tend to focus on the case of assent to particular truths.9 However, there is one aspectof Suárez’s discussion that will prove useful in understanding certain developmentsin Descartes’s treatment of human freedom. I have in mind Suárez’s view of how itis that the will can resist those judgments that lead it to pursue particular goods. Henotes that with respect to such judgments,

the will is able either to suspend its act or to divert the intellect from thinkingabout the object; or to apply [the intellect] to investigating more carefully,regarding that object, how much goodness it has and whether it has conjoined to itsome evil or inappropriateness in light of which the will is able not only not tolove it but even to hate it. (MD XIX.6, ¶14, 1:724)

Thus, in cases where I freely decide to pursue sweets, I could have resisted the judg-ment that sweets are good insofar as they satisfy my cravings for them by distract-ing myself from thinking about sweets or by considering the goods that derive from


8. Cf. the discussion in §1.2.2 (iii) of Suárez’s account of final causality. There is somequestion whether Suárez allowed that judgment is an efficient cause of volition in the casewhere, as in the case of the blessed, it determines the will with respect to both exercise andspecification. His claim that there is no reason for the cognition of a good object to be an effi-cient cause, since appetite is moved toward that object by its own nature and not by any exter-nal agent (MD XIX.18, ¶49, 1:646), may seem to reveal that he did not. To be sure, the passageindicates that the goodness involved here is not such that it determines the will. As I indicatein §5.2.1 at note 40, however, Suarez indicates that the intellect is not an efficient cause of thewill even in the case of cognition of a necessary good.

9. As we will also discover in §5.2.3, though, Descartes’s treatment of freedom in thePassions pushes to the forefront the case of the pursuit of the good.

resisting sweets. Given this view in Suárez, it is not too surprising that the emphasison the ability of the will to distract the intellect reappears in Descartes at just thepoint where he is attempting to accommodate the Jesuit account of human freedom(see §5.2.2).

(ii) Divine Providence and Middle Knowledge

In 1597 Pope Clement VIII established a Congregation on Grace (Congregatio deAuxiliis) to review the Concordia of the Jesuit Molina.10 The publication of this workignited a controversy that pitted members of the recently formed (in 1540) Society ofJesus against members of Thomas’s established Dominican order, most prominentlyDomingo Bañez (1528–1604). Most objectionable to Bañez and other Dominicanswas the denial in the Concordia that God determines free human action. For Molina’scritics, this denial threatens divine providence. When Molina died in 1600, rumor hadit that the Congregation was on the verge of declaring his views to be heretical.However, due in part to the reconciliationist efforts of the Jesuit Cardinal RobertBellarmine (1542–1621), the Congregation ultimately ended in a draw, and Pope PaulV issued a decree in 1607 that forbids the two sides from labeling the views of theiropponents heretical or even temerarious. This decree promised an official resolutionof the dispute at “an opportune time,” which time we still await.11

As a result of the 1607 decree, it is theologically acceptable for Catholics to endorseeither the Molinist views in the Concordia or the opposing views of Molina’sDominican critics. Both sides in fact agree that God creates in a providential manner,and that such providence requires that prior to the divine decision to create a particularworld God knows all future contingent propositions concerning that world, includingthose propositions concerning free human actions in that world. The dispute principallyconcerns the explanation of how God has knowledge of the latter sort of proposition.According to Dominicans such as Bañez, God is able to know in advance how humanagents would freely act in certain circumstances because his “efficacious concursus”determines those free actions that are good, whereas his failure to offer such a concur-sus results in those free actions that are evil. For Molina, however, this account of fore-knowledge is unacceptable insofar as it involves the determination of free human actionby the granting or withholding of an efficacious concursus. On the view that Molinashared with Suárez, our actions can be free only if they involve an indifference thatis incompatible with this sort of determination. Thus, God’s foreknowledge of suchactions cannot be grounded in his free decrees concerning his own concursus. Rather,such knowledge must derive from God’s simply seeing that free agents would decide toact in certain ways in certain circumstances, even though such action is determined nei-ther by the circumstances nor by the causal contributions of the divine will.12


10. For the full title of the Concordia, see chapter 1, note 71.11. For an account of the Congregation and of Bellarmine’s role in it, see Brodrick 1961.12. For a summary that is sympathetic to the Molinist side of this dispute, see the editor’s

introduction to Molina 1988. For a discussion more sympathetic to the Dominican side, seeOsborne 2006.

This is the Molinist theory of “middle knowledge” (scientia media), so calledbecause it posits a sort of divine knowledge of free human action that stands betweenGod’s “natural” knowledge of necessary features of creation and his “free” knowl-edge of contingent features of the created world that his will determines. The theoryof middle knowledge was theologically objectionable for Dominicans because itseemed to render the goodness of our action independent of God’s providentialaction. But there is also the philosophical objection that Molina has provided noground for the truth of the propositions that God foreknows concerning free humanaction. Molina must reject the claim that knowledge of God’s concursus provides thefoundation, for the reasons just indicated. But the free agents themselves cannot pro-vide the foundation, since middle knowledge is logically prior to God’s creation ofa world containing such agents. Thus, there seems in the case of middle knowledgeto be truths without truthmakers.13

Molina attempted to explain God’s ability to know truths concerning free humanaction by appealing to the “absolutely profound and absolutely pre-eminent com-prehension, such as is found only in God with respect to creatures” (Molina 1988,Disp. 52, §11, 171). However, this theory of “supercomprehension,” as it has cometo be known, was not satisfactory even for Molina’s Jesuit defenders. Most notably,Suárez rejects this theory. In his On the Knowledge That God Has of FutureContingents (De Scientia quam Deus habet de Futuris Contingentibus), he objectsthat the appeal to the fact that God “supercomprehends” (supercomprehendit) anobject cannot explain God’s knowledge of free human action, since what providesthe foundation for such knowledge is not the perfection of the mode of knowledgebut rather the perfection of the object of God’s knowledge (perfectio in cognitionesolum . . . ex parte rei cognitae). The fact that God knows the free agent perfectly isthus to be explained by an appeal to the fact that the object of this knowledge revealsall truths concerning the free action of this agent (Opera 11:366).14

What, then, is the object that explains middle knowledge? Suárez’s most explicitanswer can be found not in On Knowledge but rather in his Treatise on Divine Grace(Tractatus de Gratia Dei).15 In the Treatise he appeals to God’s intuitive knowledgeof the “immediate disposition of the cause to a future effect” (per immediatem habi-tudinem causae ad effectum ex hypothesi futurum) (Opera 7:94). In particular, theobject is the habituo, or disposition, of a possible agent that would yield a particulareffect in certain circumstances. Though it is compatible with the nature of this


13. For a contemporary formulation of this philosophical objection to middle knowledge,see Adams 1990, 114–15.

14. Even so, someone like Suárez might want to retain the view in Molina that God canhave this sort of complete knowledge only of objects he infinitely surpasses to explain the factthat he does not have middle knowledge of his own free actions. For this point, see Freddoso’sremarks in the introduction to Molina 1988, 51–53.

15. Though the claim in On Knowledge that God knows conditional contingent proposi-tions concerning free human action “by the infinite power of representation of his ideas” (exvi infinitae repraesentationis suarum idearum) (Opera 11:369) anticipates the position in theTreatise.

disposition that it not yield such an effect, the idea that God has of it allows him toknow that the disposition would in fact lead the agent to act in this manner.16

Calvin Normore has helpfully explained Suárez’s position here in terms of the claimthat God has a “perfect model” of each possible being that allows him to simulate pos-sible histories to determine how a particular agent A would behave in the simulated cir-cumstances C. Normore notes that “if there is a way in which A would behave in C, aperfect model should reflect it, so if the conditional excluded middle is valid, sucha model is possible and God knows the history of the world by knowing that model,i.e., by knowing his own intellect and his creative intentions” (Normore 1985, 15). Ofcourse, the Bañezian critic could reply that prior to God’s decree concerning his ownconcursus, there is no fact of the matter as to how A would behave in C, and so noth-ing for the perfect model to capture. In effect, then, this response involves a rejectionof the conditional excluded middle.17 But assuming that there is a fact of the matterhere, and thus that the conditional excluded middle holds, it seems that God couldknow it by intuiting the results of his simulation of the history of the agent.

Molina and Suárez appealed to the theory of middle knowledge to overcome theapparent conflict of claims concerning divine providence and foreknowledge withtheir view that our actions are free in a sense that precludes their determination. A sim-ilar sort of conflict seems to resurface in Descartes’s Principles. Toward the end ofchapter 4, I noted a worry in this text concerning divine preordination of free humanaction. This problem derives from the claim, in the title of article 40 of the first partof this text, that “everything is preordained by God.” In the article itself, Descartesnotes that “we can easily entangle ourselves in great difficulties if we attempt to rec-oncile this preordination of God with our free decision [arbitrii nostri libertate], andto comprehend both together” (AT 8-1:20). In the following article, he claims that wecan avoid this conflict by remembering that our finite mind cannot comprehend theinfinite power by which God “not only knew from eternity what is or can be [quae sunaut esse possunt], but also willed and preordained it.” In particular, we cannot com-prehend how such a power “leaves free human actions undetermined [indetermi-natas].” However, Descartes concludes that our inability to comprehend does notpreclude us from accepting “the freedom and indifference [libertatis . . . et indifferen-tiae] that is in us,” since “we are so conscious [conscios] of this freedom that there isnothing that we comprehend more evidently and perfectly” (PP I.41, AT 8-1:20).

Certainly the reference here to the undetermined nature of our “freedom and indif-ference” is significant in light of the emphasis in the work of Suárez and other Jesuitson the fact that human freedom requires an indifference that precludes determination.I have more to say presently concerning the relation of these remarks in the Principlesto the Jesuit position (see §5.2.2 (i)). For the moment, however, I want to consider thefact that the conflict with which Descartes is concerned in his text differs in some sig-nificant respects from the conflict that Molina and Suárez addressed. These Jesuit the-ologians were concerned only with the question of how, given the fact that our free


16. For the connection between middle knowledge and divine ideas, see the passage fromOn Knowledge cited in note 15.

17. For a similar objection, see Adams 1990.

actions are undetermined, God could know conditional propositions concerning theseactions prior to willing to create in a particular manner. In contrast, Descartes findsproblematic not only that God foreknew these actions but also “willed and preor-dained” them. Of course, someone like Suárez could hold that God wills and pre-ordains such actions in the sense that he intentionally creates that world which he(eternally) foreknew would include the actions as part of its history. Yet there is noprecedent in the work of Suárez and the other Jesuits for Descartes’s claim that God’sinfinite power produces all that not only is but also “can be.”18 For the Jesuits, Godknows what can be by means of his natural knowledge of necessary features of theworld. Such knowledge derives not from his will but rather from ideas in his intellect.In his 1630 letter to Mersenne, however, Descartes introduced the doctrine of the cre-ation of the eternal truths, according to which such truths “have been established byGod and depend as entirely on him as the rest of his creatures” (15 Apr. 1630, AT1:145). In a subsequent letter he told his correspondent that God established thesetruths “as efficient and total cause,” and that “he is the Author of the essence as wellas of the existence of creatures” (27 May 1630, AT 1:152; italicized text in Latin). Forthe Jesuits, God’s middle knowledge of our free action is akin to his natural knowledgeof the essences of creatures in being independent of any determination by his will. Butgiven Descartes’s created truth doctrine (as I call it for short), there can be no naturalknowledge, much less middle knowledge, of creatures. All truths concerning creaturesare subject to God’s will, and so even truths concerning our free action are so subject.

The mystery in the Jesuits of how God could have prevolitional knowledge of ourundetermined action is therefore replaced by the mystery in Descartes of how truthsconcerning such action could be founded ultimately in a divine will that is the efficientcause of all truths concerning creation. The mystery in Descartes would appear to bemore intractable insofar as it seems contradictory to say that the divine will determinestruths concerning free actions that are themselves undetermined. One easy way outwould be to claim that since the created truth doctrine allows that God can do the impos-sible, it raises no additional problems for the view that God brings about the impossi-bility of determining undetermined action. In the course of this chapter, however, I hopeto show that this doctrine yields a more sophisticated response to the problem of God’sdetermination of free action, one that can accommodate certain aspects of the Jesuit the-ory of middle knowledge that distinguish it from its Dominican competitors. To set thestage for this argument, I consider Descartes’s created truth doctrine. However, I beginwith questions concerning the scope of the doctrine that broach certain complicationsfor his theory of causation that are not restricted to the special case of free human action.

5.1.2. Descartes on Created Truth

When Descartes launched his created truth doctrine in 1630, he mentioned to Mersenneonly “the mathematical truths you call eternal” (15 Apr. 1630, AT 1:145). However, inthe 1641 Sixth Replies, he indicated that the scope of this doctrine extends far beyond


18. Cf. the claim earlier in the Principles that “God alone is the true cause of all that is orcan be [sunt aut esse possunt]” (PP I.24, AT 8-1:14).

mathematical truths, and indeed seems to be without limit, since “no good or truth,nothing worthy of belief or action or omission, can be feigned the idea of which is inthe divine intellect before his will determines to bring it about that it be such” (AT7:432). Similarly, in a 1648 letter Descartes told Arnauld in an unqualified way that“every basis of truth and goodness depends on [divine] omnipotence” (AT 5:224).

In a groundbreaking discussion, Frankfurt has argued that Descartes’s createdtruth doctrine in fact has universal scope (Frankfurt 1977, 40). He has furtherclaimed that Descartes understood this doctrine to entail a “universal possibilism” onwhich eternal truths, including truths concerning God, are “inherently as contingentas” or “no more necessary than” any other propositions (42). In Frankfurt’s view,Descartes took the apparent necessity of these truths “properly to be understood onlyas relative to the character of our minds” (45). God has created our minds such thatwe perceive the eternal truths to be undeniable, but he could just as easily have cre-ated us and the world and even himself differently.

Jonathan Bennett has proposed an alternative to Frankfurt’s interpretation thatnonetheless takes all eternal truths to fall under the scope of the created truth doc-trine. Bennett offers on Descartes’s behalf a “conceptualist” account on which thenecessity of the eternal truths is to be analyzed in terms of our own mental capaci-ties. To say that a truth is necessary is simply to say that we cannot distinctly con-ceive the opposite of such a truth (Bennett 1994, 647). Thus, the created truthdoctrine amounts to the position that God has created our minds such that we can-not conceive the opposite of any eternal truth. As evidence for this reading of the cre-ated truth doctrine, Bennett appeals, among other passages, to Descartes’s remark incorrespondence with Arnauld that “I do not think we should say of anything that itcannot be brought about by God,” but “I merely say that he has given me such a mindthat I cannot conceive” the opposite of the eternal truths, since “such things involvea contradiction in my conception” (29 July 1648, AT 5:224).19

Bennett clearly differs from Frankfurt insofar as he rejects the attribution toDescartes of any sort of universal possibilism. Indeed, in Bennett’s view Descartesaccepted rather the opposite; all eternal truths are necessary in a conceptualist sense.In one respect, however, his reading is close to Frankfurt’s. For both, the perceivednecessity of the truths is tied to the constitution of our minds. The difference is sim-ply that Frankfurt takes this sort of perceived necessity to be merely apparent,whereas Bennett takes it to be that in which Descartes’s necessity consists.

However, the implication here that even truths concerning God reduce to a cre-ated feature of our mind is difficult to square with certain claims in Descartes. Mostnotable is his insistence in the Fifth Meditation, with respect to the truth that essenceand existence are inseparable in God, that “it is not my thought that effects [efficiat]this, or imposes any necessity on anything, but on the contrary it is the necessityof the thing itself, namely, the existence of God, that necessarily determines me tothinking this” (AT 7:67). Here it is something outside of my thought—namely, God’sessence—that provides the reason for God’s existence and thus grounds the deter-mination of my thought to consider his existence as necessary.


19. Bennett 1994, 656–61.

In other writings Descartes accords eternal truths concerning God a special statusthat they cannot have on the interpretations Frankfurt and Bennett have offered. Thus,he insists in his 1630 correspondence with Mersenne that “the existence of God is thefirst and the most eternal of all possible truths and the one from which alone all othersproceed” (6 May 1630, AT 1:150). Moreover, he closes out his initial discussion of thecreated truth doctrine in this correspondence by noting that the “essence of createdthings” is nothing other than eternal truths that are “no more necessarily attached to[God’s] essence than are other created things” (27 May 1630, AT 1:152–53). Giventhese remarks, the created truth doctrine would seem to be restricted to whatever is notnecessarily attached to God’s essence, and thus not include those truths concerning thisessence that surely are so attached. Even if the necessity of created eternal truths isbound up with the manner in which God creates our mind, the suggestion in these pas-sages is that the necessity of truths concerning God’s essence does not derive from hiswill, but rather serves as the uncreated ground for the necessity of created eternal truth.

The claim that Descartes allowed at least in certain passages for uncreated truthsconcerning God’s own essence is not new.20 What I think commentators have failedto appreciate, however, are the complications this sort of restriction on the createdtruth doctrine introduces for Descartes’s theory of causation. The first complicationconcerns the issue of the cause of God’s existence. Earlier I considered Descartes’sexchanges with his critics concerning his suggestion in the Third Meditation that Godderives his existence from himself. I noted Descartes’s conclusion that God’s natureserves as a “formal” rather than an efficient cause of his existence (see §2.1.2 (ii.a)).This special treatment of the case of God’s existence is relevant also to the causalaxiom—which he introduces in the Second Replies—that “no thing exists of which itcannot be asked what is the cause why it exists.” Anticipating the objection that nocause is required in the case of God’s existence, Descartes adds that “this can be askedeven of God himself, not because he needs any cause in order to exist, but becausethe immensity of his nature is the cause or reason [causa sive ratio] why he needs nocause to exist” (AT 7:164–65).21 What is significant, with respect to the created truthdoctrine, is the implication here that though the truth that God exists has a reason inthe immensity of the divine nature, this nature does not provide the sort of efficientcausal explanation for this truth that is required in the case of all created truths.

Moreover, the fact that the truth that God exists requires an ultimate reason serves todistinguish it from created eternal truths. For Descartes notes in the Sixth Replies, in apassage cited earlier, that

it is repugnant that the will of God not be indifferent from eternity to all that hasbeen made or will be made, since no good, or truth, or believing, or doing, orrefraining from doing can be feigned [ fingi potest], the idea of which is in thedivine intellect prior to his will determining or making it to be so. I am not speaking here of temporal priority; there is not any priority or order, or nature, or“considered reason” [ratione rationcinata], as it is called, such that this idea ofgood impels God to choose one thing rather than another. (AT 7:431–32)


20. See, for instance, Wells 1982.21. For an extensive discussion of this axiom and its implications, see Carraud 2002, ch. 2.

Given the essential indifference of divine action, there can be no preexisting ideasthat provide reasons for God’s creation of the eternal truths. Not only is there no rea-son for this creation that we can comprehend; there is no reason God can compre-hend either. So whereas the divine nature provides the reason for God’s existence,there is nothing that can provide a reason for the divine production of created truth.

There is some question whether the result of the created truth doctrine that Godcan have no reasons for action is consistent with the implication in Descartes thatGod contains the reality of the creatures he creates. I have noted the requirement ofDescartes’s containment axiom that the “reality or perfection” of an effect be con-tained in its (total or adequate) cause “formally or eminently” (see §2.1.3). ButDescartes himself indicates that this axiom has unrestricted scope, and so requiresthat even in the case of divine creation the effects must be contained eminently inGod. To be sure, the specific account of eminent containment that I attributed toDescartes seems to be consistent with this requirement. Recall that on this account,a cause eminently contains its effect in virtue of possessing the power that sufficesto produce the objective reality that is present in the cause’s idea of that effect (see§2.1.3 (ii)). Surely God has the power to produce the objective reality of any ideashe may have of his effects. The problem, however, is that the reality of the effect thatis contained eminently in finite minds is supposed to provide a basis for its causa-tion of the effect. However, Descartes makes clear that there can be no idea that con-ditions divine creation. Given this position, God contains not the reality of his effects(in contrast to the case of finite minds), but rather a power that is able to produce thereality of these effects ex nihilo (a power totally lacking in finite minds).22

This difference between the divine and created minds introduces a further com-plication for Descartes’s insistence on the similarity of our will to God’s. His claimin the Fourth Meditation is that it is in virtue of his “will or free decision” (voluntassive arbitrii libertas) that “I understand myself to bear some image and likeness ofGod [imaginem quandam et similitudinem Dei]” (AT 7:57).23 The claim that our willis made in the “image and likeness” of the divine will is weakened to a considerableextent by the admission, which follows the passage from the Sixth Replies just cited,


22. In the Use of Reason. Descartes’s later follower Regis (see chapter 1, note 77) argued,on the basis of a version of the created truth doctrine, that

God does not see creatures in his perfections, because it has been proved that theperfections of God have nothing in common with creatures, and by consequence thatthey cannot represent them; we must say only that God sees creatures in his will,insofar as it is by his decrees that he produces and conserves them. (Regis 1996, 169)

Elsewhere in this text, Regis also claimed that God is an “analogous” cause that, in contrast tounivocal and equivocal causes, contains his effects neither formally nor eminently (406–07).For further discussion of Regis’s position and its relation to Descartes, see Schmaltz 2000.

23. Cf. the claim toward the end of the Third Meditation that “from the one fact that God cre-ated me, there is a strong reason to believe that I have been made in some way in his image andlikeness [imaginem et similitudinem], and that I perceive that likeness, which includes the ideaof God, by the same faculty that enables me to perceive myself” (AT 5:51), as well as the com-ment on this claim in the passage from the Conversation with Burman considered in §2.1.2 (i).

that “the indifference that belongs to human freedom is much other [longe alia] than[the indifference] of divine [freedom].” This admission prefaces Descartes’s claimthat whereas human indifference “does not belong to the essence of human free-dom,” since it involves an ignorance of reasons that leads our mind in one directionrather than another, God’s supreme perfection requires that he be indifferent in sucha way that “no good or truth . . . can be feigned the idea of which is in the divineintellect prior to his will determining or making it to be so” (AT 7:432–33). Butgiven this admission, it seems that Descartes is not warranted in claiming, as he doesin the Fourth Meditation, that the divine will is not greater than our own “consideredin itself formally and precisely” (AT 7:57).24

As we will discover in §5.2, however, these remarks reflect only one stage in theevolution of Descartes’s views on human freedom. In passages from later writings,he increasingly emphasizes the undetermined nature of our free action. Though thisshift in his thought perhaps reduces somewhat the difference between the divineand human wills just noted,25 it also introduces the problem for the compatibility ofthe created truth doctrine with human freedom that I mentioned toward the endof §5.1.1 (ii). Recall Descartes’s acknowledgment in the Principles that it is diffi-cult to comprehend how it is that the power by which God not only knows butalso wills and preordains “whatever is or can be” could leave “free human actionsundetermined.” The claim that God wills what “can be” reflects the implicationof Descartes’s created truth doctrine that eternal truths derive from the indifferentdivine will. But this doctrine requires that there be no truth, or at least no truth con-cerning the created world, that is independent of God’s will. Even contingent truthsconcerning our undetermined free actions can hold only because God has freelywilled that it be so.

This result of the created truth doctrine may seem to place Descartes on the sideof the Dominicans in their battle with the Jesuits on the issue of the relation of freehuman actions to God. For Dominicans such as Bañez held that the divine will is thesource of truths concerning such actions. What the Dominicans could not accept,however, is Descartes’s insistence in the Principles that God’s power leaves our freeaction “undetermined.” Moreover, there is his later claim in the Passions that every-thing follows necessarily from God’s immutable decree alone “except for thosethings that this same decree has willed to depend on our free will” (PS II.146, AT


24. Descartes holds in the Principles that the term ‘substance’ cannot apply “univocally”to God and creatures given that God is dependent on no other being, whereas created sub-stances depend on God (PP I.51, AT 8-1:24). Nonetheless, as I have argued in Schmaltz 2000,he allowed for the traditional Thomistic position that God is related in an analogical mannerto creatures. One might think that in a similar manner Descartes could say that even thoughour will differs fundamentally from the divine will, nonetheless an analogical relation holdsbetween the two. One problem with this proposal, however, derives from his insistence in theFourth Meditation that his will is “so great that I apprehend no idea of any greater” (AT 7:57).In contrast to the case of substantiality, then, the suggestion here is that the sort of perfectionthat pertains to his will is not different in kind from the perfection that pertains to God’s will.

25. Though Descartes never admitted that human freedom requires the sort of indifferencefrom any consideration of truth and goodness that is essential to divine freedom.

11:439).26 So Descartes seems to have wanted to combine the Dominican positionthat truths concerning our free action derive from the divine will with the Jesuit posi-tion that God leaves these truths under the control of our will.

The obvious question here is whether there is any intelligible way to combinethese seemingly conflicting positions. In the Principles, Descartes suggests that it isbeyond our power to reconcile our freedom with God’s power, and it might bethought there is nothing more he can say on this issue. As I indicate in §5.3, though,Descartes has more to say elsewhere on the issue of the relation of our freedom todivine providence. What he has to say, moreover, can be linked to Suárez’s accountof middle knowledge in terms of God’s idea of the habituo that results in free action.Before we consider Descartes’s views on this issue, though, we need to understandhis attitude toward the sort of human freedom that Suárez associated with indiffer-ence. As I have indicated, he has a complex and evolving view of our free action.Whereas he starts in the Meditations with a non-Suárezian conception of humanfreedom as determination to the true, he ends in the Passions with a Suárezianemphasis on the importance to such freedom of control over pursuit of the good.

5.2. INDIFFERENCE AND HUMAN FREEDOM

In the unfinished Rules for the Direction of the Mind (Regulae ad DirectionemIngenii), which he abandoned in 1628, Descartes distinguishes faith in what God hasrevealed from immediate knowledge both of first principles through intuition and ofremote consequences of those principles through deduction. There he says that theacceptance of revelation, as well as of “anything obscure,” is set apart by the fact thatit “is not an act of native intelligence [ingenii] but of will [voluntatis]” (RM III, AT10:370). The view here that the acceptance of what is not intellectually evident derivesfrom the will is something of a scholastic commonplace, as illustrated by Suárez’sremark that a person “who believes things that he does not see clearly . . . believesbecause he wills to,” which is why “theologians claim that the act of faith . . .depends on a pious disposition of the will” (MD XIX.5, ¶20, 1:717). However, Suárezmade clear that the will is involved only in the case of judgments concerning the good,and not judgments concerning truth. All of the latter derive from the intellect alone,and even in the case of judgments concerning the truth of something that is less thanevident, the strength of the assent of the intellect is tied to the clarity of its perceptionof its object (¶¶15–16, 1:716). Even in his early work, then, Descartes departs fromthe scholastic line in claiming that assent to the truth of claims that are obscure is anact of will rather than of intellect.27


26. For more on Descartes’s account of freedom in the Passions, see §5.2.3.27. Cf. Thomas’s distinction between intellectus and scientia, on the one hand, where

intellectual assent is moved by its object, and opinio and fides, on the other, where assent ismoved not only by its object, but also by decision. On the view here, the will can commandassent to propositions that do not themselves compel assent (ST II-II. 1.4). Nonetheless,

In his later account of judgment, however, Descartes’s differences with the scholas-tics are even more pronounced. According to this account, judgment involves an act ofwill not only in the case of assent to obscure matters, but also in the case of assent tothe results of intuition and deduction, or what he came to call the perception of “clearand distinct ideas.” This change from his position in the Rules introduces certain diffi-culties for his view of human freedom that derive from his commitment to the freedomof the will in its contribution to judgment. In particular, on his new account thereappear to be two different kinds of freedom that judgment can involve. With respect tojudgments concerning perceptions that are obscure, freedom seems to consist in theability of the will to refrain from assenting even in cases where it does in fact assent.In contrast, we will see the indication in Descartes that the nature of the will is suchthat it is impelled by clear and distinct ideas to assent. Given this view, the sort of free-dom involved in the case of clear and distinct perception seems to involve intellectualdetermination rather than the ability to do otherwise.

A related problem is that it is unclear which of these two kinds of freedom is mostprimary. In passages associated with his account of freedom in the Fourth Meditation,Descartes emphasizes that the will is most free when it is most determined by theintellect, and thus lacks any sort of indifference. However, he later makes the claimin the Passions, which I have noted already, that God leaves it to the will alone todetermine its free action. We will discover that with this shift there is a correspondingmove from an initial preoccupation with the role of the will in assent to truth to a finalacknowledgment of the importance to human freedom of the control of the will overits pursuit of the good.28

5.2.1. The Fourth Meditation

In his set of objections to the Meditations, Arnauld counsels that Descartes makeclear that his discussion in the Fourth Meditation “is dealing above all with mistakeswe commit in distinguishing between the true and the false, and not those that occurin pursuit of good and evil” (AT 7:215). In response, Descartes notes that “the entirecontext of my book” makes clear the restriction to the consideration of truth andfalsehood (AT 7:247–48). Nevertheless, on the basis of Arnauld’s remarks, he askedhis editor Mersenne to insert into the Synopsis of the Meditations the disclaimer that“it should be noted in passing that I do not deal [in the Fourth Meditation] at all withsin, i.e. the error which is committed in pursuing good and evil, but only with theerror that occurs in distinguishing truth from falsehood” (AT 7:15).29


Aquinas insisted that even in cases where the will commands assent, the assent itself is an actof intellect rather than (as Descartes would have it) an act of will.

28. The importance for Descartes’s account of freedom of the shift in his thought from afocus on the search for truth to a consideration of the pursuit of the good is highlighted inSchickel 2005.

29. Descartes made this request in To Mersenne, 18 Mar. 1641, AT 3:334–35. The restric-tion to the consideration of truth had been indicated in the First Meditation (AT 7:22) and inthe Second Replies (AT 7:149).

For scholastics such as Suárez, the restriction to the case of the consideration ofthe true and the false would involve as well a restriction to the role of the intellect injudgment. In the Fourth Meditation, however, Descartes insists that all judgmentsrequire contributions both from the “faculty of knowledge” ( facultate congnescendi),that is, “the intellect” (intellectu), and from the “faculty of election or freedom ofdecision” ( facultate eligendi sive . . . arbitrii libertate), that is, “the will” (voluntate).The former contributes merely “the ideas from which judgments can be made,” whichstrictly speaking are not themselves true or false. Truth or falsity enters only with theassent or dissent of the will with regard to these ideas (AT 7:56).30

Descartes is most concerned in the Fourth Meditation to argue that mistakes injudgment are our fault, since they derive from the fact that we have failed to restrictour will to clear and distinct ideas. In cases where our ideas are confused or obscure,31

the judgmental act of our will is not determined, and thus we are “indifferent” to theseideas. Given this indifference, we are able to refrain from judgment, and so to avoidfalling into error. Here we may seem to have a parallel to the case in Suárez of ourpursuit of an object that is not perceived to be a necessary good. For Suárez empha-sized that in the case of such a pursuit our will is indifferent in the sense that it hasthe ability in the very same circumstances to refrain from such a pursuit. Just asSuárez held that the freedom involved in our pursuit of the good requires an ability todo otherwise, so it seems that Descartes allows that the freedom involved in our judg-ments concerning the true requires such an ability.

However, the analogy is undermined by Descartes’s treatment of the role of the willin judgments concerning clear and distinct ideas. He notes that when he made suchjudgments, “a great light in the intellect was followed by a great inclination in the will,and thus the spontaneity and freedom of my belief were all the greater in proportion tomy lack of indifference” (AT 7:59). Such judgments correspond most closely to thecase in Suárez of the pursuit of a necessary good. But whereas Suárez claimed that theact of will involved in this pursuit cannot be free since indifference is lacking, and thusis merely voluntary, Descartes insists that the “spontaneity and freedom” of his will are“all the greater” because its action is determined by the intellect.

This is not merely a verbal dispute over which acts of will are to be included in theextension of the term ‘free’. Rather, it is over which acts are paradigmatic instances ofour freedom. For Suárez, as we have seen, only uncoerced acts that are indifferent have


30. In the Fourth Replies, Descartes insists that apart from judgment, ideas “have no ref-erence to the truth or falsity of their objects,” and thus cannot be said to be “formally” true orfalse (AT 7:232). The discussion here concerns Arnauld’s objections to the remarks in theThird Meditation concerning the “material falsity” of ideas (see AT 7:206–7). For more onDescartes’s somewhat obscure notion of material falsity, see chapter 2, note 44.

31. In the Principles, Descartes defines a clear perception as one that “is present andaccessible to the attentive mind” and a distinct perception as one that “as well as being clear,is so sharply separated from all other perceptions that it contains in itself only what is clear”(PP I.45, AT 8-1:22). A perception can fail to contain in itself only what is clear, and thus beobscure, even though it is present to the attentive mind, and so is clear. However, if a percep-tion fails to be present to the attentive mind, and so is confused, it is automatically obscure(see PP I.46, AT 8-1:22).

the perfection of control that is essential for human freedom. Though determined voli-tional acts have the perfection of being determined to the good, this perfection does notallow for the merit that pertains to genuinely free human action (see figure 5.1).

The distinction between coerced and uncoerced acts of will is not so clear inDescartes.32 However, his main difference from Suárez concerns the relationbetween voluntariness and freedom. Whereas Suárez emphasized that voluntarinessis distinct from a true freedom involving indifference, Descartes speaks in the FourthMeditation as if the spontaneity involved in voluntary action is equivalent to free-dom, and indeed claims in the Second Replies that it is “the essence of the will” thatit is carried “voluntarily and freely” (voluntarie . . . et libere) toward what is clearlyknown to be good (AT 7:166).33 Far from being paradigmatic instances of freedom,judgments involving indifference are presented in the Fourth Meditation as beingimperfect instances of freedom, since they involve “a defect of knowledge or a kindof negation.” It is only when we are impelled by our clear and distinct ideas toembrace the truth that we are “wholly free” in a manner that excludes all indiffer-ence (AT 7:57–58). The account of free will in the Fourth Meditation (see figure 5.2)indicates a clear alternative to what we find in Suárez.

The suggestion in the Fourth Meditation that determination to the true rather thanindifference is a defining aspect of the freedom involved in our judgment is reflected


Will

Coerced Uncoercedvoluntary

Necessary good

Determined byintellect

(Perfection ofdetermination to good)

Non-necessary good

Freedom ofindifference

(Perfection of controlrequired for merit)

FIGURE 5.1 Suárez on the Will

32. In contrast to Suárez, Descartes does not mention explicitly in the Fourth Meditation thecase of coerced action. Presumably this is because coercion does not seem to be a considerationin the case of assent to ideas. However, the definition of will in this text does mention lack ofdetermination by an external force, and it seems that one could contrast cases where such forcelimits options to a single undesirable one from cases where there is no such constraint.

33. See also the claim in the Sixth Replies that since a person has a will that is determinedto the true and the good, “it is evident that he will embrace the good and the true all the morewillingly [libentius], and thus also more freely [liberius], that he sees more clearly” (AT7:432).

in comments on this text in which Descartes emphasizes that God’s freedom differsin fundamental ways from our own. Drawing on his created truth doctrine, Descartesclaims in the Sixth Replies that divine perfection requires that nothing can determineGod to will anything, and thus “the supreme indifference to be found in God is thesupreme indication of his omnipotence.” However, he notes that in our case “not onlyare we free when ignorance of what is right makes us indifferent, but we are most freewhen a clear perception impels [impellit] us to pursue something” (AT 7:432–33). Incontrast to the case of God, then, the perfection of our freedom consists in the deter-mination to the true (and the good) rather than in an indifference to action.

One might object at this point that Descartes’s remarks in the Fourth Meditationindicate more ambivalence on the issue of what is essential to our free will thanI have allowed. In particular, there is the famous passage in which he claims that ourwill “consists only in the fact that we can do or not do (that is, affirm or deny, pur-sue or flee), or rather [vel potius] only in that when something is proposed bythe intellect to us for affirming or denying, or pursuing and fleeing, we are carried[feramur] in such a way that we sense no power external to us that determines us toit” (AT 7:57). Scholars have disagreed over what kind(s) of freedom the two clausesof this passage endorse, and what the vel potius connector is supposed to indicate.One prominent view is that the two clauses offer two different kinds of freedom,with the first clause indicating a “freedom of indifference” that requires a contra-causal ability to do otherwise, and the second clause indicating a “freedom of spon-taneity” that does not require such an ability.34 The vel potius connector is then readeither as a retraction of the definition of free will in terms of freedom of indiffer-ence,35 or as an indication that our will can exhibit either kind of freedom.36


Will

Coerced (?) Uncoercedvoluntary = free

Clear and distinctideas

Freedom ofspontaneity

(Perfect freedom =determination to the true)

Confused or obscureideas

Freedom ofindifference

(Imperfect freedom =lack of determination)

FIGURE 5.2 Free Will in the Fourth Meditation

34. See especially Kenny 1972, which distinguishes between “liberty of indifference” and“liberty of spontaneity.” Cf. Gilson 1913b and Beyssade 1994.

35. Gilson 1913b, 310; Beyssade 1994, 206.36. Kenny 1972, 18.

However, it seems that vel potius is best read as indicating not a retraction or aqualification, but rather an equivalence, as in, “or in other words.” The question, then,is whether freedom of indifference is to be understood in terms of freedom of spon-taneity or the other way around. Though some have argued that freedom of indiffer-ence is primary,37 I take the primacy of freedom of spontaneity to be indicated bypassages we have just considered in which Descartes emphasizes that indifference isnot essential to human freedom. Thus, I understand the claim that we “can do or notdo (that is, affirm or deny, pursue or flee)” to affirm our ability to determine our ownaction in a manner that precludes determination by an external power.38 This deter-mination involves indifference in cases where we lack clear and distinct ideas, buteven in cases where we possess such ideas, our will remains the source of the fact thatwe are carried inevitably toward what the intellect represents to us.39

I propose that we understand Descartes’s view that the will determines its own freeaction in terms of Suárez’s account of the relations between the faculties of intellect andwill in action. Earlier I noted the claim in Suárez’s De Anima, with respect to the pro-duction of appetition, that “apprehension is required, not as an efficient cause, but as anapplication to the object” (DA V.3, ¶8, Opera 3:760) (see §4.2.1). This claim is reflectedin his view in the Metaphysical Disputations that cognition is merely a necessary con-dition for an appetitive act directed toward the cognized object, and that this facultyrather than the cognition itself is the efficient cause of the act (MD XVIII.7, ¶49, 1:646).To be sure, the particular passage from the Disputations is concerned only with cogni-tion of nonnecessary goods (in particular, sensory goods), and it might be thought thatappetition must move itself simply because its act is not determined by cognition.However, elsewhere in this text Suárez appeals to Thomas in support of the view thatthe function of the intellect is simply “to illuminate, direct and regulate the operationsof the will,” and not to move the will as an efficient cause (MD XIX.6, ¶7, 1:721).40

The indication here is that cognition serves merely as a final cause that moves the willto act on its own. Even in the case where the intellect presents a necessary good, it onlyinstructs the will to determine itself in a particular manner.


37. See, for instance, Alanen 2003, 240–46, and Ragland 2006a and 2006b.38. For a similar view that freedom is equated in this passage with an ability of the will to

determine itself, see Hatfield 2003, 192–98. But see note 43.39. Cf. Campbell’s view that the first clause of Descartes’s definition requires that there be

genuine alternatives even in the case of clear and distinct perception, but that these alternativesare compatible with the determination of action in the actual case. This is what Campbell calls“two-way compatibilism” (Campbell 1999). See also the emphasis in Ragland 2006a on theimportance of alternative possibilities for freedom. However, toward the end of the FourthMeditation Descartes stresses that his will would be free even if God precluded the possibilityof error either by restricting his deliberation to clear and distinct perception or by impressing onhis memory the importance of making judgments only in the case of such perception (AT 7:61).There is no indication here of the requirement that alternatives be open to him. I take this samepoint to count against the view that Descartes’s definition in the Fourth Meditation indicates thenecessity for freedom of a contracausal power to do otherwise (see the work cited in note 37).

40. Suárez cites ST I-II.9.1 and 3.

Descartes admittedly does not use the notion of final causality in describing therelation of perception to the will. However, he also refrains in passages connected tohis account of freedom in the Fourth Meditation from referring to such perception asan efficient cause of the will. It is true that he speaks of perception as “impelling”(impellit) the will (Sixth Replies, AT 7:433), and this may seem to suggest an efficientcausal relation. However, in other passages he indicates that the impulsion is to beunderstood in terms of the will being “drawn” (fertur) by perception (Second Replies,AT 7:166), or of something being “proposed” (proponitur) in perception for action onthe part of the will (Fourth Meditation, AT 7:57). In general, the stress in the FourthMeditation is on the distinction between the passive intellect, which merely offers upideas for consideration, and the active will, which acts on those ideas.41 Given thisemphasis, it seems that the relation between perception and the will is better con-ceived in terms of an agent being drawn toward a perceived end, as in Suárez, ratherthan in terms of a perception being an efficient cause of an effect in a patient.42

The remarks in the Fourth Meditation also appear to allow for a Suárezian under-standing of indifference in terms of a contracausal power to do otherwise. After all,when Gassendi objects to the account of indifference in this text on the grounds thatour will is determined to judge in a particular manner even when perception isobscure (Fifth Objections, AT 7:317), Descartes responds by insisting that our willhas “the freedom of moving itself, without the determination of the intellect, in onedirection or another” (Fifth Replies, AT 7:378).43 In the case of judgments concern-ing clear and distinct ideas, however, there is no indication that the will has the power“of moving itself in one direction or another.” To be sure, Descartes could still say thateven in this case our will can do or not do in the sense that it determines itself to act.He in fact does say that our freedom is most perfect in this case given that it is the


41. See also Descartes’s claim that will and intellect “differ only as the activity and pas-sivity of the same substance. For strictly speaking, understanding is the passivity of the mindand willing is its activity” (To Regius, May 1641, AT 3:372).

42. In §3.2.1 (i), I noted the problem that the suggestion in the discussion in PP II.37 of thefirst law of nature that even minds never change “unless by external causes” (AT 8-1:62) seemsto conflict with Descartes’s claim that our will can be a source of internal change. We can nowsee that the problem derives from Descartes’s view that (setting aside the difficult case of God;see §5.3) nothing external to our will can be an efficient cause of its free action. However, onthe Suárezian position that I attribute to Descartes, perceptions could serve as final causes ofsuch action. Descartes also held that perception is required for judgment involving the assentof the will (see, e.g., PP I.34, AT 8-1:18), and it seems that he would allow that it is requiredfor any act of will. Thus, he could hold that even in the case of changes due to free action, acause external to the will is required, albeit a final rather than an efficient cause.

43. But cf. Hatfield’s claim that for Descartes, indifference “is not incompatible with thewill being determined by other factors, such as habit, to choose one way rather than another”(Hatfield 2003, 194; the reference here is to Descartes’s remarks in what I call in §5.2.2 (iii)“the Jesuit letter”). Though Descartes is concerned in his response to Gassendi only withexternal determination by the intellect, and not by internal determination by habit, I take hisclaim in this response that our will “can guard against our erring” by “moving itself in onedirection or another” to indicate a stronger sense of indifference than Hatfield suggests.

nature of the will to be carried to the true (and the good). But certainly the implica-tion here that our freedom is compatible with the fact that the nature of our will neces-sitates its being drawn to clear and distinct ideas would have been troubling to Suárez.Though we will discover that Descartes came to be worried by this implication in hislater writings, at the time of the Meditations he seems to have been content to iden-tify the freedom of our will with a self-determining activity that is not negated butrather perfected by clear and distinct perception.

5.2.2. Mid-1640s Developments

Some commentators have found in Descartes the consistent affirmation of a particu-lar account of human freedom, though without agreeing on the precise nature of thisaccount.44 In contrast, my argument here is that his view of human freedom evolvedover time.45 We have seen the emphasis in the Fourth Meditation and related texts ona kind of freedom that is most perfect when our will is determined by its nature toassent to clear and distinct ideas. However, during the mid-1640s, after the publica-tion of the second Latin edition of the Meditations in 1642, Descartes increasinglyemphasizes the importance of our control over the will in free action. Though he stillacknowledges during this time the determination of our will to clear and distinctideas, he also increasingly attempts to allow for some dependence of this determina-tion on choice. We see this development in Descartes’s thought first in his account ofhuman freedom in the Principles (1644), and then in his commentary on the FourthMeditation in two letters, the first a 1644 letter to the Jesuit Denis Mesland, which Icall the Mesland letter, and the second an undated letter most likely from around thissame time and almost certainly addressed to a Jesuit correspondent, which I call theJesuit letter.46

(i) Principles

I have noted the emphasis in the Principles on the problem of reconciling the fact that“everything is preordained by God” with the fact that we experience a “freedom andindifference” (libertatas et indifferentia) in ourselves exhibited in our “undetermined”(indeterminatas) actions (PP I.40–41, AT 8-1:20). Here already there seems to be adramatic shift from the emphasis in the Fourth Meditation on a kind of freedom thatis most perfect when indifference is completely absent. Yet there are some important


44. Cf., for instance, the view in Campbell 1999 that Descartes consistently accepted acompatibilist account and the view in Alanen 2003, 240–46, and in Ragland 2006b that heconsistently accepted an incompatibilist account.

45. Here I develop the evolutionary account of Descartes’s views on human freedom thatI first presented in Schmaltz 1994.

46. Cf. the discussion of the Mesland and Jesuit letters in Marlin 1986, which criticizesthe analysis of this correspondence in Kenny 1972. Though he anticipates my conclusionbelow (against Kenny) that the Jesuit letter marks an important break from Descartes’s earlierviews, Marlin does not mention the sort of problems that I do with the positive account of free-dom in this letter.

qualifications in the remarks in the Principles that indicate that the differences are notas great as they may first appear to be. For instance, in this text Descartes restricts hisconsideration of our freedom “to assent or not to assent at will” to cases in whichmatters are “not completely certain and examined” (PP I.39, AT 8-1:19). Moreover,he continues to insist there that our nature is such that “whenever we perceive any-thing clearly, we assent to it spontaneously [sponte], and can in no way doubt that itis true” (PP I.43, AT 8-1:21). There is thus no denial here of the determination of ourwill toward clear and distinct perception that Descartes highlighted in the FourthMeditation.47

What is missing in the Principles, however, is his earlier claim that such adetermination constitutes a perfect form of freedom. Indeed, at one point in thistext Descartes seems to indicate that genuine freedom precludes determination ofaction. Thus, he notes there that whereas automata cannot be praised for produc-ing certain motions, since “they necessarily exhibit” such motions, we can bepraised for embracing the truth, since “it is more credit to us when we embracebecause we do it voluntarily, than would be the case if we were not able not to haveembraced” (PP I.37, AT 8-1:18–19). In contrast to the view in the FourthMeditation, voluntariness is contrasted with cases where one cannot do other-wise.48 To be sure, there is no explanation in the Principles of how the fact that“we were able not to have embraced” the truth is consistent with our inability todoubt our clear and distinct perceptions. So there are difficulties here that remainto be addressed. Even so, we are closer to the view in Suárez that any sort of free-dom in us that involves merit must include an indifference that consists in a con-tracausal power to do otherwise.49

(ii) Mesland Letter

In a 1644 letter to “a Reverend Jesuit Father,” identified in another letter as DenisMesland,50 Descartes responds to questions about the Fourth Meditation that


47. For the view that these features of the Principles reveal that the account of human free-dom in this text is compatible with the account in the Fourth Meditation, see Kenny 1972,20–21, and Cottingham 1988, 250–52.

48. The French edition of the Principles replaces the reference to what is voluntarywith what derives from “a determination of our will” (une determination de notre volonté)(AT 9-1:40). For discussion of the changes in this section of the Principles, see Beyssade1996.

49. See also Descartes’s claim in a 1645 letter to Elisabeth that we experience in ourselvesan independence “that suffices to make our actions praiseworthy or blameworthy” (3 Nov.1645, AT 4:333), and his claim in a 1647 letter to Queen Christina that “it is only whatdepends on the will that is subject to reward and punishment” (20 Nov. 1647, AT 5:84).

50. In particular, Descartes indicates that Mesland is the correspondent in a later letter toa Jesuit, at AT 4:121.

concern an indifference connected to our ability to suspend judgment. He begins byagreeing that we have this ability, but explains that

it is certain, it seems to me, that a great light in the intellect is followed by a greatinclination in the will; so that seeing very clearly that a thing is good for us, it isvery difficult, and even, as I believe, impossible, as long as one continues in thesame thought, to stop the course of our desire. But because the nature of our soulis such that it hardly attends for more than a moment to the same thing, as soon asour attention turns from the reasons by which we know that this thing is good forus, and we retain in memory only that it is desirable to us, we have the power torepresent to our mind certain reasons that make us doubt, and suspend ourjudgment, and also maybe even form a contrary judgment. And so, since youregard freedom not simply as indifference but rather as a real and positive powerto determine oneself, the difference between us is merely a verbal one, for I agreethat the soul has such a power. (AT 4:116; italicized text in Latin)51

The claim here that it is impossible “as long as one continues in the same thought”to resist “a great light in the intellect” is in line with the view in both the FourthMeditation and the Principles that our will is by nature determined to assent toclear and distinct perception. In light of the suggestion in the Principles that we pos-sess a kind of “freedom and indifference” that involves merit, however, it is signifi-cant that Descartes does not merely stop in the Mesland letter with the claim that weact voluntarily even when we must follow the inclination of our will. Rather, heinsists that even in the case where we are led by a great light of intellect, it is possi-ble for us subsequently to be distracted and then consider reasons that lead us to sus-pend judgment. Thus, a continuing assent to clear and distinct perception requires acertain effort of attention that depends on the will rather than on that perceptionitself.

Admittedly, there are remnants in the Mesland letter of the suggestion in theFourth Meditation that perfect freedom consists in determined assent to clear anddistinct perception. Thus, Descartes notes there that “we may earn merit eventhough, seeing very clearly what we must do, we do it infallibly, and without anyindifference, as Jesus Christ did during this [earthly] life” (AT 4:117). Nevertheless,he also attempts there to accommodate the Jesuit view that indifference is essentialto human freedom. For instance, he emphasizes that the indifference that he took tobe accidental to freedom in the Fourth Meditation is merely one that involves somesort of defect in knowledge. He therefore allows for the view of his correspon-dent that the will is indifferent in the sense of having “a real power to determine one-self,” and concludes that any difference between them “is merely a verbal one” (AT4:116). Descartes may appear to indicate a more than verbal difference when he


51. The remarks here are reminiscent of Descartes’s claim in the Fifth Meditation thatthough “my nature is such that so long as I perceive something very clearly and distinctly Icannot but believe it to be true,” it is also the case that “I cannot fix my mental vision contin-ually on the same thing to keep perceiving it clearly,” and thus that “other arguments can noweasily occur to me which might easily undermine my opinion, if I did not possess knowledgeof God” (AT 7:69; cf. Second Replies, AT 7:144–46; Fourth Replies, AT 7:245–46).

claims that whereas Mesland restricts freedom to a power of determination accom-panied by indifference in the sense of the Fourth Meditation, he takes freedom toinclude “in the general sense what is voluntary” (AT 4:116). Here we seem to havethe sort of substantive disagreement with the Jesuit position connected to the insis-tence in the Fourth Meditation that true freedom is exhibited in our determinedassent to clear and distinct perception.52 In contrast to the case of this text, however,there is the admission in the Mesland letter that at least in the case of the continua-tion of this assent, there is a certain kind of dependence on an apparently undeter-mined effort of the will.

(iii) Jesuit Letter

The third stage of Descartes’s development during the mid-1640s is reflected in aLatin fragment of an undated letter that Adam and Tannery found in a manuscriptfrom the Bibliothèque Mazarine and presented in their edition of Descartes’s writ-ings as a continuation of a letter to Mesland written in French and dated 9 February1645.53 It seems unlikely that Descartes would have switched languages midletter,and if Mesland were the correspondent it would be difficult to explain why cross-references are missing from both the fragment and the Mesland letter. Even so, thefact that the fragment repeats many of the claims from the Mesland letter regardingindifference and its relation to free action allows us to be confident that it was com-posed around the same time. Moreover, we can be fairly certain that the fragmentwas from a letter to a Jesuit correspondent, since there is the same attempt that wefound in the Mesland letter to accommodate the Jesuit emphasis on the importanceof indifference for freedom.

In line with his procedure in the Mesland letter, Descartes begins in the Jesuitletter by distinguishing the indifference considered in the Fourth Meditation,which involves merely a lack of impulsion of the will toward truth and goodness,and a kind of indifference that is identified with “a positive faculty of determiningoneself to one or other of two contraries that is to pursuing or avoiding, to affirm-ing or denying.” He then claims that our will has such a positive faculty not onlywhen it possesses indifference in the Fourth Meditation sense, but also when it ismoved by an evident reason to act in a certain way. Descartes cites as evidence thatthe action derives from the positive faculty of determination in this latter case thefact that


52. In fact, Kenny takes the position in the Mesland letter to be consistent with the viewin the Fourth Meditation; see Kenny 1972, 23–24.

53. See the editorial comments on this letter in AT 4:172. As Kenny notes, Clerselier hadearlier provided a French translation of this letter and dated portions of it from 1630 and 1637,whereas Adam and Tannery initially presented the French version of this letter as part of aMay 1641 letter to Mersenne (Kenny 1972, 25–26). The former view of the letter is ruled outby the fact that it refers to the account of indifference in the Fourth Meditation, whereas thelatter is implausible given its attempt to accommodate a Jesuit account of indifference towhich Mersenne was not committed.

although morally speaking [moraliter loquendo] we can hardly move in thecontrary direction, absolutely [absolute] however we can. For it is always opento us to hold back from following a clearly cognized good, or from admitting aperspicuous truth, providing however that we consider it good to demonstratethe freedom of our decision by this. (AT 4:173)

Thus, the positive faculty essential for freedom requires not only that the will deter-mine itself toward an evident reason, as the Fourth Meditation suggests, but furtherthat the will have the ability “absolutely speaking” to move itself in a contrary direc-tion even in the case of its consideration of such an evident reason.

Descartes introduces a complication into his account when he continues by not-ing that before acts of will are elicited, freedom consists “either in a greater facilityin determining oneself or in a greater use of the positive power we have of follow-ing the worse although we see the better.” We more easily determine ourselves whenwe follow evident reasons, but make greater use of the positive power in turningaway from such reasons (AT 4:174). In addition to the positive faculty of determi-nation, then, there is a positive power that is revealed in our choice of the worse,which involves not only indifference but also a kind of perversion.54

Presumably we exercise our positive faculty of determination both when we deter-mine ourselves to follow evident reasons and when we exercise our perverse power.It is unclear, however, that the exercise of this power is involved in the case thatDescartes cites in the letter, where we hold back from following a clearly cognizedgood since we consider it better to demonstrate our freedom by doing so. For in thiscase we seem to think it better to act in this way, and therefore appear not to “see”that this is a worse option. In general, it appears that the ability to do otherwiseabsolutely speaking does not require the exercise of a perverse power to do what weperceive to be worse. For we could be motivated to act contrary to evident reasons incases where we (confusedly) perceive some good that follows from so acting to begreater than the good of being led by evident reasons.

In any event, the suggestion in the Jesuit letter is that our freedom involves thepossession of a positive faculty that allows us absolutely speaking to resist even clearand distinct perceptions. Ferdinand Alquié has argued that such a suggestion marksa decisive break with Descartes’s view in previous work, including the Mesland let-ter, that we cannot refuse assent to something clearly and distinctly perceived at themoment it is so perceived.55 But we could perhaps reduce the differences betweenthe Mesland and Jesuit letters by drawing on the position in the former that when itis distracted from considering a clearly known good, our will can attend to reasonsthat lead it to suspend judgment or form a contrary opinion. When Descartes saysin the Jesuit letter that absolutely we can move against evident reasons, he neednot have meant that we can so move at the moment we perceive them. He couldhave meant, rather, that it is always in our power to distract attention in a way thatallows us to act otherwise. Though the Mesland letter does not mention a power of


54. Following Kenny’s claim that this positive power involves a “liberty of perversion”(Kenny 1972, 28).

55. Alquié 1950, 288–92.

distraction, it seems to leave room for the view that such a power explains the powerwe have to resist even evident reasons.56

Alquié therefore may well distinguish the views in the Mesland and Jesuit letterstoo sharply. Nevertheless, I think that he is correct to draw attention to an importantshift in Descartes’s thought on human freedom during the mid-1640s. In particularthere is a shift away from the focus in the Fourth Meditation on what Alquiéhas called la liberté éclaire—“the enlightened freedom” that occurs when the spon-taneous assent of the will is determined by the clear and distinct perceptions ofthe intellect. What comes to be emphasized rather is what he has called la libertépositive—“the positive freedom” that occurs when the act of will is undetermined.57

This sort of freedom is indicated by talk in the Jesuit letter of a positive faculty ofdetermination essential to freedom that includes the power “absolutely speaking” toturn away even from evident reason.

Interestingly, the distinction in the Jesuit letter between what the will can do“morally speaking” and “absolutely” is present also in the Metaphysical Disputations.Suárez appeals to such a distinction in that text in particular to address the question ofwhether evil action presupposes an antecedent defect in intellectual judgment. Heclaims that though no such defect is “absolutely necessary” (absolute necessarium)for us to choose evil, still “morally speaking [moraliter loquendo] the will neverlapses unless there is some antecedent defect in the intellect” (MD XIX.7, ¶¶10–11,1:725–26). But Suárez’s use of this distinction differs in important respects from theuse of it in Descartes’s Jesuit letter. For Suárez appeals to a defect in our intellect toexplain our ability morally speaking to lapse into sin. In contrast, the indication inDescartes’s Jesuit letter is that the defect of thinking it would be good to demonstrateour freedom by turning away from evident reasons serves to explain our abilityabsolutely to resist reasons that we would otherwise follow morally speaking.

Even so, toward the end of §5.1.1 (i), I alluded to the fact that another feature ofSuárez’s account of human freedom sheds some light on the distinction in the Jesuitletter between what we can do absolutely and morally speaking. I have in mindSuárez’s claim that the influence of intellectual judgment on free action requires aprior resolution on the part of the will. Given a resolution to pursue a certain good,the judgment that particular objects are instances of such a good leads me to pursue


56. For a similar reading of the Mesland and Jesuit letters, see Kenny 1972, 28–30.However, Kenny does not draw attention to the fact that there is no mention in the Meslandletter of the power of the will to distract attention. Moreover, the suggestion in the Meslandletter that perfect freedom consists in the determined assent to clear and distinct perception ismissing from the Jesuit letter.

57. Alquié 1950, 289–90. On the basis of close readings of the texts, Michelle Beyssadehas argued that this sort of shift is reflected in changes to the Fourth Meditation in the 1647French edition of the Meditations (Beyssade 1994). For a critique of this argument, seeCampbell 1999, 187–89. I have already expressed my disagreement with Beyssade’s claim(cited in note 34) that the vel potius connector in the Latin edition of the Fourth Meditationindicates a retraction of the claim that a two-way power is essential to free action. However,my evolutionary account of Descartes’s views on human freedom conflicts with Campbell’sclaim against Beyssade that these views were consistently compatibilist in nature.

them. In the terms Descartes employs in the Jesuit letter, given this judgment I pur-sue such objects morally speaking. Since it is in my power to withdraw the resolution,however, absolutely I can resist the pursuit. As we have seen, Suárez even speaks ofthe will as having the ability to “divert the intellect from thinking about the object”that is presented as good. When read in light of the Mesland letter, the Jesuit letterindicates that our will has a similar power of diversion that can prevent us “from pur-suing a clearly known good, or from admitting a clearly perceived truth” (AT 4:173).

Of course, the reference here to the role of the will in the admission of a perspic-uous truth reflects Descartes’s deep differences with the scholastics on the nature ofjudgment. Even so, it is interesting that in the Jesuit letter, he takes the turning awayfrom a clearly perceived truth to be motivated by pursuit of an alternative good, inparticular, the good of demonstrating freedom. A further movement toward Suárezis reflected in the indication in this letter that a consideration of the pursuit of thegood can be important for an account of the role of the will in the search after truth.To be sure, I have cited several remarks above concerning the freedom of our willthat date prior to the Jesuit letter but that mention not only our assent to the truebut also our pursuit of the good. Even so, in the discussions of free will in theMeditations and the Principles, he tends to consider the former in abstraction fromthe latter. In contrast, the view in the Jesuit letter is that the freedom exhibited in theassent of our will to a clearly perceived truth involves a power to do otherwise“absolutely speaking” that is itself entangled with our pursuit of the good. To besure, the suggestion in this letter that doing otherwise absolutely speaking requiresthe exercise of a perverse power of doing the worse tends to confuse matters. In thefinal account of the freedom of our will that he provides in the 1649 Passions of theSoul, Descartes focuses on less bizarre considerations that lead us away fromthe search after truth. As in the case of the Jesuit letter, however, these considera-tions involve the role of the will in the pursuit of the good.

5.2.3. Passions of the Soul

In the Passions Descartes emphasizes that “the will is by its nature so free that it cannever be constrained [contrainte]” and that its volitions are “absolutely in its power[absolument en son pouvoir] and can be changed only indirectly [indirectement] bythe body” (PS I.41, AT 11:359). Volitions are here contrasted with passions in thestrict sense, that is, certain confused and obscure perceptions that we refer to oursoul rather than to our body or to external objects and that are “caused, maintainedand strengthened by some motion of the spirits [in the body]” (I.27, AT 11:349).58

Such passions are “absolutely dependent on the actions that produce them, and canbe changed by the soul only indirectly, except in cases where it is itself their cause”(I.41, AT 11:359–60). In cases where the passionate feelings do not have the soul astheir cause, they can come into conflict with volitions deriving from the will. Thus,


58. In this text, Descartes also uses the term ‘passions’ more broadly to include all per-ceptions that have the soul or body as their cause (PS I.19, AT 11:343).

the passion of fear that arises from battle may compete in the soul with the volitionto act in a courageous manner (I.47, AT 11:364–66).59

In this sort of competition, it cannot be assumed that the volition will win out overthe passion. Indeed, Descartes notes that “weak” souls allow themselves to be car-ried away by their strongest current passions. In contrast, “strong” souls are able tofollow “determinate judgments” concerning action that conflict with such passions.Sometimes these judgments are themselves based on passions that have previouslyinfluenced the soul. In the case of the strongest souls, however, the judgments deriveultimately not from the passions but rather from “knowledge of the truth” as deter-mined by clear and distinct perception (PS I.48–49, AT 11:366–68).

Descartes is not very explicit on how the passions overwhelm weak souls. Butthough it is tempting to think that he takes these passions to be efficient causes of thevolitions of such souls, it is important to remember the indication in the Passions thatthe body can affect volition only “indirectly.” My suggestion earlier, with respect tothe relation between intellect and will in the Fourth Meditation, was that it was moreappropriate to conceive of Descartes’s perceptions as final causes that lead the willto act in a particular manner than to conceive of them as efficient causes of effectsin the will. Likewise, I propose here that we conceive of his passions as final causesthat weak souls choose to follow. Passions can affect volition only indirectly, sincethey require the cooperation of the will, which itself is the efficient cause of its voli-tions.60 This sort of cooperation is precisely what strong souls refuse to afford theircurrent passions, since they find it best to rely on their determinate judgments asguides in their pursuit of the good.

Since Descartes’s earlier accounts of theoretical judgment tend to bracket con-siderations involving pursuit of the good, it is difficult to see how the will couldrefrain from assenting to clear and distinct ideas. Indeed, his suggestion in the Fourth


59. In this article, Descartes presents the competition as consisting in the conflict betweena volition in the soul and a motion in the pineal gland that corresponds to the passion. Herehe is resisting the traditional scholastic view that there can be conflicts internal to the soul thatinvolve its sensitive and rational parts. However, it seems that even on Descartes’s own viewthere can be a conflict between a volition and a passion in the soul that corresponds to themotion in the gland. Thus, his differences from the scholastics do not seem to be as great ashe suggests in this article.

60. Cf. Hoffman’s claim that remarks in the Passions allow for the position that percep-tions in general, and passions in particular, can cause volitions by causing the soul to will ina particular manner (Hoffman 2003, 268–72). Hoffman also cites Descartes’s claim toElisabeth that passions can cause the soul to lose its freedom as showing that he allows thatvolitions can be caused by the passions to be unfree (291–92, citing To Elisabeth, May 1646,AT 4:411). However, Hoffman’s textual evidence for the first point does not seem to me toshow that volitions are passions deriving from bodily efficient causes. When Descartes allowsin the Passions that volitions can be passions (in the broad sense), he has in mind merely thatour soul perceives its volitions, not that volitions themselves can be passive effects of some-thing other than the will (see PS I.19, AT 11:343). Moreover, the claim in the letter toElisabeth can be read as saying not that the passions can cause unfree volitions, but rather thatthey can bypass judgment entirely in causing bodily behavior.

Meditation is that the will determines itself without impediment in assenting to suchideas. However, the view in the Passions is that in the case of our pursuit of the good,the passions frequently provide a seductive alternative to knowledge of the truth.For the passions naturally move us to pursue the objects they present as beneficial toourselves as a soul-body composite, and to flee those objects they present as harm-ful to that composite (PS II.137, AT 11:430). But though we have an immediatemotivation to follow the passions, it is nonetheless the case that they “almost alwaysmake the goods as well as the evils they represent to be much greater and moreimportant than they are” (II.138, AT 11:431). Whereas it is difficult to see howDescartes could allow that we resist a clear and distinct idea once we actually per-ceive it, it is easier to understand why he would have thought it requires some effortof will to resist a reliance on the passions in favor of arduous intellectual determi-nation in attempting to distinguish good from evil. Moreover, acquiescence to thepassions does not require the exercise of anything as exotic as the perverse power ofdoing the worse while seeing the better. All that is required is the inclination that allhuman souls have in virtue of their union with a body to judge good and evil imme-diately by means of the passions.

I noted Descartes’s claim that the soul does not have direct control over the passionsexcept in cases where “it itself is their cause.” In the Passions, the most notable casein which the soul has control is that of the passion of generosity (generosité).61 Notcoincidentally, this passion is also the one most closely linked to our control over ourown will. In the Passions, Descartes prefaces his discussion of generosity by makingthe point, which we saw earlier in the Principles, that “it is only those actions thatdepend on free decision [libre arbitre] for which we can be praised or blamed with rea-son” (PS III.152, AT 11:445). Indeed, the claim in the Passions is that our control overour free actions “renders us in some manner similar to God in making us masters ofourselves” (11:445). Generosity is able to arise in someone not only when he knowsthat he can be praised or blamed only for using our freedom well or poorly, but alsowhen he feels in himself “a firm and constant resolution to use it well, by undertakingand executing all that he judges to be best” (III.153, AT 11:446).62 This passion in turnstrengthens virtuous habits that lead to the proper use of the will (III.161, AT11:453–54). In this way, generosity provides the means of overcoming the inclinationto depend on our current passions for judgments concerning good and evil.

According to the Passions, then, an attachment to clear and distinct ideas is not asimple matter of having an actual perception of them. Rather, we must have a firmand constant resolution to seek them out and to act in accord with them. And thisresolution is made possible by our recognition of the primary good of using our freewill well. Here, then, the search for truth is subordinated to the pursuit of the good.


61. Cf. the discussion in Shapiro 1999a of Descartes’s account of generosity.62. Descartes says that the passion “consists” (consiste) in these two components, but

given his view that passions in the strict sense are feelings that are referred to the soul and arecaused, maintained, and strengthened by the motion of the animal spirits, it seems that the pas-sion could consist only in the positive feelings associated with the recognition of a resolutionto use the will well.

Moreover, Descartes’s view in the Passions is Suárezian in emphasizing that ouradherence to our best intellectual judgment itself depends on decisions that are underthe control of our will. In terms, again, of the view in the Jesuit letter, whereasmorally speaking it is impossible for us to resist a clearly cognized good given ourresolution to use free will in the best manner, absolutely we are able to turn awayfrom such a good, since we are free to withdraw the resolution. In contrast to the sug-gestion in the Jesuit letter, the primary case of such a withdrawal occurs not whenwe exercise our perverse power of doing the worse while seeing the better, but whenwe fail to stand up to our current passions in deciding how to act.

Descartes held consistently to certain tenets in his mature discussions of humanfreedom in the period bounded by Fourth Meditation (1641), on one end, and thePassions (1649), on the other. For instance, he indicates throughout that we make thebest use of our freedom in adhering to clear and distinct ideas. Moreover, from startto finish he suggests that whenever we consider confused or obscure ideas, there isan indifference that allows us to refrain from judgment. It is important, however, notto overlook the very real development of Descartes’s views that occurs over thisperiod of time. In the Fourth Meditation, he insists that human freedom is perfectedin the case where the clarity of knowledge precludes any sort of indifference.Starting with the Principles, however, he begins to link our freedom to a kind ofmerit that requires an undetermined control over decision. The link is only strength-ened in the Mesland and Jesuit letters, where he emphasizes the ability of our will toconsider goods other than those revealed by clear and distinct ideas. Admittedly,these letters were written to Jesuit correspondents, and the Principles was modeledon texts used in the Jesuit schools. It might be thought, then, that Descartes wasmerely playing to the Jesuit crowd in drawing attention to the importance for our freeaction of a kind of indifference.63 However, his final word is in the Passions, and thistext was not directed to a Jesuit audience. There we find not the view in the FourthMeditation that we are most free when least indifferent, but rather the claim that theproper use of our freedom requires a resolution to resist the ever-present temptationto follow our current passions in making judgments about good and evil. The earlierfocus on the freedom of a disembodied mind is therefore replaced with a considera-tion of the more complex case of the freedom of an embodied soul, and with thisshift there is a recognition of the fact that our free assent to the truth is dependent onour exercising proper control over our will in the pursuit of the good. ThoughDescartes began in the Meditations with an account of free will that has some decid-edly anti-Suárezian elements, he ended with an account in the Passions that perhapswould not have seemed so strange to Suárez.

5.3. HUMAN FREEDOM AND DIVINE PROVIDENCE

I have noted at several points the pessimistic conclusion in the Principles that we con-front insuperable difficulties in attempting to reconcile our undetermined “freedom


63. As suggested, for instance, in Gilson 1913b.

and indifference” with the power by which God not only knew but also willed andpreordained all that “is or can be” (PP I.40–41, AT 8-1:20). In correspondence withElisabeth soon after the publication of this text, however, Descartes seems to be moreoptimistic that we can find a way around these difficulties. Thus, he tells Elisabeththat “the independence that we experience and sense [experimentons et . . . sentons]in ourselves and that suffices to render our actions praiseworthy or blameworthy, isnot incompatible with a dependence that is of another nature, according to whichall things are subject to God” (3 Nov. 1645, AT 4:332–33). As in the case of thePrinciples, there is a stress here on a freedom that involves independence and thatmakes us worthy of merit. However, the view in this earlier text seems to differ fromthe position Descartes offers to Elisabeth insofar as it lacks the proposal that we canreconcile our freedom with the complete dependence of creatures on God once werecognize the different levels at which our will and divine power operate.

Nevertheless, I believe that a closer reading of Descartes’s remarks to Elisabethreveals that the differences with the views in the Principles are not as great as theyinitially appear. In particular, the discussion in the Elisabeth correspondence allowsfor a fundamental mystery concerning God’s determination of our free action. Thismystery involves not only the problem—familiar from the Molinist account of mid-dle knowledge—of comprehending how God can foreknow with certainty actionsthat are themselves undetermined. In addition, there is the problem—deriving fromDescartes’s created truth doctrine—of comprehending how God can be the efficientcause that produces the truths that provide the foundation for this sort of foreknowl-edge. My ultimate conclusion is that an understanding of the dependence of our freeaction on God in terms of the divine creation of eternal truths provides support for aconservationist account of God’s production of that action. I therefore take Descartesto deviate from the sort of concurrentist treatment of human freedom that we find inSuárez. However, I also claim that once God’s creation of truths concerning our freeaction is posited, the remaining mystery in Descartes concerning God’s providentialcontrol over such action is just the one that Suárez admitted in response to hisDominican critics.

5.3.1. The Elisabeth Correspondence

In July 1645, Descartes proposed to Princess Elisabeth that they read togetherSeneca’s On the Happy Life (De Vita Beata) to reflect on the position that we acquirehappiness not from fortune but only from ourselves (21 July 1645, AT 4:252–53).Based on their subsequent discussion of this text, Descartes claimed that once we rec-ognize that “there is a God on whom all things depend, whose perfections are infinite,whose power is immense and whose decrees are infallible,” we will be able “to acceptcalmly all the things that happen to us as expressly sent by God” (15 Sept. 1645, AT4:291). But Elisabeth was not persuaded, arguing in response that though this recog-nition can perhaps lead us to accept what follows from “the ordinary course ofnature,” it cannot help us with regard to “those humans impose on us, the decision ofwhom appears to be entirely free.” She concluded that it is faith alone that can teachus that “God takes care to rule volitions, and that he has determined the fortune ofeach person before the creation of the world” (30 Sept. 1645, AT 4:302*).


In response, Descartes insists that not only faith but also the reasons that prove“that God exists and is the first and immutable cause of all the effects that do notdepend on human free will” reveal equally that “he is also the cause of all the effectsthat do so depend.” For natural reason reveals that God must exist as a supremelyperfect being, and “he would not be supremely perfect if anything could happen inthe world without coming entirely from him [entirement de lui].” He anticipates theobjection that God does not determine particular effects, since he is merely a “uni-versal cause” of such effects, but urges that

the distinction of the Schools between universal and particular causes is out ofplace here: because what makes the sun, for example, the cause of flowers is notthe cause of the fact that tulips differ from roses, [since] their production dependsalso on other particular causes to which they are not subordinated; but God is sucha universal cause of all that he is in the same way the total cause. (6 Oct. 1645,AT 4:314)

In an earlier discussion of this passage, I read the claim that God is a universaland total cause of all effects as indicating not that there are no other causes of sucheffects, but merely that there are no such causes that are not subordinated to God’suniversal causality (see §2.2.1). Given this reading, the conclusion that God is a uni-versal and total cause of our free action does not indicate that our will is not an effi-cient cause of such action. What it indicates is that our causation of our own freeaction must be subordinated to God’s universal causation of everything in the cre-ated world.

Even so, some commentators have concluded on the basis of the remarks toElisabeth that Descartes allowed that God determines even our free actions.64 Yetsuch a determination would seem to be incompatible with the claim in the Principlesthat such actions are “undetermined.” Elisabeth recognized the difficulty here whenshe responded to the letter containing the assertion that God is the universal and totalcause of our free action by protesting that “it seems to me to be contrary to commonsense to believe in [free will] depending on God in its action, as it is depending onhim in its being” (28 Oct. 1646, AT 4:323*). Descartes answered, in a passage quotedpreviously, that the independence that renders our free action meritorious is “of quiteanother kind” than the dependence whereby all created things are subject to God(3 Nov. 1645, AT 4:333). In a subsequent letter, he attempts to illustrate the consis-tency of the independence of our free actions with their dependence on God in termsof a story about a king who causes a free violation of his prohibition of dueling. Theking causes this violation by ordering two individuals to meet who he “knows withcertainty” would duel if they met. But this action of the king “does not prevent theirfighting when they meet from being as voluntary and as free as if they met on some


64. See, for instance, Gorham’s claim that in his correspondence with Elisabeth, Descartes“implies that our free actions are determined by God” (Gorham 2004, 416), and Chappell’sclaim that this correspondence indicates that God is “responsible for every volition by con-curring in all actions of mind” (Chappell 1994, 184).

other occasion and he had known nothing about it.” Likewise, according toDescartes, God

knew exactly all the inclinations [inclinations] of our will; it is he himself whohas given them to us, it is also he who has disposed all other things external to ussuch that such and such objects present themselves to our senses at such and suchtimes, on the occasion of which he has known that our free decision woulddetermine us to such and such a thing; and he has willed it such, but he has notwilled that it be constrained to that [thing]. (Jan. 1646, AT 4:353–54)

We seem to have here the Molinist position that God causes our free action merelyby producing a will that he knew would freely yield a certain action in certain cir-cumstances. Contrary to the remarks in the Principles, this sort of dependence ondivine knowledge seems to be perfectly compatible with the claim that the actionsthemselves are undetermined.65

But though a Molinist reading appears to be warranted by the emphasis inDescartes’s letter on divine knowledge of our inclinations,66 such a reading cannotreflect his most considered position. For as I have indicated, Descartes’s created truthdoctrine requires, contrary to the Molinist view in Suárez, that God has created eventruths concerning the actions that derive from our free will (see §5.1.1 (ii)). In theTreatise on Divine Grace, as we know, Suárez explained divine knowledge of freehuman action by appealing to the presence in God of an uncreated idea of thehabituo in us that would yield certain free actions in particular circumstances.Despite the fact that this habituo seems to be similar to the inclinations thatDescartes took to be involved in God’s knowledge of our free action, Descartes can-not hold that an uncreated idea provides the foundation for this knowledge. The cre-ated truth doctrine in fact commits him to the position that God is the efficient causeof whatever it is that allows him to know that certain actions follow from the incli-nations of our will.

Once we read the Elisabeth correspondence in light of the created truth doctrine,then, we can see that there is still something fundamentally mysterious there con-cerning the dependence of our free action on God. God must not only know whichactions follow from the inclinations in certain circumstances and produce a will withthose inclinations in those circumstances; in addition, he must be the efficient causeof the truth that a will with those inclinations would produce such actions in thosecircumstances. We can no more comprehend how God could cause this sort of con-ditional truth than we can comprehend how God can bring it about that certain eter-nal truths derive necessarily from the essence of a triangle. And just as Descartes


65. At least insofar as we can make sense of infallible knowledge of undetermined action.As I indicate below, the mystery of how such knowledge is possible is no more eliminated byDescartes than it was by Suárez. However, this mystery does not presuppose a divine deter-mination of free action.

66. As I assumed in an earlier consideration of this letter in Schmaltz 1994, 17–19. I owemy appreciation of the inadequacy of the Molinist reading to the discussion in Ragland 2005,178–86.

emphasized that the power by which God has created the eternal truths exceeds ourcomprehension, so he could say the same about the power by which he determinedconditional truths concerning our free will.

In contrast to the case of the eternal truths, however, what renders the divine powerincomprehensible in the case of our free action is the fact that it leaves such actionundetermined. There is no similar problem of comprehension on the Dominican viewthat God renders determinate truths concerning free action that natural causes leaveindeterminate. Moreover, there is a sense in which Descartes’s created truth doctrineleaves fundamental features of the Jesuit theory of middle knowledge intact. From aDominican perspective, what is particularly troubling about Suárez’s version of thistheory is that God could know with certainty that a habituo in us would result in cer-tain free actions even though those actions do not follow from the nature of that fea-ture. But then what would be equally problematic from that perspective is Descartes’ssuggestion to Elisabeth that God could know that certain actions derive from theinclinations of our will even though such inclinations leave our action undetermined.The mystery here is of the same sort whether the idea that is the source of divineknowledge is uncreated, as in Suárez, or rather derives from God’s will, as inDescartes. For the mystery concerns not the ultimate source of the object of God’sknowledge of how we would freely act, but rather the manner in which such an objectcould allow for infallible knowledge of something that is itself undetermined. Despitesiding with the Dominicans in holding that divine causality is involved in fixing thetruth of future contingent propositions concerning our free action, then, Descartesremains on the side of the Suárez and the Jesuits in embracing the view that Godleaves such action in its undetermined state.

5.3.2. Created Truth and Conservationism

I have noted the reference in the discussion of human freedom in the Principles toGod’s power over all that “is or can be.” Descartes’s account of the divine creation ofeternal truths concerning what “can be” in fact broaches difficulties that are analogousto the difficulties that confront his view of the dependence of our free actions on God.In both cases, the central problem concerns the fact that the nature of the effectappears to preclude the sort of dependence on divine power that Descartes posited.Thus, the result of the created truth doctrine that eternal truths derive from God’sindifferent will seems to be in tension with the conclusion that these truths are them-selves necessary. There is the claim in Frankfurt that Descartes denies the necessityof the eternal truths (see §5.1.2), but this claim is belied by Descartes’s own insistenceon the fact that “it is because [God] wills the three angles of a triangle to be neces-sarily [necessario] equal to two right angles that this is now true and could not beotherwise [fieri aliter non potest]” (Sixth Replies, AT 7:432).67 Nevertheless, theimplication in Descartes that God is not necessitated in creating the eternal truthsseems to support Frankfurt’s claim that such truths ultimately are contingent.


67. See also the reference in the Mesland letter to the fact that there are “things that Godcould have rendered possible, but that he nonetheless willed to render impossible” (AT 4:118).

In the case of our free action, there is an analogous problem concerning the com-patibility of the claim that our free actions are undetermined with the conclusion thatthey derive “entirely” from God as their “efficient and total cause.”68 God’s deter-mination of our actions is supposed to involve his determination of truths concern-ing how we would freely act in certain circumstances. According to the Principles,however, these free actions are by their nature undetermined. Just as the fact thateternal truths are necessary seems to show that they cannot derive from a divine willthat is not itself necessitated, so the fact that our free actions are undetermined seemsto show that truths concerning them cannot be determined by God.

There is a solution in Descartes to the problem concerning created eternal truthsthat turns on the fact that such truths derive necessarily from something else that Godhas freely created, and thus that is not itself necessary.69 Descartes claims in the FifthMeditation, for instance, that eternal truths concerning a triangle follow from the“determinate nature, or essence, or form of a triangle that is immutable and eternal,and not invented by me or dependent on my mind” (AT 7:64). He even notes in thistext that this truth can no more be separated from this essence than existence can beseparated from the divine essence (AT 7:66). Certainly this is a striking claim giventhe position in Descartes that the derivation of God’s existence from his essence isindependent of the divine will. But he could still hold that the case of this derivationdiffers from the case of the derivation of truths concerning triangles insofar as theessence of triangularity, in contrast to the divine essence, has an efficient cause. ThusGod could create truths concerning triangles that are necessary by creating anessence that it was in his power not to create, and thus is not itself necessary, but thatis nonetheless such that these truths follow from it necessarily.70

There is an analogous solution to the problem of the divine determination of truthsconcerning our free action. The suggestion I find in the Elisabeth correspondence is


68. See the similar connection between these two problems in Della Rocca forthcoming.Whereas Della Rocca holds that Descartes sets both problems aside by appealing the divineincomprehensibility, though, I find in Descartes’s post-Principles comments, at least, the sug-gestion that there is more to say in the case of our free action.

69. In his account of the modal status of Descartes’s eternal truths, Curley has emphasizedthe claim in the Mesland letter that “even if God willed certain truths are necessary, that is notto say that he has willed them necessarily; for to will that they be necessary is somethingentirely different than willing them necessarily or being necessitated to will them” (AT 4:118).According to Curley, the suggestion here is that the truths are necessary but not necessarilynecessary (Curley 1984, 581–83). In terms of this account, we could take eternal truths thatare contingently necessary to be those that derive immediately from essences that are them-selves contingent, and eternal truths that are necessarily necessary to derive immediately fromthe one essence that is itself necessary, namely, the divine essence.

70. My suggestion that Descartes distinguished between the created essence and the truthsthat necessarily and immutably derive from it may seem to conflict with his remark in the1630 Mersenne correspondence that the essence that God creates “is nothing other than [n’estautre chose que] the eternal truths” (27 May 1630, AT 1:152). It may be that the later remarksin the Fifth Meditation require a distinction between essences and the truths deriving fromthem that is not fully present in this 1630 correspondence.

that God determines such truths by making it the case that the relevant actions followin a nondeterministic manner from the inclinations of our will. Having determinedthis sort of connection between the inclinations and the action, God is then in a posi-tion to know with certainty which volitions we would freely choose in certain cir-cumstances, even though our will is not determined to choose in this manner. Just asGod’s nonnecessitated action can produce necessary truths through the mediation ofessences, so his determination can result in truths concerning something that is itselfundetermined through the mediation of inclinations (see figure 5.3).

There is, however, one important respect in which Descartes’s account of the eter-nal truths differs from his account of free human action. Though he makes clear thatGod is the efficient cause of created eternal truths, Descartes mentions only creationwhen characterizing the production of such truths, and not conservation or concur-rence.71 In contrast, he uses concurrentist language in discussing God’s causal con-tribution to our free action. Most notably, he comments in the Fourth Meditation that

I must not complain that God concurs [concurrat] with me in choosing[eliciendos] those voluntary acts, or those judgments, in which I err; for these actsare all true and good, insofar as they depend on God, and it is in some mannermore perfect that I could choose them, than if I could not. (AT 7:60)

For Suárez, the claim that God concurs in our free decision indicates that he not onlyconserves us in existence as we act, but also produces our decision by means of a“per se and immediate concursus” that is identical to the action by which we pro-duce it (see §1.2.3 (ii)).72 So the question that confronts us now is whether Descartesintended to provide room for a similar sort of additional divine concursus in the caseof our free action.

I argued in chapter 3 that given Descartes’s emphasis on the constancy of theimmediate effects of God’s “ordinary concursus” to body–body interactions, we


Non-necessary divine creationof

Divine determinationof

necessary eternal truths truths concerning undeterminedfree action

immutable essences

ground

inclinations of will

ground

FIGURE 5.3 Eternal Truths and Free Action

71. The 1630 correspondence on the created truth doctrine includes the claim that God“from all eternity willed and understood them to be, and by that fact he created them” (ToMersenne, 27 May 1630, AT 1:152). If we take creation “from eternity” to preclude any sortof temporality, then there would be no room for continued conservation or concurrence, bothof which presuppose temporal effects.

72. In addition, as we know, Suárez holds that in the case of free action, God offers a con-cursus with refraining from making a choice or with deciding differently.

must conceive of this concursus in a conservationist manner. So also, I claimed inchapter 4 that God’s causal contribution to mind–body interaction consists not in hisimmediate production of the mental and bodily effects of that interaction, but ratherin his creation and conservation of the natures that make such interaction possible.This line of argument suggests that we should accept a conservationist account of theway in which Descartes thinks God has “willed and preordained” our free action.Moreover, the story of the duelers that Descartes provides in the Elisabeth corre-spondence seems to support this sort of account. For the dueling is dependent on theking not in the sense that he directly causes it, but in the sense that he indirectly pro-duces it by giving an order that he infallibly knew would result in the decision to duelthat involves an indeterministic sort of freedom. So also, free action is not said todepend on God in the sense that he produces the same action that free agents pro-duce. Instead, Descartes’s claim is that God wills that such agents have an inclina-tion that leads them to a certain action even though they are not “constrained to” thataction in the sense of being determined to produce it. We know that given the cre-ated truth doctrine, this sort of willing must involve more than God’s producinginclinations and circumstances that he prevolitionally knows will result in certainundetermined free actions in those circumstances. For God must also be the efficientcause of the conditional truth that the inclinations would yield the actions in thosecircumstances. Having created these truths, however, there is nothing that God needsto do to produce the free actions beyond creating and conserving a world in whichagents with the relevant inclinations exist in the appropriate circumstances.

Certain remarks from writings Descartes produced toward the end of his life rein-force the impression that he did not take God to directly produce our free action.There is, for instance, Descartes’s claim in a 1647 letter to Queen Christina ofSweden that “free decision [libre arbitre] is in itself the noblest thing that can be inus, and that renders us in a certain manner [en quelque façon] equal to him andseems to exempt us from being his subjects” (20 Nov. 1647, AT 5:85). The need forthe qualification that our will makes us only “in a certain manner” like God and thatit only “seems” to exempt us from being his subjects is evident given the createdtruth doctrine. For this doctrine requires not only that the divine will have a sort ofindifference that is impossible for us, but also that God determine all that “is or canbe” concerning even our free action. Nevertheless, insofar as it is in our control tochoose how we freely act, our free action is comparable to the acts that derive fromGod’s supremely indifferent will. And insofar as what God creates does not itselfdetermine truths concerning our free action, we are distinct from subjects that Goddetermines simply by creating and conserving with their natures in certain circum-stances. We therefore retain in this late letter the position from Descartes’s earlynotebook that “free decision” is one of the most distinctive of the “miracles” thatGod has produced.

Finally, there is Descartes’s claim in the 1649 Passions—quoted previously—thatreflection on divine providence leads us to consider “everything that affects us tooccur of necessity and as it were by fate [comme fatale],” with the notable exceptionof matters that God “has willed to depend on our free decision” (PS II.146, AT11:439). As in the letter to Christina, so here there is an emphasis on the fact that ourfree decision “renders us in some manner similar to God in making us masters of


ourselves” (III.152, AT 11:445). The suggestion in the Passions is that God allowsus to be masters of ourselves by leaving it to our will to determine its free actions.In light of the remarks to Elisabeth, we can take the more complete position here tobe that God creates and conserves our will with inclinations that allow God to knowwith certainty how we would act in certain circumstances but that leave our willundetermined with respect to such action. We still have something similar to themystery, deriving from Suárez, of how God could know our free action on the basisof his idea of a habituo of our will that does not determine this action. YetDescartes’s talk of a divine decree that dictates that our free action depend on ourwill alone reveals how far he is from the view in Suárez that God produces thisaction by means of a concursus that is identical to the action itself.


217

Conclusion

According to a standard narrative concerning the history of philosophy, Descartes’stheory of causation marks a sharp break from past conceptions of causality. In par-ticular, this story has it that Descartes set out on a new path by replacing the fourAristotelian causes prominent in scholastic natural philosophy with the efficientcauses required for his new mechanistic physics. There is admittedly a residue ofscholasticism, for instance, in his claim in the Third Meditation that a cause mustcontain its effect formally or eminently. However, such a claim can and has beenunderstood to reflect a merely superficial connection to past thought. For those whotake Descartes to herald the advent of modernity, what is novel in his theory ofcausation is more significant than what is borrowed.

There can be no doubt that this theory involves a significant break with thescholastic past. Having rejected material forms and qualities distinct from extensionand its modes, for instance, Descartes simply could not speak in the way the scholas-tics did of the containment of “forms” in bodily causes either actually or “virtually”in the more noble power to produce those forms. Moreover, it is certainly importantthat in one fell swoop, he eliminated from his physics the sort of final causality thatscholastics such as Suárez took to derive from God’s concursus with natural causes.The displacement of final causes in Descartes is nowhere more dramatic than in hisclaim that God has no ends when he acts as an efficient cause in creating eternal truths.

Nevertheless, it is important to recognize that the emphasis in Descartes on effi-cient causality did not emerge from history ex nihilo, but was anticipated in ascholastic reconceptualization of causation that culminated in the work of Suárez.We also need to remember that there were issues in scholasticism concerning effi-cient causation that Descartes’s new ontology did not eliminate entirely. There is, forinstance, the scholastic problem of explaining the bodily production of changes in anincorporeal intellect. To be sure, the immaterial mind that Descartes posited was a

res cogitans with sensations and feelings rather than, as on the standard scholasticview, a purely intellectual substance. Moreover, his paradigmatic case of the actionof body on mind is the production in the mind of sensory states rather than, as in thescholastics, the production in the intellect of intelligible species. Even so, Descartesretained from scholasticism both the sense that the immaterial is more noble than thematerial, and the axiom that what is less noble cannot suffice to produce a change inwhat is more noble. He thereby also inherited the difficulty of explaining the causalcontribution of something that is material to changes in something immaterial. Atleast some causal aspects of the “mind–body problem” in Descartes were not new,then, but merely recycled versions of problems that previous thinkers confronted.

What I take to reveal most clearly the significance of Descartes’s connections toscholastic accounts of causality, however, is his emphasis on the foundational rolefor his physics of the claim that God conserves the same quantity of motion in thematerial world by means of his “ordinary concursus.” I have argued that such a claimprovides the primary clue for discerning fundamental elements of Descartes’s theoryof causation. Though the claim itself may seem to draw on a traditional scholasticconcurrentism, a more careful consideration of its scholastic context serves to revealits radical implication that God’s contribution to the natural causal order of the mate-rial world is exhausted by his creation and continued conservation of that world. Incontrast to the view in scholastics such as Suárez that divine concursus is identicalto the actions by which secondary causes produce particular changes, Descartes indi-cated that this concursus produces an effect that is constant and thus distinct fromthe changes that are attributable only to secondary causes. There is a sort of causaldivision of labor that is closer to the conservationist views of Durandus than to any-thing in the work of the scholastic concurrentists.

Of course, neither Durandus nor his concurrentist opponents confronted the prob-lem in Descartes of explaining how a mere res extensa could be an efficient cause.But though some have thought that this problem led Descartes to give up on genuinebodily causes and to embrace a kind of occasionalism, the texts seem to me to indi-cate otherwise. In the case of body–body interaction, I have found an explanation ofthe “forces for acting” in Descartes’s physics in terms of special features of the dura-tion of moving bodies. Such forces are supposed to allow the moving bodies to be(total or adequate) causes of particular changes that leave intact the total quantityof motion that God continually conserves as universal and primary cause. In con-trast, it is not bodily duration that matters most in the case of body-to-mind action,but rather the special nature of the human mind. Descartes’s most consideredposition is that this nature is such that certain bodily motions are able to affect thismind as (partial) efficient causes of its sensations. What is missing from Descartes,however, is any appeal to the immediate activity of God in the production of sensa-tion. Though the absence of such an appeal would be surprising on an occasionalistor even a concurrentist reading of his theory of causation, it is to be expected on theconservationist reading that informs this study.

The problem of attributing causal power to a merely extended thing of course pro-vides no barrier to the claim in Descartes that finite minds are real causes. Moreover,his acceptance of the scholastic axiom that the effect cannot be more perfect than itscause does not produce the sort of difficulty for mind-to-body action that it produces


for body-to-mind action. Nevertheless, Descartes’s insistence on the immateriality ofmind does broach the scholastic problem of explaining how something that is incor-poreal can be “present” to something that is corporeal in order to act on it. Descartescould have simply dismissed this problem, as Durandus did, by claiming that a causeneed not be present at a place in order to act there. It is surely significant that he didnot take this route, but instead appealed to the fact that an immaterial mind has an“extension of power” in virtue of which it is present where it acts.

The suggestion in Descartes that the mind is the cause of its free volitionsbroaches different issues from scholasticism. There is in the background the Jesuitview in Suárez that the will that produces such volitions must have the power to dootherwise. Descartes started from a position that was not entirely friendly to such aview when he emphasized in the Fourth Meditation that we are most free when ouracts follow necessarily from the orientation of our will toward the true (and good).However, he was increasingly drawn to the view that our freedom requires the mas-tery of our will over its action, particularly with respect to pursuit of the good. Hepersisted in his commitment to the presence of this sort of mastery even while hold-ing that God is not only the “universal and total cause” of our free action, but alsothe indifferent cause of truths concerning that action. Yet though these claims regard-ing God’s causal contribution to our free action in some respects go beyond anythingin Suárez, they stop short of Suárez’s concurrentist conclusion that God concurs withour will as the immediate efficient cause of such action. On the conservationist viewsuggested in Descartes, God merely leaves it to us to determine how our will freelydetermines our own volitions.

I have been concerned, then, to argue that a conservationist framework providesa means both of deciphering and of interrelating Descartes’s discussions of variousforms of causal interaction. However, I also have indicated that this frameworkallows us to appreciate certain problems internal to his system. I have mentioned twoparticular problems, the first of which concerns body–body interaction, and the sec-ond mind-to-body action. The first problem is connected to the objection, prominenteven among early modern sympathizers, that Descartes’s physics fails to provide asatisfactory account of the “force for resisting” that bodies at rest exhibit in collision.A conservationist understanding of this physics reveals that this force is problematicprecisely because it cannot be a constant result of God’s ordinary concursus. Rather,the force for resisting involves a momentary impulse that God would have to createby means of an act distinct from the continuation of his original act of creating thematerial world. This is one act too many given Descartes’s own view that particularchanges in matter derive from secondary causes rather than from God as primarycause.

The second problem for Descartes is connected to the objection, prominent inLeibniz, that the admission that finite minds can produce new motion in the worldconflicts with the principle in Cartesian physics that requires the conservation of thetotal quantity of motion. I have argued that Leibniz was wrong to suggest thatDescartes himself was led by this difficulty to conclude that finite minds are limitedto acting on body by changing the direction of its motions. Nevertheless, there is aclear philosophical motivation for this conclusion given Descartes’s understandingof God’s ordinary concursus in terms of his conservation of the quantity of matter

Conclusion 219

and motion he originally created. Just as this concursus cannot explain the additionalimpulses required by the force for resisting, so it cannot explain the conservation ofany quantity of motion that other minds add to the world subsequent to creation.

It might be thought that these problems with the force for resisting and the intro-duction of new motion provide reason to question my conservationist interpretationof Descartes. However, the problems derive directly from his own insistence thatGod’s action as primary cause consists merely in the continuation of his original actof creating the world. Far from leading to occasionalism, such an insistence shiftsthe burden for explaining particular changes in nature from God to secondary causes.There is no similar shift in scholastic concurrentism, which requires immediatedivine involvement in all causal operations. Nevertheless, we cannot understand thisshift adequately, or perhaps even recognize it in the first place, if we fail to take seri-ously the scholastic context of Descartes’s theory of causation.


221

Works Cited

Adams, Robert M. 1990. Middle Knowledge and the Problem of Evil. In The Problem of Evil,ed. M. M. Adams and R. M. Adams, 110–25. Oxford: Oxford University Press.

Alanen. Lilli. 1996. Reconsidering Descartes’ Notion of the Mind-Body Union. Synthese 106:3–20.

——. 2003. Descartes’s Concept of Mind. Cambridge, MA: Harvard University Press.Alquié, Ferdinand. 1950. La découverte métaphysique de l’homme chez Descartes. Paris:

Presses Universitaires de France.Ariew, Roger. 1999. Descartes and the Last Scholastics. Ithaca, NY: Cornell University Press.Aristotle. 1984. The Complete Works of Aristotle. Ed. J. Barnes. 2 vols. Princeton, NJ:

Princeton University Press.Armogathe, Jean-Robert. 1995. Caterus’ Objections to God. In Descartes and His

Contemporaries: Meditations, Objections, and Replies, ed. R. Ariew and M. Grene,34–43. Chicago: University of Chicago Press.

Arthur, Richard. 1988. Continuous Creation, Continuous Time: A Refutation of the AllegedDiscontinuity of Cartesian Time. Journal of the History of Philosophy 26:349–75.

Averroes (Ibn Rushd). 1969. Averroes’ Tahafut al-tahafut: Incoherence of the incoherence.Trans. S. van den Bergh. 2 vols. London: Luzak and Company.

Baker, Gordon, and Katherine Morris. 1996. Descartes’ Dualism. London: Routledge.Battail, Jean-François. 1973. L’avocat philosophe: Géraud de Cordemoy, 1626–1684. The

Hague: M. Nijhoff.Bedau, Mark. 1986. Cartesian Interaction. Midwest Studies in Philosophy 10:483–502.Beeckman, Isaac. 1939. Journal tenu par Isaac Beeckman de 1604 à 1634. The Hague:

M. Nijhoff.Belaval, Yvon. 1960. Leibniz critique de Descartes. Paris: Gallimard.Bennett, Jonathan. 1994. Descartes’s Theory of Modality. Philosophical Review 103:639–67.——. 2001. Learning from Six Philosophers: Descartes, Spinoza, Leibniz, Locke, Berkeley,

Hume. 2 vols. Oxford: Clarendon Press.Bergson, Henri. 1907. L’évolution créatrice. Paris: Alcan.

Beyssade, Jean-Marie. 1979. La philosophie première de Descartes. Paris: Flammarion.Beyssade, Michelle. 1994. Descartes’s Doctrine of Freedom: Difference between the

French and Latin Texts of the Fourth Meditation. In Reason, Will, and Sensation:Studies in Descartes’s Metaphysics, ed. J. Cottingham, 191–206. Oxford: ClarendonPress, 1994.

——. 1996. Des Principia aux Principes: Variations sur la liberté. In Descartes PrincipiaPhilosophiae (1644–1994) ed. J.-R. Armogathe and G. Belgioioso, 37–52. Naples:Vivarium.

Blackwell, Richard J. 1966. Descartes’ Laws of Motion. Isis 57:220–34.Bourke, Vernon. 1970. A History of Ethics. 2 vols. Garden City, NY: Image Books. Brodrick, James. 1961. Robert Bellarmine: Saint and Scholar. Westminster, MD: Newman

Press.Broughton, Janet. 1986. Adequate Causes and Natural Change in Descartes’ Philosophy. In

Human Nature and Natural Knowledge: Essays Presented to Marjorie Grene on theOccasion of Her Seventy-fifth Birthday, ed. A. Donagan, A. N. Perovich Jr., and M. V.Wedin, 107–27. Dordrecht: Reidel.

Broughton, Janet, and Ruth Mattern. 1978. Reinterpreting Descartes on the Notion of theUnion of Mind and Body. Journal of the History of Philosophy 16:23–31.

Campbell, Joseph. 1999. Descartes on Spontaneity, Indifference, and Alternatives. In NewEssays on the Rationalists, ed. R. Gennaro and C. Huenemann 179–99. New York:Oxford University Press, 1999.

Carraud, Vincent. 2002. Causa sive ration: La raison de la cause, de Suarez à Leibniz. Paris:Presses Universitaires de France.

Chappell, Vere. 1994. Descartes’s Compatibilism. In Reason, Will, and Sensation: Studies inDescartes’s Metaphysics, ed. J. Cottingham, 177–90. Oxford: Clarendon Press, 1994.

Chomsky, Noam. 1966. Cartesian Linguistics: A Chapter in the History of RationalistThought. New York: Harper & Row.

Clarke, Desmond. 1982. Descartes’ Philosophy of Science. University Park: PennsylvaniaState University Press.

——. 1996. The Concept of Vis in Part III of the Principia. In Descartes: PrincipiaPhilosophiae (1644–1994) ed. J.-R. Armogathe and G. Belgioioso, 321–39. Naples:Vivarium.

Clatterbaugh, Kenneth. 1980. Descartes’s Causal Likeness Principle. Philosophical Review89:379–402.

——. 1995. Cartesian Causality, Explanation and Divine Concurrence. History of PhilosophyQuarterly 12:195–207.

——. 1999. The Causation Debate in Modern Philosophy, 1637–1739. New York: Routledge.

Clauberg, Johannes. [1691] 1968. Opera Omnia Philosophica. Reprint, Hildesheim: GeorgOlms.

Clerselier, Claude. 1667. Letter to Louis de la Forge, 4 December 1660. In Lettres de Mr. Descartes, vol. 3, ed. C. Clerselier, 640–46. Paris: Angot.

Cohen, I. Bernard. 1964. “Quantum in se est”: Newton’s Concept of Inertia in Relation toDescartes and Lucretius. Notes and Records of the Royal Society of London 19:131–55.

Cohen, Sheldon. 1982. St. Thomas Aquinas on the Immaterial Reception of Sensible Forms.Philosophical Review 91:193–209.

Costabel, Pierre. 1967. Essai critique sur quelques concepts de la mécanique cartésienne.Archives internationals d’histoire des sciences 20:235–52.

Cottingham, John. 1985. Cartesian Trialism. Mind 94:218–30.——. 1988. The Intellect, the Will, and the Passions: Spinoza’s Critique of Descartes. Journal

of the History of Philosophy 26:239–57.

222 WORKS CITED

——. 1997. Force, Motion and Causality: More’s Critique of Descartes. In The CambridgePlatonists in Philosophical Context: Politics, Metaphysics and Religion, ed. G. A.J. Rogers, J. M. Vienne, and Y. C. Zarka, 159–71. Dordrecht: Kluwer.

Courtine, Jean-François. 1990. Suarez et le système de la metaphysique. Paris: PressesUniversitaires de France.

Curley, E. M. 1984. Descartes on the Creation of the Eternal Truths. Philosophical Review93:569–97.

Dalbriez, Roland. 1929. Les sources scolastiques de la théorie cartésienne de l’être objectif.Revue d’histoire de la philosophie 3:464–72.

Della Rocca, Michael. 1999. “If a Body Meet a Body”: Descartes on Body-Body Causation.In New Essays on the Rationalists, ed. R. Gennaro and C. Huenemann, 48–81. NewYork: Oxford University Press, 1999.

——. Forthcoming. Causation without Intelligibility and Causation without God in Descartes.In The Blackwell Companion to Descartes, ed. J. Broughton and J. Carriero. Oxford:Blackwell.

Des Chene, Dennis. 1996. Physiologia: Natural Philosophy in Late Aristotelian andCartesian Thought. Ithaca, NY: Cornell University Press.

——. 2000a. Life’s Form: Late Aristotelian Conceptions of the Soul. Ithaca, NY: CornellUniversity Press.

——. 2000b. On Laws and Ends: A Response to Hattab and Menn. Perspectives on Science8:144–63.

——. 2001. Spirits and Clocks: Machine and Organism in Descartes. Ithaca, NY: CornellUniversity Press.

Descartes, René. 1964–74. Œuvres de Descartes. Ed. C. Adam and P. Tannery, nouvelleprésentation. 11 vols. Paris: Vrin.

——. 1984–85. The Philosophical Writings of Descartes. Ed. and trans. J. Cottingham,R. Stoothoff, and D. Murdoch. 2 vols. Cambridge: Cambridge University Press.

——. 1989. The Philosophical Writings of Descartes. Vol. 3, The Correspondence. Ed. andtrans. J. Cottingham, R. Stoothoff, D. Murdoch, and A. Kenny. Cambridge: CambridgeUniversity Press.

Du Hamel, Jean. 1692. Réflexions critique sur le système cartesien de la philosophie deMr. Regis. Paris: Couterot.

Dubarle, Dominique. 1964. Sur la notion cartésienne de quantité de mouvement. MélangesAlexandre Koyré 2:118–28.

Duhem, Pierre. 1955. Études sur Léonard de Vinci. Paris: F. de Nobele.Durandus of Saint Pourçain. [1571] 1964. In Petri Lombardi Sententias Theologicas

Commentariorum libri IIII. 2 vols. Reprint, Ridgewood, NJ: Gregg Press.——. N.d. Does God Act Immediately in Every Action of a Creature? (On the Theological

Sentences, bk. 2, dist. 1. q. 5). Trans. A. J. Freddoso. http://www.nd.edu/~afreddos/translat/ duran215.htm.

Eardley, Peter. 2006. The Foundations of Freedom in Later Medieval Philosophy: Giles ofRome. Journal of the History of Philosophy 44:353–76.

Fontenelle, Bernard le Bovier de. 1989–2001. Œuvres completes. Ed. A. Niderst. 9 vols. Paris:Fayard.

Frank, Richard M. 1992. Creation and the Cosmic System: Al-Ghaza-li- and Avicenna.Heidelberg: Carl Winter.

Frankfurt, Harry. 1977. Descartes on the Creation of the Eternal Truths. Philosophical Review86:36–57.

Freddoso, Alfred. 1991. God’s General Concurrence with Secondary Causes: WhyConservation Is Not Enough. Philosophical Perspectives 5:553–85.

Works Cited 223

http://www.nd.edu/~afreddos/translat/duran215.htm

http://www.nd.edu/~afreddos/translat/duran215.htm

——. 1994. God’s General Concurrence with Secondary Causes: Pitfalls and Prospects.American Catholic Philosophical Quarterly 67:131–56.

——. 2001. Suarez on God’s Causal Involvement in Sinful Acts. In The Problem of Evil inEarly Modern Philosophy, ed. E. Kremer and M. Latzer, 10–34. Toronto: University ofToronto Press.

Frede, Michael. 1980. The Original Notion of Cause. In Doubt and Dogmatism: Studies inHellenistic Epistemology, ed. M. Schofield, M. Burnyeat, and J. Barnes, 217–49. Oxford:Clarendon Press.

Gabbey, Alan. 1980. Force and Inertia in the Seventeenth Century: Descartes and Newton. InDescartes: Philosophy, Mathematics and Physics, ed. S. Gaukroger, 230–320. Sussex:Harvester Press.

——. 1982. Philosophia Catesiana Triumphata: Henry More (1646–1671). In Problems ofCartesianism, ed. T. M. Lennon, J. M. Nichols, and J. W. Davis, 171–250. Kingston andMontreal: McGill–Queen’s University Press.

Galileo (Galilei). 1967. Dialogue concerning the Two Chief World Systems. Trans. S. Drake.Los Angeles and Berkeley: University of California Press.

Garber, Daniel. 1983. Mind, Body, and the Laws of Nature in Descartes and Leibniz. MidwestStudies in Philosophy 8:105–33.

——. 1992. Descartes’ Metaphysical Physics. Chicago: University of Chicago Press.——. 1993. Descartes and Occasionalism. In Causation in Early Modern Philosophy, ed. S.

Nadler, 9–26. University Park: Pennsylvania State University Press, 1993.Geulincx, Arnold. 1891–93. Antverpiensis Opera Philosophica. Ed. J. Land. 3 vols. The

Hague: M. Nijhoff. Ghaz-ali-, Abu- Ha-mid al-. 1958. Tahafut al-falasifah. Incoherence of the philosophers. Trans.

Sabih Ahmad Karmali. Lahore: Pakistan Philosophical Congress.Gilson, Étienne. 1913a. Index scolatico-cartésien. Paris: Alcan. ——. 1913b. La liberté chez Descartes. Paris: Alcan. ——. 1925. René Descartes: Discours de la méthode, texte et commentaire. Paris: Vrin.——. 1930. Études sur le role de la pensé médiévale dans la formation du système cartésien.

Paris: Vrin.——. 1986. Notes pour l’histoire de la cause efficiente. In É. Gilson, Études médiévale,

167–91. Paris: Vrin.Gorham, Geoffrey. 1999. Causation and Similarity in Descartes. In New Essays on the

Rationalists, ed. R. Gennaro and C. Huenemann, 296–309. New York: Oxford UniversityPress, 1999.

——. 2002. Descartes on the Innateness of All Ideas. Canadian Journal of Philosophy32:355–88.

——. 2003. Descartes’s Dilemma of Eminent Containment. Dialogue: CanadianPhilosophical Review 42:3–25.

——. 2004. Cartesian Causation: Continuous, Instantaneous, Overdetermined. Journal of theHistory of Philosophy 42:389–423.

——. 2006. Descartes on Persistence and Temporal Parts. In Time and Identity, ed.J. Campbell and M. O’Rourke. Cambridge, MA: MIT Press.

Gouhier, Henri. 1926. La vocation de Malebranche. Paris: Vrin.Gueroult, Martial. 1953. Descartes selon l’ordre des raisons. 2 vols. Paris: Aubier. ——. 1968. Spinoza, I: Dieu (Ethique, I). Hildesheim: Georg Olms.——. 1980. The Metaphysics and Physics of Force in Descartes. In Descartes: Philosophy,

Mathematics and Physics, ed. S. Gaukroger, 196–229. Sussex: Harvester Press.Haldane, John. 1983. Aquinas on Sense Perception. Philosophical Review 92:233–39.

224 WORKS CITED

Haslanger, Sally. 2003. Persistence through Time. In The Oxford Handbook of Metaphysics,ed. M. Loux, 315–54. Oxford: Oxford University Press.

Hatfield, Gary. 1979. Force (God) in Descartes’ Physics. Studies in History and Philosophy ofScience 10:113–40.

——. 2000. Descartes’ Naturalism about the Mental. In Descartes’ Natural Philosophy, eds.S. Gaukroger, J. Schuster, and J. Sutton, 630–58. London: Routledge, 2000.

——. 2003. Routledge Philosophy Guide Book to Descartes and the Meditations. London:Routledge.

Hattab, Helen. 2000. The Problem of Secondary Causation in Descartes: A Response to DesChene. Perspectives on Science 8:93–118.

——. 2003. Conflicting Causalities: The Jesuits, Their Opponents, and Descartes on theCausality of the Efficient Cause. Oxford Studies in Early Modern Philosophy 1:1–22.

——. 2007. Concurrence or Divergence? Reconciling Descartes’s Science with HisMetaphysics. Journal of the History of Philosophy 45:49–78.

Hoffman, Paul. 1986. The Unity of Descartes’ Man. Philosophical Review 95:339–70.——. 1990. St. Thomas Aquinas on the Halfway Status of Sensible Being. Philosophical

Review 99:73–92.——. 1999. Cartesian Composites. Journal of the History of Philosophy 37:251–70.——. 2003. The Passions and Freedom of the Will. In Passion and Virtue in Descartes, ed. B.

Williston and A. Gombay, 261–99. New York: Humanity Books.Hutton, Sarah. 2005. Women Philosophers and the Early Reception of Descartes: Anne Conway

and Princess Elisabeth. In Receptions of Descartes: Cartesianism and Anti-Cartesianismin Early Modern Europe, ed. T. M. Schmaltz, 3–23. London: Routledge, 2005.

Iribarren, Isabel. 2005. Durandus of St Pourçain: A Dominican Theologian in the Shadow ofAquinas. Oxford: Oxford University Press.

Jesseph, Douglas. 2005. Mechanism, Skepticism, and Witchcraft: More and Glanvill on the Failuresof the Cartesian Philosophy. In Receptions of Descartes: Cartesianism and Anti-Cartesianismin Early Modern Europe, ed. T. M. Schmaltz, 199–217. London: Routledge, 2005.

Jolley, Nicholas. 1987. Descartes and the Action of Body on Mind. Studia Leibniziana19:41–53.

Kaufman, Daniel. 2000. Descartes on the Objective Reality of Materially False Ideas. PacificPhilosophical Quarterly 81:385–408.

Keeling, S. V. 1968. Descartes. 2nd ed. Oxford: Oxford University Press.Kenny, Anthony. 1972. Descartes on the Will. In Cartesian Studies, ed. R. J. Butler, 1–31.

Oxford: Blackwell.Kent, Bonnie. 1995. Virtues of the Will: The Transformation of Ethics in the Late Thirteenth

Century. Washington, DC: Catholic University Press.King, Peter. 2003. Scotus on Metaphysics. In The Cambridge Companion to Duns Scotus, ed.

T. Williams, 15–68. Cambridge: Cambridge University Press.Laporte, Jean. 1928. La finalité chez Descartes. Revue d’histoire de la philosophie 4:366–96.——. 1950. Le rationalisme de Descartes. Paris: Presses Universitaires de France.Leibniz, G. W. 1969. Philosophical Papers and Letters. 2nd ed. Ed. L. E. Loemker. Dordrecht:

Reidel.——. [1875] 1978. Die Philosophischen Schriften. Ed. C. I. Gerhardt. 7 vols. Reprint,

Hildesheim: Georg Olms. ——. 1981. New Essays on Human Understanding. Ed. and trans. P. Remnant and J. Bennett.

Cambridge: Cambridge University Press.——. [1967] 1985a. The Leibniz-Arnauld Correspondence. Ed. H. T. Mason. Reprint,

New York: Garland.

Works Cited 225

——. 1985b. Theodicy: Essays on the Goodness of God, the Freedom of Man and the Originof Evil. Trans. A. Farrer. La Salle, IL: Open Court.

——. 1989. Philosophical Essays. Ed. and trans. R. Ariew and D. Garber. Indianapolis:Hackett.

Lennon, Thomas M. 2007. The Eleatic Descartes. Journal of the History of Philosophy45:29–47.

Lerner, Ralph, and Muhsin Mahdi, eds. 1995. Medieval Political Philosophy: A Sourcebook.Ithaca, NY: Cornell University Press.

Levy, Ken. 2005. Is Descartes a Temporal Atomist? British Journal for the History ofPhilosophy 13:627–74.

Lloyd, A. C. 1976. The Principle that the Cause Is Greater than its Effect. Phronesis21:146–56.

Locke, John. 1975. An Essay concerning Human Understanding. Ed. P. H. Nidditch. Oxford:Clarendon Press.

Loeb, Louis. 1981. From Descartes to Hume: Continental Metaphysics and the Developmentof Modern Philosophy. Ithaca, NY: Cornell University Press,

——. 1985. Replies to Daisie Radner’s “Is There a Problem of Cartesian Interaction?”: IsThere a Problem of Cartesian Interaction? Journal of the History of Philosophy23:227–31.

——. 2005. The Mind-Body Union, Interaction, and Subsumption. In Early ModernPhilosophy: Mind, Matter, and Metaphysics, ed. C. Mercer and E. O’Neill, 65–85.Oxford: Oxford University Press.

Machamer, Peter. 1976. Causality and Explanation in Descartes’ Natural Philosophy. InMotion and Time, Space and Matter: Interrelations in the History and Philosophy ofScience, ed. P. Machamer and R. Turnbull, 168–99. Columbus: Ohio State UniversityPress.

Maier, Anneliese. 1982. On the Threshold of Exact Science. Trans. S. Sargent. Philadelphia:University of Pennsylvania Press.

Maimonides, Moses. 1963. The Guide of the Perplexed. Trans. S. Pines. 2 vols. Chicago:University of Chicago Press.

Malebranche, Nicolas. 1958–84. Œuvres complètes de Malebranche. Ed. A. Robinet. 20 vols.Paris: Vrin.

——. 1997. The Search after Truth and Elucidations of the Search after Truth. Ed. and trans.T. M. Lennon and P. J. Olscamp. Cambridge: Cambridge University Press.

Marion, Jean-Luc. 1991. Sur la théologie blanche de Descartes. 2nd ed. Paris: PressesUniversitaires de France.

——. 1996. Questions cartésiennes II. Sur l’ego et sur Dieu. Paris: Presses Universitaires deFrance.

Marlin, Randal. 1986. Cartesian Freedom and the Problem of the Mesland Letters. In EarlyModern Philosophy: Essays in Honor of Robert F. McRae, ed. G. J. D. Moyal andS. Tweyman, 195–215. New York: Caravan Books.

Marmura, Michael E. 1981. Al-Gh-azli-’s Second Causal Theory in the 17th Discussion of theTah-afut. In Islamic Philosophy and Mysticism, ed. P. Morewedge, 85–112. Delmar, NY:Caravan Books.

——. 1995. Ghaz-alian Causes and Intermediaries. American Oriental Society 115:89–100.Mattern, Ruth. 1978. Descartes’s Correspondence with Elisabeth: Conceiving Both the Union

and the Distinctness of Mind and Body. In Descartes: Critical and Interpretive Essays,ed. M. Hooker, 212–22. Baltimore: Johns Hopkins University Press.

McLaughlin, Peter. 1993. Descartes on Mind-Body Interaction and the Conservation ofMotion. Philosophical Review 102:155–82.

226 WORKS CITED

——. 2000. Force, Determination and Impact. In Descartes’ Natural Philosophy, eds.S. Gaukroger, J. Schuster, and J. Sutton, 81–112. London: Routledge, 2000.

Menn, Stephen. 1990. Descartes and Some Predecessors on the Divine Conservation ofMotion. Synthese 83:215–38.

——. 1997. Suárez, Nominalism, and Modes. In Hispanic Philosophy in the Age ofDiscovery, ed. K. White, 226–56. Washington, DC: Catholic University Press.

——. 2000. On Dennis Des Chene’s Physiologia. Perspectives on Science 8:119–43.Molina, Luis de. 1988. On Divine Foreknowledge (Part IV of the Concordia). Trans.

A. J. Freddoso. Ithaca, NY: Cornell University Press.More, Henry. 1995. Henry More’s Manual of Metaphysics; A Translation of the Enchiridium

metaphysicum (1679) with an Introduction and Notes. Ed. A. Jacob. Hildesheim: GeorgOlms.

Nadler, Steven. 1994. Descartes and Occasional Causation. British Journal for the History ofPhilosophy 2:35–54.

——. 1996. “No Necessary Connection”: The Medieval Roots of the Occasionalist Roots ofHume. Monist 79:448–66.

——. 1997. Occasionalism and the Mind-Body Problem. In Studies in Seventeenth-CenturyPhilosophy, ed. M.A. Stewart, 75–95. Oxford: Clarendon Press, 1997.

Newton, Isaac. 1999. The Prinicipia: Mathematical Principles of Natural Philosophy. Trans.I. B. Cohen and A. Whitman. Berkeley and Los Angeles: University of California Press.

Normore, Calvin. 1985. Divine Omniscience, Omnipotence, and Future Contingents: AnOverview. In Divine Omniscience and Omnipotence in Medieval Philosophy, ed.T. Rudavsky, 3–22. Dordrecht: Reidel.

——. 1986. Meaning and Objective Being: Descartes and His Sources. In Essays onDescartes’ Meditations, ed. A. O. Rorty, 223–41. Berkeley and Los Angeles: Universityof California Press, 1986.

Norris, John. 1693. Reason and Religion, or, The grounds and measures of devotion consid-er’d from the nature of God, and the nature of man, in several contemplations. With exer-cises of devotion applied to every contemplation. London: S. Manship.

Nye, Andrea. 1999. The Princess and the Philosopher: Letters of Elisabeth of the Palatine toRené Descartes. Lanham, MD: Rowman and Littlefield.

Olivo, Gilles. 1997. L’efficience en cause: Suárez, Descartes et la question de la causalité. InDescartes et le Moyen Âge, ed. J. Biard and R. Rashed, 91–105. Paris: Vrin.

O’Neill, Eileen. 1987. Mind-Body Interaction and Metaphysical Consistency: A Defense ofDescartes. Journal of the History of Philosophy 25:227–45.

——. 1993. Influxus physicus. In Causation in Early Modern Philosophy, ed. S. Nadler.,27–55. University Park: Pennsylvania State University Press, 1993.

Osborne, Thomas. 2006. Thomist Premotion and Contemporary Philosophy of Religion. Novaet Vetera 4:607–32.

Pasnau, Robert. 2002. Thomas Aquinas on Human Nature: A Philosophical Study of Summatheologiae Ia 75–89. Cambridge: Cambridge University Press.

——. 2004. Form, Substance, and Mechanism. Philosophical Review 113:31–88.Perler, Dominik, and Ulrich Rudolph. 2000. Occasionalismus: Theorien der Kausalität im

arabisch-islamischen und im europäischen Denken. Göttingen: Vandenhoeck and Ruprecht.Pessin, Andrew. 2003. Descartes’s Nomic Concurrentism: Finite Causation and Divine

Concurrence. Journal of the History of Philosophy 41:25–49.Prendergast, Thomas L. 1972. Descartes and the Relativity of Motion. Modern Schoolman

49:64–72.Prost, Joseph. 1907. Essai sur l’atomisme et l’occasionalisme dans l’école cartésienne. Paris:

Paulin.

Works Cited 227

Quinn, Philip L. 1988. Divine Conservation, Secondary Causes, and Occasionalism. In Divineand Human Action, ed. T. V. Morris, 50–73. Ithaca, NY: Cornell University Press.

Radner, Daisie. 1971. Descartes’ Notion of the Union of Mind and Body. Journal of theHistory of Philosophy 9:159–71.

——. 1985a. Is There a Problem of Cartesian Interaction? Journal of the History ofPhilosophy 23:35–50.

——. 1985b. Rejoinder to Richardson and Loeb. Journal of the History of Philosophy23:232–36.

Ragland, C. P. 2005. Descartes on Providence and Human Freedom. Archiv für Geschichte derPhilosophie 87:159–88.

——. 2006a. Descartes on the Principle of Alternative Possibilities. Journal of the History ofPhilosophy 44:377–94.

——. 2006b. Is Descartes a Libertarian? Oxford Studies in Early Modern Philosophy 1:57–90.Regis, Pierre-Sylvain. 1692. Réponses aux Réflexions critique de M. Du Hamel sur le système

de M. Regis. Paris: Cusson.——. 1996. L’Usage de la raison et de la foy, ou l’accord de la foy et de la raison. Ed. J.-R.

Armogathe. Paris: Fayard.Reid, Jasper. 2007. The Evolution of Henry More’s Theory of Divine Absolute Space. Journal

of the History of Philosophy 45:79–102.——. 2008. The Spatial Presence of Spirits among the Cartesians. Journal of the History of

Philosophy 46.Remnant, Peter. 1979. Descartes: Body and Soul. Canadian Journal of Philosophy 9:377–86.Renault, Laurence. 2000. La réalité objective dans les premières objections aux Méditations

métaphysiques: Ockham contra Descartes. Revue de métaphysique et de morale103:29–38.

Revius, Jacobus. 1643. Suarez repurgatus, sive, Syllabus Disputationum metaphysicarumFrancisci Suarez Societatis Iesu theologi. Leiden: F. Hegerum.

——. 1650. Statera philosophiae Cartesianae qua Principorum ejus falsitas, & domatumimpuritas expenditur ac castigatur. Leiden: P. Leffen.

Richardson, Robert. 1982. The “Scandal” of Cartesian Interactionism. Mind 91:20–37.——. 1985. Replies to Daisie Radner’s “Is There a Problem of Cartesian Interaction?”: Union

and Interaction of Body and Soul. Journal of the History of Philosophy 23:221–26.Rochemonteix, C. de. 1899. Un college des jésuites aux XVIIe et XVIIIe siècles: Le Collège

Henri IV de la Flèche. 4 vols. Le Mans: Leguicheux.Rozemond, Marleen. 1998. Descartes’s Dualism. Cambridge, MA: Harvard University Press.——. 1999. Descartes on Mind-Body Interaction: What’s the Problem? Journal of the History

of Philosophy 37:435–67.——. 2005. Descartes, Mind-Body Union and Holenmerism. Philosophical Topics

31:343–68.Ruler, J. A. van. 1995. The Crisis of Causality: Voetius and Descartes on God, Nature and

Change. Leiden: Brill.Schickel, Joel. 2005. “Freedom of the Will, Passion, and Virtue in Descartes’s Theory of

Judgment.” Ph.D. diss., Duke University.Schmaltz, Tad M. 1992a. Descartes and Malebranche on Mind and Mind-Body Union.

Philosophical Review 101:281–325.——. 1992b. Sensation, Occasionalism, and Descartes’ Causal Principles. In Minds, Ideas,

and Objects: Essays on the Theory of Representation in Modern Philosophy, ed.P. Cummins and G. Zoeller, 37–56. Atascadero, CA: Ridgeview Press.

——. 1994. Human Freedom and Divine Creation in Malebranche, Descartes and theCartesians. British Journal for the History of Philosophy 2:3–50.

228 WORKS CITED

——. 1996. Malebranche’s Theory of the Soul: A Cartesian Interpretation. New York: OxfordUniversity Press.

——. 1997. Descartes on Innate Ideas, Sensation, and Scholasticism: The Response toRegius. In Studies in Seventeenth-Century Philosophy, ed. M.A. Stewart, 33–73. Oxford:Clarendon Press, 1997.

——. 2000. The Disappearance of Analogy in Descartes, Spinoza, and Regis. CanadianJournal of Philosophy 30:85–114.

——. 2002a. Radical Cartesianism: The French Reception of Descartes. Cambridge:Cambridge University Press.

——. 2002b. Review of Jonathan Bennett: Learning from Six Philosophers: Descartes,Spinoza, Leibniz, Locke, Berkeley, Hume. Mind 111:367–73.

——. 2003. Cartesian Causation: Boby–Body Interaction, Motion, and Eternal Truths. Studiesin History and Philosophy of Science 34:737–62.

——. 2006. Deflating Descartes’s Causal Axiom. Oxford Studies in Early Modern Philosophy3:1–31.

Schwartz, Michael. 1991. Who Were Maimonides’ Mutakallimun? Some Remarks on Guideof the Perplexed, Part I, Chapter 73. Maimonidean Studies 2:159–209, 3:143–72.

Scott, David. 2000. Occasionalism and Occasional Causation in Descartes’ Philosophy.Journal of the History of Philosophy 38:503–28.

Secada, J. E. K. 1990. Descartes on Time and Causality. Philosophical Review 99:45–72.——. 2000. Cartesian Metaphysics: The Late Scholastic Origins of Modern Philosophy. New

York: Cambridge University Press. Shapiro, Lisa. 1999a. Cartesian Generosity. Acta Philosophica Fennica 64:249–75.——. 1999b. Princess Elizabeth and Descartes: The Union of Soul and Body and the Practice

of Philosophy. British Journal for the History of Philosophy 7:503–20.Simmons, Alison. 2001. Sensible Ends: Latent Teleology in Descartes’ Account of Sensation.

Journal of the History of Philosophy 39:49–75.——. Forthcoming. Guarding the Body: A Cartesian Phenomenology of Perception. In Early

Modern Metaphysics: Essays in Honor of Vere Chappell, ed. P. Hoffman, D. Owen, andG. Yaffe. Peterborough, Ontario: Broadview Press.

Smith, Norman (Kemp). 1902. Studies in the Cartesian Philosophy. New York: Macmillan.Smith, Norman Kemp. 1952. New Studies in the Philosophy of Descartes: Descartes as

Pioneer. London: Macmillan.Specht, Rainer. 1966. Commercium mentis et corporis: Über Kausalvorstellungen im

Cartesianismus. Stuttgart: Friedrich Frommann.——. 1971. Über “occasio” und verwandte Begriffe vor Descartes. Archiv für

Begriffsgeschichte 15:215–55.——. 1972. Über “occasio” und verwandte Begriffe bei Zabarella und Descartes. Archiv für

Begriffsgeschichte 16:1–27.Spinoza, Benedict de. 1925. Spinoza Opera. Ed. G. Gerhardt. 4 vols. Heidelberg: Carl

Winters. ——. 1985. The Collected Works of Spinoza. Vol. 1. Ed. and trans. E. Curley. Princeton, NJ:

Princeton University Press.Spruit, Leen. 1994–95. Species Intelligibilis: From Perception to Knowledge. 2 vols. Leiden:

Brill.Stump, Eleonore. 1997. Aquinas’s Account of Freedom: Intellect and Will. Monist

80:576–97.Suárez, Francisco. 1866. Opera Omnia. Ed. D. André and C. Berton. 26 vols. Paris: Vivès.——. 1947. On the Various Kinds of Distinctions (Disputationes Metaphysicae, Disputatio

VII). Trans. C. Vollert. Milwaukee, WI: Marquette University Press.

Works Cited 229

——. [1866] 1967. Disputationes Metaphysicae. 2 vols. Reprint, Hildesheim: Georg Olms. ——. 1994. On Efficient Causality: Metaphysical Disputations 17, 18, and 19. Trans.

A. J. Freddoso. New Haven, CT: Yale University Press.——. 2002. On Creation, Conservation, and Concurrence: Metaphysical Disputations 20–22.

Trans. A. J. Freddoso. South Bend, IN: St. Augustine Press.Suppes, Patrick. 1954. Descartes and the Problem of Action at a Distance. Journal of the

History of Ideas 15:146–52.Sylwanowicz, Michael. 1996. Contingent Causality and the Foundations of Duns Scotus’

Metaphysics. Leiden: Brill.Thomas Aquinas. 1871–80. Doctoris Angelici Divi Thomae Aquinatis Sacri Ordinis F.F.

Praedicatorum opera omnia. Ed. S. Fretté and P. Maré. 43 vols. Paris: Vivès. ——. 1952. On the Power of God. Westminster, MD: Newman Press. ——. 1964–81. Summa Theologiae. Ed. Blackfriars. 61 vols. New York: McGraw-Hill.——. 1987. Truth and the Disputed Questions of Truth. Albuquerque, NM: American

Classical College.Tollefsen, Deborah. 1999. Princess Elisabeth and the Problem of Mind-Body Interaction.

Hypatia 14:59–77.Verbeek, Theo. 1992. Descartes and the Dutch: Early Reactions to the Cartesian Philosophy,

1637–1650. Carbondale: Southern Illinois University Press. ——, ed. 1993. Descartes et Regius. Autour de l’explication de l’esprit humain. Amsterdam:

Rodopi.Wells, Norman. 1982. Descartes’ Uncreated Eternal Truths. New Scholasticism 56:185–99.Westfall, Richard S. 1971. Force in Newton’s Physics: The Science of Dynamics in the

Seventeenth Century. New York: American Elsevier.Williams, Bernard. 1978. Descartes: The Project of Pure Inquiry. Atlantic Highlands, NJ:

Humanities Press.Wilson, Margaret. 1978. Descartes. London: Routledge.——. 1999a. Can I Be the Cause of My Idea of the World? (Descartes on the Infinite and

Indefinite). In Wilson 1999d, 108–25. Originally in Essays on Descartes’ Meditations,ed. A. O. Rorty, 339–58. Berkeley and Los Angeles: University of California Press,1986.

——. 1999b. Descartes on the Origin of Sensation. In Wilson 1999d, 41–68. Originally inPhilosophical Topics 19 (1991): 293–323.

——. 1999c. History of Philosophy in Philosophy Today: and the Case of SecondaryQualities. In Wilson 1999d, 455–94. Originally in Philosophical Review CentennialIssue, 101 (1992): 191–243.

——. 1999d. Ideas and Mechanism: Essays on Early Modern Philosophy. Princeton, NJ:Princeton University Press.

——. 1999e. Spinoza’s Causal Axiom (Ethics I, Axiom 4). In Wilson 1999d, 141–65.Originally in God and Nature: Spinoza’s Metaphysics, ed. Y. Yovel, 133–60. Leiden:Brill, 1991.

Yakira, Elhanan. 1994. La causalité de Descartes à Kant. Paris: Presses Universitaires deFrance.

230 WORKS CITED

231

Index

actionbody-to-mind

account of, in Comments, 149–160, 162(see also mind, innate sensory facultyof)

scholastic problem with, 145–149See also motion, as sign for sensation;

sensationdefinition of, in Suárez, 31–32at a distance, 96, 108, 112, 143 n.31, 164,

165, 166, 170–171free

vs. coerced action, 182, 194–195and generosity, 207–208God’s concursus with, 42–44, 178, 184,

214vs. voluntary, 35 n.66, 64 n.37, 182, 195See also freedom, human; volition; will

immanent, 32–33, 34, 42, 107, 155 n.55mind-to-body

change of direction account of, 131,172–173, 175–177

and conservation of motion, 99 n.33,171–177, 219–220

scholastic problem with, 163–167See also union, of mind and body, and

mind-body interactionin patient, 19, 31transeunt, 32–33, 36See also causation; cause; force

Adam, Charles, 202

Adam of Wodenham, 92 n.7Alanen, Lilli, 161–162Albert of Saxony, 92 n.7Alquié, Ferdinand, 203–204angels, 15, 16, 32 n.59, 107 n.54, 131, 134,

141, 142–143, 164–167, 169, 174Aquinas. See Thomas AquinasAristotle, 4, 20, 29, 30–31, 35, 60, 80 n.85, 94,

110 n.59, 129, 140, 145, 146, 167 n.79(see also categories, Aristotelian;motion, Aristotelian concept of; will,Aristotelian concept of)

Arnauld, Antoine, 59–60, 140–141, 193As‘arı, Abu l’Hasan al-, 12attribute. See mode, vs. attributeAugustine of Hippo, 75 n.71, 145Aureoli, Peter, 92 n.7Averroes (Abu’l-Walid ibn Rushd), 15Avincenna (Abu Ali al-Husayn ibn Sına),

14, 33axiom

conservation, 49–51, 71–84, 85–86 (see also conservation)

containment, 7, 49–71, 84–85, 112–115,128, 130, 148–149, 153–157, 190–191(see also containment)

vs. principle, 49 n.1voluntarist, 25, 26, 180

Baker, Gordon, 135 n.11, 159 n.61Bañez, Domingo, 184, 191

Baquillanı, Abu Bakr al-, 13 n.13Beeckman, Isaac, 110 n.59Bellarmine, Robert, 184Bennett, Jonathan, 50, 51, 82–83, 84, 112

n.64, 188–89Beyssade, Michelle, 204 n.57Blackwell, Richard, 89 n.5body

argument for existence of, 55–56, 66, 116,151–152

Descartes vs. scholastics on, 11–12, 44–47,87, 122, 129, 217 (see also quality,real, Descartes vs. scholastics on)

divisibility of, 78–81, 91–92, 107indefinite extension of, 70 n.57particular, mode vs. substance, 53 n.9, 68

n.53, 78, 80–81See also extension, as essence of body;

intellect, more noble than body; mind,more noble than body

Broughton, Janet, 112–114, 153–157Buridan, John, 94–95Burman, Frans, 57–58

Cajetan (Tommasio de Vio), 25 n.35, 31 n.57,147 n.38, 180 n.4

Campbell, Joseph, 197 n.39Carraud, Vincent, 3, 60 n.26, 74 n.68Cartesian interaction. See interaction,

Cartesian, scandal ofcategories, Aristotelian, 25–26, 46, 116Caterus, Johannes (Johan de Kater), 59, 65causation

occasional, 9–10, 145, 130–131, 157–161(see also occasionalism)

orders of, 17, 22, 40–42, 76, 85–86universal vs. particular, 20–21, 57 n.17,

76–77, 85–86, 90–91, 123, 210–211

See also action; cause; overdetermination,causal

causeadequate/total, 56–58, 76–77, 85–86, 114,

148–149, 153definition of, in Suárez, 29efficient, 4, 30–33, 56–61, 162

definition of, in Suárez, 30–31priority of, 4, 11, 29–36, 59–64See also God, as efficient cause

exemplary, 24 n.31extrinsic, 29, 33, 113final, 4, 33–36, 61–64, 158–159

and God, 36, 44, 61–64and intellectual agents, 34–35, 64metaphorical motion of, 29, 34, 36, 64and natural agents, 35–36, 62–63

See also intellect, as final cause of will;union, of mind and body, and teleology

formal, 4, 30, 45, 59–61, 158–159, 189 (see also form)

immediate, 20–21, 39instrumental, 17, 19, 21, 40–41, 43,

146–147, 154intrinsic, 29, 30, 61material, 4, 30modal vs. substantial, 103–104necessary, 42–43partial, 56–57, 154primary, 20, 40, 89–105principal, 17, 40, 42, 45, 90remote, 39, 151secondary, 17, 20–23, 39, 40, 42, 105–124secundum esse/secundum fieri, 18, 23, 33,

38–39, 73–75, 76–77, 81, 83–84, 86,104, 117–118, 121

as simultaneous with its effect, 74–75, 81,112

and reason, 36, 74, 189–190See also action; causation; freedom, human;

overdetermination, causal; forceClarke, Desmond, 112 n.62, 122 n.82Clatterbaugh, Kenneth, 68 n.50Clauberg, Johann, 5 n.6, 171 n.85, 173Clement VIII, 184Clerselier, Claude, 5 n.6, 172–173, 202 n.53Cohen, I. B., 107–108collision, rules for, in Descartes, 89 n.5,

98–99, 114–115compatibilism, casual, in Thomas Aquinas, 11,

16–19, 165 n.70concurrentism, 6, 7, 16–19, 40–44, 84–85, 86,

100–102, 125–127, 209, 218, 219, 220(see also concursus, ordinary)

concursus, ordinary, 8, 92, 93, 99–105,122–124 , 128, 171–172, 175–176,214–215, 219–220 (See also conserva-tion, and concursus; freedom, human,God’s concursus with; God, power of,absolute vs. ordinary)

Condemnation of 1277, 25, 180Congregation on Grace, 184conservation, 18–19, 73–77

and concursus, 40–44, 99–105, 127–128

not distinct from creation, 22, 38–39, 81–84,125–126, 178

See also action, mind-to-body, and conservation of motion; axiom, conservation; quantity, conserved

conservationism, merecritique of, in Suárez, 22–23, 41–42

232 INDEX

in Descartes, 6, 8, 23–24, 41, 85–86,125–128, 160, 175–177, 212–216, 220

in Durandus, 19–24, 86, 89containment

eminent, 3, 4, 58, 67–71, 113, 148–149,166–167, 190–191

formal, 3, 4, 55, 64–67, 84, 114–115, 149,153–155 (see also reality, formal)

See also axiom, containmentCordemoy, Geraud de, 5 n.6, 171 n.85Cottingham, John, 117 n.75, 138 n.21created truth. See truth, createdcreation, 17–19, 36–39

ab aeterno, 37, 38, 74de novo, 37, 38, 45, 75ex nihilo, 18, 19, 36–39, 58, 76, 82–83, 90,

125–126See also conservation, not distinct from cre-

ation; truth, createdCurley, Edwin, 213 n.69

Dalbriez, Roland, 65 n.42Della Rocca, Michael, 47 n.88, 213 n.68Descartes, René. See body, Descartes vs.

scholastics on; collision, rules for, inDescartes; heaviness, discussion of, inDescartes; hierarchy, ontological, inDescartes; judgment, Descartes vs.scholastics on; laws, of nature, inDescartes; motion, definition of, inDescartes; occasionalism, andDescartes; quality, real, Descartes vs.scholastics on; soul, Descartes vs.scholastics on; rest, problems with, inDescartes; truth, created, doctrine of, inDescartes; union, of mind and body,argument for, in Descartes

Des Chene, Dennis, 105 n.49, 109, 127–128Desgabets, Robert, 107 n.54distinction

formal, 26–27, 28 n.46modal, 26–28, 78–79real, 25–26, 27, 78in reason, 21, 26, 27, 61, 77, 85, 119

Dominicans, 10–11, 19–20, 179, 184, 191–192(see also freedom, human, Dominicansvs. Jesuits on; Thomism)

Du Hamel, Jean, 107 n.54Dubarle, Dominique, 97 n.25Duhem, Pierre, 94Duns Scotus, John, 26–27, 28 n.46, 31 n.57,

40–41, 42, 165, 182 n.7 (see alsoScotism)

Durandus of Saint-Pourçain, 11, 17, 19–24,36, 37–38, 40, 41, 42, 85–86, 126, 147,

165, 167, 170, 178, 218 (see also con-servationism, mere, in Durandus)

durationof God, 78 n.80modes of, 79, 118–120and substance, 77–78, 119See also force, in body–body interaction,

and duration; time

Elisabeth, Princess of Bohemia, 3, 130,131–135, 144, 167–169, 209–212

Elizabeth Stuart, Queen of Bohemia, 132esse. See cause, secundum esse/secundum fieri;

reseternal truths, creation of. See truth,

createdextension, as essence of body, 60–61, 71, 87,

88, 91–92, 116, 119, 134, 168, 169,171, 217 (see also body; God, extension of; mind, extension of; quantity, conserved, of matter)

Farabi, Abu Nasr Muhammed al-, 14Fonseca, Peter, 27Fontenelle, Bernard le Bovier de, 9–10,

122–123, 130force

in body–body interactionfor acting, 95, 97, 111–112, 115–116,

124, 127–128, 218and duration, 88, 117–121, 122, 218 (see

also duration)fictionalist interpretation of, 105,

115–116, 120–121and God, 102–105, 121–124, 218impressed vs. inertial, 98–99ontology of, 104–105, 116–121, 160–161for resisting, 97–98, 111–112, 115–116,

124, 127–128, 176–177, 219 (see alsorest, problems with, in Descartes)

in union, 3, 134, 138, 140, 144, 167–168,174–175

formaccidental/substantial, 17, 32, 41, 45, 61eduction of, 32, 45–46and matter, 45–46, 92perfect vs. imperfect, 164See also quality, real

Frankfurt, Harry, 188–189, 212Freddoso, Alfred, 22 n.27Frederick V, King of Bohemia, 132freedom, human

account ofin Descartes’s correspondence, 200–205in the Fourth Meditation, 193–199in the Passions, 205–208

Index 233

freedom, human (continued)in the Principles, 199–200 (see also

freedom, human, and divine providence, in the Principles)

in the Rules, 192in Suárez, 180–187

compatibilist view of, 197 n.39, 198 n.43,204 n.57 (see also freedom, human,incompatibilism of)

vs. divine freedom, 190–191, 196, 215 (seealso God, indifference of) and divine providence,

Domincans vs. Jesuits on, 179, 184–187,191–192

in the Elisabeth correspondence, 209–212in the Principles 177, 186–87, 199–200,

208–209See also action, free, God’s concursus

with; middle knowledgeexercise vs. specification of, 35 n.66, 42–43,

182–183incompatibilism of, 179–180, 184, 191,

199–200, 208 (see also freedom,human, compatibilist view of)

indifference of, 8, 35, 43, 179, 182, 184, 186, 192–208 (See also God,indifference of)

positive faculty of, 202–204spontaneity of, 196–197See also action, free; truth, created, and

human freedom; volition; will

Gabbey, Alan, 83 n.92, 98 n.30, 99 n.32,117–118, 119

Gabriel Biel, 92 n.7Galileo Galilei, 110Garber, Daniel, 6, 88, 98 n.30, 102–105,

115–116, 120–122, 156 n.57, 175–176Gassendi, Pierre, 62, 73–74, 198Geulincx, Arnold, 144Ghazalı, Abu Hamid al-, 14–16, 17Gilson, Etienne, 65 n.42, 83 n.92God

argument for existence of, 3, 49–50, 51–52,55–56, 71–72

as causa sui, 59–60, 189as efficient cause, 36, 44, 187, 211, 213,

214, 215, 217extension of, 163–164, 165, 169–170immutability of, 90–91, 93–94, 100, 109,

115 n.74, 116, 121–122, 124, 125–126,160, 210, 214–15

indifference of, 62, 189–191, 212, 215 (seealso freedom, human, vs. divine freedom)

power of, 10, 15, 17–18, 36, 59, 75, 177,186, 188, 192, 211–212

absolute vs. ordinary, 99 (see also concur-sus, ordinary)

See also cause, final, and God; duration, ofGod; force, in body–body interaction,and God; freedom, human, God’sconcursus with; freedom, human, anddivine providence; sensation, anddivine goodness; truth, created; will,human, in image of God

Gorham, Geoffrey, 70 n.76, 80 n.89, 81 n.90Gouhier, Henri, 158 n.59Gueroult, Martial, 50, 51, 53 n.9, 68 n.53,

79–81, 102, 109 n.58, 117

Hatfield, Gary, 105 n.48, 161 n.63, 198 n.43Hattab, Helen, 115 n.73heaviness, discussion of, in Descartes, 71, 137,

138, 140, 163, 167–169, 170–171heterogeneity, problem of, 130, 132, 133, 163

(see also interaction, Cartesian, scandalof)

hierarchy, ontological, in Descartesenhanced, 69simple, 53–54, 68

Hobbes, Thomas, 119Hoffman, Paul, 206 n.60Hume, David, 47Hyperaspistes, 69–70, 73–74

idea. See reality, objective; judgment,concerning clear and distinctperception; sensation, as mode ofmind; thought, pure vs. sensory

impetus, theory of, 94–95indifference. See freedom, human, indifference

of; God, indifference ofintellect

agent vs. passive, 32–33as final cause of will, 64, 183, 197–198, 206

(see also will, relation to intellect)more noble than body, 146, 147, 148 (see

also mind, more noble than body)See also species, intelligible; thought, pure

vs. sensoryintellectualism. See volunatism, ethical, vs.

intellectualisminteraction, Cartesian, scandal of 5–7, 8, 9 (see

also action; heterogeneity, problem of;union, of mind and body, andmind–body interaction)

James of Viterbo 146–147Javellus, Chrysostom 180 n.4

234 INDEX

Jesuits 25, 26–27, 43, 93–94, 172, 200–205,219 (see also freedom, human,Dominicans vs. Jesuits on)

judgmentconcerning clear and distinct perception,

193, 194, 195, 198–199, 200, 201–202,206–208

Descartes vs. scholastics on, 179, 183,192–195, 197–199, 200–205

concerning truth vs. goodness, 181, 183,192–194, 205, 207–208

See also intellect, as final cause of will; will,relation to intellect

Kenny, Anthony, 202 n.53Kent, Bonnie, 180

La Forge, Louis de, 5 n.6La Mare, William de, 181 n.5laws

of nature, in Descartesfirst, 73, 106–108, 198 n.42 (see also

quantum in se est)second, 108–110third, 110–116, 174–175 (see also

collision, rules for, in Descartes)psychophysical 161–162

Leibniz, Gottfried Wilhelm, 31, 74 n.68, 79n.84, 163, 172–173, 176–177, 219

light, nature of, 18 n.21, 73 n.66, 108 n.57Locke, John, 67 n.47Loeb, Louis, 54–55, 161–162Lucretius, 107–108

Maier, Anneliese, 94–95Maimonides. See Moses MaimonidesMajor, John, 92 n.7Malebranche, Nicolas, 5, 71 n.62, 93 n.11, 96

n.24, 121, 122–123, 124 n.84, 144Marion, Jean-Luc, 76Marmura, Michael, 16McLaughlin, Peter, 173–177Menn, Stephen, 25mere conservationism. See conservationism,

mereMersenne, Marin, 56, 187–188Mesland, Denis, 200–201, 202metaphysics, renovation of, in Suárez, 25–28,

31–32, 180middle knowledge, 179, 185–187, 211mind

extension of, 131, 134, 138, 169–171, 219immateriality of, 78, 132–133, 135, 146

n.35, 163–164, 219indivisibility of, 68–69, 78, 107

innate sensory faculty of, 150–155, 157–158,160–162, 177

more noble than body, 68–69, 218 (see alsointellect, more noble than body)

See also sensation, as mode of mind; soul;thought; union, of mind and body

miracles, 14, 99mode

vs. attribute, 54 n.10, 79, 117–18vs. substance, 28, 44, 68, 71, 76–77See also body, particular, mode vs.

substance; distinction, modal; duration,modes of; quality, real, difference frommode; sensation, as mode of mind

Molina, Luis de, 40 n.71, 179, 184–187More, Henry, 88, 97–98, 103–104, 113–114,

169–171Morris, Katherine, 135 n.11, 159 n.61Moses Maimonides (Moshe ben Maimon),

12–14, 16–17, 80motion

cinematic view of, 102–103 (see also time,atomistic concept of)

definition of, in Descartes, 89determination of, 109, 111, 119–120, 132

n.7, 172, 174impulse view of, 103–104inclination to, 109, 110, 111, 120–121,

123–124migration theory of, 112–114, 159rectilinear nature of, 108–111, 120,

123–124as sign for sensation, 136, 152 n.47, 156–157speed of, 95, 96voluntary, 171–177See also cause, final, metaphorical motion

of; collision, rules for, in Descartes;impetus, theory of; laws, of nature, inDescartes; quantity, conserved, ofmotion

Nadler, Steven, 145, 151 n.45, 157–161Newton, Isaac, 98–99nominalism, vs. realism, 26, 32, 92Normore, Calvin, 186Norris, John, 71 n.62

occasionalismargument against, in scholastics, 16–17, 20,

22–23and Descartes, 4–5, 9–10, 76–77, 84–86,

121–122, 126, 144, 155–156, 158–161,220

medieval Islamic, 7, 10, 12–16, 19, 47, 80

Index 235

occasionalism (continued)post-Descartes, 16, 92 n.11, 122–123, 144,

172–173See also causation, occasional

Ockham. See William of OckhamOlivio, Gilles, 64O’Neill, Eileen, 68–71overdetermination, causal, 20 n.24, 22, 165

Paul V, 184Perler, Dominic, 11Pessin, Andrew, 100 n.35, 127 n.88Peter of Auvergne, 33 n.61phantasm. See species, intelligible, relation to

phantasmPlato, 145, 146, 163 (see also Platonism)Platonism, 14, 140–141, 146–147, 148Pollot, Alphonse, 132power, extension of. See mind, extension ofProclus, 52 n.6providence, divine. See freedom, human, and

divine providence

quality, realDescartes vs. scholastics on, 28, 35, 44–45,

53 n.8, 61 n.29, 62–63, 71, 105, 113,116, 122, 167–168, 170–171, 217

difference from mode, 28See also form, accidental/substantial; heavi-

ness, discussion of, in Descartesquantity, conserved

of matter, 91–92of motion, 93–96, 171–177 (see also action,

mind-to-body, and conservation ofmotion)

of rest, 96–99 (see also rest, problems with,in Descartes)

quantum in se est, 107–108, 115, 118–119Quinn, Philip, 41 n.74

Raconis, Charles François d’Abra de, 32 n.58Radner, Daisie, 53–54, 67Ragland, C. P., 197 n.39Razı, Fakhr al’Din al-, 13realism. See nominalism, vs. realismreality

formal, 52, 65–67, 148–149, 154 (see alsocontainment, formal)

objective, 52, 65–67, 116–117, 149, 151,153, 154, 190 (see also sensation,representative nature of)

Regis, Pierre-Sylvain, 43 n.77, 104 n.47, 107n.54, 123 n.83, 127 n.87, 190 n.22

Regius, Henricus, 126, 129, 138 n.20, 149–150Reid, Jasper, 171 n.87

Remnant, Peter, 175 n.96res, 25, 27, 37 n.68, 41, 44–46, 77, 107rest, problems with, in Descartes, 120, 124,

128, 176–177, 219–220 (see alsoconservation, of rest; force, for resisting)

Revius, Jacobus, 101–102, 149 n.43Richardson, Robert, 130, 140 n.23Rozemond, Marleen, 145–146, 147Rudolph, Ulrich, 11

scholasticism, 4, 6–7, 25–28, 30, 44–45, 92,94, 101–102, 133, 145–149, 163–167,180, 192, 194, 217–218 (see alsoaction, body-to-mind, scholasticproblem with; action, mind-to-body,scholastic problem with; body,Descartes vs. scholastics on;Dominicans; freedom, human,Dominicans vs. Jesuits on; Jesuits;judgment, Descartes vs. scholastics on;occasionalism, argument against, inscholastics; quality, real, Descartes vs.scholastics on; Scotism; soul, Descartesvs. scholastics on; Thomism)

Scotism, 11, 27, 65Scott, David, 152 n.48, 152–153 n.51Scotus. See Duns ScotusSeneca, 209sensation

as arbitrary, 135–137confusion of, 138, 140, 143and divine goodness, 63 n.34internal vs. external, 141–142as mode of mind, 148, 155representative nature of, 66–67, 143See also action, body-to-mind, account of,

in Comments; mind, innate sensoryfaculty of; motion, as sign forsensation; species, intentional; thought,pure vs. sensory

Simmons, Alison, 141–142, 162 n.66Smith, Norman (Kemp), 5–6, 8, 50,

173–174Soto, Domingo de, 25, 26soul

Descartes vs. scholastics on, 129, 148, 155,217–218

weak vs. strong, 206–207See also mind

Specht, Ranier, 158 n.59species

intelligible, 32–33relation to phantasm, 146–149, 154–155

intentional, 46Spinoza, Benedict de, 107 n.55

236 INDEX

Suárez, Francisco, 11–12, 22–47, 49, 50–51,52, 53, 57, 59, 61, 62–66, 69–71,74–75, 77, 78, 82–84, 85–86, 87–89,90, 92, 93, 94–95, 99, 100, 113, 114,116, 125–126, 147–149, 154, 155–156,166–167, 178–187, 192, 194–195,204–205, 211, 214, 218, 219 (see alsoaction, definition of, in Suárez; cause,definition of, in Suárez; cause, efficient,definition of, in Suárez; conservationism,mere, critique of, in Suárez; freedom,human, account of, in Suárez;metaphysics, renovation of, in Suárez)

substance. See cause, substantial vs. modal;duration, and substance; form,accidental/substantial; mode, vs.substance; res

Tamburini, Michelangelo, 93Tannery, Paul, 202teleology. See cause, final; union, of mind and

body, and teleologyThomas Aquinas, 10, 16–19, 20–23, 37–39, 40,

41, 43, 73–75, 80, 82, 99, 145–149, 154,163–167, 181, 182 n.7, 192–193 n.27(see also compatibilism, causal, inThomas Aquinas; Thomism)

Thomism 25–26, 32, 43, 65, 74–75, 92,147–148, 180–181

thoughtas essence of mind, 129pure vs. sensory, 141–142, 150–151See also mind; soul

timeatomistic concept of, 13–15, 47, 79–81, 109

n.58 (see also motion, cinematic view of)and duration, 77 (see also duration)endurantist vs. perdurantist concept of,

80–81parts of, 78–81, 118See also cause, as simultaneous with its

effecttruth, created

doctrine of, in Descartes, 187–192, 212–216and essences, 213–214and human freedom, 211–216necessity of, 188, 212–214vs. truths concerning God, 188–189, 213

union, of mind and bodyargument for, in Descartes, 140–142in the Elisabeth correspondence, 132–135,

167–169fact vs. explanation of, 139–140, 142and mind–body interaction, 131–144,

167–169ontological status of, 138–139, 155and teleology, 63–64, 136, 162 (see also

sensation and divine goodness)two accounts of, 135–140See also force, in union; sensation; volition,

human, as confused

Vasquez, Gabriel, 65 n.43Voetius, Gisbertius, 126, 138 n.20, 149 n.42volition

human, as confused, 142–144vs. passion, 177, 205–206and psychokenesis, 143See also action, free; freedom, human;

motion, voluntary; willvoluntarist axiom. See axiom, voluntaristvoluntarism, ethical, vs. intellectualism

180–182

Walter of Bruges, 181 n.5weight. See heaviness, discussion of, in

DescartesWestfall, Richard, 117 n.75will

ability to act morally vs. absolutely,202–203, 204–205, 208 (see alsofreedom, human, positive faculty of)

Aristotelian concept of, 181first vs. second act of, 33, 34–35habituo of, 185–186, 192, 211–212, 216human, in image of God, 190–191,

215–216relation to intellect, 181–182, 192–208 (see

also intellect, as final cause of will)See also action, free; freedom, human;

judgment; volitionWilliam of Ockham, 26, 65 n.43, 92 n.7Williams, Bernard, 5, 143Wilson, Margaret, 3, 135–137, 140 n.23, 141

n.26, 153 n.52

Index 237

Date post:	08-Mar-2015
Category:	Documents
Upload:	muhammad-amirul-kamal
View:	277 times
Download:	2 times

Descartes on Causation

Documents