Leibniz: Nature and Freedom

Citation preview

Leibniz: Nature and Freedom

Donald Rutherford J. A. Cover, Editors

OXFORD UNIVERSITY PRESS

LEIBNIZ

This page intentionally left blank

LEIBNIZ Nature and Freedom

EDITED BY

Donald Rutherford and J. A. Cover

1

2005

1 Oxford University Press, Inc., publishes works that further Oxford University’s objective of excellence in 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 # 2005 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 Data Leibniz : nature and freedom / edited by Donald Rutherford and J.A. Cover. p. cm. Includes bibliographical references and index. ISBN-13 978-0-19-514374-4; 978-0-19-514375-1 (pbk.) ISBN 0-19-514374-4; 0-19-514375-2 (pbk.) 1. Leibniz, Gottfried Wilhelm, Freiherr von, 1646–1716. 2. Free will and determinism. 3. Philosophy of nature. I. Rutherford, Donald, 1957– II. Cover, J. A. (Jan A.), 1958– B2599 .F7L45 2005 193—dc22 2004050094

9 8 7 6 5 4 3 2 1 Printed in the United States of America on acid-free paper

For R. C. Sleigh, Jr. Teacher, Colleague, Friend

This page intentionally left blank

Preface

This volume of new essays is intended to provide a timely picture of the current state of scholarship on Leibniz’s metaphysics. We are grateful to the authors for contributing to the collection. We would also like to thank Daniel Garber and Robert M. Adams for their sound advice in the early stages of the project; Matthew Kisner, for his assistance in preparing the typescript for publication; and Peter Ohlin of Oxford University Press, for his enthusiasm and patience. Finally, we are especially thankful to Madeleine Picciotto and Karen Cover, for their inextinguishable good humor and support in the face of challenges to completing this project: overcommitment, physical injury, and stubbornness. The contributions of Robert C. Sleigh, Jr., to Leibniz scholarship have and continue to set a benchmark for a generation of students and researchers in the history of early modern philosophy. Those who read his work are fortunate to have him as a fellow philosopher; those who have studied with him are lucky to have him as a mentor; those who know him are blessed to have him as a friend. We dedicate this volume to Bob in fondness and gratitude.

This page intentionally left blank

Contents

Contributors Abbreviations

xi xiii

Introduction 3 Donald Rutherford and J. A. Cover 1.

The One and the Many and Kinds of Distinctness: The Possibility of Monism or Pantheism in the Young Leibniz Mark Kulstad 20

2.

Leibniz and Sleigh on Substantial Unity Christia Mercer 44

3.

Leibniz on Precise Shapes and the Corporeal World Samuel Levey 69

4.

Leibniz and Idealism Daniel Garber 95

5.

Compossibility, Expression, Accommodation Catherine Wilson 108

6. Leibniz and Occasionalism Nicholas Jolley 121 7.

Leibniz’s Two Realms Jonathan Bennett 135

8.

Leibniz on Spontaneity Donald Rutherford 156

9. Moral Necessity Robert Merrihew Adams 10.

181

Spontaneity and Freedom in Leibniz Michael J. Murray 194

x Contents 11.

Leibniz Against Molinism: Freedom, Indifference, and the Nature of the Will Sean Greenberg 217

12. Video Meliora Proboque, Deteriora Sequor: Leibniz on the Intellectual Source of Sin Jack D. Davidson 234 Bibliography Index 263

255

Contributors

ROBERT MERRIHEW ADAMS is a senior member of the Faculty of Philosophy, Oxford University. JONATHAN BENNETT is retired and lives on an island near Vancouver, B.C. J. A. COVER is Professor of Philosophy at Purdue University. JACK D. DAVIDSON is Visiting Assistant Professor of Philosophy at the University of Memphis. DANIEL GARBER is Professor of Philosophy at Princeton University. SEAN GREENBERG is Assistant Professor of Philosophy at Johns Hopkins University. NICHOLAS JOLLEY is Professor of Philosophy at the University of California, Irvine. MARK KULSTAD is Professor of Philosophy at Rice University. SAMUEL LEVEY is Associate Professor of Philosophy at Dartmouth College. CHRISTIA MERCER is Professor of Philosophy at Columbia University. MICHAEL J. MURRAY is Professor of Philosophy at Franklin and Marshall College. DONALD RUTHERFORD is Professor of Philosophy at the University of California, San Diego. CATHERINE WILSON is Professor of Philosophy at the University of British Columbia.

This page intentionally left blank

Abbreviations

(citations by page number unless otherwise indicated) A

Gottfried Wilhelm Leibniz. Sa¨mtliche Schriften und Briefe. Ed. Deutsche Akademie der Wissenschaften. Darmstadt, Leipzig, Berlin: Akademie Verlag, 1923–. Cited by series, volume, page.

AG

G. W. Leibniz. Philosophical Essays. Ed. and trans. Roger Ariew and Daniel Garber. Indianapolis: Hackett, 1989.

Ak

Immanuel Kant. Gesammelte Schriften. Ed. Akademie der Wissenshaften. Berlin: Reimer (later De Gruyter), 1910–. Cited by volume and page number.

AT

Oeuvres de Descartes. Ed. Charles Adam and Paul Tannery. Paris: J. Vrin, 1964–74. Cited by volume and page.

C

Opuscules et fragments ine´dits de Leibniz. Ed. Louis Couturat. Paris: Presses Universitaires, 1903. Reprint, Hildesheim: Georg Olms, 1966.

CSM

The Philosophical Writings of Descartes. Ed. and trans. John Cottingham, Robert Stoothoff, Dugald Murdoch, and (vol. 3) Anthony Kenny. Cambridge: Cambridge University Press, 1984, 1985, 1991. Cited by volume and page.

CWS

The Collected Works of Spinoza. Vol. 1. Ed. and trans. Edwin Curley. Princeton: Princeton University Press, 1985.

DM

G. W. Leibniz. Discours de Me´taphysique. Cited by section as in G 4:427– 463. English trans. in AG, L, and WFPT.

DSR

G. W. Leibniz. De Summa Rerum. Metaphysical Papers, 1675–1676. Ed. and trans. G.H.R. Parkinson. New Haven: Yale University Press, 1992.

xiv Abbreviations E

G. W. Leibniz. Opera Philosophicae quae exstant Latina, Gallica, Germanica omnia. Ed. J. E. Erdmann. Berlin: Eichler, 1839–40. Reprint, Aalen: Scientia Verlag, 1974.

Enquiry

David Hume. Enquiry Concerning Human Understanding. Ed. L. A. SelbyBigge and P. H. Nidditch. Oxford: Oxford University Press, 1975.

Essay

John Locke. An Essay concerning Human Understanding. Ed. Peter H. Nidditch. Oxford: Clarendon Press, 1975. Cited by book, chapter, and section.

Ethics

Baruch Spinoza. Ethica. Roman numerals refer to parts; A: axiom; P: Proposition; D (following roman numeral): definition; Arabic numerals refer to axioms, propositions, definitions etc.; D (following P and arabic numeral): demonstration of a proposition; C, S, L: corollary, scholium, and lemma respectively; Post, Exp, Pref, and App: Postulate, Explanation, Preface, and Appendix. English trans. in CWS.

G

Die philosophischen Schriften von Gottfried Wilhelm Leibniz. Ed. C. I. Gerhardt. Berlin: Weidmann, 1875–90. Reprint, Hildesheim: Georg Olms, 1978. Cited by volume and page.

GLW

Briefwechsel zwischen Leibniz und Christian Wolff. Ed. C. I. Gerhardt. Halle: Schmidt, 1860. Reprint, Hildesheim: Georg Olms, 1963.

GM

Leibnizens Mathematische Schriften. Ed. C. I. Gerhardt. Berlin: A. Asher; and Halle: H. W. Schmidt, 1849–63. Reprint, Hildesheim: Georg Olms, 1971. Cited by volume and page.

Grua

Textes ine´dits d’apre`s des manuscrits de la Bibliothe`que provinciale d’Hanovre. Ed. Gaston Grua. Paris: Presses Universitaires, 1948. Reprint, New York: Garland, 1985.

H

G. W. Leibniz. Theodicy. Trans. E. M. Huggard. New Haven: Yale University Press, 1952. Reprint, LaSalle, Ill: Open Court, 1985.

L

G. W. Leibniz. Philosophical Papers and Letters. Ed. and trans. Leroy E. Loemker. 2nd ed. Dordrecht: Reidel, 1969.

LA

The Leibniz-Arnauld Correspondence. Ed. and trans. H. T. Mason. Manchester: Manchester University Press, 1967. Reprint, New York: Garland, 1985.

LC

The Leibniz-Clarke Correspondence. Ed. H. G. Alexander. Manchester: Manchester University Press, 1956, following Clarke’s 1717 English translation. Cited by letter and section.

LBr

¨ ffentliDer Briefwechsel des Gottfried Wilhelm Leibniz in der Ko¨niglichen O chen Bibliothek zu Hannover. Ed. Eduard Bodemann. Hannover: Hahn, 1895. Reprint, Hildesheim: Georg Olms, 1966. Cited by Bodemann’s numbering.

Abbreviations

xv

LH

¨ ffentlichen Bibliothek zu HannDie Leibniz-Handschriften der Ko¨niglichen O over. Ed. Eduard Bodemann. Hannover: Hahn, 1889. Reprint, Hildesheim: Georg Olms, 1966. Cited by Bodemann’s numbering.

LOC

G. W. Leibniz. The Labyrinth of the Continuum: Writings on the Continuum Problem, 1672–1686. Ed. and trans. Richard T. W. Arthur. New Haven: Yale University Press, 2001.

LR

G. W. Leibniz. Discours de me´taphysique et correspondance avec Arnauld. Ed. Georges Le Roy. Paris: Vrin, 1970.

Mon

G. W. Leibniz. Monadology. Cited by section as in G 6:607–623. English translation in AG, L, and WFPT.

MP

G. W. Leibniz. Philosophical Writings. Ed. and trans. Mary Morris and G.H.R. Parkinson. London: Dent, 1973.

NE

G. W. Leibniz. Nouveaux essais sur l’entendement humain. Cited by book, chapter, and section as in A, ser. 6, vol. 6. Translated in RB.

OM

Oeuvres completes de Malebranche. Edited by Andre Robinet. 20 vols., plus ‘‘Index des citations’’ and ‘‘Index general.’’ Paris: Vrin, 1958–84. Cited by volume and page.

PLP

G. W. Leibniz. Logical Papers. Ed. and trans. G.H.R. Parkinson. Oxford: Clarendon Press, 1966.

PNG

G. W. Leibniz. Principes de la Nature et de la Graˆce, Fonde´s en Raison. Cited by section as in G 6:598–606. English translation in AG, L, and WFPT.

Principles

Rene Descartes. Principia Philosophiae. Cited by part and article, as found in AT, vol. 8, and CSM, vol. 1.

RB

G. W. Leibniz. New Essays on Human Understanding. Ed. and trans. Peter Remnant and Jonathan Bennett. Cambridge: Cambridge University Press, 1981. Cited by page that corresponds to A, ser. 6, vol. 6.

S

G. W. Leibniz. Monadology and Other Essays. Ed. and trans. P and A. M. Schrecker. Indianapolis: Bobbs-Merrill, 1965.

ST

St. Thomas Aquinas. Summa Theologiae. Madrid: Biblioteca de Autores Cristianos 1955–1958. Cited by part, question, and article.

T

G. W. Leibniz. Essais de Theodice´e. Cited by section number as in G 6:21– 471; ‘‘a’’ refers to the ‘‘Abrege’’ and is cited by objection number; ‘‘pd,’’ ‘‘h,’’ ‘‘k,’’ and ‘‘cd’’ precede section numbers of the ‘‘Discours Pre´liminaire,’’ the ‘‘Reflexions’’ on Hobbes, ‘‘Remarques’’ on King’s volume The Origin of Evil, and ‘‘Causa Dei,’’ respectively. Translated (minus ‘‘Causa Dei’’) in H.

xvi Abbreviations Treatise

David Hume. A Treatise on Human Nature. Ed. David Fate Norton and Mary J. Norton. Oxford: Oxford University Press, 2000.

WF

Leibniz’s ‘‘New System’’ and Associated Contemporary Texts. Ed. and trans. R. S. Woolhouse and Richard Francks. Oxford: Clarendon Press, 1997.

WFPT

G. W. Leibniz: Philosophical Texts. Ed. and trans. R. S. Woolhouse and Richard Francks. Oxford: Oxford University Press, 1998.

LEIBNIZ

This page intentionally left blank

Introduction DONALD RUTHERFORD

AND

J. A. C O V E R

or much of the twentieth century the dominant interpretation of Leibniz’s

F philosophy was that framed in the pioneering studies of Bertrand Russell

(1900) and Louis Couturat (1901).1 Although differing in details, the accounts of Russell and Couturat both presented Leibniz’s philosophy as a development of his logic, in particular, his formulation of the principle of sufficient reason as a necessary and sufficient condition for truth: in every true proposition, necessary or contingent, the concept of the predicate is contained in the concept of the subject. From this principle, Couturat argued, follow all of the main theses of Leibniz’s metaphysics: the identity of indiscernibles, the complete concept theory of substance, the doctrine of expression, the hypothesis of preestablished harmony, the ideality of space and time.2 Russell expressed greater skepticism than Couturat about the success of Leibniz’s philosophical program. Where Couturat found in Leibniz’s writings a consistent expression of a rationalist metaphysics he labeled ‘‘panlogism’’ (Couturat 1901: xi), Russell stressed the way in which Leibniz’s rationalism led to a conclusion that Leibniz himself could not countenance: the necessary existence of all finite things (Russell 1900: 183). Despite regarding this conclusion as decisive grounds for rejecting Spinoza’s philosophy, Leibniz was—in Russell’s judgment—trapped by his own starting points into a similar form of necessitarianism. The Russell-Couturat interpretation spurred the investigations of several generations of scholars who endeavored to elaborate the details of Leibniz’s system and to respond to the difficulties that Russell had located in it. Foremost among these is the problem of contingency, which goes to the heart of Leibniz’s metaphysics and has decisive consequences for his moral philosophy and theology. The latter areas of Leibniz’s thought, however, received comparatively little attention from scholars during this period.3 For the most part their energies were devoted to analyzing metaphysical problems—concerning necessity and contingency, truth, identity, and substance—in isolation from the historical context in which they engaged Leibniz and their systematic connections to other parts of his philosophy. 3

4 Introduction

Representative is the comparative neglect of the topic of freedom, identified by Leibniz as one of the ‘‘two famous labyrinths’’ in which reason is prone to become entangled (G 6:29/H 53). Although the problem of contingency was primarily of interest to Leibniz because of its ramifications for divine and human freedom, the latter topics did not become a focus of English-language scholarship until the 1970s, and only recently have historians of philosophy begun to explore the complex theological landscape within which freedom was such a hotly contested issue during the seventeenth century.4 The past thirty years have witnessed a renaissance in Leibniz studies.5 With increasing regularity scholars have recognized the importance of adopting a synoptic approach to Leibniz’s philosophy, interpreting its various branches (logic, metaphysics, ethics) in relation to each other and to other areas of his thought (mathematics, physics, jurisprudence, theology). At the same time there has been a newfound sensitivity to developmental questions. Rejecting the assumption that Leibniz defends a single, unchanging set of philosophical views throughout his life, commentators have distinguished an early, middle, and late Leibniz, and have debated the consistency—synchronic and diachronic—of the positions he articulates within these periods.6 Finally, there has been a growing awareness of the need to interpret Leibniz’s views within the larger context of seventeenth-century intellectual life and to accommodate within such readings Leibniz’s own understanding of his relationship to the history of philosophy. To a degree unusual among seventeenth-century philosophers, Leibniz was eager to highlight the affinity between his views and those of his predecessors— ancient, medieval, and modern. Basic to his approach is the goal of establishing agreement between his positions and those of other philosophers, pursuing in this way a vein of truth that has been only slowly and incompletely extracted over millenia. Leibniz’s embrace of this goal has its roots in theology. He is convinced that there is a single source of truth, the divine intellect, to which all philosophers have had access, and he is committed to locating agreement among rival factions wherever possible. In a late letter to Nicolas Remond, he writes: I have tried to uncover and unite the truth buried and scattered under the opinions of all the different philosophical sects, and I believe I have added something of my own which takes a few steps forward. . . . I have found that most of the sects are right in a good part of what they propose, but not so much in what they deny. (G 3:605/L 654–655)7

Leibniz emphasizes that his own philosophy has been able to unite elements of truth from apparently irreconcilable alternatives (e.g., Aristotelianism and mechanism, or the Jansenist and Jesuit positions on freedom). Yet there are limits beyond which even he is not prepared to go. In the case of two prominent contemporaries, Hobbes and Spinoza, Leibniz finds their doctrines supported by principles that are objectionable at such a fundamental level that there can be no hope of accommodation, even if evidence suggests that in his youth Leibniz was strongly influenced by their writings. This anomaly offers an important clue to the deeper motivations of Leibniz’s philosophy, and a point around which to organize a historically informed account of his views.

Introduction

5

The Threat of Naturalism The philosophies of Hobbes and Spinoza exemplify a trend in modern philosophy that Leibniz labels ‘‘naturalism.’’ In a late commentary on Spinoza’s writings, he singles out Proposition 16 of Part 1 of the Ethics as a source for this doctrine: ‘‘From the necessity of the divine nature there must follow infinitely many things in infinitely many ways (that is, everything which can fall under infinite intellect).’’ ‘‘This view,’’ Leibniz argues, ‘‘is quite false, and makes the same mistake that Descartes insinuated, that matter successively accepts all shapes. Spinoza begins where Descartes leaves off: in naturalism’’ (AG 277).8 As Leibniz characterizes it, the fundamental error of naturalism is its conception of God as an infinite power whose effects are expressed necessarily, without the exercise of wisdom or choice. ‘‘As far as one can understand [Spinoza],’’ Leibniz writes in the Theodicy, ‘‘he acknowledges no goodness in God, properly speaking, and he teaches that all things exist through the necessity of the divine nature, without any act of choice by God’’ (T 173). Such a conception of God threatens the heart of Leibniz’s philosophical project, which seeks to comprehend the existence of all finite things in terms of their creation by a supremely wise and good God, whose justice is exhibited in his realization of the best of all possible worlds. The contingency of the world’s existence is an essential feature of Leibniz’s picture. If there were only one world that could exist, or if God’s power entailed the existence of all possible worlds, then the foundations of Leibniz’s theodicy would be undermined. There would be no way to defend God as a creator who exhibits justice in his actions, and who serves as a model for the agency of created rational beings. Furthermore, if the existence of the world and its constituents were necessitated by God’s action, then the claim of human beings to act freely would be compromised. The price to be paid by naturalism, therefore, is high: not only do we lose a conception of God as a fundamentally moral agent, but we also sacrifice the conditions for moral agency on the part of created rational beings. An early text by Leibniz fleshes out this picture of naturalism by linking it to the revival of the teachings of the two principal schools of Hellenistic philosophy: ‘‘There are two sects of naturalists fashionable today which have their source in antiquity; the one revives the opinions of Epicurus and the other is, in fact, composed of Stoics’’ (G 7:333/AG 281). Leibniz associates the revival of Epicureanism with the views of Hobbes: ‘‘Epicurus in times past, and Hobbes today, who hold all things to be corporeal, have given enough evidence to show that, on their view, there is no providence’’ (G 7:333/AG 282). According to Epicureans and their modern followers, any substance, including the soul and God himself, is composed of extended matter.9 If this is so, then God can be neither all-powerful nor allknowing; hence the case for God’s providence and justice cannot be made. The materialism of Hobbes and the Epicureans is one of the earliest targets of Leibniz’s criticism and a stimulus for his own efforts to preserve a place in nature for incorporeal, form-like principles. His objections to the teachings of what he calls ‘‘the sect of the new Stoics,’’ associated with Descartes and Spinoza, take a different tack. For these philosophers, he allows, there may be incorporeal

6 Introduction

substances, including God and human souls. Nevertheless, these ‘‘new Stoics’’ reduce God’s status to that of the soul of the world, or its ‘‘primary power,’’ which is determined to act by a blind necessity. Thus, in their case, too, divine justice is jeopardized. Against Spinoza, this charge is well founded.10 But how, one wonders, can Leibniz justify it in the case of Descartes, who goes to such lengths to defend the idea of God as a transcendent, supremely perfect being? Here Leibniz is certainly taking some liberty, but he reasons that the inner logic of Descartes’ philosophy leads him to a position that is—in its theological and moral consequences— indistinguishable from that of Spinoza. While Descartes affirms God’s omnipotence and supreme goodness, his understanding of the former precludes viewing God as a just creator. In support of this conclusion Leibniz cites Descartes’ doctrine of the creation of eternal truths, which he takes to imply the absence of objective standards against which to evaluate the justice of God’s action, and his pronouncement in Part 3 of the Principles of Philosophy that matter successively takes on all possible forms, which Leibniz takes to preclude the claim that nature always operates in the best, or ‘‘fittest,’’ manner.11 For these reasons, Leibniz concludes, Descartes’ God, or perfect being, is not a God like the one we imagine or hope for, that is, a God just and wise, doing everything possible for the good of creatures. Rather, Descartes’ God is something approaching the God of Spinoza, namely, the principle of things and a certain supreme power or primitive nature that puts everything into motion and does everything that can be done. (G 4:299/AG 242)

The fundamental threat of naturalism, as Leibniz sees it, lies in its rejection of an understanding of God as a wise and just creator, acting for the sake of intelligible ends. By reducing God’s role to that of an ultimate causal power, whose effects are expressed necessarily, naturalists reject the search for features of harmony and design in nature. Nature is as it is necessarily and not as a creator has intended it to be. Naturalism thus poses a direct challenge to traditional theological conceptions of the order of nature and of the place of human beings in nature. If God is not a wise and just creator in whose image human beings have been made, then there seems to be no special place reserved for human beings in nature, nor can human beings look to God for moral principles on which they may rely in governing their actions.

Nature and Natures These are conclusions that Spinoza openly embraces in the Ethics: human beings are to be understood as a part of nature, obeying the same necessary laws as other finite things. In the Preface to Part 3, Spinoza complains that ‘‘most of those who have written about the affects, and men’s way of living, seem to treat, not of natural things, which follow the common laws of nature, but of things which are outside nature. Indeed they seem to conceive man in nature as a dominion within a dominion’’ (CWS 491). The target of Spinoza’s criticism is the common view of human beings as both a part of nature and hence subject to natural laws, and at the same time superior to nature and capable of acting in ways that are not constrained

Introduction

7

by its laws. The view finds its sharpest expression in Descartes’ philosophy where human nature is cleanly dissected into the ontologically distinct essences of mind and body. Bodies act according to necessary laws of motion. Minds, which are present only in human beings, possess essential powers of reason and freedom on the basis of which it is concluded that ‘‘man disturbs, rather than follows, the order of nature, that he has absolute power over his actions, and that he is determined only by himself ’’ (CWS 491). Leibniz’s philosophy can be read as an attempt to find a middle way between the extremes of Spinoza’s naturalism, which limits human nature to the effects of necessary universal laws, and Descartes’ dualism, which preserves the uniqueness of human beings by positing within them a substance, res cogitans, whose actions ‘‘disturb, rather than follow, the order of nature.’’ In staking out this middle ground, Leibniz is led to defend his own version of a naturalistic theory of human beings, albeit one that is consistent with their creation by a wise and good God. According to Leibniz, divine wisdom requires creation of a world that is intrinsically intelligible, a condition that is met by ensuring that every being is endowed with a nature whose powers are sufficient to account for all of its effects. Leibniz’s naturalism is specifically intended to rule out the possibility that physical or psychological phenomena are in any way miraculous, that is, that they occur in a way that could be explained only by appeal to a direct intervention by God, or to ‘‘occult powers’’ that lie beyond the reach of reason. Instead, all natural phenomena can be explained in terms of the action of powers inherent in the natures of created substances. Like Spinoza, Leibniz finds Descartes’ dualism beset by conceptual problems, particularly Descartes’ failure to explain the mind’s ability to ‘‘disturb’’ the universal laws that govern the motion of matter. By virtue of what does mind have the power to move matter, and to do so in a way that routinely violates fundamental laws such as the conservation of momentum? Leibniz’s answer is that such problems can be avoided only if it is assumed from the start that mind and body do not stand in real causal relations to each other: mind cannot causally interact with matter, nor matter with mind. Up to this point Leibniz agrees with the conclusion drawn by Nicolas Malebranche, who develops Descartes’ philosophy in a system in which no finite thing operates as a true cause. As Malebranche would have it, causal power is exercised by God alone, who produces successive coordinated changes in the states of minds and bodies that leave the impression of direct causal interaction between them (both body-body and mind-body causation). Leibniz, however, rejects Malebranche’s occasionalism, seeking instead to preserve the identity of created substances as causal agents capable of bringing about changes in their own states. Against the thesis that God is responsible for all real change in the world, Leibniz defends what he calls the ‘‘system of concurrence,’’ or ‘‘preestablished harmony,’’ according to which God creates minds and bodies such that each, through the exercise of its own powers, in accordance with laws proper to its nature, brings about successive effects that are coordinated with one another. The impression that the mind is capable of directly causing changes in the body is, as Malebranche holds, an illusion; however, against Malebranche, Leibniz insists that subsequent changes in the states of

8 Introduction

both mind and body are the product of actions internal to each—actions that God, as a supremely wise creator, has chosen to harmonize with each other. As Leibniz makes clear in his 1695 essay New System of the Nature and Communication of Substances, it is basic to his theory that the agreement of soul and body and the changes internal to each can be explained naturally, in terms of the exercise of powers inherent in the substances themselves: [W]hy should God not be able to give a substance, from the beginning a nature or an internal force that can produce in it, in an orderly way (as would happen in a spiritual or formal automaton, but free in the case where it has a share of reason), everything that will happen to it, that is, all the appearances or expressions it will have, without the help of any created being? This is especially so since the nature of substance necessarily requires and essentially involves progress or change, without which it would not have the force to act. And since this nature that pertains to the soul is representative of the universe in a very exact manner (though more or less distinctly), the series of representations produced by the soul will correspond naturally to the series of changes in the universe itself, just as the body, in turn, has also been accommodated to the soul for the situations in which the soul is thought to act externally. (G 4:485/AG 144)

Leibniz’s efforts to distance himself from Descartes’ interactionism place him on a course parallel to that charted by Spinoza in the Ethics. Accepting that the fundamental difference between the properties of mind and body precludes their causal interaction, Spinoza defends a position that anticipates Leibniz’s preestablished harmony.12 Bodies and minds do not influence each other’s states causally. Modes of extension (bodies) interact causally with other modes of extension in accordance with laws of motion; modes of thought (ideas/ minds) interact causally with other modes of thought in accordance with psychological laws. Nevertheless, Spinoza insists that these two independent causal orders agree perfectly with each other: ‘‘The order and connection of ideas is the same as the order and connection of things’’ (IIP7). His demonstration of this proposition, however, proceeds very differently from Leibniz’s defense of the superficially similar thesis of preestablished harmony. Where Leibniz regards preestablished harmony as the result of God’s intelligent design of a universe in which the autonomous causal powers of soul and body are providentially aligned, Spinoza’s doctrine of parallelism is a consequence of the necessary and unitary order of reality, in which God and nature are one in substance. The causal order of bodies is God’s own infinite power expressed through infinite modes of extension, and the causal order of ideas is that same power expressed through infinite modes of thought. For Spinoza, all causal power is ultimately the power of God, the one substance from whose necessary existence there necessarily follow ‘‘infinitely many things in infinitely many ways.’’ Malebranche’s insistence that God is the sole true cause of all that happens in the world brings him, in Leibniz’s opinion, perilously close to Spinoza’s position. Although Malebranche is adamant that God is a transcendent being, who creates and conserves a world of separate, finite substances, Leibniz questions whether Malebranche’s occasionalism allows him to uphold such a view. As Leibniz sees it, in denying to finite things an internal

Introduction

9

principle of action sufficient to explain their identity through change, Malebranche in effect undermines the claim of finite things to be separate substances. Thus, Malebranche’s occasionalism is one step away from Spinoza’s monism.13 Leibniz’s disagreements with Malebranche and Spinoza over the correct account of the mind-body relation are rooted in deeper issues about the metaphysics of causation. In the case of all three philosophers, their explanations of the apparent causal relation between the human mind and the human body—responses to the perceived failure of Descartes’ interactionism—emerge as special cases of general theories of the causal relations that unite finite things with God and with each other. Leibniz’s theory posits a complex array of harmonies linking the states and actions of finite things—harmonies that are ascribed to God’s choice of this as the best of all possible worlds.14 With this he advances his broader program of turning back Spinoza’s theological naturalism: the limitation of God to the substance of the world, or an infinite and necessary causal power. At the same time, Leibniz upholds—against Descartes and Malebranche—his own brand of philosophical naturalism, which insists on the intelligibility and causal autonomy of nature: whatever happens in nature can be explained in terms of the exercise of finite causal powers.

Harmony and the ‘‘Two Natural Kingdoms’’ The fundamental principles of Leibniz’s philosophy remained unchanged through most of his career. Until the end of his life, however, Leibniz continued to refine the system he constructed around these principles, often framing his views in response to the probing of insightful correspondents.15 One of the fullest summaries of Leibniz’s final philosophy is contained in the famous Monadology, composed two years before his death. Leibniz begins by positing a single basic entity, the monad: a simple, soul-like substance, endowed with essential properties of perception and appetition. Every monad, according to Leibniz, is confusedly omniscient: it represents the universe from the perspective of a living body, and is distinguished from other monads by the distinctness of its perceptions, or the adequacy of its knowledge (Mon 60). As ontologically independent substances, monads stand in no real relations to each other; there is no common external space in which they are located. Nevertheless, as a condition of their forming a single world, monads stand in ideal relations to each other. God creates each monad such that its states are coordinated with those of every other monad. Since each monad represents the same universe from a different perspective, the ‘‘interconnection or accommodation of all created things to each other, and to all the others, brings it about that each simple substance has relations that express all the others, and consequently, that each simple substance is a perpetual, living mirror of the universe’’ (Mon 56). This constitutes the first and most basic level of harmony in Leibniz’s system: a ‘‘universal harmony, which results in every substance expressing exactly all the others through the relations it has to them’’ (Mon 59). By coordinating infinitely many diverse perceptual representations of an infinitely complex universe, God is

10 Introduction

able to assure that individual monads are related to each other in a world and that together these monads make up the best of all possible worlds: ‘‘this is the way of obtaining as much variety as possible, but with the greatest order possible, that is, it is the way of obtaining as much perfection as possible’’ (Mon 58). In conceiving of the world’s harmony, however, Leibniz does not stop here. Every monad also is endowed with its own body, together with which it constitutes an organism, or living being, such as a plant or an animal. The body of each organism is infinitely complex, consisting of smaller organisms within smaller organisms to infinity (Mon 64). And Leibniz insists that nature as a whole consists of bodies organized in this fashion: [T]here is a world of creatures, of living beings, of animals, of entelechies, of souls in the least part of matter. Each portion of matter can be conceived as a garden full of plants, and as a pond full of fish. But each branch of a plant, each limb of an animal, each drop of its humors, is still another such garden or pond. (Mon 66–67)

Leibniz’s descriptions of the microstructure of matter rely heavily on metaphor, yet they support the basic point that over and above the minimal ontology of soul-like monads, he defends a second level of ontological commitment, which regards the world as consisting of an infinite envelopment of living beings: soul-like monads together with their organic bodies. Leibniz’s explanation of the operation of these living beings generalizes from his hypothesis of preestablished harmony. Advanced in the New System as a solution to the Cartesian problem of mind-body communication, the theory now is extended to Creation as a whole: nature everywhere consists of soul-like monads whose activities are harmonized with those of their individual organic bodies, and thereby with the states of all other monads. Leibniz characterizes this scheme as involving a ‘‘perfect harmony between two natural kingdoms, the one of efficient causes, the other of final causes’’ (Mon 87): Souls act according to the laws of final causes, through appetitions, ends, and means. Bodies act according to the laws of efficient causes or of motions. And these two kingdoms, that of efficient causes and that of final causes, are in harmony with each other. (Mon 79)

Leibniz maintains that the ‘‘two natural kingdoms’’ are causally autonomous and universal. Thus, for him, nature everywhere consists of souls (or soul-like monads) acting in accordance with laws of final causation and bodies acting in accordance with laws of efficient causation. The interpretation of Leibniz’s ‘‘two natural kingdoms’’ brings with it some of the most difficult and hotly contested issues in recent Leibniz scholarship. On the face of it, the view is consistent with two very different ontological positions. Most straightforwardly, it might be thought, Leibniz’s final ontology is one of ‘‘living beings,’’ specified as ‘‘corporeal’’ or ‘‘composite substances’’ (Mon 61), to which simple substances contribute as organizing forms. On this reading, to the extent that corporeal substances appear to interact with other corporeal substances, they do so as bodies in accordance with laws of efficient causation. To the extent that

Introduction

11

they act purposively, for the sake of specific ends, they do so as souls in accordance with laws of final causation. The attractions of this reading are many, and it is surely correct to think that it is Leibniz’s position at one level of analysis. It may even have been his preferred theory prior to the emergence of the doctrine of monads in the 1690s.16 What is less clear is whether it is his ultimate metaphysical theory—his bedrock account of reality, as created by God. Weighing against this conclusion are numerous texts in which Leibniz explicitly limits reality to monads alone. As he writes in a 1704 letter to Burcher de Volder: [C]onsidering the matter carefully, we must say that there is nothing in things but simple substances, and in them, perception and appetite. Moreover, matter and motion are not substances or things as much as they are the phenomena of perceivers, the reality of which is situated in the harmony of the perceivers with themselves (at different times) and with other perceivers. (G 2:270/AG 181)

On this reading of Leibniz’s position, soul-like monads alone are fully real, or entia per se. They alone are genuine substances, each requiring nothing for its existence except God’s creation and conservation. All other things, including composite corporeal substances, are only ‘‘entities by aggregation [Entia per aggregationem], and to that extent phenomena, existing, as Democritus put it, by convention and not by nature’’ (G 2:252/AG 177). With this account in place, Leibniz’s ‘‘two natural kingdoms’’ end up having very different statuses, with the kingdom of efficient causes subordinated, via relations of ontological dependency, to the kingdom of final causes. As it is in itself, reality (i.e., monads) operates according to laws of final causation. However, that same reality is represented by monads as a world of bodies, interacting through collisions in space and time. Thus, the laws of efficient causation—laws of the motion and collision of bodies—apply only at the phenomenal level, anticipating the claim of Kant’s transcendental idealism. It remains an open question whether an idealism that recognizes monads alone as substances is Leibniz’s ultimate position. The textual evidence in support of this view is not unambiguous, even in writings from the final decade of his life.17

The Kingdoms of Nature and Grace Leibniz regards human minds as a subclass of monads that are distinguished by their capacity for reflection and understanding. This brings with it significant epistemic, moral, and theological consequences. While ‘‘souls, in general, are living mirrors or images of the universe of creatures,’’ Leibniz writes, ‘‘minds are also images of the divinity itself, or of the author of nature, capable of knowing the system of the universe, and imitating something of it through their schematic representations of it, each mind being like a little divinity in its own realm’’ (Mon 83). Minds are characterized most basically by their capacity to comprehend intellectual truth: the eternal and necessary truths of reason and contingent truths concerning the order and harmony of the created world. In virtue of this knowledge, they are able to grasp reflectively the ends that have guided God in creation,

12 Introduction

and they are able to structure their own actions around the pursuit of these same ends. For this reason, minds stand in a unique relation to God: This is what makes minds capable of entering into a kind of society with God, and allows him to be, in relation to them, not only what an inventor is to his machine (as God is in relation to the other creatures) but also what a prince is to his subjects, and even what a father is to his children. (Mon 84)

Collectively, rational minds comprise what Leibniz calls ‘‘the city of God’’: ‘‘the most perfect possible state under the most perfect of monarchs’’ (Mon 85). This ‘‘truly universal monarchy,’’ he writes, ‘‘is a moral world within the natural world, and the highest and most divine of God’s works’’ (Mon 86).18 The designation of rational minds as ‘‘a moral world within the natural world’’ returns us to Spinoza’s complaint about those who ‘‘conceive man in nature as a dominion within a dominion.’’ In one sense, this is an apt description of Leibniz’s position: rational minds stand in a fundamentally different relation to God, through whom they are constituted as a ‘‘dominion within a dominion.’’ In Leibniz’s view, however, the ‘‘moral world’’ to which minds belong can be fully explained in naturalistic terms. In contrast to Kant, the capacity of minds for moral agency does not presuppose a ‘‘transcendental’’ freedom. Instead, freedom is conceived as a power that is exercised under the constraints of natural causation. As Leibniz understands it, the highest form of freedom is ‘‘freedom of mind,’’ wherein an agent acts in accordance with reason’s judgments concerning the greatest good. At the same time, understood metaphysically, that choice is a spontaneous action, caused by the operation of a power inherent in the agent’s soul or substance. The final harmony that Leibniz invokes in the Monadology relates ‘‘the physical kingdom of nature and the moral kingdom of grace,’’ or ‘‘God considered as the architect of the mechanism of the universe, and God considered as the monarch of the divine city of minds’’ (Mon 87). The harmony of the kingdoms of nature and grace addresses the primary topic of Leibniz’s theodicy: the enactment of God’s justice with respect to the merit and welfare of human beings. A just God, Leibniz believes, must ensure that virtue is properly rewarded with happiness, and vice punished with unhappiness. Yet our experience of our earthly existence suggests that this balance often is not achieved: the virtuous suffer unaccountably, and the wicked appear to profit from their wickedness. The orthodox Christian response to this paradox is that a proper reckoning will come only in another life, in which those justified through Christ’s grace will enjoy the bliss of eternal happiness, while the rest will suffer eternal damnation. While professing allegiance to the Christian doctrine of justification, Leibniz presents a different picture of the operation of divine justice in the Monadology—one consistent with the broader features of his naturalism. On this account, the balance between merit and happiness is preserved by natural means within the earthly existence of human beings: God the architect pleases in every respect God the legislator, and, as a result, sins must carry their penalty with them by the order of nature, and even in virtue of the mechanical structure of things. Similarly, noble actions will receive their rewards through mechanical means with regard to bodies, even though this cannot, and must not, always happen immediately. (Mon 89)

Introduction

13

In response to the objection that the balancing of merit and happiness often does not seem to occur within our earthly existence, Leibniz argues that this existence cannot be measured by what we take to be a natural lifetime. Living beings do not, properly speaking, die. What we take to be death is merely a stage in life, an enfolding and diminution of the body, which is followed by a later regeneration, development, and growth (Mon 73). Divine justice, Leibniz claims, is expressed with respect to this succession of life-stages, which are united in the experience of a single rational agent. And, crucially, he insists that all of this is encompassed within God’s plan for the best of all possible worlds, in which ‘‘harmony leads things to grace through the very paths of nature’’ (Mon 88).

Leibniz: Nature and Freedom Leibniz’s voluminous writings, spanning more than half a century, contain a heady mix of rigorous argument and imaginative speculation. The views they defend, a cross section of which we have outlined above, have occupied commentators for more than three centuries. The essays gathered in this book are important new contributions to the Leibniz literature. The book opens with two studies that tackle different aspects of one of the primary topics of Leibniz’s metaphysics—the relation of unity and multiplicity—as it is explored in his early writings. Mark Kulstad’s essay, ‘‘The One and the Many and Kinds of Distinctness,’’ offers a fresh take on an issue that has been the subject of much recent scholarship: Leibniz’s reputed brief embrace of monism and pantheism—doctrines associated with Spinoza—during his formative Paris period (1672–76). Commentators have been divided on whether Leibniz’s metaphysical inquiries ever led him to a position resembling Spinoza’s. Kulstad sharpens the debate by framing it in terms of the kinds of distinction that might be held to exist between the unity of God and the multiplicity of created things. On his reading, in at least one text from the period Leibniz advances the thesis that there is no ‘‘real distinction’’ between God and created things. Given this, Kulstad argues, there is a sense in which Leibniz can be understood to defend versions of monism and pantheism, although not ones that immediately license the charge of Spinozism. In ‘‘Leibniz and Sleigh on Substantial Unity,’’ Christia Mercer examines the basis of Leibniz’s views on the unity of corporeal substance. Mercer begins from the influential analysis of Robert Sleigh, who links the unity of Leibnizian substances to their possession of an identity through change. For Leibniz, Sleigh argues, this identity is guararanteed by the presence of a substantial form: an internal force or power by virtue of which any substance satisfies the condition of spontaneity: ‘‘each noninitial state of an individual has as its real cause some predecessor state of that individual’’ (Sleigh 1990: 131). Mercer acknowledges the significance of Sleigh’s analysis, yet she argues that important questions about unity remain unanswered by it, above all, the problem of how a substantial form—a self-sufficient active principle—produces a unity in its associated passive principle, or body, whose components are in constant flux. Mercer contends that the key to this problem lies in an historically informed account of Leibniz’s early writings, which reveal his

14

Introduction

underlying assumptions about unity. According to Mercer, these assumptions are broadly Platonist in character, and include the idea that a more perfect being (soul) has the power to ‘‘emanate’’ certain of its essential properties (unity, self-sufficiency, indestructibility) to a less perfect being (body). Mercer draws on this background in outlining the suggestive theory of the ‘‘hypostatic union’’ of mind and body, which Leibniz defends in an essay from 1669–70. The next two contributions, by Samuel Levey and Daniel Garber, focus on the vexed issue of the ontological status of body in Leibniz’s middle period (roughly 1680–1700). Commentators have long held that Leibniz’s commitment to an idealist metaphysics—one that restricts an independent existence to mind-like substances and relegates material things to mind-dependent phenomena—can be traced at least as far back as 1686. One of the central planks in their argument is Leibniz’s frequent denial that there is any precise shape in corporeal things, and his related claim that the properties of extension, shape, and motion involve something ‘‘imaginary.’’ In ‘‘Leibniz on Precise Shapes and the Corporeal World,’’ Samuel Levey takes issue with this conclusion. On Levey’s account, Leibniz’s denial is restricted to the ‘‘precise shapes’’ of traditional geometry, and thus is consistent with bodies possessing ‘‘fractal shapes’’ of infinite complexity. Levey shows that Leibniz’s views on the fine structure of matter and motion (developed in conjunction with his resolution of the ‘‘labyrinth of the continuum’’) support such a position, which parallels his reductive analysis of the property of motion. In both cases, Levey concludes, Leibniz’s analyses allow for ‘‘a nuanced realism about shape and motion as properties that belong to bodies as they are in themselves’’— and this is consistent with (though it does not entail) a realistic interpretation of corporeal substance. Daniel Garber is well known for his defense of Leibniz’s realism about corporeal substance during his middle period (Garber 1985). In ‘‘Leibniz and Idealism,’’ Garber revisits this reading, responding to criticisms raised by Sleigh (1990) and Adams (1994), and offering a broader perspective on Leibniz’s metaphysical outlook during the 1680s and 1690s. It is a mistake, Garber argues, to insist on a nascent idealism in Leibniz’s middle period based on views he will come to espouse later. Texts from the middle period offering prima facie support for an idealist interpretation can be reconciled, according to Garber, with Leibniz’s contemporaneous account of causal relations among animated bodies. At the same time, Garber allows that there is little evidence of Leibniz defining himself as a realist about corporeal substance during the period. In the course of a concerted attack on the Cartesian account of body, Leibniz relies implicitly on the supposition of corporeal substances, conceived as form-matter composites on analogy with living bodies. But he largely ignores the troubling metaphysical issues— central to his later thought—that such a supposition raises. The concept of harmony is fundamental to Leibniz’s understanding of the relations that unite a plurality of substances in a single world. Substances harmonize in their perceptions of a common universe, and changes in individual souls harmonize with changes in their respective bodies. The essays by Catherine Wilson and Nicholas Jolley both raise challenges to features of Leibniz’s hypothesis of preestablished harmony. Wilson’s ‘‘Compossibility, Expression, Accommodation’’

Introduction

15

responds to a puzzle Sleigh (1990) has posed about Leibniz’s account of the worldconstituting relations among substances. A number of texts suggest that Leibniz is inclined to regard the relations of compossibility, expression, and accommodation (or harmony) as entailing one another: any collection of compossible substances will be one in which each member of that collection both expresses and harmonizes with every other. But Sleigh objects that this is hardly obvious: compossibility is a necessary fact about the substances comprising a single world; by contrast, expression and harmony appear to add increasing degrees of perfection to a world. In Wilson’s words, they are ‘‘value-adding relations.’’ Wilson proposes a novel way of connecting these concepts. Universal expression is a feature of any Leibnizian possible world and is defined in terms of the correspondence among the contents of the perceptions of soul-like substances. Accommodation, on the other hand, is restricted to those worlds in which a set of more complex agreements obtains among the interacting, spatiotemporal bodies ‘‘projected’’ by those substances. In ‘‘Leibniz and Occasionalism,’’ Nicholas Jolley addresses another perennial topic of Leibniz scholarship: Leibniz’s defense of the system of preestablished harmony against Malebranche’s occasionalism. One line of criticism, going back to Arnauld, has challenged whether Leibniz succeeds in distinguishing preestablished harmony as a genuine alternative to occasionalism. Other critics have charged that Leibniz’s two main arguments against Malebranche—that occasionalism requires a God who ‘‘disturbs’’ the laws of nature, and that it presupposes the occurrence of ‘‘perpetual miracles’’—both miss the mark. According to Jolley, Malebranche has an easy response to the first of these arguments. The second, however, hinges on a much-discussed issue in recent philosophy of science: whether laws of nature constitute a ‘‘rock bottom’’ in scientific explanation, or whether such laws themselves require an explanation in terms of causal powers. Malebranche and Leibniz agree that God governs nature through general laws, and that miracles are to be understood as exceptions to these laws. Their disagreement turns on a further condition Leibniz imposes on laws of nature. Against Malebranche, Leibniz holds that a law-like principle is not truly a law of nature unless it is grounded in the causal powers of created substances. Since Malebranche denies the existence of finite causal powers, by Leibniz’s lights he is committed to a denial of the existence of laws of nature; hence the charge of perpetual miracles. Jolley goes on to show how Leibniz’s late metaphysics has the resources to overcome some of his disagreement with Malebranche, by restricting his grounding requirement to monads and settling for laws at the phenomenal level of bodies. However, the evidence suggests that Leibniz was unwilling to make this move, requiring that laws of impact be grounded in the physical forces of matter, which in turn are grounded in the primitive forces of monads. For Leibniz, the laws of nature include not only physical laws of efficient causation, governing the motions and collisions of bodies, but also laws of final causation, or appetite, which govern changes in the states of monads. In ‘‘Leibniz’s Two Realms,’’ Jonathan Bennett surveys the range of claims Leibniz makes on behalf of teleological concepts in the explanation of nature, and the harmony he posits between parallel orders of final causation and efficient causation. Bennett argues that in almost every case Leibniz’s assertions of the importance of teleology

16 Introduction

do not withstand scrutiny. The teleological nature of monadic appetitions is ‘‘dubious at best’’; and Leibniz fails to formulate the notion of a ‘‘teleological pattern of behavior’’ that is required to frame an informative thesis about the harmony between ‘‘teleological events in the mind and mechanistic ones in the body.’’ In the end, Bennett concludes that the ‘‘teleological load’’ in Leibniz’s philosophy is carried entirely by reason, to which Leibniz assigns both a cognitive and a conative function. Reason teaches human beings what the good is and disposes us to pursue it because it is good. Thus, there is at least one case in which Leibniz’s claims for teleology are borne out. Leibniz’s conception of rational action as action motivated by reason’s perception of the good is the starting point for the remaining essays, all of which are concerned with Leibniz’s treatment of the human will and its freedom. In ‘‘Leibniz on Spontaneity,’’ Donald Rutherford explores the conception of spontaneity, or self-determination, that is presupposed by Leibniz’s analysis of freedom. Traditionally, spontaneity has been understood to imply an absence of constraint or compulsion: a spontaneous act is one in which an agent moves herself as opposed to being moved by external causes. In the Theodicy, Leibniz makes it clear that by ‘spontaneity’ he means the power of self-determination that is the source of all the actions of every monad, whether rational or not. But if this so, then his account of spontaneity seems too broad to tell us anything informative about freedom; in particular, the notion of monadic spontaneity fails as a basis for distinguishing free acts from those in which an agent is physically constrained. Rutherford argues that Leibniz’s writings also offer support for a narrower notion of agent spontaneity, which characterizes cases in which an individual can be understood as acting for the sake of the greatest apparent good. Rutherford shows how the latter notion is defined within the framework of monadic spontaneity, and why Leibniz believes that both sorts of spontaneity are relevant to the understanding of freedom. The contributions by Robert Adams and Michael Murray target Leibniz’s doctrine of moral necessity, a species of determination that he takes to be consistent with the contingency of free acts. In ‘‘Moral Necessity,’’ Adams argues that the distinctive feature of this modality is that the notion of value—good or perfection— plays an essential role in the determination and explanation of choice. As such, Leibnizian moral necessity is opposed both to metaphysical necessity (that whose denial implies a contradiction) and to ‘‘any value-free mechanism determining choice.’’ Nevertheless, the concept of moral necessity is consistent with Leibniz’s efforts to interpret contingency as a species of hypothetical necessity, or determination in accordance with universal laws. On Adams’s account, ‘‘moral laws,’’ according to which ‘‘the value or apparent value of an end, or some analogue of apparent value, is sufficient of itself to determine the action of a substance,’’ ‘‘are strong enough to determine choice infallibly in advance.’’ The crucial point is that there is no deeper explanation of how such final causation operates—in particular, no explanation in terms of mechanical causes. Adams concludes by considering how this reading comports with the defense of contingency that Leibniz offers in connection with his conceptual containment theory of truth. That an agent in acting freely could have acted otherwise is supported by the thought that the agent’s

Introduction

17

complete concept contains as primary constituents not the free choices themselves but ‘‘value-laden reasons’’ from which those choices follow by moral necessity, and that no finite analysis would suffice to derive the choices from their reasons. Adams expresses skepticism, however, as to whether any further mitigation of Leibniz’s determinism is possible. The topic of moral necessity is approached from a different direction by Michael Murray in ‘‘Spontaneity and Freedom in Leibniz.’’ Building on an interpretation defended in earlier writings, Murray argues that Leibniz’s conception of moral necessity is indebted to accounts of the notion developed by late-sixteenthcentury Spanish Jesuits, and that, according to such accounts, moral necessity entails the denial of psychological determinism, or the view that volitions are ‘‘causally necessitated by psychological antecedents.’’ Like Adams, Murray interprets moral necessity teleologically: in acting freely, an agent acts for the sake of a perceived good. In contrast to Adams, however, Murray does not see moral necessity as defined simply by the rejection of a ‘‘blind’’ or ‘‘value-free’’ mechanism determining choice. In Murray’s view, an assertion of moral necessity presupposes the denial that an act of choice is either physically or causally determined by antecedent conditions, although such conditions may nonetheless be sufficient for choice. In support of this reading, Murray appeals to texts in which Leibniz endeavors to defend the independence of human freedom from divine determination. In order for creatures to be held responsible for sin, God, through his willing, cannot be understood to be the cause of sin. For a free action to be genuinely a creature’s own, the creature must have ‘‘an active will, which is not psychologically determined.’’ Murray concludes that theological worries of this sort offer ‘‘powerful evidence of anti-compatibilist sympathies in Leibniz’s thought,’’ and that these sympathies favor the adoption of Jesuit moral necessity as a ‘‘buffer’’ between God and creatures. The final two essays, by Sean Greenberg and Jack Davidson, focus on the role of ‘‘intelligence’’ in Leibniz’s account of freedom. In ‘‘Leibniz against Molinism: Freedom, Indifference, and the Nature of the Will,’’ Greenberg analyzes Leibniz’s rejection of the doctrine of freedom promulgated by the sixteenth-century Jesuit Luis de Molina and elaborated by his successor Francisco Sua´rez. Central to the Molinist account is the notion of the will’s indifference: an agent is free if, with all the requisites for action posited, she is able to act or not act. It has been argued that Leibniz’s opposition to this conception of freedom hinges on its incompatibility with the principle of sufficient reason. For the Molinist, the will is capable of determining itself to one of a pair of alternatives, independent of antecedent grounds that represent one as better than the other. Arguably, this would be a choice for which there was no sufficient reason. Greenberg contends that Leibniz’s case against the Molinist goes beyond this, resting ultimately on their diverging conceptions of mind and will. According to Greenberg, in Leibniz’s account of freedom, as against that of the Molinist, the ‘‘will has no distinct functional role to play.’’ Rather than being an independent ‘‘self-determining power,’’ the will is a tendency or effort that depends entirely upon the mind’s representational states. Thus, for Leibniz, the issue of freedom turns not on what the will can or cannot do but on the distinctness of the mind’s perceptions. The highest ‘‘freedom of mind’’

18 Introduction

is the condition in which choice is determined by reason’s representations of the good (and not by the passions); hence Leibniz’s claim in the Theodicy that intelligence is the ‘‘soul’’ of freedom. Leibniz’s intellectualism is further explored by Jack Davidson in ‘‘Video Meliora Proboque, Deteriora Sequor: Leibniz on the Intellectual Source of Sin.’’ As Davidson defines it, intellectualism is the view that all volitional error (or sin) is preceded by cognitive error, or, conversely, that knowledge is sufficient for virtue. Davidson contends that Leibniz subscribes to an intellectualist position throughout his career. In later writings, however, Leibniz realizes that a potent form of intellectualism to which he had been attracted—one identifying volition with intellectual judgment—‘‘faces grave difficulties in explaining cases of incontinence and in assigning culpability for sin.’’ Davidson argues that Leibniz’s considered view is a ‘‘weak intellectualism,’’ which holds that sin always involves cognitive error but does not depend solely on cognitive error. The additional factor incorporated in Leibniz’s account is the influence of the passions: appetitive and aversive states that ‘‘unseat the rule of reason, without themselves being cognitive.’’ As Davidson explains it, the passions’ effectiveness in this regard depends upon the ‘‘essential temporality of practical reasoning.’’ Subsequent to the intellect’s judgment that a particular sinful act is to be avoided, the passions may ‘‘distract or misdirect’’ the intellect’s focus from the wrongness of the act to the pleasure anticipated in doing it. For Leibniz, the choice of the pleasurable over the virtuous presupposes a cognitive error—a judgment that the former promises a greater good than the latter. Davidson’s point, however, is that this can occur in a way that is outside the intellect’s control: an agent may retain the knowledge that a certain act is wrong and yet choose to perform the act, because of inattention and the nonrational force of the passions. Accordingly, Davidson concludes, Leibniz’s program of moral education depends as much on the habituation of virtue as it does on the acquisition of intellectual knowledge. In this introduction we have attempted to convey a sense of the breadth of Leibniz’s philosophical program and the diversity of topics and approaches that characterizes recent scholarship on it. It is hoped that the essays gathered here provide the inspiration for many more studies of Leibniz’s philosophy.

Notes 1. See also Cassirer 1902, which has had less influence on the literature. 2. Couturat (1901: x–xi) took Leibniz’s position to be expressed most clearly in the study known as Primae Veritates (‘‘First Truths’’), now dated to 1689 (A 6.4:1643–1649/AG 30–34). 3. This applies only to the English-language literature. On the Continent, there appeared inter alia the pioneering studies of Grua (1953, 1956) and Heinekamp (1969). 4. Parkinson 1970 was the first significant study of the topic in English. The other of Leibniz’s two labyrinths, that of the continuum, did not fare significantly better in the literature. See now, though, Levey 1998 and Richard Arthur’s introduction to LOC. 5. Adams (1994) links this renaissance to contemporary developments in modal logic and metaphysics, which have relied heavily on the notion of possible worlds. Against the Russell-Couturat interpretation, Adams argues that Leibniz’s view of problems concerning modality and identity ‘‘was

Introduction

19

largely shaped by his theology, and his use of logical doctrines in dealing with them cannot be understood apart from such metaphysical conceptions as that of substantial form’’ (4). 6. See, e.g., Garber 1985 and C. Wilson 1989. 7. For further texts and discussion, see Rutherford 1996 and Mercer 2001. 8. ‘‘Spinosa incepit ubi Cartesius desinit: In naturalismo’’ (Beeley 2002: 10). 9. In the main text of Leviathan Hobbes stops short of asserting the corporeal nature of God (see chap. 12.7). Such an assertion does appear, however, in chap. 3 (‘‘On certain objections against Leviathan’’) of the Appendix to the 1668 Latin edition, which Leibniz may have known. It is reproduced in Curley’s edition (Hobbes 1994: 540–542). 10. Arguably the charge fails in the case of the ancient Stoics, who uphold divine providence and, in one sense, divine justice. For further discussion, see Rutherford 2001, 2003. 11. For Descartes’ statement of the first view, see his letters to Mersenne of April 15, May 6, and May 27, 1630 (CSM 3:20–26), and his Replies to the Sixth Objections (CSM 2:291–292, 293–294); for the second, see Principles 3.47 (CSM 1:257–258). 12. See, in particular, paragraph 85 of the Treatise on the Emendation of the Intellect, which Leibniz knew from Spinoza’s 1677 Opera posthuma: ‘‘[O]bjective effects proceed in the soul according to the formal nature of its object. This is the same as what the ancients said, i.e., that true knowledge proceeds from cause to effect—except that so far as I know they never conceived the soul (as we do here) as acting according to certain laws, like a spiritual automaton’’ (CWS 37). For Leibniz’s appropriation of the term ‘spiritual automaton,’ see the passage quoted earlier from the New System and T 52. 13. See On Nature Itself, sec. 15 (G 4:515/AG 165–166). 14. For an account of how the maximization of harmony underwrites this being the best of all possible worlds, see Rutherford 1995a. 15. See the articles collected in Lodge 2004. 16. For an influential account of Leibniz’s shifting views during the 1680s and 1690s, see Sleigh 1990, chaps. 5 and 6. 17. For arguments supporting an idealist reading, see Adams 1994 and Rutherford 1995a. For arguments against, see Hartz 1998. 18. Leibniz’s description of the ‘‘city of God’’ as a ‘‘truly universal monarchy’’ is at odds with Augustine’s limitation of its membership to the Christian elect. Leibniz’s city resembles more closely the Stoic cosmopolis: ‘‘a single city and state shared by humans and gods’’ (Cicero, De finibus 3.64).

1

The One and the Many and Kinds of Distinctness The Possibility of Monism or Pantheism in the Young Leibniz MARK KULSTAD

The One and the Many It would be no doubt an exaggeration to say that Leibniz was obsessed by the various forms of the one-many problem. But the issue certainly pervaded his philosophical thought. From universal harmony to perception, from corporeal substance to theology, from philosophy of science to nominalism, the problematic of the one and the many played an important role in his reflections throughout the course of his philosophical life. If many individual aspects of this emphasis are well known, their combined weight has perhaps been insufficiently appreciated. Consider, to begin, an issue currently being examined in great depth: (1) the issue of corporeal substance in the philosophy of Leibniz.1 Is a corporeal substance, somehow involving infinitely many substances, itself also strictly one substance, strictly an unum per se? And if so, exactly how do the many substances become one? Arnauld was one of those who pressed Leibniz on this (see, for example, LA 65–66, 108–109). Is the corporeal substance perhaps—as Hans Poser has suggested recently—something strictly less than an unum per se but also something more than simply an unum per aggregationem, some sort of derivative unity?2 If so, the exact sense of this unity or oneness would be of considerable interest. Commencing from the issue of corporeal substance, one might readily proceed in either of two opposite directions. (2) One might withdraw into the interior life of a single Leibnizian substance to consider the case of perception. Perception for Leibniz is simply the expression of the many in the one,3 thus providing as it were a psychological perspective on our problem.4 Bayle was one of those who pressed Leibniz for a clearer solution in the case of a simple substance, somehow 20

The One and the Many

21

one despite the multiplicity within it; Leibniz conceded to him that the issue was ‘‘one of those which are most in need of clarification’’ in his philosophy.5 (3) Alternatively, one might turn outward to the exterior complexity of the whole world to consider the case of Leibnizian universal harmony. This universal harmony is, for Leibniz—as of course for him all harmonies are—a unity in multiplicity, only in this case among all things (A 2.1:333). Both of these examples bring the issue of the one and the many into the spotlight, though at very different levels of analysis, the psychological and the cosmological—perhaps not surprising in a philosophy where the microcosm mirrors the macrocosm. With only a passing reference to (4) Leibniz’s nominalism, which of course provides the German philosopher with a response to one of the most historically familiar versions of the problem of the one and the many,6 let us turn to philosophical theology and its relation to the one and the many in Leibniz’s thought. Here we find, first, the question (5) how the one God gives rise to the infinitely many of this world (Kulstad 1999). Second, we find the complicating feature in Leibniz’s philosophical theology that (6) the one God is really triune, that is, a God that is one but somehow involves three persons—the God of the Trinity.7 (Note that this is, atypically for Leibniz, an example of the problem of the one and the finitely many.) Finally, in an issue that has recently gained prominence in connection with discussions of infinity in Leibniz, we find Leibniz subject to change on the question (7) whether God is the soul of the universe.8 Early in his life, at least briefly, Leibniz endorsed the view that God is the soul of the universe (A 6.3:474/DSR 25). In his maturity, he fought against the view with some passion (Carlin 1997). The topic is relevant to the current discussion in that such a relation between God and the world would be one way to bring unity, or oneness of some kind, to the multiplicity of God and the infinitely many items of this world. Although this is not the specific topic that will be taken up in this essay, there is some relation between it and the main issues I will discuss here. While one could provide further examples in this vein, issues (1)–(7) suffice to make it understandable why Heinrich Schepers has said recently that ‘‘one can view Leibniz’s metaphysics as the work of rationally explicating the different expressions of the polarity of the one and the many’’ (2000: 171).9 It is not, to be sure, the project of this essay to canvass this whole topic. Rather I shall aim to contribute something in connection with an intriguing and important aspect of it, not appearing in (1)–(7) just sketched. It is an issue highlighted by recent discussions of the question whether Leibniz ever adopted the monism (some would say pantheism) of Spinoza, that is, whether Leibniz ever seriously considered a philosophical solution of the problem of the one and the many, which made the multiplicity of the world so many different aspects or manifestations or modes of one single divine substance, or of a divinity uniting, and in some sense not really distinguished from, all existing things. In continuing this discussion, my ultimate purpose is to focus attention on issues important not only for the question of the one and the many but also for clarifying a number of critical concepts in Leibniz’s philosophy, early and late.10 Since ‘pantheism’ and ‘monism’ are both controversial and laden with multiple meanings, a preliminary discussion of the matter of terminology is in order.

22 Leibniz: Nature and Freedom

The monism we are interested in is, roughly speaking, the doctrine that there is only one substance. A relevant variant of this view is one in which the one substance is God. (Obviously, Spinoza’s system is a paradigm of monism in this sense.) Pantheism in its etymological sense is the view that all things are God. The basic idea here is quite frequently put in a pair of statements, as in ‘‘God is All and All is God,’’ and is often linked to, or equated with, the idea that God is identical to the world rather than distinct from it.11 It is typically held also that pantheism requires a special unity of all things. Obviously, more could and should be said about the identity and distinctness just mentioned—and it will be, later in this essay. But for the present, these preliminary definitions and remarks may serve as a reasonable basis for discussion. Turning from the definitions of these terms to the relation between them, we should note that, especially in connection with Spinoza, the terms ‘pantheism’ and ‘monism’ are sometimes used interchangeably. While this procedure is perhaps justifiable in some contexts, the associated concepts should in general be distinguished. I will here follow Michael Levine, the author of Pantheism and useful related articles,12 in saying that not all forms of pantheism are monistic, and not all forms of monism are pantheistic (1992: 95–98). The former point allows for pantheistic systems that are not based on a substance-metaphysics at all, and also for ones that are so based, but involve many substances rather than just one, substances nonetheless integrated into the pantheistic divine unity that Levine emphasizes. The latter point allows for a monism of one substance, but one in which ‘‘‘Unity’ in a relevant pantheistic sense is absent’’ (Levine 1992: 98). Such a system would not, for Levine, count as pantheistic. It is worth noting that in the end Levine endorses the idea of Spinoza as both a monist and a pantheist, thereby acknowledging that unity in a relevant pantheistic sense is present in Spinoza.13 Since Spinoza holds that the one substance, God, is indivisible,14 this suggests, what I shall in fact assume for purposes of this essay, that a monism in which the one substance is both divine and indivisible falls under the relevant pantheistic sense of ‘unity’ and hence counts as pantheistic. This still leaves some question as to the exact boundaries of unity in a ‘‘relevant pantheistic sense’’ in other contexts. But I shall not pursue this detail further here.15 With these terminological points addressed, and returning to the subject at hand, we note that a natural first response to one aspect of the topic proposed earlier is likely to be simply that since Leibniz is so obviously opposed to monism—a fortiori, to monistic pantheism—there is little point in examining any claimed Leibnizian endorsement of these two positions. Leibniz, some might insist, positively reveled in the multiplicity of the infinitely many individual substances, the world of monads, that he made famous. He clearly opposed himself to Spinoza on the point of monism versus monadism, indeed, so strongly that it is implausible that he ever took a supportive view. The following passage might serve as a basis for such a sentiment: [I]t is through these very monads that Spinozism is destroyed, for there are just as many true substances, as many living mirrors of the universe which subsist always, or as many concentrated universes, as there are monads; according to Spinoza, on the contrary, there is only one substance. He would be right if there were no

The One and the Many

23

monads; then everything except God would be of a passing nature and would vanish into simple accidents or modifications, since there would be no substantial foundation in things, such as consists in the existence of monads. (to Bourguet, December 1714; G 3:576)

Now I have no intention of opposing the view that the mature Leibniz was indeed a pluralist, that is, that he believed in his maturity that there are many substances rather than just one. But despite the forcefulness of passages such as this one, there is profit to be made by looking into passages that some (myself included) have taken to suggest that at certain points in his youth Leibniz was, or may have been, a monist and/or a pantheist.16 The payoff, again, is that it helps bring into sharper focus certain important issues for Leibniz’s philosophy, youthful and mature, that might not otherwise be sufficiently understood or appreciated.

Kinds of Identity and Distinctness in Relation to Monism and Pantheism Let us begin by picking up the thread of identity and distinctness in connection with some puzzling—some might even say mystifying—accounts of the doctrine of pantheism. What could it mean, for instance, to say that God is identical to the world, as opposed to distinct from it? Or what could it mean to say that God is all or is everything? Leibniz himself appeared to see serious difficulties here, for he expressed himself quite strongly to Eckhard as follows: ‘‘it seems impossible for there to be a Being that is everything; for it could be said of such a Being that it is you and it is me, which I think you will not admit’’ (28 April 1677; A 2.1:323, quoted in Adams 1994: 130–131). Conceivably, Leibniz had a simple argument in mind: if a Being were identical to everything, then it would be identical to you, and identical to me; in this case, by the transitivity of identity, you and I would be identical; but we are not identical; so the supposition is absurd.17 Clearly, conceptions of identity and distinctness are fundamental here, as they clearly were in our initial formulations of pantheism. But if the foregoing reconstruction of Leibniz’s thought correctly identifies the specific sense of identity that those who have claimed that God is all and all is God had in mind, then the claim is indeed puzzling. This suggests that we need to look harder at notions of identity and distinctness in connection with pantheism (and, as we shall see, in connection with monism as well). Some remarks of G.H.R. Parkinson will serve to motivate further this closer look. While in the end coming down on the side of those who see no pantheistic tendencies in the philosophy of Leibniz in 1675–76, Parkinson points out ‘‘a remarkable passage . . . which may be thought to be clearly pantheistic. Leibniz says, ‘Just as the form of the number three differs from all its parts, so do things differ from God. God is all things; creatures are some things.’ ’’18 Parkinson’s comment on this is that ‘‘Leibniz seems to be saying here that God is all things, and that creatures—i.e. what are normally called God’s creation—are parts of God. It is unlikely, however, that this really was Leibniz’s view’’ (Parkinson 1978: 88). I won’t take issue with the interpretation itself, here. Rather, I want to focus on the issue of identity and distinction that is imbedded in Parkinson’s formulation

24 Leibniz: Nature and Freedom

that God is all things and that things, or what are normally called creatures, are parts of God. It seems clear enough that Parkinson here offers a part-whole formulation of a proposed (pantheistic) God-world relationship: God is a whole in which the things of this world are parts, and in this sense God is the whole thing. But notice that the objection Leibniz raised in philosophizing with Eckhard does not seem to apply to this formulation. There, Leibniz to all appearances interpreted the idea of God being everything in terms of God being numerically identical with each thing. Via the transitivity of numerical identity, he would then be able to derive what he took to be the absurdity that one person would be numerically identical to another. Such an absurdity does not follow from a partwhole formulation of the God-world relationship of the sort just presented. On this formulation, it is not the case that God is numerically identical to each thing of this world. Rather, God is numerically identical to a whole, and each of the things of this world is a part of this whole. Since it is not the case that God is numerically identical to these parts taken individually, the inference Leibniz presented to Eckhard is blocked. So far, in discussing Parkinson’s ‘‘remarkable passage’’ and Parkinson’s commentary on it, we have focused on identity. But looking at the same issue with our focus turned instead toward distinctness is also worthwhile.19 Consider the following question: assuming the part-whole formulation of pantheism, is a thing of this world distinct from God? Neither yes nor no is a completely satisfying answer. Either answer seems to require further explanation. Yes, some will be tempted to say, any thing of this world is distinct from God given the part-whole formulation because no thing (e.g., Eckhard) simply is God. Eckhard, for instance, would indeed be a part of God on this formulation, but he would not be numerically identical to God. Others, however, may be tempted by a different answer. No, they might say, on the part-whole formulation a thing of this world is not truly distinct from God, for each such thing is a part of God, is within God. To be sure, no such thing is numerically identical to the whole, to God. But the thing is numerically identical to something that is a part of God. In this sense, it is not distinct from God. It is not something that exists over and above God, something that exists in addition to God. It is something that exists as part of God. What is suggested here is a sense of ‘distinctness’ different from the sense of ‘numerical distinctness.’ A proper part, we might say, is indeed numerically distinct from the whole of which it is a part, but it is not distinct from the whole in another sense, a sense that we can here specify as follows, under the label ‘mereological distinctness’: x and y are mereologically distinct iff (i) x and y are numerically distinct and (ii) neither x nor y is a part of the other.

While this modestly formal statement may seem foreign or strained, I submit that it captures a fairly standard aspect of ordinary human communication about things, about whether or not some thing a and some thing b are distinct from, other than, or separate from one another. But whether or not this point meets with agreement, I contend that in connection with issues of monism and pantheism a good argument that this sense of distinctness is valuable is that it affords a basis from which to

The One and the Many

25

give a defense of Spinoza’s argument for monism (Kulstad 1996). This encourages the thought that the examination of this specific sense of distinctness, and, more generally, various specific senses of identity and/or distinctness involved in various forms of monism or pantheism, can be quite important in thinking clearly about these issues. Continuing in this vein, I would claim that another sort of distinction is significant in this connection, a sort well known to students of seventeenth-century philosophy. This is real distinction (as opposed to a modal distinction or a distinction of reason), the form of distinction crucial in Descartes’ argument for mind-body dualism and also taken up by Spinoza in his attempted defense (very much in opposition to Descartes) of the idea that the attributes of thought and extension might both be attributes of the same substance.20 It is here that we will find the main textual focus of this essay. In approaching the texts, the following senses of ‘monism’ are available to us.21 The list is not meant as exhaustive. (A) MonismND is true iff there exists an x such that x is substance and there exists no y such that y is substance and x is numerically distinct from y. (B) MonismMD is true iff there exists an x such that x is substance and there exists no substance y such that y is mereologically distinct from x.22 (C) MonismCD is true iff there exists an x such that x is substance and there exists no substance y such that y is constituentially distinct from x.23 (D) MonismRD is true iff substance exists and there exist no x and y such that x and y are substances and x is really distinct from y.24

A similar list of senses of ‘pantheism’ is available to us. Note that in all of these definitions, it is assumed that the God that is said to exist has unity in the relevant pantheistic sense, whatever that sense might be exactly. The definitions involve various ways of spelling out the rough idea that God is all, or that nothing is ‘‘distinct’’ from God. (A) PantheismND is true iff God exists and nothing is numerically distinct from God. (B) PantheismMD is true iff God exists and nothing is mereologically distinct from God. (C) PantheismCD is true iff God exists and nothing is constituentially distinct from God. (D) PantheismRD is true iff God exists and nothing is really distinct from God.

If not complete, the latter list captures a good bit of the logical and historical space occupied by discussions familiar to most of us. Pantheism of form (A) seems to be approximately what Leibniz had in mind as a target when he ridiculed a pantheism-like doctrine in discussions with Eckhard (see earlier). Assuming the truth of pantheism in the sense of formulation (A), that is, the truth of PantheismND, each thing—including Leibniz and Eckhard—would be numerically

26 Leibniz: Nature and Freedom

identical to God. The consequence that each human (along with each rock along the road, and each atom in the void, etc.) would count as the full (not partial) divinity has fascinated (and/or repulsed) thoughtful people for centuries. The consequence that Leibniz seems to have highlighted, however, was that in this case he and Eckhard would not be numerically distinct (as they had of course been assuming); rather they would be numerically identical. This was a consequence Leibniz did not think Eckhard would accept.25 Formulation (B) covers the sort of thing that many have in mind in more popular contemporary discussions of pantheism, namely, the vaguely scientificreligious idea that considers the whole universe of contemporary science (assume it as materialistic and finite, for present purposes) to be somehow ultimately unified and divine, with each material thing in the universe (including human life forms) being a part of the pantheistic whole, a pantheistic God. On this view, God is All (the universe, or Weltall, as the Germans would say), and individual material things are part of the All. Putting this a bit more technically, on this view nothing in the universe is mereologically distinct from the whole, which is God, even though each individual material thing in the universe is numerically distinct from the whole (a point that clearly separates this example from the previous one). Putting this in less formal terms, nothing in the universe is something over and above God, nothing is external to God.26 Formulation (C), with its emphasis on constituency rather than parts, would allow for a pantheism in which all things go to make up God in some sense, but not strictly in the sense of being parts of God—perhaps because (as in the case of Spinoza’s system) God would somehow not be taken to be divisible into parts.27 Of course, in terms of specific cases, the example just given would already count as an example of PantheismCD, since in it God exists and nothing is mereologically distinct from God. It follows from this that God exists and nothing is constituentially distinct from God,28 which is just PantheismCD. But if one wants at least the suggestion of a more concrete example of the latter type of pantheism that is not also an example of the former, consider, from a specific perspective, the example I have already mentioned, provided by Leibniz in a passage that Parkinson calls ‘‘remarkable’’: ‘‘God is all things; creatures are some things’’ (Parkinson 1978: 88).29 Now one could (and Parkinson himself does, as shown earlier) interpret this in terms of whole and parts. But such a reading is not forced by the passage.30 Leibniz, or someone else propounding a verbally similar view, might have in mind a doctrine in which the relation implied between God and things was not that of whole and parts, but rather something less strict, something nonetheless falling under some notion of constituency.31 Formulation (D) can also be related to a philosophically familiar sort of view, though one that is clearly quite different from the three views just considered. Simplifying a bit, we can discuss this on the model of Descartes’ idea that only two substances are really distinct, while a substance and its modes are not really distinct, but only modally distinct (Principles 1.60–61). Analogously, in a system in which all the things of this world were taken as modes of the one substance, God, one might say that the things of this world, that is, the modes, are not really distinct from God; rather each mode is only modally distinct from God. Nothing else in

The One and the Many 27

this system, we assume, would be really distinct from the existing God either; hence we would have a case of PantheismRD.32 I think it will be agreed that the various doctrines of monism and pantheism that have been considered through the ages (and not infrequently vilified) have often been obscure and confusing, difficult to understand and distinguish. I hardly claim that the modest approach offered here can cut through all the haze. But I do claim that focusing on identity and distinctness, in their various senses and manifestations—some of which have been put on display here—can help remove some of the mystification and confusion surrounding the doctrines of monism and pantheism.33 Hereafter I shall try to do just this sort of thing in a more limited way, in connection with recent discussions of the young Leibniz as a possible monist and/or pantheist.

Real Distinction and Leibniz’s Argument That All Things Are One With these preliminaries out of the way, let us turn directly to the young Leibniz, for a look at what appears to be—I will not here say definitely is—a youthful expression by Leibniz of monistic pantheism. The advantage of the previous discussion is that we can now put the hypothesis in a much more specific form: the present hypothesis is that the passage in question can be read as an expression of a monistic pantheism of the specific types, Monism RD and PantheismRD. Although the passage has been discussed by others as well, notably Adams and Mercer,34 my discussion is distinguished by its emphasis on the central role of the notion of real distinction in this text, as opposed to the other kinds of distinction on display in the second section herein. The quotation in question (which I shall refer to as the ‘‘all things are one’’ passage), is drawn from a paper dated by the Academy editors to November 1676(?), Quod Ens Perfectissimum Sit Possibile: It can easily be demonstrated that all things are distinguished, not as substances . . . but as modes. This can be demonstrated from the fact that, of those things which are radically distinct, one can be perfectly understood without another; that is, all the requisites of the one can be understood without all the requisites of the other being understood. But in the case of things, this is not so; for since the ultimate reason of things is unique, and contains by itself the aggregate of all requisites of all things, it is evident that the requisites of all things are the same. So also is their essence, given that an essence is the aggregate of all primary requisites. Therefore the essence of all things is the same, and things differ only modally. . . . If only those things are really different which can be separated, or, of which one can be perfectly understood without the other, it follows that no thing really differs from another, but that all things are one. (A 6.3:573/DSR 93, 95)35

This is, in my view, a fascinating passage, well worth the attention it has received in recent years. The argued thesis, that all things are one, is of interest in its own right, even if all mention of monism and pantheism were set aside. (I shall return to this specific question later.) But in fact the suggestion of monism36 and of pantheism37 has been a focal point for discussion to date. This examination will

28 Leibniz: Nature and Freedom

not be an exception to this, even if it goes beyond previous discussions (including my own) in a variety of ways. I cannot here deal with all the useful commentary that this passage has received. My more limited aim is to offer a reading of this passage that articulates the argument present in it and that makes clear how, and in just what sense of distinctness, one could see this passage as a statement of monism and/or pantheism. I shall be concerned as well to show the relation between Leibniz’s use of real distinction in this passage and Descartes’ important use of the same distinction. I argue that Leibniz’s use of the notion of real distinction (and related concepts, primarily modal distinction) is close enough to Cartesian uses38 to make it reasonable to look to material from Descartes as an occasional aid in interpreting Leibniz’s meaning in this passage. Let us examine things a bit more closely. In the ‘‘all things are one’’ passage, Leibniz uses the terms ‘really differ’ and ‘modally differ’ (realiter differunt, realiter . . . differe, and differunt . . . modo) at important places in his argument, and also suggests both real and modal distinctions in his opening statement of the thesis to be proved: ‘‘all things are distinguished, not as substances . . . but as modes [Res omnes non ut substantias sed modos distingui]’’ (A 6.3:573; emphasis added). He gives an account (perhaps only a partial account) of real distinction as well, namely, that ‘‘only those things are really different which can be separated, or, of which one can be perfectly understood without the other’’ (A 6.3:573). To all appearances this is the same account given in connection with things that are said to be ‘‘radically distinct’’ earlier in the passage. This use of the two terms ‘really’ and ‘radically’ in connection with distinction is a bit confusing, but in fact Leibniz at first wrote not radicaliter distincta but rather realiter distincta (really distinct) earlier in the passage. So it appears both from Leibniz’s terms and from content he assigned them that we have one concept at work in these two portions of the passage, but some uncertainty and inconsistency as to labeling.39 Certainly the label realiter distincta would have been more familiar to contemporaries, and in fact, the concept referred to would have been quite familiar as well: for key points from Descartes on real distinctness are quite similar to Leibniz’s points on real (or ‘‘radical’’) distinctness in the passage above. First, Descartes says, ‘‘Real distinction properly exists only between two or more substances’’ (Principles 1.60). This makes sense of Leibniz talking about things not being ‘‘distinguished as substances,’’ and then turning immediately to discussion of real/ radical distinction, without ever explaining what justifies the transition. For the transition was familiar enough to those who knew Descartes not to require explanation: the distinction that applies to substances is real distinction, not modal distinction. Second, Descartes says, ‘‘we perceive these to be really distinct from one another from the sole fact that we can clearly and distinctly understand one without the other’’ (Principles 1.60). Leibniz’s talk of perfectly understanding one without the other is certainly very close to this, and consequently Leibniz’s (admittedly wavering) choice of the term ‘really distinct’ to connect up with this would not have seemed out of place to contemporaries.40 Finally, although it will not play a central part in my discussion of real distinction, it is worth mentioning that Leibniz includes in this passage another idea about his real distinctness that is

The One and the Many 29

close to Descartes’ position. Late in the passage Leibniz gives two apparently equivalent clauses (seu is the connective): the first is the one I have just considered— one can be understood perfectly without the other; the other is, roughly, one can be separated (separari) from another. The latter is quite close to what Descartes gives in Definition 10 of his ‘‘synthetic’’ exposition of a part of his system in the Objections and Replies: ‘‘Def. X. Two substances are said to be really distinct when each of them can exist apart from the other’’ (AT 7:162/CSM 2:114). Leibniz talks about things that can exist in separation from one another; Descartes about things that can exist apart from one another.41 These seem to be terminological variants of the same basic idea.42 As for modal distinction, Leibniz does not provide us with much beyond the label in the ‘‘all things are one’’ passage. This makes things more difficult than they were in connection with real distinction, where he offered specific conditions of real distinction. Still, given the similarities to Descartes on real distinction—and certain textual clues that I will mention in a moment—it seems not implausible that Leibniz had in mind something similar to Descartes’ position when talking about modal distinction. For my purposes, the key point from Descartes is that modal distinction exists both ‘‘between a mode . . . and the substance of which it is a mode, and . . . between two modes of the same substance’’ (Principles 1.61). It is the first of these that might be unexpected. Although Leibniz uses different terminology in other passages in the same time period, he seems to have a similar twofold distinction involving nonsubstantial items clearly in mind, in a context not dissimilar from that of the ‘‘all things are one’’ passage: ‘‘So just as these properties differ from each other and from essence [by a modal difference?], so do things differ from each other and from God’’ (A 6.3:519/DSR 77; emphasis added). A final point before I consider the argument from the ‘‘all things are one’’ passage in greater depth: what does Leibniz mean by the key term ‘requisite’ (requisitum)? In a paper written at about the same time as the ‘‘all things are one’’ passage, he says, ‘‘A requisite is that without which a thing cannot exist’’ (December 1676?; A 6.3:587/DSR 113).43 Generally, in this time period the ultimate or primary requisites were taken by Leibniz to be the primary attributes, or the unanalyzable perfections or forms of God.44 Adams’s statement in this regard is helpful. He says that in the ‘‘all things are one’’ passage, ‘‘the ‘first requirements’ [primary requisites] of a thing are presumably the simplest, purest attributes occurring in the analysis of the thing. Leibniz is supposing that the conjunction of all the simplest, purest attributes constitutes the essence of God’’ (Adams 1994: 129). With these points better understood, let us turn to a reconstruction of Leibniz’s argument in the ‘‘all things are one’’ passage, one highlighting real and modal distinction: 1. Things are really distinct only if ‘‘one can be perfectly understood without another.’’ (text) 2. ‘‘[O]ne can be perfectly understood without another’’ just in case ‘‘all the requisites of the one can be understood without all the requisites of the other being understood.’’ (text) 3. ‘‘[T]he ultimate reason of things is unique’’ and it ‘‘contains by itself the aggregate of all requisites of all things.’’ (text)

30 Leibniz: Nature and Freedom 4. If ‘‘the ultimate reason of things is unique’’ and it ‘‘contains by itself the aggregate of all requisites of all things,’’ then ‘‘the requisites of all things are the same.’’ (text)45 5. ‘‘[T]he requisites of all things are the same.’’ (from 3, 4) 6. If the requisites of all things are the same, then to understand all the requisites of one thing is to understand all the requisites of another. (implicit)46 7. To understand all the requisites of one thing is to understand all the requisites of another. (from 5, 6)47 8. No thing can be perfectly understood without another. (from 2, 7) 9. ‘‘[N]o thing really differs from another.’’ (from 1, 8) 10. ‘‘[A]ll things are one’’ if ‘‘no thing really differs from another.’’ (implicit)48 11. ‘‘[A]ll things are one.’’ (from 9, 10) 12. ‘‘[A]n essence [of a thing] is the aggregate of all . . . requisites [of the thing].’’ (text) 13. If ‘‘the requisites of all things are the same,’’ and ‘‘an essence [of a thing] is the aggregate of all . . . requisites [of the thing],’’ then ‘‘the essence of all things is the same.’’ (text) 14. ‘‘[T]he essence of all things is the same.’’ (from 5, 12, 13) 15. If ‘‘the essence of all things is the same,’’ then ‘‘things differ only modally.’’ (text)49 16. ‘‘[T]hings differ only modally.’’ (text) 17. ‘‘[N]o thing really differs from another’’ and ‘‘things differ only modally.’’ (from 9, 16) 18. If ‘‘no thing really differs from another’’ and ‘‘things differ only modally,’’ then ‘‘all things are distinguished, not as substances . . . but as modes.’’ (implicit) 19. ‘‘[A]ll things are distinguished, not as substances . . . but as modes.’’ (from 17, 18; Q.E.D.)

A few comments on the reconstruction are in order here, not least because of possible questions concerning lines 12–19. First, with regard to lines 12–16: why have I changed the order of presentation, placing the points about essence, which in the passage earlier occur well before the conclusion that all things are one, after that conclusion? The reason for this is that, on my reading of the argument, the points about essence (along with the linked point about modal distinction) are not logically necessary for the conclusion that all things are one. Placing those points after that conclusion makes this feature of my reconstruction clearer, and it serves as well to emphasize the natural link between lines 12–16 and the final segment of the argument, lines 17–19. A second question in fact deals with just that final segment of the argument. It is connected with a curiosity of Leibniz’s argument that might be introduced via the following challenge: why are lines 17–19 even present in the reconstruction, given that in the formal demonstration portion of Leibniz’s ‘‘all things are one’’ passage, no such subargument appears explicitly? Strictly, the assumption here is correct, taking the ‘‘formal demonstration’’ to begin immediately after the initial statement of what can be ‘‘easily demonstrated.’’ But if we consider the challenge as a modest question, with a bit more flesh on the bones, a response suggests itself.

The One and the Many

31

Isn’t it curious, one might ask, that Leibniz says that a certain thesis ‘‘can be easily demonstrated,’’ and then, when proceeding to demonstrate it (starting with ‘‘This can be demonstrated from the fact that . . . ’’), he never again mentions the thesis in question, namely, that ‘‘all things are distinguished, not as substances . . . but as modes’’? Now an important element of my reconstruction—and the basis for my answer to the challenge just posed—is that Leibniz never lost sight of his aim to demonstrate this thesis; he simply thought that the connection between what he explicitly demonstrated and the initial thesis was obvious enough that it did not need to be spelled out. In a more formal reconstruction, of course, such economy and understatement are not permitted: one looks for the final ‘‘Q.E.D.’’ Lines 17–19 make possible that expected culmination—and not, it is hoped, in an implausible way. For the promised conclusion was that things are not distinguished as substances but as modes. Given what I have discussed about real and modal distinction, and given the interim conclusions of the foregoing reconstruction about real distinction and modal distinction (that is, lines 9 and 16, ones based firmly on what is explicitly in the demonstration), the final steps to the initially stated thesis should come as no surprise, even if they are not explicitly stated by Leibniz. Although a great many further comments would be in order in connection with this argument, concerning, for instance, possible rival reconstructions, or the critical evaluation of this reconstruction,50 I shall limit myself here rather strictly. I shall focus on the relation of this argument to monism and pantheism, or, more specifically, given the foregoing distinctions and hypothesis, to Monism RD and PantheismRD. To be sure, a first look at this argument may well suggest to some that, since no things are really distinct (distinct, that is, as multiple substances are), and since moreover all things have the same essence, the divine essence,51 and are only modally distinct (distinct, that is, as modes are), this means that for Leibniz, in this passage, at least, all things are taken as modes of the one substance God, which amounts to the position of Spinozistic monism, which, it is often held, is a form of monistic pantheism. But I am trying to be more careful here than going with a ‘‘first look’’ answer. For one thing, are we really sure we know what sense of ‘mode’ is being used here?52 More generally, now that we have seen some of the quite diverse forms monism and pantheism can take, can such an answer be taken seriously if it does not specify more exactly just what type of monism Spinozistic monism is? And even if that were done, would we have enough evidence from the few brief points mentioned in the last sentence to be confident that Leibniz’s text commits him to that same type of monism, whatever that should turn out to be? I suggest that we not try to encompass anything as grand as Spinoza’s system, and instead restrict ourselves to work with the definitions or types of monism and pantheism already laid out, and proceed in terms of them, even if the text does not always provide a completely definitive answer in terms of correspondence. Let us begin with PantheismRD, the doctrine that God exists and nothing is really distinct from God. Does Leibniz espouse this position in the ‘‘all things are one’’ passage? To be sure, the answer is not altogether obvious, in part because this passage does not include an argument for the existence of God, or even an explicit statement that God exists. But the context of the passage is a paper that argues that the most perfect being is possible and that proceeds from this to a proof that ‘‘a

32 Leibniz: Nature and Freedom

being of this kind . . . is necessary’’ (A 6.3:572/DSR 93). Accordingly, it seems that we can take the first clause as satisfied and turn to what in any case is the distinctive and controversial clause of the definition, the second clause having to do with the exact relationship of God and what other traditions call ‘‘creatures.’’ Does Leibniz in this passage endorse the view that nothing is really distinct from God? There is an easy, but perhaps not fully satisfying, answer to this question. It is that, yes, Leibniz does endorse this. In fact, he claims to demonstrate that no thing is really distinct from any other, and it follows immediately from this that, in particular, no thing (or nothing) is really distinct from God.53 And so, given our assumption, it follows that Leibniz endorses PantheismRD in the ‘‘all things are one’’ passage. What about the question whether the passage commits Leibniz to Monism RD, which is the doctrine that substance exists and there exist no x and y such that x and y are substances and x is really distinct from y? Here two things are not quite as obvious as they might at first seem. The ‘‘all things are one’’ passage is altogether silent, explicitly at least, on the matter of substance. But assuming for the moment that God’s existence is the existence of a substance, then the first clause would be satisfied, and we would again reduce things to the second clause of the definition in question, which this time is that ‘‘there exists no y such that y is a substance and y is really distinct from x.’’ Here line 9 of the foregoing reconstruction, Leibniz’s intermediate conclusion that ‘‘no thing really differs from another,’’ looms large. Given this point, it follows immediately that there are no substances S1 and S2 such that S1 is really distinct from S2. And so, given the assumption, it follows that Leibniz endorsed MonismRD in the ‘‘all things are one’’ passage. But it might be objected that, even if it is clear that the mature Leibniz took God as a substance, given that the ‘‘all things are one’’ passage seems in general to be quite at odds with the positions of the mature Leibniz, why must we assume that here? To mention just one point, it is typically granted that an individual substance is a paradigm case of a thing, but in the group of writings (the so-called De Summa Rerum) of which the ‘‘all things are one’’ passage is a part, Leibniz says: ‘‘I use the word ‘thing’ readily, for we are accustomed to say, ‘God is a being,’ but we are not accustomed to say, ‘God is a thing’ ’’ (A 6.3:519/DSR 77). If Leibniz were not taking God as a thing in the ‘‘all things are one’’ passage,54 then presumably he would not be taking God as an individual substance in that passage either, and so presumably the first clause of the definition of MonismRD would not be satisfied. To cover even this possibility, I note that Leibniz used the term ‘substance’ in two senses, for both an individual substance (as in the foregoing discussion) and for an essence.55 It is clear that in the ‘‘all things are one’’ passage Leibniz is talking about the divine essence, the essence of ‘‘the ultimate reason of things’’ (A 6.3:573/ DSR 93; see also Adams 1994: 129). And it is clear moreover that in that passage Leibniz contends that ‘‘the essence of all things is the same’’ (A 6.3:573/DSR 95). In this light, let us consider again the definition of Monism RD in connection with the ‘‘all things are one’’ passage. The definition is that substance exists and there are no substances S1 and S2 such that S1 is really distinct from S2. Let us read this now in terms of essences rather than individual substances. With respect to the first clause, I contend, as in the case of my discussion of Pantheism RD, that God, as the most perfect possible being, is said by Leibniz in the writing in question to be

The One and the Many

33

necessary, and hence to exist, and accordingly, the divine essence (substance, in this reading) exists. With respect to the second clause, if we again read in not individual substances but essences, we find that since the essences of all things are the same, one thing cannot be perfectly understood without another, and so no two essences (substances) are really distinct. To be sure, what would result would be a rather different Monism RD from the one in terms of individual substances. We would be left with the oneness of the divine essence rather than the oneness of the divine individual substance. But it would be a monism defined in terms of real distinction nonetheless.56

Comparative Testing: Adams and Mercer One way to test the usefulness of a new approach in the history of philosophy is to compare it with important prior treatments of the same issue. That is what I shall undertake in this final section. The two scholars who have most clearly devoted attention to the kinds of issues highlighted in this essay are Robert M. Adams and Christia Mercer. While I cannot on this occasion give their remarkable research the full and thorough attention it deserves, I can focus on a few key issues relevant to my reading of the ‘‘all things are one’’ passage and the delineation of different kinds of monism and pantheism, especially Monism RD and PantheismRD, with their emphasis on real distinction. Robert Adams’s views on monism, pantheism, and the ‘‘all things are one’’ passage are set out in section 3 of chapter 4 of his Leibniz: Determinist, Theist, Idealist. A challenging question is posed in the title of section 3: ‘‘Is Leibniz’s Conception of God Spinozistic?’’ (Adams 1994: 123). To be sure, the question is ultimately answered in the negative with respect to Leibniz’s mature philosophy, in the sense that, according to Adams, Leibniz’s root conception of God does not have ‘‘inescapably, Spinozistic implications’’ (Adams 1994: 134). However, what is most interesting and novel in Adams’s discussion is that the title’s question is answered in the affirmative with respect to one early period of Leibniz’s development. Specifically, Adams believes that at the time of the writing of the ‘‘all things are one’’ passage, Leibniz developed his conception of God in a way that did have Spinozistic implications. According to Adams, in that passage, ‘‘Leibniz flatly affirms the Spinozistic idea that finite things are only modes’’ (Adams 1994: 129). Moreover, although no specific definition is given, Adams refers to the conclusion of the argument of the passage as a ‘‘monistic conclusion’’ (Adams 1994: 129). For the moment I will pass over the intriguing mention of monism and the question of what specific sense of monism might be involved here. Instead I will focus on the question whether the Spinozistic conception of God Adams talks about is pantheistic. The answer is not perfectly clear, although Adams often uses the term ‘pantheism’ in the section and has certainly left the impression among some commentators57 that he ascribes a brief pantheistic phase to Leibniz, followed by a rejection of pantheism. Using Adams’s words as a suggestive guide, I will set out in brief the position I think Adams adopts on this question. At the time of Leibniz’s ‘‘all things are one’’ passage, finite things are taken by Leibniz as

34 Leibniz: Nature and Freedom

modes of the attributes of God. Accordingly, at ‘‘this time, ontological externality of the resulting entities [finite things] seems explicitly rejected’’ (Adams 1994: 130). To all appearances,58 this means that the ‘‘all things are one’’ passage is read by Adams as being pantheistic. But, according to Adams, by 1678 the treatment of finite things as modes of God is definitely rejected by Leibniz (Adams 1994: 130). A key to the rejection of this view is the developing Leibnizian position that ‘‘[t]he relation of the limited attributes of creatures to the absolute attributes of God is not the relation of mode to attribute, since they exclude each other from any one subject, by virtue of the negation involved in the creaturely attributes’’ (Adams 1994: 134). In line with this, Leibniz then believed that ‘‘creatures are not ‘in’ God because their limited attributes cannot be ‘in’ God’’ (Adams 1994: 133–134).59 Accordingly, although Adams does not say this in exactly these words, it seems clear that on his view Leibniz has, at the later stage, come to endorse the ‘ontological externality’ of finite things produced by God.60 Accordingly, at this later stage a pantheistic conception of God is not adopted by Leibniz.61 There is much that I agree with here, and, to repeat, I do not propose a fullscale treatment of Adams’s position, only a few remarks related to the main results of this essay. Consider how Adams initiates the section (‘‘Leibniz’s Later Years,’’ Adams 1994: 130) in which he examines the move away from the apparently Spinozistic position of the ‘‘all things are one’’ passage. Among other things, he quotes a passage from 1677 that I discussed earlier, the one in which Leibniz critiques the idea of ‘‘a Being that is everything.’’62 Adams presents this as a rejection of the pantheisticsounding Leibnizian quotation of 1676, ‘‘God . . . is everything. Creatures are some things’’ (A 6.3:512, as quoted in Adams 1994: 130),63 and seems to present it also as a more general turning away from the denial of ontological externality that Adams sees in the ‘‘all things are one’’ passage. But assuming the correctness of my earlier analysis of Leibniz’s critique of a Being that is everything, that critique presupposes a definition of pantheism along the lines of (A) earlier: PantheismND is true iff God exists and nothing is numerically distinct from God. Meanwhile, the Leibnizian quotation of 1676 has been interpreted by Parkinson, as we have seen, along the lines of PantheismMD. Now to offer a critique of PantheismND is not necessarily to offer a critique of PantheismMD. Nor would offering a critique of PantheismND necessarily be to offer a critique of PantheismRD, which on my interpretation earlier is precisely what Leibniz affirms in the ‘‘all things are one’’ passage. Please note that I am not saying here that Adams is doing either of these things. But I am saying that the distinctions I have drawn in the second section herein should be kept clearly in mind in attempting to chart Leibniz’s possibly shifting stances on pantheism (and, for that matter, monism). In fact, the point against ‘‘Being as everything’’ is not a main element in Adams’s account of the move away from the Spinozistic position of the ‘‘all things are one’’ passage. Other issues, such as those concerning subjects, powers, limited and absolute attributes, and being ‘‘in God,’’ play a more central role. But the points just discussed are nonetheless useful as a base from which to evaluate these other issues. As we have seen, Adams puts considerable emphasis on the idea that, in the more mature Leibniz, ‘‘[t]he relation of the limited attributes of creatures to the absolute attributes of God is not the relation of mode to attribute, since they

The One and the Many

35

exclude each other from any one subject, by virtue of the negation involved in the creaturely attributes’’ (Adams 1994: 134). Here, then, we have subjects, limited and unlimited attributes, nicely integrated. But is not the presupposition here that the younger Leibniz, the Leibniz of the 1676 ‘‘all things are one’’ passage, did not see some of these points, while the somewhat older Leibniz (correctly) did? And given the nature of the points, might not that be contrary to the principle of charity, and so (some might say) should not be adopted until other initially plausible alternatives have shown themselves to be implausible? I suggest an alternative, which I believe is plausible. Suppose that Leibniz never doubted (not even when he wrote the ‘‘all things are one’’ passage) that the possession of an absolute or unlimited attribute by some x, and the possession of a limited form of the same attribute by some y, means that x and y are distinct in some sense, in particular, are numerically distinct. In terms of Leibniz’s law, plausibly applied to the ‘‘all things are one’’ passage, God and a finite thing could not be numerically identical, since God has all attributes in unlimited or absolute form and a finite thing does not. But, the alternative reading continues, the main point of the ‘‘all things are one’’ passage concerns real distinction, not numerical distinction. For purposes of argument, let me put the alternative interpretation another way: the ‘‘all things are one’’ passage does not have to do with numerical distinction. If an argument is desired, here is a rough-and-ready one. Consider a Cartesian finite mental substance at a moment, a substance with one attribute and some modes of this attribute. Consider one of these modes. Is it numerically identical to the mental substance? No, it is numerically distinct. For the substance ‘‘can exist without other things, while needing only the ordinary concurrence of God’’ (Principles 1.51), while the mode cannot so exist. So the test of Leibniz’s law is not met. But what about real distinctness? Is the mode really distinct from the mental substance? The answer is clear. This sort of situation is a paradigm case of Descartes’ modal distinction, as opposed to real distinction. The mode and the substance are not really distinct, even though they are numerically distinct. Returning to the case of God and a finite thing, the question relevant to the ‘‘all things are one’’ passage is not whether God and the finite thing are distinct in some sense. They are. They are numerically distinct. The question is whether they are really distinct or only modally distinct.64 And the answer to this turns on the specific criterion of real distinctness, not on issues of numerical distinctness (and not necessarily, I would contend, on issues of subjects, powers, and being in God either—a connection here would need to be argued more thoroughly). Christia Mercer deserves much credit for her pathbreaking work on issues connected with the young Leibniz, including those connected with the topics of this essay.65 For example, in a recent article, ‘‘God as Both the Unity and Multiplicity in the World’’ (Mercer 2000: 71–95), she discusses texts thought by some to suggest pantheism in the young Leibniz, and she connects this with the problem of the one and the many—or, more exactly, with the issue of unity and multiplicity. Her ultimate view, however, is quite different from those who have claimed to see pantheistic leanings in the young Leibniz. As she puts it in a related article, ‘‘Leibniz was never a pantheist of any sort’’ (Mercer 1999: 273). Still, and interestingly, she not

36 Leibniz: Nature and Freedom

only strongly affirms that Leibniz did argue for a cosmological solution of the problem of the one and the many, at least under the heading of unity and multiplicity, but she also clearly endorses the view that it is God who is ‘‘both the Unity and Multiplicity in the World,’’66 going so far as to conclude her article with the statement: ‘‘That God is the unity and multiplicity in the world is a fact about Leibniz’s metaphysics that has gone unnoticed for too long’’ (Mercer 2000: 95). She takes pains, backed by serious philosophical and scholarly work, to (1) explain the sense in which she thinks Leibniz means this, (2) provide texts and historical analysis to support her reading, and (3) distinguish her reading from pantheistic readings. Both Adams’s text discussed earlier and my own work figure among the targets of Mercer’s critique in connection with pantheism and associated issues (Adams 1994: 123–130; Kulstad 1994: 424–428). Again, as was the case with Adams, I will only be able to take on a small portion of Mercer’s work. Again my focus will be on important points in her work that stand out when viewed from the perspective of the results of this essay. And the most important of these is the issue of distinction. Mercer is concerned to emphasize, and to document textually, Leibniz’s commitment in his early writings to the distinction that obtains between God and finite things. This has obvious relevance to my work earlier on the concepts of identity and distinction in the second section herein. At one point, for example, Mercer says, ‘‘there are a number of passages throughout the Paris period that imply a distinction between God and creatures’’ (Mercer 1999: 277). It is fairly clear from the context of this and similar statements that she introduces such Leibnizian passages to oppose pantheistic readings of the young Leibniz.67 But this point is particularly clear in the case of another passage, in which nonidentity and nonpantheism are immediately juxtaposed. She says, ‘‘each creature is an inferior image or expression of that essence [the essence of the supreme being, God] and is not identical to it. Leibniz was not a pantheist during the period of 1676–77 and what sounds like pantheism is seventeenth-century Platonism’’ (Mercer 2000: 93). Put simply, it appears that for Mercer, if creatures and God are distinct, then there cannot be that denial of ontological externality that she takes as the defining characteristic of pantheism.68 My own view on this, to anticipate a bit, is that distinctness comes in a number of different varieties, and that these need to be distinguished in relation to various sorts of pantheism before one comes to the conclusion that Leibniz was never a pantheist ‘‘of any sort.’’ To be sure, I think there is room for accommodation between Mercer’s reading and my own on the matter of distinctness—but not, I think, as much as Mercer might desire. For example, when Mercer is speaking about created things having their own power of acting, she says ‘‘they are in some obvious sense distinct from God’’ (Mercer 1999: 278). I agree, and in this there is room for accommodation. There is indeed ‘‘some obvious sense’’ in which Mercer and I can agree that even the ‘‘things’’ of the ‘‘all things are one’’ passage are distinct from God: they are at least modally distinct from God. And, given the discussion earlier, we can agree further that they are numerically distinct as well. Thus, we can agree that PantheismND can be ruled out as a Leibnizian position at this time. Where agreement breaks down is in relation to the question whether the

The One and the Many 37

things of this world are held by Leibniz to be distinct in some sense that makes it possible to say that ‘‘Leibniz was never a pantheist of any sort’’ (Mercer 1999: 273). On my view, the answer to this question is no, at least at the time of the ‘‘all things are one’’ passage, for in that passage, Leibniz argues that ‘‘no thing really differs from another’’ (A 6.3:573/DSR 95). Why does this not count as a denial of a specific sort of distinctness, and an affirmation of a specific sort of pantheism, namely, PantheismRD? (I have, of course, already presented my argument that it does.) In short, I think that one of the lines of argument in Mercer’s arsenal, that Leibniz would insist on a distinction between God and creatures throughout his youthful writings, would be enhanced by greater attention to the kinds of distinctness involved. Without such attention—and without a corresponding look at the kinds of distinctness claimed or denied in various sorts of pantheism—it would seem to be difficult to get a firm grip on which philosophers at which moments are not pantheists ‘‘of any sort.’’ Let’s look at one more passage in connection with distinctness, again linked with ontological externality. It provides another chance to argue how the perspective of my analysis of the ‘‘all things are one’’ passage in terms of real and modal distinction may change the way one sees passages near the center of the discussion of monism, pantheism, and the young Leibniz. Here is Mercer’s statement: As Robert Adams puts it in his recent book, Leibniz denied ‘the ontological externality’ of creatures.[69] But Leibniz is no more a pantheist than is Plotinus. That is, this is not pantheism, it is Platonism. He [Leibniz] writes, ‘‘It seems to me that the origin of things from God is of the same kind as the origin of properties from an essence. . . . So just as these properties differ from each other and from essence, so do things differ from each other and from God.’’ (Mercer 2000: 93, quoting from A 6.3:518–519/DSR 77; her emphasis)

A closer inspection of Mercer’s reading of Leibniz’s words sheds light on how two interpretational strains can diverge so markedly. Mercer’s target is Adams’s claim that Leibniz, in 1676, argued for pantheism in the sense of a denial of the ‘‘ontological externality’’ of creatures. Mercer clearly believes that the Leibnizian text she gives tells against a pantheistic interpretation, presumably on the grounds that (1) it asserts that ‘‘things differ . . . from God,’’ and (2) Adams’s interpretation of the denial of ontological externality rules out such difference between God and creatures. But viewed from the perspective of real and modal difference, Mercer’s choice of texts is problematic. For if the criterion of ontological externality is real difference (as it is in PantheismRD), so that creatures fail to be ontologically external to God if they are not really distinct from God, then the considerations adduced earlier suggest strongly that in this passage things are not said to be ontologically external to God. For those considerations pointed, as we saw, to the conclusion that the difference of the passage earlier is modal difference rather than real difference.70 An assertion of the modal difference of God and creatures would not prove that God and creatures are really different. And of course, a key point in the ‘‘all things are one’’ passage is precisely that ‘‘things differ only modally . . . [and] no thing really differs from another’’ (A 6.3:573/DSR 95).

38 Leibniz: Nature and Freedom

What, to conclude, are the results of our comparative testing? Having looked at important prior treatments of Leibniz on pantheism and monism, namely, those of Adams and Mercer, I hope one can see how the careful delineation of kinds of distinction—and most notably attention to the special cases of modal and real distinction highlighted in the ‘‘all things are one’’ passage—is essential in gaining full clarity on the subtle and difficult issues associated with unraveling Leibniz’s developing thought in this area. This essay has attempted to provide such distinctions and to provide an interpretation of Leibniz’s texts in light of them. The key interpretive claim is that in the ‘‘all things are one’’ passage of 1676, Leibniz endorsed and argued for the positions of Monism RD and PantheismRD.

Notes I thank Brenda Connor, a Ph.D. student in philosophy at Rice University, for her intelligence and hard work in carrying out the research assistance associated with this project. 1. Here I can mention only a few who have contributed to this discussion: Garber 1985; Sleigh 1990, especially chaps. 5 and 6; Adams 1994, especially pt. 3, ‘‘Idealism: Monads and Bodies’’; Rutherford 1995a, especially pt. 3, ‘‘Nature’’; Rozemond 1997; Hartz 1998; Phemister 1999; Woolhouse 2000; and Garber, chap. 4 herein. 2. Poser 2000: 274, discussing the correspondence with Des Bosses, especially G 2:457–458. The discussion of corporeal substance is Leibniz’s more complicated way of dealing with (among other things) the mind-body problem, understood in something like a Cartesian way. This problem can itself be viewed as a version of the one-many problem. Descartes’ mind-body problem involves, as he writes to Elizabeth, explaining how the two substances, mind and body (which may be taken as the many in this case), can constitute one human being. As he famously says to her: ‘‘It does not seem to me that the human mind is capable of conceiving very distinctly, and at the same time, both the distinction between mind and body, and their union. To do that, it is necessary to conceive them as a single thing and at the same time consider them as two things, which is self-contradictory’’ (to Elizabeth, June 28, 1643; AT 3:690/Wilson 1969: 379). 3. G 2:311. For similar statements of the definition of perception see G 2:112 and 121; G 3:581; G 6:608; G 7:566. For discussion, see Kulstad 1977, 1982. 4. As Leibniz says in section 13 of the Monadology, ‘‘[t]his diversity [in the changing perceptions of a monad] must involve a multitude in the unity or in the simple’’ (G 6:608/AG 214). At the same stage of his life, as I will show in a moment, this unity signaled the presence of what is one. 5. For their exchange on this, see Clarification of the Difficulties Which Mr. Bayle Has Found in the New System of the Union of Soul and Body (G 4:521–523/L 495–496). 6. For more on this, see Mates 1986, especially chap. 10. 7. Especially worthy of note in this connection is the recent book by Maria Rosa Antognazza (1999). 8. Among those discussing Leibniz’s early views on this topic are Carlin 1997: 1–24, and Mercer 1999: 278. Carlin also takes up Leibniz’s later opposition to the view of God as the world soul. Some of Leibniz’s arguments on this point, as discussed by Carlin, sparked a recent, serious discussion of the infinite in Leibniz. See Brown 1998; Arthur 1999a; Brown 2000; Arthur 2001. For more on the details of the mature Leibniz’s opposition to God as soul of the world, consult also Vailati 1997: 42–46, 153–156. 9. Schepers, who gives a longer, and somewhat different, list of examples in his important study of this topic, is one of the sources of inspiration for this study, along with R. C. Sleigh, Jr., who discussed closely related matters with me during my semester at the University of Massachusetts, Amherst, in the fall of 1996. I owe him a debt of gratitude for making that research association possible, as well as for much more. 10. Perhaps the clearest examples of important related issues are the range of Leibniz’s uses of the concepts of identity and distinctness, his understanding of the doctrine of universal harmony, his theory of substance, and his views on the exact relation of God and the finite things of this world.

The One and the Many 39 11. See, for example, Owen 1971: 74; Angeles 1981: 111; Gentz 1986: 774. 12. The book is Levine 1994. For the articles, see Levine 1990, 1992. 13. Levine 1992: 98. Interestingly, as Jan Cover has reminded me in discussion, some view Spinoza as only a panentheist (one holding the view that whatever is, is in God), and not a pantheist. This emphasizes the point that there are different definitions of ‘pantheism,’ and associated distinctions, waiting in the wings. 14. See Ethics IP13, P14. 15. For more on the matter of the pantheistic sense of unity, see Levine 1990: 211–236. I do not claim that Levine would endorse my assumption for this essay about the pantheistic sense of unity. 16. See both Adams 1994: 123–134 and Kulstad 1994: 424–428. Christia Mercer has been a staunch opponent of a pantheistic reading of the young Leibniz, and she has done extensive research on many related issues. Among her works that should be carefully considered in this regard are Mercer 1999, 2000, 2001. The relations of Leibniz and Spinoza, particularly in connection with monism and/or pantheism, are hardly of recent interest only. Among the other studies that should be consulted in this area are: Stein 1890, especially chap. 5; Friedmann 1962; Parkinson 1978: 73–89. For further studies, the reader is referred to the succinct survey of almost three hundred years of reflection on the question whether Leibniz was a Spinozist provided by Morfino 1996 (who notes that as early as 1714 Bourguet had accused Leibniz of Spinozism [65]). 17. On another reading of the argument, the absurdity would lie simply in the idea that any human could be numerically identical to God. 18. Parkinson 1978: 88, quoting a passage from the paper of April 15, 1676, now available at A 6.3:512 and in Parkinson’s own newer English translation at DSR 67. 19. From this point onward, kinds of distinctness will take center stage in my discussion. But for each kind of distinctness I discuss, there will be a corresponding kind of identity. I will not typically emphasize or develop these, but it is worthwhile keeping the correlation in mind. 20. For Descartes, see the Sixth Meditation and Principles 1.51–56, 60–61. For Spinoza, see Ethics IP10S. 21. It is not built into any one of these that there exists a substance that is God. Thus, what is defined here is monism in a sense that is not necessarily theological. But if theological versions were desired, they could easily be supplied via small revisions of the definitions. 22. The definition of ‘mereological distinctness’ is given earlier, but for ease of application, here is a restatement of (B) with that definition applied to the present case: (B) Monism MD is true iff there exists an x such that x is substance and there exists no substance y such that y is numerically distinct from x and y is not a part of x. Note that this definition allows that there may exist a substance (or substances) numerically distinct from the one substance in terms of which monism is true. But such numerically distinct substances would be parts of the whole substance in virtue of which monism is true. That is, they would be numerically but not mereologically distinct from that substance. 23. The term ‘constituentially distinct’ has not yet been defined, but the basic idea is simple enough once one takes mereological distinctness as the model. Here is a reformulation of (C), with a clarification of ‘constituentially distinct’ built in: (C) MonismCD is true iff there exists an x such that x is substance and there exists no substance y such that y is numerically distinct from x and y is not a constituent of x. Although I will not define ‘constituent’ here, I can at least clarify it, thus: any part of a thing counts as a constituent of that thing, but one thing can be a constituent of another without being a part of that other. For an example of this notion at work in the context of a full discussion of one type of historical monism (Spinoza’s), see Kulstad 1996. 24. A formulation more exactly parallel to the others could have been chosen. Complexities in the ‘‘all things are one’’ passage that I will explain hereafter account for deviation. Recall that we are not trying to give all possible senses of ‘monism.’ To anticipate a bit, albeit cryptically, the present formulation handles both individual and nonindividual substances, and simplifies the treatment of other aspects of the ‘‘all things are one’’ passage at points where interpretive questions would otherwise significantly lengthen this essay. 25. One can link this to MonismND , although with no greater plausibility for the resulting doctrine than Leibniz allowed to the variant of PantheismND . If we assume that God in the system just sketched out is a substance, then the claim that nothing is numerically distinct from God implies that no

40 Leibniz: Nature and Freedom substance is numerically distinct from God. Thus, on this assumption, the present example of PantheismND would also be an example of MonismND . (If, on the other hand, God is assumed not to be a substance—for example, if the metaphysics of the case were taken to be somehow nonsubstantial generally—then this case would remain an example of PantheismND while not being an example of MonismND .) 26. As far as Monism MD is concerned, the present case provides a useful starting point. If we assume that the whole of the universe in this case counts as a single substance, then there would exist a substance and no substance would be mereologically distinct from it. That is, on this assumption, Monism MD would be true in this case. (If, on the other hand, the whole universe is assumed not to be a substance, then this case would remain an example of PantheismND but would not be an example of MonismND .) 27. Ethics IP13 in conjunction with ID6. 28. This is so because if something were constituentially distinct from God, then it could not be numerically identical to God or to a part of God (not a part, since a part is one kind of constituent). But since we are assuming an example in which PantheismMD is true, everything is either numerically identical to God or to a part of God. 29. Following his translation of what is now available in A 6.3:512, and also (in a slightly different English translation) in DSR 66–67. 30. To mention only one, confessedly strained, reading, one could let God be the set of all things and creatures be the members of this set. Since members of sets are not parts of the sets they are in, creatures would not, on this reading, be parts of God, although on some readings of ‘constituent’ they would still be constituents of God. 31. Such a system, which would count as a case of PantheismCD , would also count as a case of MonismCD , if certain assumptions about the case were made. We already have a case with God and constituents of God and nothing constituentially distinct from God. Assume that God counts as a substance here. Since there is nothing constituentially distinct from God, there is no substance constituentially distinct from God. But then MonismCD holds. Basically, this is the sort of monism that, on my view, is held by the historical Spinoza—although I limit myself to the claim that for Spinoza there is no substance constituentially distinct from God, that is, I do not ascribe to Spinoza the view that nothing is constituentially distinct from God, as was the case in the example just sketched out. See Kulstad 1996. (Of course, to return to my now familiar proviso, if we assume that God does not count as a substance here, then this case would remain an example of PantheismCD but it would not count as an example of MonismCD .) 32. This also counts as an example of MonismRD . It might be thought that the reasoning for this is straightforward, but it is not. A simplistic line of reasoning would go as follows: we have here the assertion that God is a substance, that God exists, and that nothing is really distinct from God; it follows that no substance is really distinct from God; so the case is an example of MonismRD . This is not strictly correct (although the reason why is a function of the somewhat unusual features of (D) in comparison with the other accounts of monism). What is required in the case of MonismRD is not simply that there be no substance that is not really distinct from the substance God, but also that there be no substances that are really distinct from each other. In the simplistic line of reasoning, this clause is not explicitly covered (although it might be argued that the resources are there to cover it, if the simplistic line of reasoning were made more sophisticated, with no new premises added). Fortunately, the further stipulation on the foregoing case, not included in the simplistic line—namely, that all the things of this world are modes—is sufficient to yield the result there are no substances other than God in the system described, a fortiori, no substances that are really distinct from one another. However, if, in the foregoing description of the system, it were assumed that we were dealing with a nonsubstantial metaphysical system, in which neither God nor anything else would be a substance, but it were nonetheless the case that God exists and that nothing is really distinct from God, then in this—admittedly abstractly described—case, PantheismRD would be true, but not Monism RD . 33. For another effort to clarify issues in a related vein, see Cover and O’Leary-Hawthorne 1999, chap. 7, ‘‘The threat of one substance.’’ They make clear that their main focus is a slightly different onesubstance threat from what might have been expected starting from the standpoint of this essay. What they emphasize is the question whether there is pressure in the system of the mature Leibniz away from

The One and the Many 41 his official position of ‘‘many created substances’’ and toward a position of ‘‘just one.’’ Still, the authors make clear that our problem, concerning monism with respect to all substance, not just created substance, is related: ‘‘the reader,’’ they say, ‘‘will see analogous pressures toward collapsing the distinction between God and the created world’’ (all three quotations from Cover and O’Leary-Hawthorne 1999: 255; the authors refer the reader also to n. 27 of chap. 7). 34. Adams 1994: 129–1230; Mercer 1999: 276. My first discussion of it in print was Kulstad 1994: 425–428. 35. Where I place the first set of ellipses, Parkinson’s translation adds an important phrase that in the Academy edition is in the footnotes, not in the main text. This is discussed in the text hereafter. 36. In a Spinozistic form, with the things of this world being modes of God rather than substances. 37. With all things being unified, indeed, one, and having as their single essence ‘‘all primary requisites,’’ which in turn appear, from comparison with other texts, to be simply the attributes or forms of God. See, for instance, De Formis Seu Attributis Dei (A 6.3:513–515), and A 6.3:520. (The latter suggests more simply that the collection of forms or attributes is the essence of God. A 6.3:514 brings in a subject as well.) Thus, we would seem to have a one or a unity which is all things and which is divine. 38. There is not space in this essay to document the point in a suitable way. However, I have attempted to document it in related essays. The fullest exposition appears in ‘‘Descartes, Spinoza, and Leibniz on Real Distinction, Pluralism, and Monism,’’ unfortunately not yet published. But a briefer statement is included in Kulstad 2001: 664–665. Occasionally, I have made use of bits of these essays in the present one. 39. This is not too surprising in a paper that was never prepared for publication. 40. Hans Burkhardt has reminded me, in discussion, that many would label this Descartes’ ‘‘epistemological criterion’’ of real distinctness. The corresponding ontological criterion is the one given in Definition 10 of the ‘‘Geometrical Exposition’’ in the Second Replies (AT 7:162/CSM 2:114). 41. One possible difference between Leibniz and Descartes is suggested by a strict reading of some of the earlier passages from both on real distinction. Whereas Definition 10 from Descartes is clearly meant to give a necessary and sufficient condition of real distinctness, a strict reading of, on the one hand, Descartes’ statement on clearly and distinctly understanding one thing without another and, on the other, of Leibniz’s statement on perfectly understanding one without another would seem to lead to the conclusion that the former offers a sufficient condition of real distinction; the latter a necessary condition. Of course, if both intended to be giving a necessary and a sufficient condition of real distinction (and Descartes’ definition certainly encourages the thought that he, at least, had in mind a necessary and sufficient condition of real distinction), then the difference would be only apparent. 42. In any case, Descartes uses the language of separation also: see the end of Principles 1.60. 43. From a paper entitled De Existentia. 44. Even in the one paper we are dealing with here, Quod Ens Perfectissimum Sit Possible, the talk of ‘‘all affirmative attributes’’ in ‘‘a most perfect being’’ (A 6.3:572/DSR 91) and of the ‘‘ultimate reason of things,’’ which ‘‘contains by itself the aggregate of all requisites of all things,’’ suggests that the requisites of the ‘‘all things are one’’ passage may be identified with the attributes of God. For more on requisites, see Rutherford 1995a: 112–115 (although he is dealing with a later time period and with a slightly different definition) and Adams 1994: 117–118 (who considers texts from Leibniz’s early years until the year of his death). 45. Leibniz’s term quia (because/since) provides key textual support for the claim that lines 3 and 4 are expressed in the ‘‘all things are one’’ passage. 46. This seems not only implicit in Leibniz’s reasoning but also true. 47. Although this precise conclusion may not seem obvious in Leibniz’s text, I think it is there. The specific passage I have in mind in support of this—‘‘in the case of things, this is not so’’—is a bit opaque when taken out of context. But I think reflection on the statement, in context, will make it clear that Leibniz affirms the proposition expressed in line 7 of the reconstruction. 48. Strictly, the text has the two clauses joined by conjunction rather than by any kind of definitional connection. Thus, line 10 must be taken as something implicit in Leibniz’s argument rather than explicit. But the suggestion is strong that Leibniz thinks there is such a connection between these two clauses.

42 Leibniz: Nature and Freedom 49. Strictly, Leibniz puts this as a conjunction rather than a conditional. The reconstruction gives my best guess as to what Leibniz intends here. This is one of the interesting points connected with this passage that unfortunately will not be taken up here for special discussion. 50. For more on this important matter, see Adams 1994: 129. 51. Leibniz doesn’t say this specifically in the paragraph in question, but ‘‘the aggregate of all requisites of all things’’ that he does specifically talk about there, as being contained in ‘‘the ultimate reason of things,’’ clearly suggests it. Adams (1994: 129) apparently reads it the same way. And elsewhere in writings of the same time period Leibniz confirms this. See, for example, A 6.3:519–20/DSR 79. 52. In an earlier version of this study, Mercer’s important research on this question (Mercer 1999: 283–295) had been presented and evaluated. Unfortunately, space limitations of this book precluded its inclusion in the final version. Very briefly put, my response to her suggestion that ‘mode’ is here used in a Platonistic sense is that the closeness of Leibniz’s language and concepts (especially real distinction, which is defined in this passage) to Descartes’ concepts of mode and of the real distinction of substances suggests at least a Cartesian substance-mode relationship, if not necessarily a Spinozistic substancemode relationship. 53. That is, it follows if we take it that Leibniz includes God in the class of things talked about in the ‘‘all things are one’’ passage. But this is not necessarily certain. See Kulstad 2001: 667–668. So, to cover all possibilities, assume for the moment that God is not to be counted as a thing in the sense of the ‘‘all things are one’’ passage. Then the inference just given, in the most obvious way of drawing the conclusion, via simple instantiation of the general principle that no things really differ, would be blocked. But there is another way to proceed. Given the specific account of real distinction given by Leibniz in the ‘‘all things are one’’ passage, one can still draw the conclusion that no things are really distinct from God, though not as directly. Taking God as ‘‘the ultimate reason of things,’’ which ‘‘contains by itself the aggregate of all requisites of all things’’ (A 6.3:573/DSR 93), (and perhaps adding a principle which both Adams and Mercer endorse, that for Leibniz at this time the analysis of all things involves all attributes that are in the divine essence [Adams 1994: 129; Mercer 1999: 282]) we may conclude, using lines 1 and 2 of the foregoing argument, that no thing is really distinct from God. For (perhaps oversimplifying a bit—adding the principle given a moment ago may be necessary), for any thing (excluding God) it is not the case that all the requisites of that thing can be understood without all the requisites of God being understood, taking these latter requisites to be all the attributes contained in the divine essence. So it is not the case that the thing can be perfectly understood without God (via line 2). So the thing and God are not really distinct (via line 1). Since the same argument applies for any thing, nothing is really distinct from God, which means that the second clause of the definition of PantheismRD is satisfied. 54. See the preceding note. 55. See A 1.7:248–249, drawn from Sleigh 1990: 97. 56. I shall return to something like this possibility in my discussion of Adams hereafter. 57. For example, see Mercer 1999: 273, 275–276, 279, and 293. 58. Mercer also reads Adams this way (Mercer 1999: 279). 59. Adams 1994: 133–134. Recall that Adams had suggested earlier that in the ‘‘all things are one’’ passage, Leibniz is saying, ‘‘finite things are in God as ‘modes’ of the divine attributes’’ (126, with a relation to 129). In Leibniz’s later position, as Adams sees it, things have changed: ‘‘creatures are not ‘in’ God’’ (133). This point is challenged in Mercer 1999: 279–90. 60. Adams 1994: 130 provides evidence for this. 61. He is quite clear, as we have seen, that a ‘‘Spinozistic’’ conception of God is not adopted by Leibniz at this later time. 62. A 2.1:323; translation from Adams 1994: 130. 63. A 6.3:512; translation from Adams 1994: 130. 64. Or, if you wish, return to a case of two finite minds. Are they really distinct or only modally distinct? 65. See Mercer 1999, 2000, and 2001. 66. Mercer 2000: 71 (the title) and 88–92. 67. In the case of the first passage, it appears in a paragraph that follows immediately after the laying out of a pantheistic ‘‘developmental story’’ of Leibniz’s youth, a paragraph that begins with the statement that this story ‘‘is false’’ (Mercer 1999: 277).

The One and the Many 43 68. Mercer 1999: 279. 69. She cites Adams 1994: 128, but the reference seems to be to 130. 70. In Descartes’ theory of modal distinction, as presented in Principles 1.61, two modes of the same substance differ modally, not really, from each other and from the substance. Although Leibniz’s case of properties of an essence does not fit this exactly, it is certainly related. And insofar as it does fit, the analogy points to things differing from each other and from God by a modal, not a real, distinction. This is just the sort of thing that would be involved in PantheismRD and is the sort of thing we found in the ‘‘all things are one’’ passage.

2

Leibniz and Sleigh on Substantial Unity CHRISTIA MERCER

obert C. Sleigh Jr.’s Leibniz and Arnauld: A Commentary on Their Correspon-

R dence is wonderfully Leibnizian. Sleigh creates a harmonized series of tightly

related components out of the complicated details that passed between Arnauld and Leibniz in their correspondence (April 1686 to March 1690). Because Sleigh’s book is a commentary on the letters that exchanged hands between Leibniz, Arnauld, and their intermediary, Ernst von Hessen-Rheinfels, Sleigh restricts his analysis of Leibniz’s metaphysics primarily to the philosophical material contained in the letters. There are virtually no serious philosophical questions raised in the correspondence that are left unexamined. There are, however, some questions left unanswered. The reason for this is not hard to identify: in the correspondence, Leibniz was neither clear nor explicit about some of his most important views and there are no related texts in which he offers neat answers to the relevant questions. In other words, there are some questions that Sleigh left unanswered about Leibniz’s views, and these are the questions to which Leibniz himself seemed either unable or unwilling to give answers to Arnauld. In this essay, I would like to turn to one of the most important—and surely one of the most intractable—of these difficulties, namely, the question of substantial unity. I do not pretend to have greater insight into the underlying notion of substance and unity as presented in the correspondence between Leibniz and Arnauld and the Discourse on Metaphysics than did Sleigh. In fact, I consider it a truth that if Sleigh works carefully through a text and does not make good sense of a topic, then there is no sense to be made of the topic as presented in that text. However, I would like to show here that once we place our topic within a wider textual scope, and once we position the relevant texts within a broader intellectual context, we can construct a satisfactory answer to some of the questions that arise about substantial unity. The cluster of problems surrounding the notion of unity have plagued Leibniz scholars for a very long time. Between the correspondence with Arnauld (1686–90) 44

Leibniz and Sleigh on Substantial Unity 45

and the Monadology of 1714, Leibniz is clear about the central role that unity plays in his metaphysics, but he is rarely more than suggestive about how we are supposed to explain this crucial feature of substance. As with many of his core tenets, however, Leibniz is much more explicit about his motivating assumptions in his early works. When we turn our attention to the early period (roughly, 1666 through 1676) in which he is developing his ideas and working out the details of his system, it is much easier to glimpse his underlying views about unity. In the first section that follows, I summarize the most important parts of Leibniz and Arnauld: A Commentary on Their Correspondence concerning substantial unity. Such a summary affords significant insights into Leibniz’s ideas about unity. But there are a few important questions left unanswered. Before turning to the early works for help with these, it will be useful to situate the early writings in their proper historical setting. The second section offers a brief outline of the relevant Platonist background, and the third section turns to an analysis of the most important early texts on the topic of substantial unity. More textual study needs to be done before a full account of Leibniz’s views is available, but the materials of these early works offer at least tentative answers to our questions about substantial unity.

Leibniz and Sleigh on Unity, Identity, and Substance According to Sleigh, it is in the correspondence with Arnauld that Leibniz ‘‘first worked out in detail his conception of an individual substance and what he took to be its philosophical consequences’’ (Sleigh 1990: 95). One of the assumptions that underlies this conception of substance is that a substance is what is a unity per se.1 Although in the correspondence with Arnauld and the related Discourse on Metaphysics, Leibniz says a great deal about substantial unity, some very important questions remain insufficiently answered. Let’s consider the most significant of these. As Sleigh makes clear, ‘‘the most stringent and exacting standards of substantial unity’’ include incorruptibility, ingenerability, and ‘‘most significantly’’ indivisibility (Sleigh 1990: 104). Leibniz writes to Arnauld in his letter of December 8, 1686: ‘‘Substantial unity requires a being that is complete, indivisible, and naturally indestructible,’’ and, moreover, this unity derives from ‘‘soul or substantial form. These are the only true complete beings’’ (LA 76). As Sleigh is right to note, for Leibniz, it is the soul or substantial form that confers the relevant sort of unity. Leibniz is however frustratingly unhelpful about the exact means by which a soul or substantial form produces a unity that is indivisible, ingenerable, and naturally indestructible. Nor does he explain precisely why such beings are ‘‘the only complete’’ ones, or even exactly why this is important. We have here a set of questions that concern both the nature and power of a substantial form: what is it about a soul or substantial form F in a substance S that confers unity on S; how does this unity guarantee the indivisibility, indestructibility, and ingenerability of S; in what sense is S a complete being and how is this completeness related to its other features (e.g., the indivisibility)?

46 Leibniz: Nature and Freedom

In his analysis of substantial forms in chapter 6, Sleigh offers important help with these questions. He explains: There are two central theses concerning the composition of corporeal substances— one negative, the other positive—to which Leibniz held fast in the Discourse and the correspondence. The negative thesis is that nothing whose essence is extension is an individual substance. The positive thesis is that each created individual substance, hence each corporeal substance, includes a substantial form. Leibniz saw these theses as connected. He often wrote as if establishing the negative thesis were tantamount to establishing the positive thesis. (Sleigh 1990: 116)

Since, for Leibniz, one of the main goals of the correspondence is to tempt ‘‘the great Arnauld’’ away from Cartesianism and toward the metaphysics of preestablished harmony, it is not surprising that Leibniz is keen to show Arnauld the various weaknesses of the Cartesian account of corporeal substance. As a means to this goal, Leibniz is very concerned to show that something whose entire essence is constituted by extensa is inadequate as a substance. Because Leibniz spends so much time on this topic, so does Sleigh. Although many of these details are enormously interesting, we can cut to the point most relevant to our topic, namely, that anything (whether extended or not) that is divisible is a being by aggregation, and that a being by aggregation is not a substance (Sleigh 1990: 119). Leibniz wants to convince Arnauld that, whereas beings by aggregation admit of degrees, substances or beings with substantial unity do not. The former are a matter of convention; the latter are not (Sleigh 1990: 121). The former change constantly; the latter remain the same unified thing throughout the course of their existence. That is, it is a fundamental view of Leibniz that, in Sleigh’s words, ‘‘[n]o substance is such that its unity and identity conditions are a matter of degree, a matter of convention.’’ As Sleigh nicely puts it: When [Leibniz] waxed most eloquent about substance, in contrast to various pretenders, it is this fundamental intuition that is generating the steam: ‘‘I maintain that one cannot find a better way of restoring the prestige of philosophy and transforming it into something precise than by distinguishing the only substances or complete entities, endowed with true unity . . . all the rest is merely phenomena, abstractions or relations.’’ (Sleigh 1990: 121; LA 101)

For Leibniz in the correspondence with Arnauld, therefore, there are fundamental individuals and there are nonfundamental individuals. The latter are aggregates, which are divisible, destructible, and temporary. They admit of degrees in the sense that they can be more or less unified and more or less divisible (e.g., a pile of rocks is more divisible than a piece of marble). Sleigh intends to articulate the difference between the fundamental individuals and the aggregates, and thereby to gain clarity on the nature of substantial unity. To this end, the question to which Sleigh turns is ‘‘what did Leibniz take to be the relation between an entity through aggregation and the entities that compose it, in virtue of which the entity through aggregation can be said to be ‘a state of being of those entities from which it is composed’?’’ In an attempt to answer this question, Sleigh cites an essay related to the correspondence entitled General Notations (Notationes

Leibniz and Sleigh on Substantial Unity

47

Generales), in which, as Leibniz explains, ‘‘[t]he chief point is this: an army accurately considered is not the same thing even for a moment, for it has nothing real in itself that does not result from the reality of the parts from which it is aggregated’’ (Sleigh 1990: 123; Grua 323). Sleigh’s gloss on this point is as follows: an aggregate is a state of being of those entities that compose it, in the sense that any truth about the aggregate can be expressed in propositions that ascribe modes and states to the composing entities without any need to refer to the aggregate itself. In other words, I take Leibniz to be claiming that aggregates are logical constructions from modes and states of the entities aggregated. Given this view of aggregates, it is easy enough to see why Leibniz would accept the grounding principle. (Sleigh 1990: 123–124)

The grounding principle claims: ‘‘For any x, if x is a being through aggregation, then there exists a decomposition D of x such that, for any y, if y is an element of D, then y is not a being through aggregation’’ (Sleigh 1990: 121). Leibniz explains to Arnauld part of the motivation behind the principle: ‘‘What constitutes the essence of an entity through aggregation is only a state of being of those entities from which it is composed; for example, what constitutes the essence of an army is only a state of being of the men who compose it’’ (LA 96–97). For Leibniz, then, a being that is divisible is one that is an aggregate, and an aggregate is something constituted of nonaggregates, that is, things with substantial unity. A nonaggregate or something with substantial unity is not divisible. To drive home this important point, Sleigh offers the following passage written by Leibniz in an essay entitled Definitions of Metaphysical and Logical Concepts: [N]o entity that is truly one [Ens vere unum] is composed of parts. Every substance is indivisible and whatever has parts is not an entity, but only a phenomenon. From these considerations the ancient philosophers correctly attributed substantial forms, such as minds, souls, or primary entelechies, to those things that they said made up an Unum per se. And they denied that matter by itself is a single entity [Unum Ens]. Certainly those things that lack these [substantial forms] are no more a single entity [Unum Ens] than a pile of sticks. . . . Certainly, these things do not remain the same more than a moment, whereas, by contrast, true substances remain through changes. (Sleigh 1990: 124)

There is, then, an intimate connection between entities that are divisible and entities through aggregation. As Sleigh explains: ‘‘An entity through aggregation is an entity whose existence depends upon those entities from which it is aggregated, in such fashion that a change in entities aggregated means a different entity through aggregation’’ (Sleigh 1990: 124). As Sleigh summarizes the point: ‘‘The feature common to divisible entities and entities through aggregation, to which Leibniz wished to draw our attention, is this: such entities are wholes composed in such a fashion that their identity conditions require a different whole for every change in composition’’ (Sleigh 1990: 125). Substances, however, are not like aggregates. They are truly one in the sense that they are indivisible and remain the same thing through time. Leibniz insists that each thing that is truly one has its own substantial form. An entity through aggregation does not have its own substantial form. The underlying assumption

48 Leibniz: Nature and Freedom

here is that a substantial form F of a substance S confers both unity and identity on S. The problem is that Leibniz does not explain how this is supposed to work. That is, nowhere in the Discourse or the correspondence with Arnauld does he explain exactly how substantial forms perform this double metaphysical task. So, the important question to address is: what exactly is it about substantial form F of a substance S such that F confers identity on S while it also confers unity? Sleigh is perturbed by the unity- and identity-making powers assigned to substantial forms, and for good reason. On the one hand, Leibniz endorses a notion of corporeal substance according to which a substantial form (somehow) constitutes a unity with its passive principle or body, which is itself constituted of other corporeal substances; on the other hand, he does not explain how this account is consistent with the indivisibility requirement. In his letters to Arnauld, Leibniz makes clear that substances are capable of remaining the same unified things although components of them come and go. Leibniz explains, for example, that ‘‘fire can transform an animal and reduce it in size,’’ although the animal retains its identity (LA 156). In an attempt to make sense of Leibniz’s position, Sleigh defines the term ‘composite entity’ as what refers to any individual, one proper component of which is another individual; and he distinguishes between what is divisible and what is deconstructible component-wise. While, for Leibniz, substances are not divisible, they are deconstructible component-wise. But this is all very odd. As Sleigh wonders: ‘‘What is the operative difference between being divisible and being deconstructible component-wise on which Leibniz wished to build so much metaphysics?’’ (Sleigh 1990: 126). It is in an attempt to answer this question that Sleigh turns his attention to Leibniz’s account of identity, which he considers to be the underlying issue. Sleigh justifies the transition from a discussion of unity and divisibility to one about identity in the following way: The real test is this: does the composite entity in question depend on each and every one of its components for its existence? If the answer is yes, then we have no substance; if the answer is no, we may have a substance. So divisibility is not really the vital matter here; the vital matter is whether the particular entity in question can remain the same entity over time while undergoing change of components. Leibniz’s claim amounts to this: given a substantial form suitably related to various components, we have a composite entity that can pass the test of remaining the same through change of components; absent the form, we do not. (Sleigh 1990: 126)

According to Sleigh, our ‘‘task’’ is to unearth ‘‘the explanation’’ behind the distinction between the sort of composite entity whose identity changes when its components do and the sort whose identity persists even through changes in its components. Sleigh’s point here is important. According to Leibniz in his correspondence with Arnauld, it is the substantial form of a substance that confers both the unity and identity on the substance. As noted earlier, one of the underlying questions here is: what exactly is it about mind-like or soul-like substantial form F of a substance S such that F confers identity on S while it also confers unity? In the light of Sleigh’s distinction between divisibility and component change, we might revise this question as follows: what exactly is it about mind-like or soul-like substantial form F of a substance S such that F confers identity and unity on S despite changes in the components of S?

Leibniz and Sleigh on Substantial Unity

49

In his attempt to uncover Leibniz’s views about identity, Sleigh calls attention to some of the passages in which Leibniz seems to explain identity in terms of the causal autonomy of substances. For Leibniz, explains Sleigh, ‘‘remaining numerically identical over time is a basic condition that the metaphysically ultimate individuals of an acceptable substance ontology must satisfy’’ (Sleigh 1990: 128). Here the underlying assumption is that, in those cases when an entity persists despite component changes, the entity is one that ‘‘contains’’ all its predicates past, present, and future. In a letter of December 1686, Leibniz calls such entities ‘‘complete’’ and explains that they ‘‘express’’ all of their states, ‘‘as the concept of a substance must do’’ (Sleigh 1990: 126; LA 72). As Sleigh points out, Leibniz intends to convince Arnauld that ‘‘any alteration in the properties’’ of a person ‘‘would yield a different person’’ (Sleigh 1990: 128). One of my original questions concerned what it meant for a substance to be complete. As it turns out, Leibniz’s notion of completeness is more robust than one might first assume. He seems to identify completeness with causal autonomy where the idea is that F confers causal autonomy on S, along with unity and identity. The odd thing here is that F is supposed to confer unity, identity, and completeness on S despite changes in the components of S. Clearly, for Leibniz, substantial unity, identity, completeness, and causal autonomy are all closely related and are grounded in the substantial form. ‘‘Much is at stake here,’’ and so Sleigh sets about analyzing more thoroughly the notion of an individual substance that Leibniz presents to Arnauld (Sleigh 1990: 127–128). From Sleigh’s painstaking analysis, we can draw the following conclusions. First, in the correspondence with Arnauld and related Discourse on Metaphysics, Leibniz is prepared to embrace a view very like his later position, according to which each substance has a ‘‘law of order’’ that ‘‘constitutes’’ the substance. As Leibniz writes to Arnauld: ‘‘Each substance contains in its nature the law by which the series of its operations continues and all that has happened or will happen to it’’ (Sleigh 1990: 129; LA 136). The important point is that despite substantial changes, the substance remains numerically the same as long as those changes follow, in Leibniz’s words, ‘‘from its own nature’’ (Sleigh 1990: 129; Grua 323). According to Sleigh, this ‘‘account of substance’’ is ‘‘fresh in our period’’ (Sleigh 1990: 129) and was motivated by the rigorous conception of ‘‘what is involved in being an individual substance’’ (Sleigh 1990: 130). The second conclusion to draw from Sleigh’s analysis of Leibniz’s account of substantial identity is that the latter is consistent both with ‘‘the doctrine of spontaneity’’ and with (what Sleigh calls) ‘‘the doctrine of superintrinsicalness.’’ The doctrine of superintrinsicalness describes how a property or state is related to the individual substance that has it; the doctrine of spontaneity tells us how that property or state was produced. Sleigh considers Leibniz’s views about spontaneity to be helpful in explaining how the ‘‘law of order’’ is supposed to guarantee the identity of a substance S. According to the doctrine of superintrinsicalness, ‘‘every individual has all its properties intrinsically.’’ As Sleigh puts it: ‘‘Consider an individual x and a property f that x has; if f is such that, for any y, were y to lack f then y would not be x, then let us say that x has f intrinsically’’ (Sleigh 1990: 57). According to the doctrine of spontaneity, ‘‘the series of states, constituting the

50 Leibniz: Nature and Freedom

history of an individual substance, must be generated by a relation of causality, applied to its initial state’’ (Sleigh 1990: 129). It is important to Sleigh that ‘‘every property included in the concept of an individual substance is intrinsic to that substance’’ and ‘‘[t]hat an entity is an individual substance only if its series of states may be generated by a relation of real causality applied to its first state.’’ In the correspondence with Arnauld, Leibniz’s original formulation of the doctrine is: ‘‘Every present state of a substance occurs to it spontaneously and is only a consequence of its preceding state’’ (Sleigh 1990: 134; LA 47). Given all this, Sleigh then ‘‘speculates’’ that Leibniz regarded his doctrine of spontaneity as preserving intrinsicalness ‘‘so that whenever some state f of x is a real cause of state g of x, then if x has f intrinsically, x has g intrinsically’’ (Sleigh 1990: 130). For Sleigh, such claims are ‘‘a consequence’’ of Leibniz’s rigorous conception of ‘‘what is involved in being an individual substance’’ (Sleigh 1990: 130). Once we put these claims together with the plausible assumption that ‘‘Leibniz supposed that the initial state of any substance is intrinsic to it,’’ Sleigh proposes that we ‘‘attribute the following view to Leibniz: every noninitial state of a substance has as its real cause some preceding state of that substance; and, since every state of an individual substance thereby turns out to be intrinsic to that substance, every state of that substance (including its initial state) is included in, and in that sense a consequence of, its concept’’ (Sleigh 1990: 130). As we approach the climax of this part of Sleigh’s analysis, it is worth quoting him at length. He writes: Let us take stock of what ground has been covered and remains to be covered. Our first concern was how Leibniz differentiated between composites that are divisible and those that are not. Careful scrutiny of Leibniz’s arguments in the correspondence suggested that that distinction turned on another—the one between composite individuals that remain numerically the same through changes over time and those that do not. So our question became: What conditions must an entity satisfy according to Leibniz in order to remain numerically the same over time? Our answer is that this condition obtains, according to Leibniz, if each noninitial state of an individual has as its real cause some predecessor state of that individual. On this interpretation, the doctrine of spontaneity is built into the notion of an individual substance, according to Leibniz. Given the thesis that the relation of real causality preserves intrinsicalness and the thesis that the initial state of a substance is intrinsic to it, we reach the doctrine of superintrinsicalness. Attempting to derive so much of Leibniz’s doctrine concerning individual substances from his conception of what is required for real unity—unity consistent with remaining numerically identical through change—may seem excessive. But I take heart from a letter from Leibniz to l’Hospital, dated July 1695. After rehearsing the doctrines of spontaneity, world-apart, and marks and traces, Leibniz concluded, ‘‘The key to my doctrine on this subject consists in this consideration of what is properly a real unity, Monas.’’ (Sleigh 1990: 130–131)

Sleigh’s analysis of Leibniz’s correspondence with Arnauld shows that Leibniz’s demands about unity are closely related to other core tenets. We begin to see, for example, that identity and completeness are intimately related to one another, and to the law of the series, which itself may be seen to imply the doctrines of spontaneity

Leibniz and Sleigh on Substantial Unity 51

and superintrinsicalness. I will return to the interrelations among these doctrines later. For now, it is noteworthy that on the basis of his thorough study of the correspondence with Arnauld and Discourse on Metaphysics, Sleigh is inclined to see the doctrines of superintrinsicalness and spontaneity as ‘‘built into the notion of an individual substance’’; and, moreover, he suggests that Leibniz’s understanding of that notion is itself grounded in his assumptions about ‘‘real unity.’’ Sleigh next turns to the question of ‘‘the internal generator, in virtue of which a substance produces its states in accord with its developmental law.’’ He identifies this internal generator with the substantial form of a substance, where the substantial form is ‘‘construed as a soul-like entity.’’ And he proposes that the ‘‘deep thesis’’ that underlies the developmental law is ‘‘the doctrine of marks and traces.’’ As Leibniz writes in a letter of 1686: ‘‘Every individual substance always contains traces of everything that has ever happened to it and indications of what will ever happen to it’’ (Sleigh 1990: 131–132; LA 39). Sleigh proposes that we understand the doctrine of marks and traces in terms of the intentions of a mind-like object: ‘‘Hence, only an entity that is a soul, or at least contains a soul, can satisfy the condition imposed by the doctrine of marks and traces, and hence, reach the level of an individual substance’’ (Sleigh 1990: 132). As Sleigh explains it: ‘‘Since past and future states of a substance (or anything else) do not literally exist in the present, they can be contained in the present state of a substance only intentionally, past states as the objects of memory, future states, as the objects of desire, or at least expectation’’ (Sleigh 1990: 132). Moreover, the intentionality of mind-like forms offers some help with their unity. As Leibniz writes in General Notations about the topic of how a composite can have true unity: ‘‘What makes these parts a unity in the case of man has attributes that cannot be made known without something that binds them together, namely, the faculty of perceiving and desiring’’ (Sleigh 1990: 132; Grua 323). On the basis of Sleigh’s discussion of the numerical identity of substance, we may conclude the following: for each individual substance S, there is a developmental law that itself is somehow generated by a soul-like substantial form F and that ultimately depends on the marks and traces of F; moreover, the doctrine of marks and traces is itself to be explained intentionally, that is, by the memories and desires of the mind-like F.2 In this section, I have summarized the discussion of substantial unity in Leibniz and Arnauld: A Commentary on Their Correspondence. Sleigh’s analysis of the notion as Leibniz discusses it in the correspondence with Arnauld and the related Discourse on Metaphysics makes genuine progress in uncovering the interconnections among Leibniz’s ideas about substantial unity, identity, and completeness. But despite this success in excavating Leibniz’s views about substantial unity, some fundamental questions remain unanswered. As noted, Sleigh’s account is based entirely on texts in which Leibniz avoids explicit acknowledgment of some of his underlying assumptions about unity. Therefore, the lack of resolution of some significant problems should not come as a surprise. But the unresolved problems do come as a disappointment. We yearn to learn more about the underlying nature of substance such that it produces its own unity, identity, and completeness. From Sleigh’s analysis, we have discovered a good deal about what substantial unities are not (they are not aggregates whose state of being is constituted of the

52 Leibniz: Nature and Freedom

entities from which it is composed), about what substantial unities do (they change components and yet remain the same), and about how their activities are directed (given the initial state of a substance S, S follows its ‘‘law’’). Perhaps most dramatically of all, Sleigh makes it perfectly clear that the substantial form is the real workhorse here: it is the mind-like substantial form that generates the completeness and causal autonomy, that confers unity and identity on its substance, and that contains the developmental law. Sleigh’s significant suggestion is that underlying all this is the mind-like substantial form with its memories and desires. This is all genuinely important. But we remain in the dark about the underlying nature of the mind-like substantial form such that it performs all these extraordinary tasks. That is, we want to know more about how the substantial form F in a substance S acts to accomplish so much. There are a number of related questions concerning both the underlying nature of F and the various metaphysical feats that it achieves. The most obvious of these are: by what means does F act and how does its activity guarantee the indestructibility and ingenerability of S; how exactly does F contain its developmental law and to what extent does the law constitute its nature; how exactly does F bind the components of S together so that component change does not affect the identity of S; can we clarify how the features of unity, self-sufficiency, and causal autonomy are related in F; can we understand more about how the features of completeness, self-sufficiency, and causal autonomy are related in F; and finally, is Sleigh right to attach so much importance to the intentionality of F? That the answers to some of these questions may significantly affect the coherence of Sleigh’s interpretation seems clear. Consider, for example, his explanation of identity. For each individual substance, says Sleigh, there is a developmental law that itself is somehow generated by a soul-like substantial form and that ultimately depends on the marks and traces of the individual; moreover, the doctrine of marks and traces is itself to be explained intentionally, that is, by the memories and desires of the soul-like substantial forms. While this helps to explain the identity of the substantial form over time, its contribution to the problem of unity is unclear if we take unity to be the relation that is supposed to exist between a substantial form and its body or passive principle. For a substance S that is constituted of a substantial form F and a passive principle P where P is a collection of corporeal substances, the unity formed between F and P remains mysterious. To use the example Leibniz offered Arnauld, an animal that is submitted to fire can decrease greatly in size while retaining its identity (LA 156). In such a case, Sleigh’s account explains how the soul of the animal remains the same, but it does not explain the unity between the soul and the body. Or to approach the problem in another way, Sleigh’s discussion of the relation among the doctrines of spontaneity, superintrinsicalness, and marks and traces is extremely helpful in revealing the subtleties and complexities of Leibniz’s thinking about substance, but his analysis also makes clear how intractable Leibniz’s views about unity are: as long as the substantial form is taken to be a soul-like entity whose thoughts and intentions arise entirely out of its own nature—that is, are intrinsic—it remains extremely difficult to see how such a self-sufficient object could create a unity with its passive principle. What exactly is the source of its metaphysical glue, and how are we to understand the constancy of it? Minds are supposed to be eternal and indestructible. There is

Leibniz and Sleigh on Substantial Unity

53

little help here in understanding Leibniz’s assumptions about the apparent indefatigability of minds. In conclusion, despite Sleigh’s thoughtful analysis of Leibniz’s views as they are presented to Arnauld, we remain unenlightened about how the mind-like substantial form is supposed to perform some of its most fundamental metaphysical tasks. Nor do other scholars succeed where Sleigh has failed. Recent accounts of Leibniz’s views of substantial unity are no more enlightening.3 We therefore find ourselves sadly befuddled about this crucial part of Leibniz’s thought. But we need not despair. As I said in the introduction to this essay, if Sleigh works carefully through a text and does not make good sense of a topic, then there is no sense to be made of the topic as presented in that text. Because Sleigh has not been able to discern Leibniz’s underlying assumptions about unity in the Discourse on Metaphysics and correspondence with Arnauld, we are fully justified in looking to other texts for answers to our questions. I propose that we turn to some of Leibniz’s early writings in which he is working out his ideas about substantial unity, and that we situate these early texts in their rightful philosophical context.

Platonist Assumptions about Being and Unity In his discussion of some of Leibniz’s assumptions about unity, Sleigh identifies an idea that, as he puts it, ‘‘has had a remarkable hold on Western thought,’’ namely, that ‘‘the fundamental individuals of an acceptable metaphysical system must have unity and identity conditions independent of human convention; otherwise, our metaphysical analysis has not reached bedrock’’ (Sleigh 1990: 121). This observation about the fundamental significance attached to substantial unity in the history of philosophy is important, and worth pursuing in greater detail. Leibniz himself encourages us in this pursuit. As quoted by Sleigh, Leibniz points out in Definitions of Metaphysical and Logical Concepts that ‘‘the ancient philosophers correctly attributed substantial forms, such as minds, souls, or primary entelechies, to those things that they said made up a Unum per se’’ (Sleigh 1990: 124). As with so many other aspects of Leibniz’s thought, a survey of the right historical texts helps to identify some of the assumptions that underlie Leibniz’s own thinking about unity, identity, and substantial form. A few choice examples from the history of Platonism should be sufficient to situate the young man’s views about these matters.4 In Plato’s Phaedo, Socrates explains to some of his friends why he is eager to die. He argues, among other things, for the immortality of the soul. It is in this context that Socrates explains to Cebes ‘‘that the soul is most like the divine— deathless, intelligible, uniform, indissoluble, always the same as itself—whereas the body is most like that which is human—mortal, multiform, unintelligible, soluble and never consistently the same.’’5 Underlying Socrates’ argument is the view that the soul is like the divine in that it is ‘‘pure and invisible’’ (80e). Or as Socrates puts it later in the dialogue: ‘‘the soul must be proved to be indestructible and immortal . . . it has great vitality and a godlike nature’’ (95c). For my purposes, it is particularly important that the soul is divine-like, is indestructible, and remains ‘‘always the same as itself.’’

54 Leibniz: Nature and Freedom

At the end of the first section, I listed a number of questions about substantial unity that needed to be addressed. One of these concerned how the features of unity and self-sufficiency are supposed to be related in the mind-like substantial form; another asked how completeness, self-sufficiency, and causal autonomy are related in F. These passages from the Phaedo suggest that the divine-like nature of soul guarantees its self-sufficiency and constitutes its completeness in that that nature will never cease to be ‘‘the same as itself.’’ And it would seem to follow from this sort of self-sufficiency that the soul would be indissoluble and ‘‘deathless.’’ The idea that the soul—at least the human soul—stands between the mortal and the divine becomes standard fare in the history of Platonism. Many Platonists assumed that the soul’s divine-like status entailed its unity and vitality. Among the cluster of intuitions here is the striking idea that the soul is the kind of thing that always remains self-sufficiently itself. Its indestructibility and immortality are supposed to follow from this. Before turning to one of the leading proponents of this idea, it will be helpful to review some points about the relation between unity and selfsufficiency. For many ancient thinkers, ontological priority was to be explained mainly in terms of self-sufficiency. For Platonists, there was a hierarchy of self-sufficiency and being such that each of the lower strata in the hierarchy was supposed to depend on and be caused by the higher. In Plato’s Republic the sensible things depend on the Ideas, which themselves depend on the Good. For many of the philosophers who followed Plato, it was taken as obvious that unity and perfection were intimately related to self-sufficiency and being, so that the more reality something has, the more unified and perfect it would be. Both Christian and non-Christian Platonists assumed that there is a supremely perfect, wholly simple, and unified being on which all else depends. The implication was that only the highest being was wholly perfect, self-sufficient, simple, and real and that the beings in the lower strata had diminishing degrees of these features. What is less a unity, for instance, is less real and what is less real is constituted and explained by what is more unified and hence more real. For many Platonists, unity was the key metaphysical notion around which a number of beliefs clustered: that eternity and immutability are the marks of true being and perfection; that utter unity or simplicity excludes the possibility of parts and the possibility of change; that simplicity implies independence and self-sufficiency. From such assumptions it is supposed to follow that a wholly unified being is eternally, immutably, and independently itself, but also that whatever partakes of unity has an equal share of self-sufficiency.6 Even these brief comments7 about the proposals of Plato and other Platonists constitute significant help with the list of questions asked about substantial unity at the end of the first section. Within the Platonist tradition, we discern a cluster of assumptions about the souls or active principles in nature: they are fundamentally unified and self-sufficient; their self-sufficiency constitutes a kind of metaphysical completeness; this self-sufficiency and completeness entail indestructibility, indivisibility, and vitality; and they are somehow capable of sharing their unity and selfsufficiency with ‘‘lesser’’ beings. The great fifteenth-century Platonist Marsilio Ficino explains in his Platonic Theology how the soul ‘‘causes life to be diffused among bodies.’’ For Ficino, the

Leibniz and Sleigh on Substantial Unity

55

soul, which is ‘‘always alive,’’ shares its unity and self-sufficiency with the body and thereby creates a harmony of components.8 In an attempt to explain the unifying powers of the soul, Ficino writes that the soul ‘‘diffuses its vivifying shadow through the division of body.’’ Although it remains ‘‘integral and simple,’’ the soul spreads ‘‘its indivisible power’’ to every single part of the body and thereby is ‘‘wholly’’ present to every part. It thereby gives its body its essence as an indivisible thing (43v). That is, according to Ficino, because the soul is a constantly active divinelike power, it is fundamentally unified, self-sufficient, and complete, and moreover it communicates its ‘‘vivifying’’ and ‘‘indivisible power’’ to anything it diffuses. But questions remain. The soul is supposed to be unextended (i.e., not in space) and yet it is supposed to diffuse its power to every part of its body. In order to explain the position of soul here, we need to turn briefly to the Platonist doctrine of emanation. Oversimplifying somewhat, we can say that: if A has an attribute f, then A can emanate f-ness to a being B. In the emanative relation, A loses nothing while B comes to instantiate f-ness. The emanative process is assumed to be continual so that B will participate in f-ness and have f if and only if A acts or emanates f-ness. It is important to emphasize the fact that, in the emanative causal relation, the f of A is greater and more perfect than that of B and yet the f in B resembles its cause. For early Platonists like Plotinus and Proclus, any act of production in the created world is a case of ‘‘imitating the One.’’9 These brief comments about emanative causation offer some assistance with one of the questions asked about substantial unity at the end of the first section, namely, how exactly does a mind-like substantial form F bind the components of a substance S together so that component change does not affect the identity of S? The answer suggested by Ficino is that S is constituted by F and the body over which F has emanative power, so that regardless of the changes in its body, S remains constantly itself. F binds together the components of S by emanating its power to them; and it follows that the changes in S are merely the result of F expanding or shrinking its emanative range. Although the soul effortlessly offers its powers to all the parts of its body, the range of its diffusion may be more or less expansive. It would be interesting to analyze the views of the Platonists in greater detail, but it remains my goal to discern Leibniz’s underlying assumptions about substantial unity. Despite the brevity of this historical material, we have made real headway with our questions. We are well prepared to discern Leibniz’s underlying ideas about unity and to construct more complete answers to our questions.

The Young Leibniz on Substantial Unity From the very beginning of his long philosophical career, Leibniz endorsed the Platonist assumptions just displayed. Although one of his main goals as a young philosopher was to construct an account of substance that was recognizably Aristotelian, Leibniz was perfectly happy to combine Platonist assumptions about God and the soul with Aristotelian ideas about substance. Briefly put, Leibniz intended to construct his theory of corporeal substance on the Aristotelian model, where a

56 Leibniz: Nature and Freedom

substantial form or active principle combined with a material or passive principle to constitute the fundamental entities of the created world.10 However, as so many medieval and early modern Aristotelians had done before, Leibniz borrowed heavily from the Platonist tradition for his views about God and the active principle in nature. For my purposes, it is particularly important to recognize that Leibniz assumed that the mind-like substantial form would possess the whole range of metaphysical powers described by Ficino. That is, for Leibniz—as for Plato, Proclus, Plotinus, Ficino, and many others—the active principle in nature is divine-like and therefore self-sufficient, complete, and unified. It is as unified and self-sufficient as anything other than God can be. From the outset of Leibniz’s philosophical career, we discern an intimate connection between unity, self-sufficiency, and completeness, and we find these features firmly rooted in the activity and vitality of the active principles in nature. Consider his first account of the nature and activities of substance. In On Transubstantiation of 1668, he defines substance as ‘‘a being that subsists per se’’ and then defines ‘‘a being that subsists per se’’ as ‘‘one that has a principle of action within itself [in se]’’ (A 6.1:508/L 115). That is, in On Transubstantiation, we find the following significant metaphysical commitment: the Principle of Substantial Activity assumes that a being S is a substance if and only if it subsists per se and S subsists per se if and only if it has a principle of activity within itself (in se). Also, in this essay, Leibniz equates mind and substantial form and implies that the principle of activity is in the latter (e.g., see A 6.1:509). It follows that both the mindlike substantial form and the corporeal substance that it activates are self-sufficient, and therefore that the substantial form is itself a substance. Thus, from the beginning of Leibniz’s philosophical reflections, it is a mind-like substantial form that is responsible for the activity and self-sufficiency of substances. In other essays of the late 1660s, Leibniz confirms this account of the activity and self-sufficiency of substance, and offers more details. Significantly, Leibniz’s belief in the essential connection between the activity of mind on the one hand and its unity, self-sufficiency, completeness, and indestructibility on the other dates from his early postgraduate days. For example, in a published text of 1664, he discusses the problem of the identity of individual things and considers a solution that assumes all of these features of the active principle in nature. He speculates in Specimen of Collected Philosophical Questions Concerning Law that the source of identity is the vivens unum, the living unity or one, which is indivisible and acts as ‘‘a fountain of life.’’ He notes that, ‘‘as the Rabbis maintain,’’ the soul is ‘‘like a little house in a certain part of the body, which no power can destroy.’’11 The implications of this early text are striking. First, the eighteen-year-old Leibniz suggests that vitality, selfsufficiency, and identity are all grounded in this living unity. Second, he commits himself to the connection between these features and indestructibility. That is, the assumption is that the vivens unum cannot be destroyed by any natural means. When we combine these assumptions with the Principle of Substantial Activity, the implication is that, for each individual substance S, there is a soul-like substantial form or vivens unum that guarantees the identity, self-sufficiency, indivisibility, and indestructibility of S. Finally, there is the provocative suggestion that this vivens unum somehow resides in a body, in which it is eternally based and to which it gives life.

Leibniz and Sleigh on Substantial Unity

57

That Leibniz thinks of the human soul as an indefatigable and eternal fortress is confirmed in other texts of the period. Consider, for example, the second part of the Conspectus, entitled ‘‘Demonstration of the Immortality of the Soul, and of Incorporality.’’ In this text of 1668–69, Leibniz lists several topics that at first glance are not related to immortality. Of the six subjects listed, only the final one, a discussion of the immortality of the soul, explicitly mentions immortality. The other five topics concern the activity of mind where the assumption is that the immortality of the soul is supposed to follow from the fact that only God can destroy an active thing. Given the material of the second section, one of these is particularly interesting: Leibniz intends to argue for the immortality of the soul based ‘‘on self-motion, following Plato’’ (A 6.1:495). Or, as Leibniz further explains in some notes that he took on the Phaedo in 1676: whatever ‘‘participates in life is not able to be extinguished’’ (A 6.3:290). For the young Leibniz, therefore, only God can destroy the active principles in nature, and once created, such principles retain vitality. No natural thing can either corrupt or destroy them. It would be very helpful to know more about the nature of this vital activity that is somehow constant and indestructible. As noted earlier, Leibniz takes the active principles in nature to be mind-like, and he often claims that they act constantly.12 Although the young man makes few explicit comments about exactly what this activity is, the evidence strongly suggests that he models the activity and thinking of created minds on the activity and thinking of God, who acts by emanation. In brief, Leibniz follows his Platonist predecessors in modeling the nature and activity of created minds on God. For example, in a note of 1671, he explains: ‘‘Just as God thinks things . . . because they follow from his nature, so does Mind. . . . Mind and God do not differ except that one is finite and the other infinite.’’13 Although previous scholars have not paid adequate attention to this important aspect of Leibniz’s thought, throughout his long life, he conceives of the relation between God and creatures as emanative. Therefore, it should not come as a surprise that he follows Platonists like Ficino in thinking of the activity of divinelike minds as one of emanation. As Leibniz explains in section 14 of Discourse on Metaphysics: ‘‘it is evident that created substances depend on God, who preserves them and who even produces them continually by a kind of emanation, just as we produce our thoughts.’’14 Given that created minds act by emanation, it is not surprising that they are ‘‘never depleted.’’ Or, as Leibniz makes the point to Oldenburg, secretary of the Royal Society, in September 1670, the activity of mind is ‘‘perpetual’’ (A 2.1:64). It is this perpetual activity of mind-like substantial forms that guarantees the self-sufficiency and completeness of substances. In a text written in conjunction with On Transubstantiation and also of 1668, Leibniz articulates his original understanding of substantial self-sufficiency. Against mechanical philosophers like Gassendi and Descartes, who consider figure, motion, and magnitude to be the primary qualities of bodies, Leibniz asks in Confession of Nature against the Atheists: ‘‘[W]hat if I should demonstrate that the origin of these very primary qualities themselves cannot be found in the nature of body? Then, indeed, I hope that these naturalists will admit that body is not self-sufficient [sibi non sufficere] and cannot subsist without an incorporeal principle.’’15 In fact, Leibniz’s original

58 Leibniz: Nature and Freedom

rejection of the mechanical conception of body as res extensa rests on the assumption that extended stuff is insufficient and incomplete by itself to explain corporeal features. Leibniz argues that because ‘‘the same matter is indeterminate as to any definite figure . . . no complete reason [plena ratio] for the figure will ever be given’’ (A 6.1:490/L 110–111). The only way to avoid this unacceptable conclusion is to insist that there is ‘‘an incorporeal being’’ that acts to organize the matter of the body and thereby constitutes a ‘‘complete reason’’ for such corporeal features. For my purposes, Leibniz’s argument against the ‘‘naturalists’’ is less important than his promotion of his own views of self-sufficiency and completeness. We can extract the following two significant metaphysical commitments from the argument in the Confession of Nature. First, the Principle of Substantial Self-Sufficiency assumes that a being S is a substance if and only if S is self-sufficient and moreover S is selfsufficient if and only if the complete reason (ratio) for its features can be discovered in the nature of S. Second, the notion of a complete reason (ratio) assumed in these principles, may be put as follows: for some state or feature f, a complete reason (ratio) of f constitutes the necessary and sufficient condition for f. Among the questions left unanswered at the end of the first section were those that concerned the relation between the completeness, self-sufficiency, and causal autonomy of the substantial form F. The conjunction of the Principle of Substantial Self-Sufficiency and the Principle of Substantial Activity suggests a good deal about Leibniz’s original understanding of the relation between activity, selfsufficiency, completeness, and causal autonomy. The principles display the two separate tasks that the substantial form F in a substance S is supposed to perform. First, F has an internal source of activity or principle of activity, which keeps F constantly active and therefore self-sufficient. As we have seen, this self-sufficiency guarantees the indivisibility and (natural) indestructibility of F. Second, the substantial form seems to possess something that directs its activities. Whereas the principle of activity acts constantly, the directive device (as yet unspecified) instructs the principle on how to behave. It is important that the activities of F must somehow involve its body or passive principle and create a substantial nature with it. It is surely noteworthy that, from the beginning of Leibniz’s ruminations about metaphysical matters, he seems to assume that (1) substances are complete in the sense that they offer a complete reason or explanation for (at least) their (primary) features, and moreover that (2) this completeness is to be grounded in the nature of the substance, that is, in the unity formed by the active and passive principles. In his analysis of Leibniz’s notion of substantial unity, Sleigh worried about ‘‘[a]ttempting to derive so much of Leibniz’s doctrine concerning individual substances from his conception of what is required for real unity’’ (Sleigh 1990: 130–131). In particular, Sleigh’s analysis of identity and completeness led him to posit spontaneity and superintrinsicalness. As a confirmation and extension of Sleigh’s point, the early works show that the young Leibniz intended to construct a tightly unified substantial nature out of a mind-like substantial form and a passive principle, and by such means to offer a complete reason for every natural occurrence. In a work of 1669–70, Leibniz offers details about how this substantial nature is supposed to be formed. In On the Incarnation of God, or On Hypostatic Union,

Leibniz and Sleigh on Substantial Unity 59

which is surely the most important text for my purposes, Leibniz presents an account of substantial unity. Although the explicit goal of the essay is to explain how the divine and human natures of Christ can form one substance, Leibniz displays a theory of substantial unity more generally. The text is worth quoting at length. According to Leibniz, the things that ‘‘are able to be unified hypostatically’’ are: 1) God and mind, 2) Mind and Body, 3) Body and Body through a common mind. Body and Body are not able to be unified in themselves hypostatically, because no Body subsists in itself. Mind and Mind are not able to be unified hypostatically, unless as perfect and imperfect because imperfect mind [i.e., created mind] does not act outside of itself unless through Body. . . . Moreover, created Mind . . . is not unified with every body, but just to the one in which it has been rooted and from which it cannot be separated. E.g. in the human body it should not be thought that the soul is unified hypostatically to all the little bodies which are in it, because they change perpetually, but [the soul] inheres in the center of the brain in a certain fixed and inseparable flower of substance, most subtly mobile at the center of the animal spirits, and [the soul] is unified substantially so that it may not be separated by death.16

Leibniz makes five claims in this passage that are especially relevant to my present concerns. They are: (1) if x and y are unified hypostatically, then either x or y subsists per se; (2) created mind cannot act outside itself except though body; (3) if x and y are unified hypostatically, then either x or y acts outside itself (through the other); (4) every created mind has a body to which it is unified hypostatically; (5) created mind is unified hypostatically with a body if and only if it is rooted in that body and cannot be separated from it.

Leibniz continues his essay by asserting the following: (6) ‘‘there is no hypostatical union except by means of the activity of the one on the other’’; i.e., if x and y are (presently) unified hypostatically, then one is (presently) acting on the other; (7) minds ‘‘have in themselves a principle of acting’’; (8) ‘‘every action [of God] on body is one of creation’’; (9) x and y are unified hypostatically if and only if (a) ‘‘one of them acts constantly by a special ratio of action [actus] on the other’’ and (b) ‘‘one of them is the other’s immediate instrument of acting’’ (A 6.1:533–534).

God is hypostatically unified with created minds in that the perfect mind acts constantly on the imperfect ones so that each of the latter is God’s ‘‘instrument.’’ God is not, however, hypostatically unified with bodies: although God constantly creates bodies, the divine mind is not the principle of the activity in an individual body nor is the body its immediate instrument of acting. Leibniz writes: ‘‘For truly the instrument of God is Mind, unified with God by means of which God acts on bodies other than by creating’’ (A 6.1:534). For my purposes, it is significant that, besides giving each mind its own principle of activity, Leibniz also suggests in this essay that God imposes on each mind ‘‘a ratio of action’’ so that the mind may act

60 Leibniz: Nature and Freedom

as ‘‘the instrument’’ of God.17 Leibniz offers the following summary of some of the claims made previously: ‘‘if A is [that which does] the unifying and B is that which is said to be unified, then (a) A is a thing subsisting per se, (b) A acts through B in C, (c) A acts immediately in B or [seu] not through another.’’18 Thus, according to Leibniz in On the Incarnation of God, for some mind-like substantial form F and some body or passive principle P, F and P are hypostatically unified just in case: F subsists per se (claim [1]), but only acts outside itself through the other; the passive principle need not subsist per se, but is the means by which F acts when it acts outside itself (claim [3]). Although God does not need a passive principle through which to act, F does (claim [2]). This means that all the activity in the natural world reduces to that of minds and bodies in hypostatic union. Moreover, it is not enough that the substantial form acts some of the time, it must act constantly on the passive principle. The idea seems to be that when the acting stops, so does the union (see claims [6] and [9]). Thus, Leibniz asserts in claim (9) that x and y are hypostatically unified if and only if the active principle acts constantly on the passive principle and that the latter is its ‘‘immediate instrument’’ of acting. By such means, we have arrived at two of Leibniz’s most basic assumptions about substantial form, namely, that it acts constantly and moreover that it only acts outside itself through its passive principle. In the remainder of the essay, Leibniz goes on to make an extraordinary claim, namely, that ‘‘there is no thought [cognitio] without a union because to render that which is thought [cognitum reddere] is itself an action of the one on the other.’’19 Thus, according to Leibniz: (10) created mind always thinks (cognoscere); (11) thought (cognitio) requires a union because to render the thing thought is itself an action of the one on the other (i.e., the formation of a thought requires an action of mind on body); (12) therefore, created mind must always be hypostatically unified with its body.

The main part of On the Incarnation of God offers a thorough account of the substantial unity forged by the mind and its body. Before mind will succeed at its assigned task, namely, to forge a unity with its body, four conditions must be met. First, according to Leibniz, each individual created mind has its own principle of activity (claim [7]) by which it acts constantly. Second, it has its own ‘‘special ratio’’ (claim [9a]). That is, the active principle or mind-like substantial form in a corporeal substance is fundamentally an active thing with its own set of instructions or ‘‘special ratio’’ in terms of which it acts. Third, when the mind acts, which it always does, it acts on its body. Finally, the result of each of these activities is a thought (although there may also be other results). It will be helpful to consider these features of mind and its relation to body in more detail. The mind-like substantial form has its own principle of activity by means of which it acts constantly and its constant activities are always on its body or passive principle. By combining mind and body in this way, Leibniz has cleverly managed to create a single unit out of active and passive principles. His strategy is straightforward: a real substantial union between the principles depends on the constant activity of the

Leibniz and Sleigh on Substantial Unity 61

one on the other because the constancy of the union of the two depends on the constancy of the connection between them. Since the two principles will cease to be a union when they cease to be connected and since constant activity assures constant connection, Leibniz’s account of substantial union requires constant activity. Thus, the hypostatic union of the principles critically depends on two features of mind: that mind constantly acts and that each mind cannot act outside of itself except through the body in which it is rooted. A comparison to organic unities may be helpful at this point. If one understands an organic unity to be composed of substantial form and matter, then it is easy to see why unity requires constant activity: if the activity involved in maintaining the organic unity stops, so does the unity. We would generally agree that when the maintenance of the organization ceases (e.g., the heart stops, the liver no longer functions), the unity of the substantial form and matter does so as well (e.g., the entity dies, the formerly organized body becomes a heap of decaying flesh). The nature of organic unities also helps us to understand what Leibniz means when he says that the active principle cannot act outside itself except through the passive: in order to act externally, the source or cause of the organization has to act through the passive principle that it organizes. The mind-like substantial form, besides having a principle of activity by means of which it acts constantly, also has ‘‘a special ratio’’ that (somehow) directs its actions. Leibniz explains at least part of his motivation: ‘‘For truly the instrument of God is mind.’’ The suggestion is that God constructs individual substantial forms so that they act according to their divinely arranged instructions. The proposal here seems to be a more developed version of the idea found in Specimen of Collected Philosophical Questions Concerning Law. In that text of 1664, Leibniz intended to offer the vivens unum as the source of substantial identity. In On the Incarnation of God, he is prepared to give every mind-like substantial form a set of instructions by which it acts and maintains its identity. Among the questions left unanswered by Sleigh’s analysis of substantial unity, one concerned the matter of how the mind-like substantial form F was supposed to contain its developmental law and to what extent the law constituted the nature of F. On the Incarnation of God offers significant help with this topic. I said earlier that Leibniz has assigned F two tasks: to act constantly and to direct its activities. While the principle of activity easily fulfills the first task, we now see that it is the developmental law or ‘‘special ratio’’ that accomplishes the second. It would seem to follow then that each substantial form is the same insofar as it contains a divinelike vivens unum. What individuates one substantial form from another and what makes it complete in the relevant sense is its ‘‘special ratio.’’ That is, each substance will be (in Plato’s terms) ‘‘always the same as itself ’’ because of the law by which it acts. There are two more points to emphasize about the features of mind and its relation to body as presented in the main part of On the Incarnation of God. Both concern the nature of the unity of the substantial nature. In the Confession of Nature against the Atheists, we witnessed Leibniz’s demand that each substantial nature offer a complete reason or explanation for its features (at least the primary ones). The account of substantial union presented in On the Incarnation of God

62 Leibniz: Nature and Freedom

easily attains that goal. Leibniz has constructed the substantial union between the mind and body so that, for any feature f of the corporeal substance S, f results from the organization of the passive principle in S and moreover this organization occurs if and only if the mind acts through the passive principle. While the mind is the source of activity, the body is what mind organizes: each is necessary and both are sufficient for the corporeal feature f. By combining mind and body in this way, Leibniz has cleverly managed to create a single corporeal nature. Finally, the nature of the union forged between the active and passive principles in On the Incarnation of God is such that the result of the activities of this substantial unity are thoughts. That is, according to Leibniz in this essay, when minds act constantly on their bodies, what they produce are thoughts.20 This point brings us to another question raised at the end of the first section. We wondered there whether or not Sleigh is right to attach so much importance to the intentionality of the mind-like substantial form. It would appear that in Leibniz’s original attempt to work out the details of his views about substantial unity, he was already thinking about unity in intentional terms.21 The proposals in On the Incarnation of God are important and contain significant details about the young Leibniz’s views about substantial unity. The essay also makes some provocative remarks about the mind’s relation to its body. Among the questions left unanswered at the end of the first section, the one I have yet to consider concerns the thorny matter of component changes. As I asked the question there, how exactly does the substantial form F in a substance S bind the components of S together so that component change does not affect the identity of S? Or, to pose the more general question: what exactly is it about mind-like or soul-like substantial form F of a substance S such that F confers identity and unity on S despite changes in the components of S? As noted at the end of the first section, this is a serious problem and one that poses potential difficulties for Sleigh’s account. For each individual substance, says Sleigh, there is a developmental law that itself is somehow generated by a soul-like substantial form. While the developmental law helps to explain the identity of the substantial form over time, its contribution to the problem of unity is unclear if we take unity to be the relation that is supposed to exist between a substantial form and its body or passive principle. For a substance S that is constituted of a substantial form F and a passive principle P where P is a collection of corporeal substances, the unity formed between F and P remains mysterious. As Leibniz pointed out to Arnauld, an animal that is submitted to fire can decrease greatly in size while retaining its identity (LA 156). Sleigh’s account helps to explain how the soul of the animal remains the same, but it does not account for the unity between the soul and the body. Moreover, as long as the substantial form is taken to be a soul-like entity whose thoughts and intentions arise entirely out of its own nature—that is, are intrinsic—it remains extremely difficult to see how such a self-sufficient object could create a unity with its passive principle. In On the Incarnation of God, Leibniz offers some clues about how we might answer the question about component change. He writes in a passage already mentioned: ‘‘it should not be thought that the soul is unified hypostatically to all the little bodies which are in it, because they change perpetually, but [the soul]

Leibniz and Sleigh on Substantial Unity 63

inheres in the center of the brain in a certain fixed and inseparable flower of substance . . . so that it may not be separated by death’’ (A 6.1:533). But what is this ‘‘flower of substance’’ and how is the soul permanently attached to it? According to Ficino, the soul is related to its body by means of emanation where the idea is that its power diffuses all the parts of that body. For Platonists like Ficino, God emanates the divine power to souls, which accordingly have that power (though in an inferior manner) and can then emanate it to their bodies. When the soul diffuses its body, it loses none of its emanative causal power and yet enlivens its body. For the young Leibniz, both the activity of mind and the relation between mind and body is to be understood along similar lines.22 Although previous scholars have not noticed this feature of the early texts, there is clear evidence that, for the young Leibniz, the activity of mind is one of emanation and moreover that minds diffuse bodies in much the same way that God diffuses creatures. That is, whether the mind in question is the infinite mind of God or the finite minds of nature, mind acts through emanation and ‘‘without being diminished.’’23 Fortunately, there is ample evidence of this account of substantial unity. Leibniz summarizes his position in a letter to Johann Friedrich of May 1671: ‘‘I am of the opinion that in a body, whether of a human being or animal, vegetable or mineral, there is a core [Kern] of its substance. . . . This core is so subtle that it remains also in the ashes of burned things and can, so to speak, draw itself into an invisible center’’ (A 2.1:108). In a fascinating essay that Leibniz attached to this letter, he offers crucial details about this ‘‘core’’ of substance. In this text, entitled On the Resurrection of the Body, Leibniz employs the same Latin verb used by Ficino (i.e., diffundere) to describe the relation between the soul and its body. He writes: ‘‘in everything there is a certain seminal center that is diffused throughout the thing.’’ This center is ‘‘the fountain of life’’ and that ‘‘in which the very soul is implanted.’’ The ‘‘subtle spirit or substance’’ cannot be destroyed but will survive through fire and other changes as ‘‘the flower of substance’’ (A 2.1:116). In the theological context of resurrection, it is not surprising that Leibniz should be concerned to show that the soul remains eternally attached to its body. Since the Christian doctrine demands that the human soul will be resurrected with its body, the believing Christian wants to rest assured that come Judgment Day, it will be her body (and not someone else’s) that accompanies her soul through the Pearly Gates. Leibniz has constructed the core of substance to satisfy exactly these demands. In fact, Leibniz is proud of his theory and insists that it has many benefits. Besides solving the problem of resurrection, it explains ‘‘the generation of plants from seeds,’’ the ‘‘development of the seed in the uterus,’’ and ‘‘the essences of chemicals’’ (A 2.1:116); it is also consistent with the radical theological demands of the doctrine of the Eucharist.24 Moreover, Leibniz is proud that his theory agrees with ‘‘the Jews.’’ He writes: ‘‘Indeed, the Jews maintain that, in a certain little bone, which they call Luz, the soul with this flower of substance remains unconquered by anything that happens’’ (A 2.1:117; see also A 6.1:91). For the young Leibniz, whether it is the development of a crystal, the generation of a plant, the movement of an object, or the resurrection of the body, the same process occurs: there is a core of substance that diffuses the thing.

64 Leibniz: Nature and Freedom

Leibniz’s notion of a core of substance is constructed so that each human substance will retain both its soul and its body from its birth to its death and for all eternity.25 Roughly speaking, the idea is that the soul remains eternally unified with a passive principle through which it acts. That is, for every substance S, there will be a substantial form F and a passive principle P. Although P can be more or less expansive, there will be some part of it, say, Pd, that remains invariably attached to F and through which F always acts. Following Leibniz’s use of the verb ‘to diffuse’ (diffundere), let’s call the relation between the active and passive principles one of diffusion. From Leibniz’s comments, we can interpret the relation as follows. For a substantial form F of a substance S, F diffuses its passive principle P just in case F has constant causal power over P. That is, like his Platonist predecessors, Leibniz intends the mind-like substantial form F in a substance S to diffuse the components of its passive principle. In the discussion of Ficino’s views about substantial unity in the second section, I showed that the soul is supposed to diffuse its unifying and vivifying powers to various parts of its body. For Ficino, the changes in the components of a substance S are merely the result of the substantial form F expanding or shrinking its emanative range. Leibniz’s view is strikingly similar except that he constructs a core of substance through which this emanation of unifying power must occur. That is, as Leibniz has constructed the core of substance, F remains eternally rooted in Pd and the diffusion of P by F will always occur through Pd. It is in this sense that the core is ‘‘the flower of substance.’’ In the letter to Johann Friedrich, Leibniz explains that the core is like ‘‘an embryo or seed of an animal [dem foetu oder frucht der Thiere],’’ which contains ‘‘the core of the whole body.’’ He insists that ‘‘this core of the substance of a human being neither increases nor decreases although its clothing and casing [Kleidt und Decke] are in constant flux.’’ These fluctuations can be extreme. Not only is ‘‘the core of the whole body’’ able to spread throughout the body, it is also able to ‘‘retract itself back to its source and fountain’’ where it is in a state of such subtlety that ‘‘no force . . . is able to damage it’’ (A 2.1:108–109). In the discussion of the resurrection of the body, it is not surprising that Leibniz is keen to explain some of the more dramatic changes that, as Christian doctrine insists, every human being will suffer in the eternity of existence. Leibniz’s theory of a core of substance is enormously clever: it explains how a substance is able to remain fundamentally the same and yet undergo the changes of natural growth and supernatural resurrection. And the diffusion relation constitutes a neat account of the unity formed between the active and passive principles, regardless of the changes in components. In every diffusion relation between a mind-like substantial form F and a body or passive principle P, each of the components of P contributes to the unity of the whole by acting according to the emanations of F through Pd. For example, in the case of an individual human substance, the person grows from infant to adult, then dramatically shrinks and expands between the moments of death and resurrection. Underneath these variations in the passive principle stands the core. Because the core of the substance is constituted by a self-sufficient soul and a passive principle through which the soul constantly acts, the core of the human is ripe for life, death, and even resurrection. For Leibniz in 1671, whether it is the generation of a plant, the growth of an adolescent, or the resurrection of the body, the same process occurs (A 2.1:116).

Leibniz and Sleigh on Substantial Unity 65

Conclusion According to Sleigh, in Leibniz and Arnauld: A Commentary on Their Correspondence, it was in the correspondence with Arnauld and the Discourse on Metaphysics that Leibniz ‘‘first worked out in detail his conception of an individual substance and what he took to be its philosophical consequences’’ (Sleigh 1990: 95). The material of the third section herein suggests, however, that Leibniz had a conception of substance in his early period, and that the young man was keen to unpack the weighty metaphysical implications of Platonist ideas about mind, activity, completeness, self-sufficiency, and unity.26 It is not surprising therefore that the texts of Leibniz’s early period (roughly, 1666–76) offer significant help with questions left unanswered in the correspondence with Arnauld and the Discourse on Metaphysics. Once we place Leibniz’s early ruminations about substantial unity in their rightful philosophical context, we can begin to discern his underlying assumptions about the power of the active mind-like principles of nature to confer unity, self-sufficiency, and completeness on substances. A brief review of these assumptions is in order. One of the most difficult questions left unanswered by Sleigh’s analysis involved how the mind-like substantial form F in a substance S could confer unity. The conferral of unity on S was supposed to guarantee the indivisibility, indestructibility, and ingenerability of S, and to allow S to retain its identity despite changes of components. Following his Platonist predecessors, the young Leibniz assumes that the divine-like nature of the active principle constantly acts through emanation so that it remains (in Plato’s words) ‘‘constantly itself ’’ and thereby is incapable of division, generation, and destruction (by anything but God). Leibniz also takes it for granted that this mind-like form can act more or less expansively and that it is eternally rooted in a passive principle from which it cannot be severed. Although from my philosophical perspective, the divine-like nature of F remains (mostly) unfathomable, there may be some comfort in the fact that, for Leibniz and many of his contemporaries, the nature of mind was unfathomable exactly because it was divine-like (e.g., A 2.1:113; A 6.1:492–493). Despite the absence of an entirely plausible account of the activity of mind-like forms, the materials of Leibniz’s early period contribute significantly to our understanding of the indivisibility, ingenerability, indestructibility, and constancy of substantial unities. The early works have also helped with questions involving the interrelations among self-sufficiency, completeness, causal autonomy, and the developmental law. What we discovered was that each substantial form contains its own principle of activity by which it acts and a ‘‘special ratio’’ or set of instructions that directs its activities. The former guarantees the self-sufficiency of S; the latter constitutes the completeness of S because it directs the substantial form F as F creates and maintains a unity with its passive principle. By such means, S retains its identity and self-sufficiency. In conclusion, Sleigh’s analysis of substantial unity in the correspondence with Arnauld and the Discourse on Metaphysics contributes significantly to our understanding of many of the subtleties of Leibniz’s metaphysics. Once we place the

66 Leibniz: Nature and Freedom

questions that remain from that study in an appropriately broad textual and contextual scope, we begin to discern Leibniz’s underlying assumptions about the power of the active mind-like principles of nature to confer unity, self-sufficiency, and completeness on substances. Although more textual study needs to be done before a full account of Leibniz’s later views is available, the analysis presented here constitutes a first attempt to construct a thoroughgoing account of Leibniz’s theory of substantial unity.

Notes I thank Don Rutherford and Jan Cover for helpful comments on this essay, and the National Endowment for the Humanities for support during my research. 1. Among scholastic philosophers, there is a standard distinction between unity per se and unity per accidens. Although there is some disagreement about how to account for each, the Latin per se implies that the feature (in this case, unity) has its source in the nature of the substance and is essential. It was common for seventeenth-century German philosophers to define unity per se in terms of ens (being) where the assumption is that every substance has ens and everything that has ens has unity per se. As Leibniz puts it to Arnauld: ‘‘I cannot conceive of any reality without true unity’’ (LA 97). It is in this sense that a unity per se is a substantial unity. Unity per se is contrasted with unity per accidens because the latter does not arise from the being of the substance, but from something else (say, the coordination of parts). For Leibniz’s rough contemporaries on this topic, see, e.g., Weidling 1696: tab. 2; Stier 1641: cap. 3; Stahl 1655: tab. 14. In the correspondence with Arnauld, Leibniz insists that an accidental unity or a unity per accidens is ‘‘a phenomenal unity’’ and exists ‘‘by opinion, convention’’ (LA 101). See Sleigh 1990: 120–121. 2. At the end of his discussion, Sleigh poses a few questions that arise about his conclusions. One of these concerns how memory is supposed to work. Sleigh’s analysis of the doctrine of marks and traces requires that every substantial form ‘‘remember’’ its past states. A question arises about how nonconscious souls can do this. Sleigh’s answer to the question is based on a distinction that Leibniz sometimes makes between ‘‘the sheer reoccurrence of a previous perception, unaccompanied by consciousness of past perception’’ and a memory ‘‘which is remembrance accompanied by consciousness of past perception’’ (Sleigh 1990: 133). It would seem that memory of the former sort is available to nonconscious minds. 3. See, e.g., Adams 1994; Rutherford 1995a. In ‘‘Leibniz to Arnauld: Platonic and Aristotelian Themes on Matter and Corporeal Substance,’’ Martha Bolton raises the question that I have noted here about substantial form and substantial unity. I would like to thank Bolton for a prepublication copy of that article (Bolton 2004), which takes a different approach to the question from mine. 4. It is virtually impossible to trace the precise sources of Leibniz’s ideas. A number of previous scholars have noticed vaguely ‘‘Platonist themes’’ in Leibniz’s philosophy, but no one has analyzed these with any precision. See, e.g., Belaval 1962; and most recently Bolton 2004. However, as I have recently argued, a (primarily) Plotinian form of Platonism contributed significantly to the development of some of Leibniz’s fundamental tenets. For more on Leibniz’s Platonism and its role in his philosophical evolution, see Mercer 2001: chaps. 5–6. 5. Plato 1997: 80a–e. Besides Plato, Leibniz was thoroughly familiar with the entire history of Platonism. Jakob Thomasius, who was the young man’s ‘‘master’’ during his studies in Leipzig, wrote extensively on the history of philosophy. For Thomasius’s erudition on Platonism, see, e.g., his Schediasma Historicum and Exercitatio de Stoica Mundi Exustione. For more about Thomasius, see Mercer 2001: chaps. 1, 3; Mercer 2004; Bode´u¨s 1993. 6. The great third-century Platonist Plotinus (204/5–270) is particularly helpful on these points. He writes about the supreme being, e.g., ‘‘[f ]or if it is not to be simple, outside all coincidence and composition and really one, it could not be a first principle, and it is the most self-sufficient, because it is simple.’’ Moreover, ‘‘what is not simple is in need of its simple components so that it can come into existence from them’’ (Plotinus 1990: 5.4.1.6–15). Also see Plotinus 1990: 3.8.10.20–26; 6.2.11.9–18; 6.9.1.14.

Leibniz and Sleigh on Substantial Unity 67 7. The material just summarized is discussed at greater length in chap. 5 of Mercer 2001. The other material of this section is new. 8. Bk. 3, chap. 2; Ficino 1559: 43r–43v. There is a translation of some of this material by Luc Deitz in Kraye 1997: 30–36. It is interesting that the verb used by Ficino to describe the relation between the soul and the body, namely, diffundere, is the same one used by the young Leibniz in his account of the relation between the active and passive principles in a corporeal substance. For more, see hereafter. A longer version of this essay included a more thorough discussion of the interesting views of Ficino. A discussion of the relation between Ficino’s and Leibniz’s views on unity will appear in an essay tentatively entitled ‘‘The Power of Unity: Leibniz and His Predecessors.’’ 9. For some of Plotinus’s comments on emanative causation, see Plotinus 1990: 5.1.6.37–39; 4.3.10.32–42; 5.5.9.1–10; 2.3.18. For some of Proclus’s comments, see Proclus 1963: props. 18, 26. 10. For more details about Leibniz’s eclectic tendencies and his commitment to the philosophy of Aristotle, see Mercer 2001: chaps. 1–3. 11. A 6.1:91. The belief that some bones are ‘‘deathless’’ is mentioned by Plato in Phaedo 80d. 12. Between roughly mid-1670 and 1672, Leibniz considered the mind-like active principles in nonhuman substances to be momentary minds which produced themselves constantly ‘‘by traduction.’’ In this case, each active principle was constituted of a series of momentary minds. For more on this, and reference to other literature, see Mercer 2001: chap. 4. 13. A 6.2:287–288. See also A 6.2:490; A 6.1:285–286, 495–496; A 2.1:97, 113. 14. For a fuller account of the emanative relation between God and the created world, see Mercer 2001: chap. 5, secs. 4–5; chap. 6, secs. 1, 3; chap. 9; chap. 10, sec. 3. 15. A 6.1:490/L 110. Leibniz’s argument here, in the Confession of Nature against the Atheists, is a bit more complicated that I am suggesting; and he has not characterized the position of mechanists like Gassendi and Descartes with complete accuracy. For more details about these and related matters, see Mercer 2001: 70–82. 16. The Latin in the latter part of this passage reads: ‘‘Porro Mens creata . . . non unitur omni corpori, sed ei tantum in quo radicata est, et a quo separari non potest. V.g. in Corpore humano non putandum est animam omnibus quae in eo sunt corpusculis hypostatice uniri, cum perpetuo transpirent, sed in ipso centro cerebri flori cuidam substantiae fixo et inseparabili, subtilissime mobili in spirituum animalium centro inhaeret et substantialiter unitur ita ut nec morte separetur’’ (A 6.1:533). 17. Notwithstanding the fact that minds are instruments of God, Leibniz insists that minds (at least human minds) are ‘‘free.’’ See A 6.1:533. 18. A 6.1:534. I have substituted lower-case letters for the numerals Leibniz uses in this passage in order to distinguish the claims he makes from the ones I have listed earlier. 19. A 6.1:534–535. For a more thorough account of some of the details of this essay, see Mercer 2001: chap. 4, sec. 3. 20. Also see, e.g., A 6.2:283. When minds act on their bodies, they also produce other things, like the organization of the body. 21. I argue elsewhere that On the Incarnation of God and related texts provide strong evidence for the conclusion that by 1671 Leibniz was prepared to construct the passive principle in nature out of mind-like substances. In brief, my argument is that otherwise these essays imply that matter is involved in thinking. For a discussion of this point, see Mercer 2001: 283–284. 22. See, e.g., A 6.1:285–6; A 2.1:97. 23. A 2.1:113. I am speaking roughly here because the theory of preestablished harmony complicates things enormously. As I argue elsewhere, the diffusion relation between the active principle or mind-like form F in a substance S and the passive principle P of S is one of preestablished harmony. That is, although F and P do not causally interact, P acts in perfect coordination with the thoughts (and instructions) of F. For more on this, see Mercer 2001: 334–340, 364–373, 376–381, 407–409, 411–43, 442–443. 24. In the development of his metaphysics, Leibniz was concerned to construct a theory of substance that would be consistent with the theological doctrines of resurrection and the Eucharist. See Mercer 2001: chap. 2; chap. 8, sec. 2. 25. At the end of his discussion of substantial unity, Sleigh poses a question: because Leibniz sometimes writes as though it were a matter of convention as to whether or not ‘‘an ontology of persisting substances’’ is better than ‘‘an ontology of transitory individuals,’’ Sleigh asks: ‘‘Did Leibniz take the

68 Leibniz: Nature and Freedom scheme of transitory individuals to be metaphysically impossible, or did he have other grounds for favoring the scheme of created substances persisting through changes?’’ Although Sleigh suspects that Leibniz had certain theological concerns that inclined him to prefer persisting substances (e.g., that moral agency had to persist), he concludes that a clear answer to the question is not forthcoming (Sleigh 1990: 133). The material that I offer here suggests that Leibniz had several reasons, many of which were theological, for the preference of persisting substances. 26. A more thoroughgoing study of the early writings suggests that Leibniz worked out the details of his philosophy much earlier than previous scholars have thought. See Mercer 2001.

3

Leibniz on Precise Shapes and the Corporeal World SAMUEL LEVEY

n many of his ‘‘middle-years’’ writings, especially those around 1679–89, Leibniz

I advances a rather puzzling argument for the claim that there are no precise

shapes in things, and suggests that shape, motion, and extension are not in things outside us but involve something imaginary (see A 6.4:1465, 1612–3, 1622). He often concludes that if there were nothing but shape, motion, and extension in them, bodies would be only phenomena ‘‘like rainbows and mock suns’’ (A 6.4:1648; see A 6.4:1464). Such claims about the status of extension and its modes certainly amount to a rejection of the Cartesian theory of corporeal substance. One might also suspect that they amount to a rejection of the very idea of corporeal being tout court.1 Leibniz’s denial of the existence of precise shapes in things has sometimes been taken to imply an antirealist or idealist reading of his own views about the nature of body during this same period, and thus to imply an idealist reading of his view of the corporeal world in general. I think this is a misunderstanding of Leibniz’s philosophy concerning the status of shape, however, and that the argument for an idealist reading of his metaphysics that relies on it is unsound. A correct understanding of the content of Leibniz’s views about shape can be recovered by considering their philosophical origins in his writings of the second half of the 1670s, and with that understanding in mind the dispute between realist and idealist interpretations of the middle-years metaphysics of the corporeal world can be properly framed.

Prelude to the Critique of Precise Shapes: Leibniz on Motion Leibniz’s critique of the Cartesian modes of extension, and in particular his critique of precise shapes, appears at least in outline in many texts from the 1680s. In the document Specimen of Discoveries of the Admirable Secrets of Nature in General, tentatively dated to 1688 by the Akademie editors,2 one finds the following synopsis: 69

70 Leibniz: Nature and Freedom Indeed, even though this may seem paradoxical, it must be realized that the notion of extension is not as transparent as is commonly believed. For from the fact that no body is so very small that it is not actually divided into parts excited by different motions, it follows that no determinate shape can be assigned to any body, nor is an exactly [exactum] straight line, or circle, or any other assignable shape of any body, found in the nature of things, although certain rules are observed by nature even in its deviation from an infinite series. Thus shape involves something imaginary, and no other sword can sever the knots that we tie for ourselves by a poor understanding of the composition of the continuum. (A 6.4:1622)3

Nearly all the characteristic elements of the critique of precise shapes are in play already in that passage: the ‘‘paradoxical’’ nature of Leibniz’s thesis about the Cartesian notion of extension; the actually infinite division of matter by motion; the lack of ‘‘exact’’ and ‘‘assignable’’ shapes in the nature of things; the adherence of nature to ‘‘certain rules’’; the point that ‘‘shape involves something imaginary’’; and the reference to the composition of the continuum as the source of the difficulties. There are a few additional elements of his attack on the Cartesian philosophy that often surface with his discussion of shape. First, Leibniz typically accompanies his remarks about precise shapes with a summary of his critique of motion; second, he often explicitly ranks shape (and motion) alongside ‘‘sensible qualities’’ such as color, heat, sound, and so on. Then, of course, there is the general conclusion of Leibniz’s argument concerning the nature of body: that its essence consists not in extension and its modes, as the Cartesians would have it, but in active and passive forces. All those points Specimen of Discoveries delivers as well, for Leibniz continues: The same thing should be said about motion, since, like place, motion too consists only in a relation, as Descartes correctly recognized. Nor is there any way of determining precisely how much absolute motion should be assigned to each subject. But motive force or the power of acting is something real, and can be discerned in bodies. Thus the essence of body must be located not in extension and its modifications, namely shape and motion—which involve something imaginary no less than heat and color and other sensible qualities—but solely in the force of acting and resisting, which we perceive not by the imagination but by the intellect. (A 6.4:1622)

There is perhaps no better single vantage point over Leibniz’s critique of precise shapes than this summary in Specimen of Discoveries. Still, the lines of argument driving the critique are left mostly unarticulated here; indeed, it is also characteristic of Leibniz’s remarks, in the 1680s, about the Cartesian modes of extension for his discussion to proceed in a didactic tone, as though the demonstrations of his claims have long since been completed. This turns out to be true of the philosophy itself, for the main lines of Leibniz’s critique of the Cartesian theory of corporeal substance were in place before the close of the 1670s; references to it in his later writings tend to advertise the results without disclosing the methodology. For the moment, however, I mean only to have taken note of what the central points of that critique are; they will be considered in more detail hereafter. This essay is primarily concerned to address Leibniz’s critique of precise shapes, yet some discussion of his critique of motion will be in order first. For a

On Precise Shapes and the Corporeal World 71

number of points from his critique of motion are routinely interwoven in the main texts to be discussed, and one strand of Leibniz’s thought about motion offers a nice parallel to his discussion of precise shapes and can serve as something of a prelude to it. In a dialogue on the first philosophy of motion, Pacidius Philalethi, written in late 1676, Leibniz offers up for scrutiny the following analysis of motion: I believe motion to be change of place, and I say that motion is in that body which changes place. (A 6.3:534)

The simple definition of motion as ‘‘change of place’’ in the first clause of the analysis is taken from Aristotle (see Physics 3.201a7–15), and it remains intact throughout the dialogue’s inquiry and indeed through the rest of Leibniz’s career.4 Although the metaphysical account of the nature of change of place that is to be articulated in the Pacidius is quite radical,5 Leibniz evidently takes the simple definition to be straightforwardly true, perhaps even a truism about motion. The second clause of the analysis, according to which ‘‘motion is in the body which changes place,’’ will eventually prove more problematic, as Leibniz will discover a difficulty for assigning change of place uniquely to individual bodies. But in the Pacidius the question whether change of place is itself absolute or only relative to something else is never raised (Leibniz confesses as much in an appended note: see A 6.3:529); so too the dialogue is silent about whether the attribution of motion or change of place to a body is the attribution of an absolute intrinsic property to it or only a relational one. It is a few months later, in February 1677, that Leibniz announces his view that motion is a ‘‘relative entity,’’ in a document now titled Motion Is Something Relative.6 There he also offers a look at the argument: A wonderful fact: motion is something relative, and we cannot distinguish just which of the bodies is moving. Therefore if it is an affection, its subject will not be any one body, but the whole world. Hence, it is necessary that all its effects are also relative. The absolute motion we imagine to ourselves, on the other hand, is nothing but an affection of our soul while we regard ourselves or other things as immobile, since we are able to understand the whole thing more easily when these things are regarded as immobile. (A 6.4:1970)

On Leibniz’s account we imagine certain bodies that are changing place to display an absolute property of motion. But that appearance of absolute motion in fact results from our automatic tendency to regard ourselves or other things as being at rest while change of place occurs. Yet since any given body can equally be taken to be moving or at rest, it follows that ‘‘we cannot distinguish just which of the bodies is moving’’ if one or more of them is supposed to be moving absolutely.7 That equivalence of hypotheses yields the difficulty that Leibniz discovers for assigning motion to an individual body, and it poses a problem for the earlier claim from the Pacidius that ‘‘motion is in that body which changes place.’’ Once we withhold the (subjective) choice of a particular frame of reference that the imagination would ordinarily impose upon our experience, we are left only with the observation that x is changing place with respect to y. And this fact by itself will not imply that x is moving; nor will it imply that y is moving; nor that both move.

72 Leibniz: Nature and Freedom

Leibniz’s argument from the equivalence of hypotheses shows that no unique assignment of motion and rest to particular bodies in the system can be determined from change of place. This is a narrowly circumscribed result concerning what we might call the epistemology of motion. It does refute a naive absolutist conception of motion according to which observed change of place actually reveals, or even could reveal, just which body is a subject of absolute motion. But it does not imply that there is no such thing as motion (for the equivalence-of-hypotheses argument is not challenging the occurrence of change of place). Nor does it imply that there can be no absolute distinction between motion and rest among bodies that are changing place relative to one another. Indeed, despite the equivalence of hypotheses, Leibniz’s own account of motion postulates an intrinsic property of individual bodies that institutes an absolute distinction between those that move and those that are at rest. In Motion Is Something Relative, he writes: It should be noted, however, that when we consider motion not in itself formally, but with respect to cause, it can be attributed to the body by whose change it occurs. (A 6.4:1970)

The change in a body by virtue of which change of place occurs is not something we can distinguish, and so our naive (imagined) assignment of absolute motion to individual bodies is in itself unfounded and can at best be only accidently correlated with the underlying facts. Nonetheless, the cause of motion is an absolute and intrinsic property belonging to individual bodies. The revised conception of motion that Leibniz offers in the wake of his equivalence-of-hypotheses argument thus remains, in a critical respect, an absolutist one. Though motion considered ‘‘in itself formally’’—that is, as change of place of things within a system of bodies—has only the status of a relative entity, the property by virtue of which change of place occurs is an absolute feature of individual bodies; and motion is to be attributed to the body whose intrinsic change causes the occurrence of change of place within a system of bodies. This appeal to the cause of motion becomes Leibniz’s standard line on the analysis of motion. A few years later he makes the same proposal: When two bodies approach each other, it can only be decided from the cause of motion, not the motion itself, which of the two is at rest or moves, or whether both are moving. It is the same with several bodies. (A 6.4:1400)

Although motion as it appears to us—as an ‘‘absolute’’ change of place—involves something imaginary; although the relational fact of change of place alone determines no unique assignment of motion and rest; nonetheless, there exists a real property that belongs absolutely to individual bodies, explains one body’s moving relative to another, and in fact sets apart those bodies that are truly undergoing a change, or are in ‘‘true motion,’’8 from those that are at rest. Objections can be raised against this proposal of Leibniz;9 but for my purposes the point to be taken here is simply that this is a feature of Leibniz’s account of motion. The appearance to us of absolute motion is only something we imagine; yet the reality behind the appearances contains an absolute distinction between motion and rest. Leibniz further maintains that the underlying causal feature that explains the occurrence of

On Precise Shapes and the Corporeal World 73

relative motion is the very quantity that is conserved in collisions, a conservation which, he holds, ‘‘must be asserted of the action, i.e. relative motion, by which one body is related to another or acts on another’’ (April 1676; A 6.3:493). Thus the particular quantity that is conserved is quantity of ‘‘action,’’ or as Leibniz will often say, ‘‘power’’ or ‘‘force’’10 (which in January 167811 Leibniz will identify as the quantity f ¼ mv2), and this belongs to individual bodies as an absolute, intrinsic feature and not merely as something we imagine to be in bodies. Thus Leibniz has a two-sided view of motion. As it appears to us, motion has to be understood to involve something imaginary; as it is behind the appearances, relational change of place is caused by absolute changes that are intrinsic to the bodies themselves and constitute ‘‘true motion.’’ And the account reductively identifies that true motion with force. This will prove to be Leibniz’s longstanding view of the nature of motion, readily found in texts of the 1680s (see A 6.4:1465, 1559); a 1694 letter to Huygens offers a good example as well: As for the difference between absolute and relative motion, I believe that if motion, or better motive force of bodies, is something real, as it seems that one must recognize, it is necessary that it should have a subject. For if a and b move near to one another, I confess that all the phenomena would be all the same, no matter which one is assumed to possess motion or rest; and even if there were 1000 bodies, I still hold that the phenomena could not furnish us (or even the angels) an infallible basis for determining the subject or the degree of motion, and that each can be conceived separately as being at rest; and it is also all that I believe that you demand of me. But you will not deny (I believe) that in truth each has a certain degree of motion, or if you will, of force, in spite of the equivalence of hypotheses. It is true I draw from it this consequence, that there is in nature something besides what Geometry can determine in it. (GM 2:184)

This two-sided view of motion will also prove to be very suggestive in understanding Leibniz’s critique of precise shapes. For in that case as well, his account will uphold some claim of reality for bodily forms behind the appearance of precise shapes while also writing off some of what appears to us as due to the imagination. And Leibniz will likewise suggest a parallel reduction of shape to force. Or so, at any rate, I shall argue. With this brief sketch of the critique of motion now in place,12 it is time to consider Leibniz’s account of precise shapes.

Precise Shapes and the Composition of the Continuum In article 12 of the Discourse on Metaphysics (1686), Leibniz launches an assault on the foundations of the Cartesian metaphysics of body. Descartes had defined body exclusively in terms of extension, seeking to put the theory of corporeal substance on the sound conceptual footing of geometry and thereby to free it from the Scholastic detritus of the concepts of form and matter. On the new Cartesian approach, the whole nature or essence of body was claimed to be nothing more than extension and its modes, that is, size, shape and motion (see AT 8A:25, 30– 31).13 While this approach was widely criticized for failing to account completely for the physics of bodies14—having no explanation of antitypy, for example, or

74 Leibniz: Nature and Freedom

inertia—it fell to Leibniz alone to suggest that, far from being the ‘‘whole essence’’ of body, the Cartesian notion of extension could not even be absolutely true of corporeal beings in the first place. The reason for this, Leibniz holds, is that extension and its modes involve something imaginary. Here is article 12: 12. That the notions involved in extension contain something imaginary and cannot constitute the substance of body. But, to resume the thread of our discussion, I believe that anyone who will meditate about the nature of substance, as I have explained it above, will find either that bodies are not substances in metaphysical rigor (which was in fact the view of the Platonists) or that the nature of body does not consist merely in extension, that is, in size, shape, and motion, but that we must necessarily recognize in body something related to souls, something we commonly call substantial form, even though it makes no change in the phenomena, any more than do the souls of animals, if they have any. It is even possible to demonstrate that the notions of size, shape, and motion are not as distinct as is imagined and that they contain something imaginary and relative to our perception, as do (though to a greater extent) color, heat, and other similar qualities, qualities about which one can doubt whether they are truly found in the nature of things outside ourselves. That is why qualities of this kind cannot constitute any substance. And if there were no other principle of identity in body other than the one just mentioned, a body could not subsist for more than a moment. (A 6.4:1545)

Leibniz’s attitude toward size, shape and motion is exposed more fully in the latter half of the passage: not only does the nature of body not consist ‘‘merely’’ in extension, but also the modes of extension themselves—or more precisely, as he puts it in this case, the notions of those modes—‘‘contain something imaginary and relative to our perception.’’ The modes of extension are to be ranked with the sensible qualities and thus are not of the right kind to constitute any substance in the first place. Once again the argument for ranking size, shape, and motion with the sensible qualities is not presented, and one is left only with Leibniz’s assurance that it is ‘‘possible to demonstrate’’ this. Even in his very early writings on physics (c. 1671), Leibniz already suggests that the figures of geometry are never to be found exactly in nature (see A 6.2:255, 269), where this claim seems to express the modest sentiment that the descriptions of nature worked with in mathematics are only idealizations of its actual structure. (Realism about the corporeal world is clearly not under discussion in those passages.) But the reported ‘‘demonstrations’’ concerning precise shapes come in somewhat later writings and are something of a very different order from the modest sentiment. And in fact Leibniz appears to have two ways of demonstrating that precise shapes involve something imaginary, one based on an account of the structure of change that individual bodies undergo across time, and the other based on an account of the division of individual bodies into parts.15 The argument from the structure of change seems to have been advanced only once, in a 1686 document, There Is No Perfect Shape At All in Bodies (A 6.4:1613), though its elements are all derived from Leibniz’s earlier writings on the continuum. The argument from the division of bodies into parts also arises from Leibniz’s work on the nature of the continuum, but by contrast with the argument from change, it is advertised repeatedly across the texts of the 1680s. The argument from change is

On Precise Shapes and the Corporeal World 75

not for that reason any less important—indeed, it may well be far the more radical of the two—but having commented on it elsewhere,16 I shall concentrate here only on the argument from the division of bodies. Leibniz follows Descartes in supposing the material universe to be a plenum and in holding that intraplenary motion demands the actual division of bodies into parts. In particular, Leibniz endorses Descartes’ demonstration from Principles of Philosophy 2.33–35 that the special case of the circulation of matter through unequal spaces in the plenum requires an infinite division of matter into parts—a demonstration that Leibniz praises as ‘‘most beautiful and worthy of his genius’’ (G 4:370).17 The hypothesis of infinite division then presents a puzzle about how to understand the structure of matter, since this infinite division holds not only of the infinite universe in the whole but also of any finite part of it. Yet by late 1676 Leibniz is convinced that no part of matter—and indeed, no interval of space, time or motion, nor in general any continuum—could be composed out of points (‘‘minima’’), on pain of the paradoxes that arise concerning the infinite and the continuum (A 6.3:549). (Nor does he allow that matter could be made up of an infinity of infinitesimal quantities, since he finds infinitesimals to be only ‘‘fictions’’ of his calculus and not something that could truly exist in nature.) Thus the ‘‘continuum’’ of matter cannot be properly understood as an undivided uniform whole like a geometrical line, since this would preclude the possibility of motion; nor can it be taken to resolve into a discrete succession of points as if it were just so many grains of sand. The true account of the continuum as it exists in nature has finally to reconcile the discrete and the continuous; but it is unclear how this could be possible. Leibniz solves this puzzle by ascribing to matter an infinitary scaling structure of divisions into ‘‘parts within parts,’’ in which every part of matter is further subdivided into finer parts, and so on ad infinitum, but in which there is no finest division into parts and thus no ultimate resolution of the continuum into points. Although there are always infinitely many divisions within any part of matter—and infinitely many points actually assigned in any interval of space, time or motion—the parts are always finite and the points of division remain ‘‘extrema’’ or endpoints of the extended parts and are not independent parts of things in their own right.18 This new conception of the structure of the continuum is unveiled in a passage—and what discussion of the continuum could be complete without it?—in Pacidius Philalethi: Accordingly the division of the continuum must not be considered to be like the division of sand into grains, but like that of a sheet of paper or tunic into folds. And so although there occur some folds smaller than others infinite in number, a body is never thereby dissolved into points or minima. . . . It is just as if we suppose a tunic to be scored with folds multiplied to infinity in such a way that there is no fold so small that it is not subdivided by a new fold. . . . And the tunic cannot be said to be resolved all the way down into points; instead, although some folds are smaller than others to infinity, bodies are always extended and points never become parts, but always remain mere extrema. (A 6.3:555)

In Leibniz’s view, the key to this new account of the continuum is to understand that the continuum is nowhere uniform in its structure (see A 6.3:547, 563). He articulates the basic idea in a statement specifically about motion:

76 Leibniz: Nature and Freedom [T]he motion of a moving thing is actually divided into an infinity of other motions, each different from the other, and does not persist the same and uniform through any stretch. (A 6.3:564)

Thus an interval of motion is always composed of submotions and not of points, but those submotions are themselves nowhere undivided and uniform wholes; rather, every part of motion is always further variegated by submotions—‘‘no motion stays the same through any space or time however small’’ (A 6.3:565)—and so there is always further complexity to be found within motion upon every scale. The account is not restricted to motion of course. Leibniz takes there to be a ‘‘harmony of matter, time and motion,’’ in which ‘‘there is no portion of matter which is not actually divided into further parts, so that there is no body so small that there is not a world of infinitary creatures in it’’ (A 6.3:566). But still ‘‘this does not mean either that a body or space is divided into points, or time into moments, because indivisibles are not parts, but the endpoints of parts; which is why, even though everything is subdivided, it is still not resolved all the way down into minima’’ (A 6.3:566). Just as is true for motion, no part of matter is the same and uniform throughout, but rather every part is broken up into still finer pieces, each different from the others, and so on ad infinitum. The new conception of the continuum, I believe, plays a leading role in Leibniz’s critique of precise shapes. In particular, the point that no motion and no part of matter stays ‘‘the same and uniform through any stretch’’ has immediate implications for shape. In nature there will be no part of any body whose shape can be perfectly described by straight lines or even by ‘‘smooth’’ curves. Since every part of matter is in fact infinitely broken up into separate parts each of which is moving with its own individual motion, to describe a given part of matter as having a shape constructible from straight lines or smooth curves is necessarily to neglect the variegation of the finer parts, and thus to neglect infinitely many details of its actual structure. Take as a simplified case of a body a finite line segment. On Leibniz’s account, this line segment cannot be exactly straight; rather it is divided by motion into many distinct subsegments—folded, as it were, into many smaller pieces, each of which is turned in a slightly different direction from its neighbors. Adjacent pieces of the line are strictly discontinuous from one another and only contiguous (see A 6.3:537), for the points at which they ‘‘touch’’ are in fact points of discontinuity and preserve the continuum only in the sense that no empty space can be assigned in between the touching endpoints (see A 6.3:546). Those points of discontinuity in the line (or singularities in the function) will preclude the line as a whole from forming a smooth curve, for lines containing such points could be at best ‘‘piecewise’’ smooth curves. But the smaller pieces of the line cannot be perfectly described by straight lines or smooth curves any more than could the original, for each of them is likewise further divided into smaller folds, and so on ad infinitum. Yet now it appears that no shape could perfectly describe the structure of the ‘‘folded line,’’ since by the lights of traditional geometry, all the shapes there are are definable only by reference to straight lines or smooth curves. The scaling structure of folds within folds that Leibniz holds to be instantiated throughout nature is recognizable, from a contemporary perspective, as a fractal

On Precise Shapes and the Corporeal World 77

structure—or at least as a natural prototype to the fractals of later mathematics.19 Taking the image of the folded tunic in the most straightforward way, it will indeed be impossible to assign any traditional geometrical shape to it, since every such shape necessarily is finitely constructible from smooth curves or straight lines. The Leibnizian tunic and the simplified case of the folded line are structures that might be approximated by infinite series of geometrical figures of increasing complexity, but every shape of the sort that can be rigorously defined in traditional geometry must fail to describe the tunic or the folded line exactly—and fail to do so at infinitely many different points. Thus the structure of the corporeal world envisaged by Leibniz’s metaphysics in 1676 simply outstrips the early modern limits of precise mathematical or geometrical description. And indeed the technical resources necessary to offer mathematically precise descriptions of fractal structures will not be suitably developed for nearly another three hundred years. Leibniz’s new vision of the continuum will have to be tempered by a second strand in his thought about the mathematical structure of things, however. If his fractal account introduces complexity at every scale of inspection in a way that prevents any actual body from having a precise shape of the sort definable in early modern mathematics, he yet also insists that no assignable error will arise from the use of that mathematics to describe the properties of actual bodies. Leibnizian bodies are supposed at once to possess fractal complexity on all scales and to converge on being mathematically continuous objects. It is as if, seen from the inside, the structure of these bodies displays a scaling array of infinite complexity, while seen from the outside their structure approximates to continuity—as if the discontinuity of the fractal somehow approaches the continuity of a traditional precise shape with arbitrary closeness without ever actually achieving it. It is far from clear that the ‘‘fractal’’ idea can be reconciled with the ‘‘convergence toward continuity’’ idea (in contemporary fractal mathematics, those two ideas are normally set in contrast to one another in order to help convey the radically discontinuous nature of fractal objects). But it is perfectly clear that Leibniz means to have it both ways. In a revealing passage from De Organo Sive Arte Magna Cogitandi (c. 1679), Leibniz writes: Even if no straight lines or circles can exist [dari possent] in nature, it is nonetheless sufficient that shapes can exist [dari posse] that differ from straight lines and circles so little that the error is less than any given error—which is sufficient in order to demonstrate certainty as well as usage. (A 6.4:159)

The straight line and the circle are, for obvious reasons, Leibniz’s stock examples of precise shapes, and this is one of the earliest texts of which I am aware that records Leibniz using the language and phrasing that will be so characteristic of the critique of precise shapes in the texts of the 1680s. By contrast, the phrase ‘‘the error is less than any given error’’ is already a well-oiled component of his discussions of the calculus and its application to physical phenomena.20 The close presence of the two in this passage suggests that by 1679 Leibniz has seen through to at least some of the consequences of his fractal account, in particular to the consequence that no precise shapes can exist in nature. And he may well be worrying about the potential difficulty that a fractal theory of body will pose to a

78 Leibniz: Nature and Freedom

philosophy that aims to uphold an account of nature as fully mathematizable— hence the (unargued) insistence that the shapes that can exist in nature will always differ ‘‘so little that the error is less than any given error’’ from the exactly mathematically definable shapes of straight lines and circles that cannot. A last element to observe in Leibniz’s early thought about the division of matter involves the link between the abstract considerations about the continuum and the more concrete topic of impulse dynamics. The hypothesis of the plenum fits naturally with the idea that motion produced in any one place will be transmitted throughout the universe, and thus that every disturbance will have effects on all bodies. Since on Leibniz’s early view the division of matter is due to differences in motion within the plenum, the hypothesis of an infinite division of matter aligns in an intuitive way with the idea that the universe is an infinite assemblage of distinct bodies each moving with its own motion, for each such motion will then have some effect, however subtle, on every other body. From the point of view of any single body, there will be an infinite fusillade of impulses coming in from every side and transmitting to it effects from all the distinct motions arising in every corner of the world. Thus the motion of any single body will be constantly agitated by the motions of its neighbors and indeed by the motions of all the other bodies in the universe. Leibniz is quick to see the connection between the division of the continuum and the continual action of an infinity of bodies upon one another. The nonuniformity of motion in every interval that is postulated to solve the kinematical puzzles concerning the structure of the continuum has as its natural dynamical correlate the idea that accelerative forces are acting on bodies in every interval of space and time (‘nonuniform motion’ is also standard language for accelerated motion), and Leibniz writes in the Pacidius that the hypothesis of the nonuniformity of the continuum ‘‘is also consistent with reason, for there is no body which is not acted upon by those around it at every single moment’’ (A 6.3:565). This idea will become quite prominent in his later discussions of shape as well.

Imagination and the Manifest Image of Nature Leibniz’s 1676 inquiry into the composition of the continuum provides a central line of argument for his later critique of precise shapes—truth in advertising for Leibniz. A second line of argument that appears in the critique of precise shapes also emerges during this early period. Leibniz repeatedly asserts precise shapes to involve something ‘‘imaginary,’’ and in writings composed around the same time as his dialogue on the continuum, one finds some very illuminating comments about the role of the imagination in constructing perceptual experience and about what Leibniz calls imaginary notions. Most significant in this connection is a document written in April 1676, about seven months before the Pacidius, and titled On Infinite Numbers. Among the several topics under discussion in On Infinite Numbers is the nature of the limit entities postulated by Leibniz’s mathematics. In his calculus, curves are represented as sections of infinitary polygons, so that a circle, for example, would

On Precise Shapes and the Corporeal World 79

be represented as a regular polygon with infinitely many infinitesimal sides. Successive approximations of a circle by a series of finite polygons of increasing numbers of sides appear to tend toward a ‘‘limit polygon’’ with infinitely many sides, which is identified with the circle itself. Yet by the time he composes On Infinite Numbers Leibniz has already ruled that infinitesimals are only fictions and not real quantities of any kind, and thus infinitary limit polygons will at best be fictions also. The question before Leibniz at this point is how to understand the role of these limit entities and the propositions of his infinitary mathematics. As he explores the issue Leibniz is careful to disavow any direct commitment to infinitary limit entities, and he suggests that his limit-analysis methods can be interpreted as shorthand for the classical ‘‘method of exhaustion’’ developed by Archimedes.21 Such suggestions will come to characterize one line of defense for the foundations of the calculus. What is especially striking in this document for my purposes, however, is the discussion Leibniz offers of the role of the imagination in mathematical reasoning: The circle—as a polygon greater than any assignable, as if this were possible—is a fictive entity, and so are other things of that kind. Accordingly, when something is said about the circle we understand it to be true of any polygon such that there is some polygon in which the error is smaller than any assigned amount a, and another polygon in which the error is smaller than any other definite assigned amount b. However, there will not be a polygon in which this error is smaller than all assignable amounts a and b at the same time, even if it can be said that polygons somehow approach such an entity in order. And so if certain polygons are able to increase according to some law, and something is true of them the more they increase, our mind imagines some ultimate polygon; and whatever it sees becoming more and more so in each single polygon, it declares to be perfectly so in this ultimate one. And even if it did not exist in the nature of things, an expression for it could nevertheless be produced, for the sake of giving propositions in abbreviated form. (A 6.3:498)

The mind passes from the series of finite polygons that increasingly approximate the properties of a circle to imagine an ‘‘ultimate polygon,’’ though no such limit entity could actually exist in nature. Leibniz explores this line of thought further: Even though these entities are fictitious, geometry nevertheless exhibits real truths which can also be expressed in other ways without them. But these fictitious entities are excellent forms of abbreviation, and for this reason extremely useful. For entities of this kind, i.e. polygons whose sides do not appear distinctly, are made apparent to us by the imagination, whence there arises in us afterwards the suspicion of an entity having no sides. (A 6.3:499)

There the step is being taken from polygonal figures to a continuous, uniformly curving one; the imaginary polygon whose sides ‘‘do not appear distinctly’’ yields to the ‘‘suspicion’’ of an entity with no sides at all—the circle. But Leibniz will deny that what the imagination conjures before the mind is an image of a perfect circle; rather there is only a thought of a uniform and continuous entity:22 Instead, what must be said is that in the mind there is a thought of uniformity, yet no image of a perfect circle: instead we apply uniformity to this image afterwards,

80 Leibniz: Nature and Freedom a uniformity we forget we have applied after sensing the irregularities. Were we then conscious at some time that we had sensed them? For consciousness is necessary for forgetting. But this is not the case. Therefore it must be said, rather, that when we sense a circle or polygon, we never sense uniformity in it, but neither do we even sense a nonuniformity, that is to say, we do not remember having sensed anything nonuniform in it, since the inequality did not immediately strike us. And because of this memory we now ascribe the name of uniformity to it. It must be seen whether we might not be conscious for very small intervals of time of many things we do not remember, or about which we are unable to speak or write, which we cannot express in characters on account of their extremely small size, since they would have little relation to such things. But they are not on this account any less sensed by our consciousness. Rather, we forget about these things, just as we forget about the things we dream about. (A 6.3:499)

It is not altogether clear whether Leibniz is talking about mathematical reasoning and images in the mind of polygons whose irregularities are gradually forgotten and replaced by the thought of uniformity, or whether he means to be discussing sensory perception and the shapes of things in the external world. Indeed, he appears to be slipping from the one to the other. I believe the doctrine of the role of the imagination that he is offering is in fact meant to cover both cases; at any rate, the point of his discussion can be made out either way. What we take to be uniformity of shape is an appearance generated by a two-stage activity of the mind: the ‘‘irregularities’’ or ‘‘inequalities’’ of the shapes actually presented to the mind are neglected or forgotten, and then the idea of uniformity is applied to fill out the thought of the object. Thus the imagination plasters over much of the fine structure of the objects under consideration—indeed, infinitely many details—to produce a cognitive appearance of uniformity. Still, the irregularities of the objects exist and are even ‘‘sensed by our consciousness,’’ but only so subtly or fleetingly that they fail to be retained in consciousness long enough to be manifest in our experience of things. Where memory fails, imagination takes up the slack. Leibniz’s account of the continuum from the Pacidius combines with his remarks on the imagination from On Infinite Numbers to suggest an intriguing picture of the relation between mind and world. Sense perception presents to the subject of experience the appearance of a world of precise shapes; the manifest image (to borrow Sellars’s term) of corporeal reality is thus very much a Cartesian image of geometrically precise bodies. But the scientific image, itself only fleetingly sensed in consciousness, is of an infinitely more complex world. Corporeal reality behind the appearances is an infinitely divided fractal world, and it is only the activity of the mind, and of the imagination in particular, that smoothes over its rough edges and presents the world in experience as if it were a Cartesian geometrically uniform one. Of course closer inspection, say perception aided by microscopes, might bring to light more of the underlying irregularities in things; but at any given scale of inspection, ‘‘most’’ of the infinitely many details of the division of matter will fall below the threshold of sustained conscious experience. The precise shapes that appear in experience are thus imaginary, or at least involve something imaginary, insofar as the uniformity they appear to contain is not actually present in the things in nature but is the result of the activity of the imagination.

On Precise Shapes and the Corporeal World 81

Still more insight into Leibniz’s thought comes in a 1678 piece, Chrysippus’ Heap, where he explicitly defines imaginary notions as follows: I call those notions imaginary which are not in the things outside us, but whose essence it is to appear to us. (A 6.4:70)23

This point about its being of the essence of imaginary notions to appear to us is carried forward into later writings, and Leibniz’s view of the manifest image can be observed in connection with it in a passage about sense perception from the New Essays (1704): [W]hen the swift rotation of a cog-wheel makes us perceive an artificial transparency, as I have noticed on visits to clock-makers, we are not able to discern the idea of the cause of this, i.e. the idea of the teeth on the wheel. The wheel’s rotation makes the teeth disappear and an imaginary continuous transparent [ring] appear in their place; it is made up of successive appearances of teeth and of gaps between them, but in such rapid succession that our imagination cannot distinguish them. So the teeth are encountered in the distinct notion of this transparency, but not in that confused sensory perception of it. It is the latter’s nature to be confused and to remain so; for if the confusion ceased (e.g. if the motion slowed down enough for us to be able to observe the parts in succession) it would no longer be this same perception, i.e. it would no longer be this image of transparency. (A 6.6:403)

It is of the essence of the sensory perception of the rotating cog-wheel that the finer details, the teeth and the gaps between them, cannot be distinguished. The notion of this appearance presented by sense perception is that of a continuous transparent ring, but it is only an imaginary notion. Note, however, that Leibniz’s holding a certain notion involved in the content of experience to be imaginary does not automatically mean that there is nothing in the world behind the appearances to correspond to the experience. For by contrast with the imaginary notion of a continuous transparent ring, there is the distinct notion of this transparency, and in it are contained those details of the reality that the appearances neglect. Presumably this distinct notion of the transparency, like those of the true nature of motion or shape, is one ‘‘which we perceive not by the imagination but by the intellect’’ (A 6.4:1622)—by virtue of a proper understanding of underlying facts. When in later texts the notions of shape, motion and extension are classified as imaginary, or involving something imaginary, those qualities are thereby being placed in the same category with sensible qualities, including those of color, sound, and heat, and even those of the illusory quality of transparency that appears to belong to the moving cog-wheel.24 In almost any context the proposition that the geometrical qualities of shape, motion, and extension are on a footing with sensible qualities would be surprising; in the era of Cartesianism, it would strike the reader as especially hard to believe. This is not lost on Leibniz, whose draft of the April 30, 1687, letter to Arnauld prefaces a few comments on shape and motion with the warning: ‘‘Perhaps one will have cause to be surprised that I place shape and motion in the same rank with phenomena.’’ After a summary of his critique of motion, the draft continues:

82 Leibniz: Nature and Freedom And with respect to shape I uphold another paradox, namely, that there is no shape exact and real, and that neither sphere, nor parabola, nor other perfect shape will ever be found in body. . . . One will always find inequalities there to infinity. That comes about because matter is always subdivided to infinity. (Draft of letter of April 30, 1687; LBr 16, Bl. 81v, quoted in Sleigh 1990: 112)

Leibniz clearly takes a degree of satisfaction in this ‘‘paradox,’’ and perhaps he finds some irony in it as well. For the implication of his analysis of precise shapes and the structure of corporeal reality is that the austere and anti-Scholastic scientific image proposed by the Cartesians, far from being a purified intellectual account of the world, is itself guilty of mistaking the sensory appearances for reality. So it is clear that the early writings on the continuum contain the elements necessary to formulate the idea that the corporeal world is infinitely complex in its actual structure but nonetheless appears to consist in bodies with mathematically definable precise shapes due to the natural limits on sense perception and the activity of the imagination. It is clear as well in the later writings that those elements have been woven together in just that way. In a 1705 letter to Sophie, Leibniz writes: It is the imperfection and fault of our senses that makes us conceive of physical things as Mathematical Beings, in which there is indeterminacy. It can be demonstrated that there is no line or shape in nature that gives exactly and keeps uniformly for the least space and time the properties of a straight or circular line, or of any other line whose definition a finite mind can comprehend. (G 7:563)

And again, with particular clarity, in a letter to Burcher de Volder, around 1704–5: A spherical body complete in all respects is nowhere in nature; the soul forms such a notion by concealing aberrations. And it is the same for any other shape that a finite mind can grasp, namely, such a shape can never exist exactly. (G 2:277)

Thus the fractal conception of the world and the account of sense perception and imagination seem to be at work in Leibniz’s thought about the appearance of precise shapes and the reality behind them both early and late in his career. They are the foundations upon which he rests his critique of precise shapes. With Leibniz’s early writings on the continuum and the imagination in mind, the critiques of motion and of precise shapes can be easily recognized in later writings as well—writings that have sometimes appeared rather enigmatic otherwise. For instance, in Wonders Concerning the Nature of Corporeal Substance (c. 1683), Leibniz writes: Even though extension and motion are more distinctly understood than other qualities, since all the rest have to be explained using them, it must in fact still be acknowledged that neither extension nor motion can be understood distinctly by us at all. This is because on the one hand we are always embroiled in the difficulties concerning the composition of the continuum and the infinite, and on the other, because there are in fact no precise shapes [certae figurae] in the nature of things, and consequently no precise motions [certi motus]. And just as color and sound are phenomena, rather than true attributes of things containing a certain absolute nature

On Precise Shapes and the Corporeal World 83 without relation to us, so too are extension and motion. For it cannot really be said just which subject the motion is in. Consequently nothing in motion is real besides the force and power vested in things. (A 6.4:1465)

Though the critiques of motion and precise shape appear only in summary form here, the outlines of his earlier views remain evident: from the analysis of the composition of the continuum and the argument from the equivalence of hypothesis, the Cartesian concepts of extension, shape, and motion can be seen to contain something imaginary; they are thus to be superseded by a metaphysics of corporeal beings developed in terms of force or power. Not all of Leibniz’s presentations of his critique of precise shapes are organized to display the role of the paradoxes of the continuum. The ‘‘merely’’ physical considerations deriving from the kinematics of intraplenary motion are often allowed to stand on their own, and some of Leibniz’s clearest statements of his view of precise shapes are given in this mode. A well-known discussion of this sort occurs in the 1689 piece Primary Truths.25 There Leibniz begins by pointing out that in the plenum all motions will be transmitted to all bodies at least to some degree: In a vessel filled with liquid—and the whole universe is just such a vessel—motion made in the middle is propagated to the edges, although it is rendered more and more insensible, the more it recedes from its origin. (A 6.4:1646–1647)

To sustain all these effects of near and remote motions, it will follow that in the plenum ‘‘there is no body so small that it is not actually subdivided,’’ and each particular body ‘‘is acted upon by everything else in the whole universe and receives some effect from everything’’ (A 6.4:1646–1647). The actions of other bodies on a given body x are expressed in x by its internal structure of parts and its changes of shape; these changes consist in its being divided into parts, and thus to express an infinity of distinct effects impressed upon it by the outer world, the division of x into parts will likewise be infinitary. The idea that the exposure of a finite body to the actions of an infinite plenum of other moving bodies causes that body and all of its parts to be divided into further parts becomes a staple of Leibniz’s thought by the mid 1680s (see A 6.4:1613–1614, 1622). In Primary Truths Leibniz presses it into its far-reaching consequences for the nature of bodily shape: There is no determinate shape in actual things, for none can be appropriate [satifacere] for an infinite number of impressions. And so neither a circle, nor an ellipse, nor any other line we can define exists except in the intellect, nor do lines exist before they are drawn nor parts before they are separated off. Extension and motion, as well as bodies themselves, insofar as only motion and extension are placed in bodies, are not substances, but true phenomena, like rainbows and mock suns. For there are no shapes in things. (A 6.4:1648)

Such examples could be further multiplied, but Primary Truths, Wonders, and passages quoted earlier from Discourse on Metaphysics and Specimen of Discoveries should suffice as evidence of the content of Leibniz’s account in the middle years.

84 Leibniz: Nature and Freedom

Shape, Force, and Realism about the Corporeal World The interpretation of Leibniz’s account of shape developed in this essay has it that for Leibniz actual bodies are in fact infinitely complex or fractal in structure and thus not definable by the precise shapes of traditional geometry. One way to say this is to say that actual bodies, as they are in themselves, possess fractal shapes of infinite complexity although they do not possess what Leibniz calls ‘‘precise shapes.’’ It is thus in accord with this reading of Leibniz to uphold (in Sleigh’s term) a ‘‘realistic construal’’ of the theory of corporeal substances and aggregates that is also on display in Leibniz’s writings. Much of that will be controversial, however, for on some important interpretations of Leibniz’s metaphysics of this period, the critique of precise shapes is to be understood as falling within the umbrella of an antirealist reading of his theory of corporeal beings—a reading according to which Leibniz does not really hold there to be corporeal beings at all (and in particular no corporeal substances, unities per se). On the antirealist reading Leibniz’s stated theory of corporeal substances and aggregates is rather to be construed ‘‘idealistically’’ so that, in the end, it is reduced to a theory countenancing only soul-like incorporeal substances. The critique of precise shapes has a special place in this line of interpretation since it is sometimes used as evidence against a realistic construal of his theory of corporeal beings and so as part of the case for the antirealist or idealist interpretation of Leibniz’s middle-years metaphysics of the corporeal world in general. The principal idealist readings of the critique of precise shapes have been advanced by Robert Sleigh, Jr. (1990), and by Robert Adams (1994). Concerning the positive view arising from the critique of precise shapes, their accounts are significantly the same: Leibniz means to hold that substances simply have no shape at all and likewise no extension, and this is to be understood as incompatible with a realistic construal of corporeal substance. Sleigh’s and Adams’s respective discussions of Leibniz’s critique and their criticisms of the sort of account this essay offers differ in some details, however, and it will be worth considering the two separately. I shall turn first to Adams’s account, and then take up Sleigh’s with its more extensive attack on the realistic interpretation of Leibniz’s view of shape. Adams explores Leibniz’s view of precise shapes with the idealist reading of Leibniz’s metaphysics assumed as a premise, and not primarily with an eye towards using the issue to argue for that reading (his defense of the idealist interpretation is given on different grounds). In particular Adams assumes that the view implicit in Leibniz’s discussion of shape is this: ‘‘Bodies—organic or living bodies in particular—are appearances of monads’’ (Adams 1994: 230). Monads are of course the simple incorporeal souls or soul-like substances that Leibniz explicitly postulates in later writings. On this account, bodies are appearances or phenomena by which we perceive monads, and the body is the expression of the soul or monad. As Adams sees it, the critique of precise shapes trades on the doctrine of soul-body expression and the capacity of a single body to express the monad. For according to Leibniz, the monad itself expresses or ‘‘mirrors’’ the whole universe, and since the universe is infinitely complex (containing an actual infinity of simple substances),

On Precise Shapes and the Corporeal World 85

a body of only finite complexity would not be adequate to express the soul or monad of which it is the appearance. The question is then what to make of Leibniz’s claim that there are no precise shapes in those bodies. On Adams’s view, ‘‘The shapes thus excluded from nature are finitely complex shapes’’ (1994: 230), and he offers the following suggestion about Leibniz’s view of bodies: [A] body conceived as a phenomenon having a certain definite extension, shape and motion is not complex enough to be an adequate expression of any real thing, according to Leibniz. It is not complex enough to express something that expresses the whole universe as a monad does. It is a mathematical abstraction. Perhaps the body that expresses a monad is an infinite series of such abstractions, each more complex than its predecessors. (1994: 230–231)

Adams places more emphasis on the doctrine of mutual expression between body and soul than I would; though that doctrine is indeed an important feature of Leibniz’s metaphysics already during the middle years (even apart from the question whether the soul is itself a monad or merely a substantial form of a corporeal substance), I do not see that it is actually playing a role in Leibniz’s arguments against the existence of precise shapes in things. At any rate, it would at best figure only in the ‘‘physical’’ versions of the argument; the argument from the composition of the continuum simply makes no reference at all to soul-body expression. And of course the role of the monad eo nomine in the argument is entirely a feature of its reconstruction along idealist lines, as the relevant texts all predate Leibniz’s first use of the term ‘monad’ in the metaphysics of this period. Still, Adams’s view that the shapes excluded from nature are finitely complex shapes and the suggestion that a body, conceived as a phenomenon, might be an infinite series of increasingly complex abstractions are both congenial to the reading I offer, at least up to a point. On the view I recommend of the middle-years metaphysics, there remains a distinction to be drawn in Leibniz’s philosophy between the phenomenon of bodily shape as it appears in perception and corporeal reality as it is in itself.26 Perception, due to its natural limits, can provide increasingly exact representations of corporeal reality through an infinite series of approximations, though at no single scale of inspection can all of the structure of the corporeal world, or of any part of it, be fully manifest to us. Still, there is a corporeal world whose structure is approximated in our sense perception of it. By contrast, Adams’s idealist Leibniz places nothing corporeal behind the infinite series of phenomena that appear in perception; rather, corporeal shape is exhausted by these mathematical abstractions. The infinitely complex reality behind all the veils of appearance is simply incorporeal and in no further respect does it possess any shape at all. The conclusion that I think Leibniz draws is not that real shapes are infinitely complex, though some things he says (e.g., in DM 6) might leave us with that impression. Rather, he concludes that shape as such is only a phenomenon. I suppose that an infinitely complex shape would involve a line segment of finite length that changes not merely its curvature but also the direction of its change of curvature infinitely many times, and that Leibniz would have thought that

86 Leibniz: Nature and Freedom an absurd and impossible monstrosity. What I assume he would say, instead of postulating infinitely complex shapes, is that for every finitely complex shape that might be ascribed to a body there is another still more complex that more adequately expresses reality. (Adams 1994: 230)

It is indeed Leibniz’s preferred way to characterize the infinite ‘‘syncategorematically,’’ so to speak, by employing the quantificational form ‘‘for every x there is a y . . . ’’ (for every p, there is a p . . . ),27 and in one sense Adams may be right about whether Leibniz would postulate infinitely complex shapes. For even if, as I suggest, Leibniz ascribes something like fractal structure to corporeal beings, he might well take this to transcend the concept of shape, or at any rate to transcend the traditional geometrical conception of shape that is supposed to be one of the modes of the Cartesian attribute of extension. But if an attribution of fractal structure is consistent with the denial of there being such a thing as an infinitely complex shape, then the dispute at this point would be only about a word. What matters here is whether Leibniz’s philosophy of body countenances the idea of infinitely complex structure that might coherently (and realistically construed) be ascribed to a corporeal being. Against this, Adams supposes that Leibniz would have thought the idea of an infinitely complex shape to be ‘‘an absurd and impossible monstrosity.’’ Since the definition Adams offers of infinitely complex shape is actually weaker than what I have been prepared to ascribe to Leibnizian bodies—on the fractal conception one could add that the infinitely complex line segment changes the direction of its change of curvature infinitely many times in every finite subinterval—his supposition about Leibniz’s hostility toward the idea of infinitely complex shape certainly would, if correct, count strongly against my interpretation. But I think Adams is mistaken in what he supposes here about Leibniz. For Leibniz was well acquainted with mathematical structures that appear to display the very properties Adams mentions, and it seems in accord with Leibniz’s own work in mathematics to see him as entirely receptive to them. Take as a simple example the expression for the value p/4 that is given by an oscillating infinite series of rational fractions: p/4 ¼ 1/1 – 1/3 þ 1/5 – 1/7 þ 1/9 – 1/11 þ etc. ad infinitum. The series of fractions—called Leibniz’s series for his discovery of it in 1673—can naturally be graphed by a curve that passes, say, above and below the line y ¼ p/4 infinitely many times between 0 and 1 on the x-axis. It is unlikely that this would have been lost on Leibniz. Another example to be drawn from Leibniz’s texts, and one with which he was clearly quite taken, is the idea of packing a finite space with nonoverlapping spheres.28 A simplified case fills a polygonal piece of the plane with an infinity of nonoverlapping circles whose radii decrease geometrically without bound (the interstices between the larger circles being filled by ever smaller ones ad infinitum). Expand the circles into spheres, and the bubbled surface (top or bottom) that now results from the initially flat polygon will contain an infinity of changes of direction of curvature despite its being a finitely bounded piece of the plane.29 Would Leibniz really have thought examples like this to be absurd monstrosities? It seems more typical of his attitude toward uncovering such instances of infinite

On Precise Shapes and the Corporeal World 87

complexity within finite objects to find them wondrous—further testimony to the marvelous nature of the secrets hidden in mathematics. And indeed if ‘‘everything is full of globes’’ (A 6.4:1933), such infinitely complex fractal shapes will be found in every portion of matter however small. Sleigh, like Adams, understands Leibniz to argue against the presence of finitely complex shapes in nature on the ground that such shapes are insufficiently complex to allow bodies to represent the infinite world. Sleigh’s account does not rely on the doctrines of soul-body expression and monadic mirroring, however, but rather places in the central role of the argument the physical accommodation of a finite body to the infinity of distinct motions of the other bodies in the universe. As he sees it, the argument from (1) the claim that bodies are actually infinitely subdivided and that each body contributes to the motion and shape of every other body to (2) the thesis that there are no precise shapes in bodies runs as follows: Let x be any material object (corporeal substance or aggregate thereof). Let S be any shape of any degree of complexity. S cannot be the shape of x because, however complex it is, it represents the influence on x (via impressed motion) of only finitely many other bodies. Given the actual infinite division of matter, whatever shape we assign to x, it is not complex enough, because it fails to take into account the influence (via impressed motion) of infinitely many other bodies. Thus baldly put, Leibniz’s reasoning may appear to need further buttressing. That project I leave to others. My point is that whatever the merits of Leibniz’s argument, his conclusion was that finite material objects strictly speaking lack shape, not that they have (per impossibile) shape of infinite complexity. (Sleigh 1990: 113)

Thus on Sleigh’s account, an argument concerning the relations between moving bodies in the universe, and expressed in terms neutral between realist and idealist interpretations of body in Leibniz, yields a conclusion that threatens the ontological status of corporeal beings: strictly speaking, they lack shape. Unlike Adams, Sleigh is expressly concerned to address the dispute over realism about corporeal beings in Leibniz’s middle-years metaphysics by reference to the critique of precise shapes. ‘‘The issue before us,’’ he writes, ‘‘is whether Leibniz then held a realistic conception of extension and its modes, or whether he looked to an idealistic, monadological analysis’’ (Sleigh 1990: 110–111). And he has in full view the basic strategy of a realistic interpretation: Those who ascribe to Leibniz a realistic interpretation of the relevant modes in our time period construe matters as follows: Leibniz was engaged in arguing that extension is not the essence of corporeal substance, that substantial form is, that the substantial form of a corporeal substance is its primitive active force, and that reference to force, active and passive, primitive and derivative, permits a realistic interpretation of size, shape and motion. (Sleigh 1990: 111)

As Sleigh goes on to note, motion and shape present two separate challenges to a realistic interpretation. Can both of them be met? In the first section of this essay I considered a sketch of a realistic interpretation of motion according to which Leibniz takes a two-sided view. Motion is change of place and, in our experience of it, appears to be an absolute or real quality of individual bodies. In fact, however, the appearance to us of absolute motion involves something imaginary and relative

88 Leibniz: Nature and Freedom

to our perception, since we imagine ourselves or some other bodies in a given system to be at rest. Yet in reality motion springs from ‘‘something real and absolute’’ (G 2:133) by virtue of which (relative) change of place occurs, namely, the physical forces that are intrinsic to individual bodies and that support an absolute distinction between rest and motion, though only God might be able to draw it correctly. This was to serve, in outline, as a model for thinking about Leibniz’s view of shape; and the analysis of shape I have offered is likewise two-sided. As it appears to us in sense perception, an individual body is finitely complex. This appearance, however, involves something imaginary due to the activity of the imagination in covering over infinitely many fine details of the body’s actual structure that are not held long enough in perception to be remembered, and it does so by filling out our sensory experience of the body with the thought of uniformity. Still, the reality behind the appearances is that of a corporeal world of infinitely complex corporeal beings whose fractal structure is at best only approximated by the precise shapes of the sort definable in traditional geometry. About motion, Sleigh acknowledges the merits of the realistic interpretation (‘‘there is much to be said in its favor’’), and he does a significant part of the philosophical and textual spadework on behalf of the Leibnizian reduction of motion to force. He is less sanguine, however, about the prospects of a realistic construal of Leibniz’s discussion of shape: No such strategy will do with respect to the problem posed by Leibniz’s remarks concerning shape. No one is likely to suppose that with the injection of form or force into shapeless mass, it literally ‘‘shapes up.’’ (1990: 112)

In the absence of a parallel reduction of shape to force, Sleigh suggests, ‘‘the realistic interpretation must be largely deflationary’’ (1990: 113). The deflationary reading he has in mind is one that trades on Leibniz’s repeated proviso to the effect that in the nature of things there are no precise shapes definable by us. This is on display in many of the major texts concerning precise shapes: ‘‘any assignable shape’’ (A 6.4:1622); ‘‘whose definition a finite mind can comprehend’’ (G 7:563); ‘‘that a finite mind can grasp’’ (G 2:277); ‘‘that we can define’’ (A 6.4:1648), and so on. Those provisos do seem to express a real element of Leibniz’s view, but one might wonder how much can finally be made of them. Sleigh describes the deflationary reading in this way: According to the deflationary reading, when Leibniz asserted, somewhat mysteriously, that material objects lack determinate shape, all he had in mind was this innocuous doctrine: take your stock high-school geometrical figures, or any others ‘‘definable by us’’ or that ‘‘possess in full force the properties that we learn in geometry’’ (G 4:568/L 583); none of them is exemplified by real objects in nature. Of course, for any given material object, including any given corporeal substance, there exists some shape such that the object has that shape. But given the intricacies of nature, that shape will be very complex, so complex that it is not ‘‘definable by us.’’ (1990: 113)

This reading is deflationary indeed, and, as Sleigh notes, ‘‘it deserves to be deflated and rejected’’ (1990: 113).

On Precise Shapes and the Corporeal World 89

Still, it appears that a realistic interpretation can take advantage of a less deflationary reading of Leibniz’s remarks about the definability of shape. If Leibniz’s point about definability is that precise shapes are definable by us because they are in principle finitely definable or constructible—for instance, if they are supposed to be constructible from algebraic curves or ones expressible by polynomial equations of finite degree—then he is making a nondeflationary point with his claim that there are no shapes ‘‘definable by us’’ in the nature of things. For not all shapes are even in principle finitely definable in that sense. Suppose that nature is everywhere fractal in structure; it could then be true that its ‘‘real shapes’’ cannot be exactly described by finitely definable shapes and that no equation of any finite degree can perfectly express the actual boundaries of any actual bodies. Thus under that reading, there is a robust sense in which no shapes definable by us—no finitely definable shapes, no ‘‘precise shapes’’—exist in nature. The idea of such a distinction between shapes that are finitely definable and those that are not was available to Leibniz. The distinction between algebraic and transcendental curves,30 for example, was emerging during Leibniz’s time (Descartes was aware of the possibility of nonalgebraic curves as well, and he excluded them from geometry proper on the grounds of their intelligibility to us).31 Likewise, the related distinction between algebraic and transcendental numbers was gradually being recognized as well. Leibniz himself even gives an argument that can now be appreciated as a (partial) proof of the fact that p is transcendental— that is, that p is not only an irrational number but also one that cannot be expressed as the root of any polynomial equation of finite degree (see Arthur 1999b). And of course the question whether there is a rigorously true and precise mathematical expression for incommensurable quantities certainly occupied his thought. It is not news that Leibniz was deeply impressed by the distinction in mathematics between finitely definable terms and those that transcend finite definition: it would inspire him to propose a similar distinction to hold between necessary and contingent truths.32 If he does hold the fractal view of corporeal beings that I have been suggesting, it would be natural for him to see a distinction of that sort between the precise shapes of geometry and the kinds of structures that will exist in nature, and to mark this distinction in terms of what is definable in such a way that it can be comprehended by a finite mind. And this, I think, is just what we find him doing in those many passages where he sharpens his thesis about precise shapes with that proviso. To continue to fill out the realistic intrepretation of shape, some defense might also be given for the idea that corporeal shape, like motion, is to be reduced to forces in Leibniz’s account. At least once, in the course of discussing the idea that the notion of form in his theory of corporeal substance is to be understood as primitive force, Leibniz mentions: this primitive force or [seu] form of substance—which, indeed, fixes shapes in matter at the same time as it produces motion. (G 4:397)

This suggests that Leibniz thinks shape and motion are both results of force, and thus the strategy of a realistic interpretation that finds a reductive analysis of the modes of extension—now reconstrued along Leibnizian rather than Cartesian

90 Leibniz: Nature and Freedom

lines—to force can be effective in both cases. The idea that force fixes shape in matter may make little sense if this has to presuppose a prior notion of ‘‘shapeless mass’’ or matter existing entirely without forces or forms. Leibniz is sometimes given to talking in that way when he imagines the compounding of form and matter to yield individual bodies (see the 1669 letter to Thomasius at A 6.2:435), but his considered view is more likely simply that the concept of force identifies the source or cause of the particular shapes in things. Matter is not some preexistent stuff that comes to be arranged in a certain way by the superaddition of certain forces; rather, what it is for there to be matter arranged into a certain shape is for there to be a certain nexus of forces acting in a given space. The existence of a corporeal being with its actual modes of (Leibnizian) extension, so to speak, consists in the existence of a particular cluster of those forces. Extension itself is to be understood as a diffusion of resistance,33 and thus the specific diffusion of it into a given space constitutes a body of a specific shape. It simply makes no sense to suppose that there might be a body of a given shape that yet has no forces acting within it, for that would not be a real body at all, but at best only an illusion of a body—like a rainbow or mock sun. The Cartesian view of the modes of extension and the conception of body defined in terms of them suffer from precisely such a defect, however, by supposing that the basic properties of corporeal beings can be conceived apart from forces. Thus if there were only such qualities of shape, motion and extension in bodies, then bodies would be nothing more than appearances in the mind—merely imaginary phenomena no more substantial than the appearances of colors or other sensible qualities. In September of 1687 Leibniz writes to Arnauld in defense of his theory of corporeal substance that identifies form or entelechy with force, and against the Cartesian account: [E]xtended mass considered without entelechies, consisting only of these qualities [i.e., shape, motion, extension] is not bodily substance, but an entirely pure phenomenon like the rainbow. Therefore, philosophers have recognized that it is form which gives determinate being to matter, and those who do not pay attention to that will never emerge from the labyrinth of the composition of the continuum, if they once enter it. (G 2:119)

The fantasy of a world in which geometry defines the essence of corporeal being has to be dispelled in order to establish the true conception. The arguments stemming from the analysis of the composition of the continuum can refute the geometrical conception of body and thus clear the way to introduce the dynamical account of body as consisting in force; thus in a text from 1683–85/6, Leibniz remarks: Concerning bodies I can demonstrate that not only light, heat, color and similar quantities are apparent, but also motion, and shape, and extension. And if anything is real, it is only the force of acting and being acted upon, and so the substance of body consists in this. (A 6.4:1504)

The realistic interpretation of shape, motion, and extension in Leibniz’s metaphysics of body should hold this view of the nature of body as fundamentally constituted by forces to be Leibniz’s great insight into the nature of corporeal

On Precise Shapes and the Corporeal World 91

beings; and it is an insight that has a pretty fair claim to being the truth about corporeal beings as well. A reductive analysis of shape and motion to force allows for a nuanced realism about shape and motion as properties that belong to bodies as they are in themselves. In the manifest image of nature, bodies appear to us to be moving absolutely and to be finitely complex geometrical objects; neither of those appearances can be absolutely true of the world outside us, as can be demonstrated by an analysis of the paradoxes of the composition of the continuum and by the argument from the equivalence of hypotheses. Yet behind the manifest image is a corporeal world constituted by forces that yield infinitely complex fractal bodies and determine those bodies to be in absolute states of rest and motion. This world is too finely grained and too rapidly changing for a finite mind’s perception of it be sustained in consciousness, and so the imagination rounds out a partial representation of the world and thereby furnishes us with a sensory experience that leaves most of the details of corporeal reality obscure. Thus the phenomena of motion and shape that appear in our perception involve something imaginary. But by means of careful philosophical inquiry, we can avoid being taken in by the illusions that ensnared the Cartesian account and can form a proper understanding of how the corporeal world is in itself. This broadly sketched account of shape and motion—and of the corporeal world in general—does not vindicate a realistic interpretation of Leibniz’s theory of corporeal beings. It only defends that sort of interpretation against a line of criticism that would put Leibniz’s critique of precise shapes at odds with realism about corporeal beings in his metaphysics. One lesson of this account of the critique of precise shapes is that antirealists about corporeal beings in Leibniz’s middle-years metaphysics will have to look elsewhere to make their case. But they may not have to look far. The reductive analysis of shape and motion to force means that the dispute over realism about corporeal beings will now be resolved into the question of the nature and reality of force itself. On this point Sleigh remarks: In numerous passages Leibniz identified the substantial form of a corporeal substance with its primitive active force. Thus, in Specimen (G 7:317/MP 84–5), he wrote: ‘‘This principle of actions, or primitive active force, from which a series of states follows, is the form of the substance.’’ But, of course, we need to know whether a realistic, nonmonadological interpretation of primitive active force is Leibniz’s last word on that subject in our time period. (1990: 111)

In later writings, such as the correspondence with Wolff, Leibniz’s ontology of forces seems to be very much an idealist one.34 In the middle years, by contrast, it is less than clear that idealism is on Leibniz’s mind at all. Even by the middle of the 1690s—which strikes me as a transitional period during which the theory of simple substances is gradually taking hold in the foundations—the case for an idealist analysis of forces is not yet certainly in place. In a passage from the 1695 New System, Leibniz writes of substantial forms: I found that their nature consists in force and that from this there follows something analogous to feeling and to appetite; and that therefore it was necessary to form a conception of them resembling our ordinary conception of souls. (G 4:479)

92 Leibniz: Nature and Freedom

The issue of realism here would turn on the character of the analogy with feeling and appetite. Does Leibniz mean in that passage to rest the account of force on purely idealist foundations, or does he intend only to illuminate the idea of force by reference to ideas we have already about the nature of the soul? I think this is still an open question. Looking ahead to the later writings, it may be hard not see the end of the corporeal world in Leibniz’s metaphysics. But it is another matter altogether how the ontology of forces in the middle years is to be understood, and no doubt an inquiry into the subject is in order. That project, however, I leave to others.

Notes My thanks to Christie Thomas, Ric Arthur, Jan Cover and Don Rutherford for helpful comments and discussion. 1. Here and hereafter I use ‘corporeal being’ to cover both sides of Leibniz’s distinction—one important to Leibniz but not to this essay—between corporeal substances such as an animal or a man, on the one hand, and mere aggregates such as a pile of stones, on the other. See A 6.4:627, 1506; G 2:59, 72. 2. Parkinson (MP 249) and Arthur (LOC 423, n. 2) instead suggest 1686 as a date for Specimen of Discoveries, which is plausible given the content of its discussion. 3. My translations of Leibniz’s writings generally follow RB, AG, and LOC, though I have sometimes modified their translations without comment. For some passages I have also consulted Russell 1900, L, MP, Sleigh 1990, Adams 1994, and Rutherford 1995a. Inaccuracies are my own. 4. In the New Essays of 1704, for instance, Leibniz writes: ‘‘I have already pointed out during one of our previous conversations that you treat as simple many ideas which are not so. Motion is one of them: I believe it to be definable, and the definition which says that it is change of place deserves respect. Aristotle’s definition is not as absurd as it is thought to be by those who do not understand that for him the Greek kineˆsis did not signify what we call ‘motion’ but rather what we would express by the word ‘change,’ which is why he gives it such an abstract and metaphysical definition; whereas what we call ‘motion’—and he would call phora, carrying—is one of the kinds of change (teˆs kineˆseoˆs)’’ (A 6.6:297). See A 6.4:1559, C 590, GM 6:327. 5. For discussion of Leibniz’s theory of motion in this dialogue, see Arthur 1989, White 1992, and Levey 2003. 6. In fact Leibniz has reached this conclusion already in notes on Descartes’ Principles of Philosophy—evidently written at the end of 1675—in which he records the following observation, clearly stating the main elements of the argument for the relativity of motion: ‘‘When two bodies are moved on the globe of the Earth, the one from east to west, the other from west to east, the Earth can be understood to be immobile. But also nothing prevents me from imagining in turn that I am on one of the smaller bodies, which will thus be regarded as being at rest. Therefore it is clear that any body can be imagined to be at rest, if only a sentient being is understood to be on it. . . . Of course we humans are accustomed to attribute rest to larger bodies, and this for the sake of abbreviation or ordering of thought. Just as it is useful to reduce two fractions to a common denominator, or the sines of two angles to the same radius; so it is useful to suppose as immobile that which is of a broad and enduring cast’’ (A 6.3:214). 7. Leibniz will frequently repeat this argument against absolute motion (for instance, in DM 18); for a relatively early example (c. 1682/3), see A 6.4:1463: ‘‘Thus concerning motion I have shown elsewhere that it is not possible to determine which subject it is in. . . . ’’ 8. See G 4:400. Also, from the fifth letter to Clarke in 1715: ‘‘I grant there is a difference between the absolute true motion of a body and a mere relative change of situation with respect to another body’’ (G 7:404). 9. Russell 1900: 86–89 advances a few pointed criticisms; see also Earman 1989: 130–134. 10. As Richard Arthur observes: ‘‘Leibniz had first written ‘of the relative speed,’ and then ‘of the force or a[ction],’ before finally settling on ‘of the action or relative motion’ ’’ (LOC 395, n. 9).

On Precise Shapes and the Corporeal World 93 11. See Garber 1995 and Fichant 1974. 12. Obviously the topic deserves closer study than can be offered here. For some discussion of Leibniz on motion see: Russell 1900: chap. 7; Garber 1982, 1995; C. Wilson 1989: 205–217; Rutherford 1995a: chap. 9. 13. For discussion of Descartes on the nature of body, see Garber 1992a: chap. 3, especially 63–69 and 75–93. See also M. Wilson 1978: 83–88, 166–169. 14. For discussion see Garber 1985: 30, 78; Sleigh 1990: 166; Williams 1978: 228; Woolhouse 1993: 94–96, 102–115. 15. Loosely those arguments might be distinguished as addressing ‘‘diachronic’’ and ‘‘synchronic’’ considerations, but the distinction itself tends to undermine the synchronic side, since Leibniz will ultimately deny that there is any purely synchronic or purely momentary structure in the world. 16. Levey, ‘‘Dans les corps il n’y a point de figure parfaite: Leibniz on Time, Change and Corporeal Substance,’’ forthcoming. See also Adams 1994: 230and Sleigh 1990: 211. 17. For some discussion of this point about intraplenary motion, see Arthur 1989; Garber 1992a: chap. 5; Levey 1998. 18. This is of course Leibniz’s enduring view of points: see A 6.4:1648; G 4:491; A 6.6:152; G 2:370, 520; etc. 19. For further discussion, see Levey 2003. 20. For example, in the Pacidius itself we find: ‘‘Quemadmodum polygonum regulare infinitorum laterum pro circulo metaphyisce haberi non potest, tametsi in Geometria pro circulo habeatur, ob errorem minorem quam ut numero ullo a nobis exprimi possit’’ (A 6.3:569; in the critical apparatus, reconstructed by Beeley 1999: 140, n. 70). 21. See GM 4:95–96. For discussion, see Mancosu 1996: chaps. 2, 6, and Arthur 1998: 36–37. 22. Leibniz is evidently worried about the paradoxes of the continuum taking hold for mental images of mathematical objects as well: ‘‘But what if that image does not represent any polygons at all? Then the image presented to the mind is a perfect circle. Here there is a surprising and subtle difficulty. For even if the image is false, the entity in it is nevertheless true; and so it follows that in the mind there is a perfect circle, or rather, there is a real image. Therefore everything else will also exist in the mind: and in it everything that I denied to be possible will now be possible’’ (A 6.3:499). 23. Leibniz goes on to say: ‘‘all those notions in which the heap or Sorites of the Stoics applies, such as wealth, poverty, baldness, heat, cold, tepidness, white and black, big and small, taken absolutely, are vague imaginary notions’’ (A 6.3:70). 24. More precisely, the notions of shape and motion are ranked near to those of the sensible qualities; Leibniz typically qualifies this classification by pointing out that the sensible qualities are even more imaginary than the Cartesian modes of extension. See A 6.4:1545. 25. Previously dated to 1686, it is now also retitled Principia Logico-Metaphysica in the Akademie edition; see A 6.4:1643. 26. There is also a distinction to be drawn between purely imaginary phenomena—phantasms or pure illusions—and what Leibniz will call real phenomena, or the appearances of real things to us; Leibniz discusses this distinction in some detail in De modo distinguendi phaenomena realia ab imaginariis (c. 1683–85/6, A 6.4:1498–1504). For some discussion, see Adams 1994: 256 27. Leibniz in several places holds that ‘infinite’ and its cognates are to be understood as syncategorematic terms; for instance in the New Essays he writes: ‘‘It is perfectly correct to say that there is an infinity of things, i.e. that there are always more of them than one can specify. But it is easy to demonstrate that there is no infinite number, nor any infinite line or other infinite quantity, if these are taken to be genuine wholes. The Scholastics were taking that view, or should have been doing so, when they allowed a ‘syncategorematic’ infinite, as they called it, but not a ‘categorematic’ one’’ (A 6.6:157). See A 6.3:503 and G 2:304–305. For discussion of Leibniz’s view of the infinite as syncategorematic see Ishiguro 1990: 78–100; Bassler 1998: 855, n. 15; Arthur 2001: 107. 28. One context of the appearance of the idea of sphere-packing is Leibniz’s use of the hypothesis of infinite division to make sense of the possibility of motion within a plenum. Leibniz writes: ‘‘Now if we suppose that every body is actually divided into parts, the objections against the plenum are easily resolved. And it is truly evident that if everything is full of globes, then between the interstices new globes can be placed again to infinity without violating motion, for it is only necessary for the smaller

94 Leibniz: Nature and Freedom globes to move more swiftly. Now if it is possible for everything to be a plenum, then everything will be a plenum, for it is absurd for any place to be left useless in which there could be an infinity of creatures.’’ Metaphysical Definitions and Reflections, c. 1678–80/1 (A 6.4:1399). Compare also Leibniz’s remarks in On the Plenitude of the World at A 6.3:525: ‘‘A wonderful plenum of the kind am I expounding is reasonable, even though it consists only of spheres. For there is no place so small that we cannot imagine a smaller sphere to exist in it. If we suppose this to be so, there will be no assignable place that is empty. . . . [G]iven the plenitude of the world, it is necessary that there exist some globules smaller than others to infinity.’’ 29. In the letter to Des Bosses of March 11, 1706, Leibniz describes a more complex packing. ‘‘Imagine a circle, and inscribe in it three other equal circles of maximum size; and in turn inscribe within each of these new circles and within the interstices between the circles three equal circles of maximum size, and imagine the progression to be thus to infinity . . . ’’ (G 2:306). A natural expansion of this case from families of nested circles to families of partially overlapping spheres would yield a finitely bounded, infinitely bubbled surface with further bubbles on the bubbles ad infinitum, so that there are infinitely many changes of the direction of curvature in every piece of the surface however small—thus satisfying even the much stronger condition on the sort of infinitely complex shape that I would ascribe to Leibnizian corporeal structure. 30. A (real) number is algebraic if it is the root of a polynomial, i.e., satisfies an equation of the form: a0xn þ a1xn
1 þ a2xn
2 þ . . . þ an
1x þ an ¼ 0, where ai is rational. Real numbers that are not roots for such equations are transcendental. A curve is algebraic if it can be expressed by an algebraic equation. 31. For some discussion see Grosholz 1991, especially chaps. 1 and 2, and Mancosu 1996: chap. 3. In a 1679 letter, apparently to Molanus, Leibniz remarks briefly on the limitation of Descartes’ approach to geometry—noting how even Descartes conceded that his method could not solve Diophantine problems or find inverse tangents—and says: ‘‘since then we have discovered things in geometry that Descartes thought were impossible’’ (A 6.2:504); however, it is not clear just which discoveries Leibniz has in mind. 32. Leibniz discusses this doctrine in many places; just as a sample one might consider The Source of Contingent Truths from an Infinite Process, Compared with the Example of Proportions between Incommensurable Quantities (1689?, A 6.4:1661–1664); see texts N. 325, 326, 3271, at A 6.4:1649–1661. The ‘‘infinite analysis’’ theory of contingency has also been widely discussed in the literature on Leibniz; for some recent examples, see Hacking 1974; Mates 1986: chap. 6; Castan˜eda 1990; Sleigh 1990: 83–89; Carriero 1993; Adams 1994: 25–30; Cover and Hawthorne 2000a. 33. The idea that extension is diffused resistance is suggested in many passage (see G 2:120; G 4:394; G 6:294, 584; G 7:398–399; GM 4:235); for some discussion see Hartz 1992: 529and Garber 1995: 291. See also Rutherford 1995a: chap. 9. 34. For discussion, see Rutherford 2004, and Adams 1994: chap. 13.

4

Leibniz and Idealism DANIEL GARBER

ntil recently, it was uncontroversial to call Leibniz an idealist. And for good

U reason. As he wrote to De Volder in a characteristic passage, ‘‘considering the

matter carefully, we must say that there is nothing in things but simple substances, and in them, perception and appetite’’ (G 2:282/AG 185).1 While the term ‘idealism’ was coined only after Leibniz’s death, so far as I can tell, the fit is pretty good. In Baumgarten’s Metaphysica (1739), for example, the idealist is characterized as ‘‘solos in hoc mundo spiritus admittens’’ (‘‘admitting only spirits in this world’’) (sec. 402). One must be somewhat careful here. Leibniz’s monads come in infinite gradations of perfection, and at the lower level, are quite different from anything he (or anyone else) would want to call a mind, strictly speaking. Furthermore, it is quite important for Leibniz that the world of monads have an elaborate structure. Though everything, in a sense, reduces to these nonextended entities and their perceptions and appetitions, it is important to Leibniz that they be structured into complexes with dominant and subordinate monads, which structures continue to infinity. But granting those complexities, insofar as everything, for Leibniz, reduces to monads, and monads are in a broad sense spiritual substances, then Leibniz would seem to be an idealist. In recent years, though, this claim has become controversial. The disagreement is not about texts such as the Monadology or the great correspondences of his later years with De Volder and Des Bosses. But some scholars have raised serious questions about Leibniz’s views in the 1680s and 1690s, what have come to be called the ‘‘middle years.’’ Scholars have suggested that during these years, the foundation of Leibniz’s metaphysics was not the monad, or simple substance, but the corporeal substance, matter united with form, understood on the model of the animal. The suggestion is that in the middle years, Leibniz was not an idealist, and that he may have recognized something in the world outside of simple and nonextended substances, something outside of mind, form, or what he later came to call monads.2 95

96 Leibniz: Nature and Freedom

The debate continues. While I used to be among those arguing against Leibniz’s idealism, I no longer think that this is entirely correct. At the same time, I also don’t think that Leibniz was an idealist during this period either. I think that there is a third position that has been overlooked in the debate. What I shall attempt to argue is that in the middle period, Leibniz simply did not hold a position on the question of idealism. Neither an idealist nor an antiidealist, he was, well, just a Leibnizian. I shall begin by trying to articulate what Leibniz’s project was concerning the nature of body in the middle years. Then I shall turn to some passages that, I think, have misled scholars into thinking that Leibniz was more sympathetic to idealism in his middle years than he really was. For Leibniz, of course, substances do not interact with one another directly. Leibniz argues that the apparent causal relations among substances must be understood in terms of the clarity or confusion of their expression. I will try to show that these passages, suggestive as they are of his later idealism, should be understood quite differently, in a way that doesn’t require the idealistic interpretation that they are usually given. I shall conclude with some reflections on the real nature of Leibniz’s project during these years, and why I think that the issue of idealism is simply not Leibniz’s worry in this period.

Substance in the Middle Years Let me begin with a crude outline of the position that Leibniz seems to have taken in the 1680s, at least. While there are lots of texts one might cite here, let me begin with some familiar ones from the correspondence with Arnauld. As we go along, I will add some less familiar texts, and some texts that cloud the waters a bit and suggest more complex views. Let us begin with the following text, from the letter to Arnauld, April 30, 1687: I believe that where there are only entities through aggregation, there will not even be real entities; for every entity through aggregation presupposes entities endowed with a true unity. . . . I do not grant that ‘‘there are only aggregates of substances,’’ and if there are aggregates of substances, there must also be genuine substances from which all the aggregates result. One must therefore necessarily arrive either at mathematical points from which certain authors make up extension, or at Epicurus’s and M. Cordemoy’s atoms (which you, like me, dismiss), or else one must acknowledge that no reality can be found in bodies, or finally one must recognize certain substances in them that possess a true unity. (G 2:96)3

But what are these substances like? What is it that makes something a true unity for Leibniz? Leibniz writes in a letter to Arnauld, November 28/December 6, 1686: [E]very part of matter is in fact divided into other parts as different one from another as the two diamonds [of the Grand Duke and the Grand Mogul]; and since it continues endlessly in this way, one will never arrive at a thing of which it may be said: ‘Here really is an entity,’ except when one finds animate machines whose soul or substantial form creates substantial unity independent of the external union of contiguity. And if there are none, it follows that apart from man there is apparently nothing substantial in the visible world. (G 2:77)

Leibniz and Idealism

97

In this way, the human being serves as a model for the genuine unities that are supposed to ground the reality of the world. As Leibniz writes in a letter to Arnauld, October 9, 1687: [M]an . . . is an entity endowed with a genuine unity conferred on him by his soul, notwithstanding the fact that the mass of his body is divided into organs, vessels, humors, spirits, and that the parts are undoubtedly full of an infinite number of other corporeal substances endowed with their own forms. (G 2:120)4

In this way, the corporeal substances whose unity grounds the reality of body are themselves made up of other, smaller corporeal substances, also unities of substantial form and matter, and so to infinity. Thus Leibniz writes in the same letter: I am very far removed from the opinion which states that animate bodies are only a small part of the others. For I believe rather that everything is full of animate bodies, and to my mind there are incomparably more souls than there are atoms for M. Cordemoy, who makes a finite number of them, whereas I maintain that the number of souls or at least forms is quite infinite, and that since matter is endlessly divisible, one cannot fix on a part so small that there are no animate bodies within, or at least bodies that have forms [informe´s], that is, with corporeal substances. (G 2:118)5

Leibniz’s view here is very nicely summed up in a note he wrote in connection with a conversation he had with Michelangelo Fardella in March 1690, during his trip to Italy: A body is not a substance but an aggregate of substances, since it is always further divisible, and any given part always has another part, to infinity. Hence, it is contradictory to hold that a body is a single substance, since it necessarily contains in itself an infinite multitude, or an infinity of bodies, each of which, in turn, contains an infinite number of substances. Therefore, over and above a body or bodies, there must be substances, to which true unity belongs. . . . (A 6.4:1668–1669/ AG 103)6

Fardella, too, wondered about whether Leibniz had idealist leanings. In his comment on this passage, he wrote: When dealing with a multitude of stones ABC, either stone A or B or C must be understood first. But it is not the same with a soul which, with other souls, does not constitute body. (A 6.4:1670/AG 104)

Leibniz replied: I do not say that the body is composed of souls, nor that body is constituted by an aggregate of souls, but that it is constituted by an aggregate of substances. Moreover, the soul, properly and accurately speaking, is not a substance, but a substantial form, or the primitive form existing in substances, the first act, the first active faculty. Moreover, the force of the argument consists in this, that body is not a substance, but substances or an aggregate of substances. Therefore either there is no substance, and therefore there are no substances, or, there is something other than body. (A 6.4:1670–1671/AG 105)

98 Leibniz: Nature and Freedom

This is, as I have said, a rather crude first approximation of Leibniz’s view in the middle period, and we will have to come back to it again for some additional details and nuances.7 But there is one feature of it that I would like to emphasize for the moment. As Leibniz understands them, the substances that ground the reality of the world are complex entities, entities that contain other substances and so to infinity. In this way, there are apparently no ultimate elements in the analysis of bodies: any level one reaches, however tiny the corporeal substances we may find at that level, contains levels tinier still, which also contain corporeal substances, and so on to infinity. In this way it is very significant that Leibniz’s term of art in this period is almost always ‘‘individual substance’’ and not ‘‘simple substance.’’ The classic definition of a simple substance is given at the beginning of the Monadology: The monad, which we shall discuss here, is nothing but a simple substance that enters into composites—simple, that is, without parts. (Mon 1)

Simple substances have no parts, and because they have no parts, they are unextended and must be understood on analogy with Cartesian souls. They are the ultimate constituents of the created world: ‘‘These monads are the true atoms of nature and, in brief, the elements of things’’ (Mon 3). But the individual substances of the middle period are not simple in this way. Alexander the Great, to take the example of an individual substance Leibniz gives in the classic passage, Discourse on Metaphysics 8, is a corporeal substance, one that contains other corporeal substances as constituents. Leibniz does, to be sure, worry a great deal about whether these constituents, the smaller corporeal substances that make up the body of Alexander, for example, constitute genuine parts or not. But however that comes out, Alexanderthe-corporeal-substance is not simple in the way a monad will later be. Indeed, even the term ‘simple substance’ doesn’t really enter Leibniz’s philosophical vocabulary until somewhat later. According to the index of A, series 6, volume 4, it appears only three times in the philosophical writings of the 1680s as a designation for the soul or substantial form of a corporeal substance, though in none of these passages does Leibniz claim that the simple substance is the ultimate constituent of reality. This is as opposed to roughly 130 appearances in the seven volumes of the Gerhardt edition.8 Though a very few of those appearances are in the mid1690s, simple substances don’t seem to replace individual substances as Leibniz’s central metaphysical entities, I would claim, until the beginning of the 1700s, perhaps 1702 or 1703. A survey of the texts very recently published in A, series 6, volume 4, makes it quite clear that Leibniz’s world in the 1680s is a world of animate substances in animate substances, all the way to infinity. This, by itself, doesn’t altogether exclude the possibility of idealism. But it is significantly different in inspiration from the kind of monadological idealism announced at the beginning of the Monadology, where Leibniz begins with simple spiritual elements from which all reality is built.

Matter and Confusion So far in my account there seems little motivation to think Leibniz an idealist in these texts, except for the fact that he was to become one later. But there are texts

Leibniz and Idealism

99

from the period that have led people quite reasonably to the view that his middle years are not so different from the mature writings where monads dominate. It is time to turn to them, and figure out what exactly they tell us about his views. Let us begin with a complexity in Leibniz’s account of corporeal substance that I ignored in the simplified account I gave earlier. Consider a passage from On the Method of Distinguishing Real from Imaginary Phenomena, an essay written between summer 1683 and winter 1685–86: Concerning bodies, I can prove that not only light, heat, color and such qualities are apparent, but also motion, shape, and extension. And if something is real, it is only the force of acting and being acted upon [vim agendi et patiendi]. And therefore the substance of body consists in this (as matter and form). Moreover, those bodies which don’t have substantial form are only phenomena, or, at least, only aggregates of true [substances]. (A 6.4:1604)9

This is a familiar theme from Leibniz’s writings in this period, that what is real in bodies is not extension, but force. In this way, it is strictly speaking wrong to see Leibniz as simply adding substantial forms to Cartesian bodies. Strictly speaking, he wants to replace a characterization of the world in terms of extension and its modes, characteristic of the Cartesian (and more generally, the mechanist) worldview, with another, deeper characterization in terms of form and matter interpreted as active and passive force. (This obviously connects with the science of dynamics, the science of force, that occupied Leibniz’s work in physics during these years. But that is another subject for another essay.) This, by itself, seems relatively unproblematic on the corporeal substance view. A familiar Leibnizian theme is the fact that extension isn’t all it’s cracked up to be, at least not all it’s cracked up to be by the Cartesians. Leibniz often complains that if bodies were just extended, as the Cartesians claim, then there would be nothing substantial in body, and that bodies would have the same status as dreams.10 Sometimes, as in some of the quotations I gave earlier, he seems to suggest that the remedy to this problem with the Cartesian view is simply the addition of form. It is only a small step further to suggest the view that extension is, itself, a derivative notion, one that must be grounded in force, passive force in particular. This is the view that is suggested in the following passage from an essay written between early 1684 and winter 1685–86: Matter is the principle of being acted on [principium passionis], that is, the primitive force of resisting, which is commonly called bulk or antitypy, from which flows the impenetrability of body. The substantial form is the principle of action or the primitive force of acting. Furthermore, there is in every substantial form a certain knowledge [cognitio] that is an expression or representation of external things in a certain individual thing, in accordance with which a body is per se one, namely in the substantial form itself. This representation is joined with a reaction or conatus or appetite which follows this thought of acting. This substantial form must be found in all corporeal substances which are per se one. (A 6.4:1507–1508)

Here it seems that in reducing corporeal substance to active and passive force, he isn’t exactly eliminating extension, but merely reducing it to what is, in his view,

100 Leibniz: Nature and Freedom

more fundamental, namely the passive force by which a body excludes another body. (Note here also that in this passage, Leibniz is careful to say that it is substantial form, the principle of action that expresses or represents external things, and not the matter or primitive force of resisting. I will come back to that.) But all is not quite that simple. The passage from On the Method of Distinguishing Real from Imaginary Phenomena I quoted earlier continues as follows: Substances have metaphysical matter or passive power to the extent that [quatenus] they express something confusedly, and active power to the extent that they express something distinctly. (A 6.4:1504)11

This passage suggests that active and passive power are to be understood in terms of distinct and confused expression. This, in turn, would seem to suggest that matter is to be understood as a feature of the perception of an entity that is capable of mental states. That is, that activity and passivity are themselves features of mentality. That is, the individual substance composed of form and matter is itself mindlike. Or, at least, so this passage suggests on a very natural reading. This passage is rather unusual for the period of the 1680s; it is the only one of its kind that I know of. However, it does resonate with another feature of Leibniz’s doctrine in the 1680s that is much more widespread, that is, Leibniz’s account of causal relations. The doctrine is nicely summarized in the Specimen of Discoveries (1688?): That whose expression is the more distinct is judged to act, and that whose expression is the more confused is judged to be passive, since to act is a perfection and to be passive is an imperfection. Again, that thing is thought to be a cause from whose state a reason for changes is most easily given. (A 6.4:1620/MP 79)

It is set out at greater length in section 15 of the Discourse on Metaphysics (1686): [W]e ascribe to ourselves—and with reason—the phenomena that we express most perfectly and we attribute to other substances the phenomena that each expresses best. Thus a substance, which is of infinite extension insofar as it expresses everything, becomes limited in proportion to its more or less perfect manner of expression. This, then, is how one can conceive that substances impede or limit each other, and consequently one can say that, in this sense, they act upon one another and are required, so to speak, to accommodate themselves to one another. For it can happen that a change that increases the expression of one diminishes that of another. Now, the efficacy [vertu] a particular substance has is to express well the glory of God, and it is by doing this that it is less limited. And whenever something exercises its efficacy or power, that is, when it acts, it improves and extends itself insofar as it acts. Therefore, when a change takes place by which several substances are affected (in fact every change affects all of them), I believe one may say that the substance which immediately passes to a greater degree of perfection or to a more perfect expression exercises its power and acts, and the substance which passes to a lesser degree shows its weakness and is acted upon [patit]. (A 6.4:1553–1554/AG 48)

Or, as Leibniz summarizes his account in a passage that appears both in a letter to Simon Foucher in 1686 and in Leibniz’s letter to Arnauld, July 4/14, 1686:

Leibniz and Idealism

101

One [substance] acts on another, since one expresses more distinctly than the other the cause or reason of the changes, a bit like we attribute motion to a boat rather than to the entire sea, and correctly so. (G 1:383; G 2:57)

From these passages, it would seem as if causal relations in the world of bodies for Leibniz are simply a matter of the distinctness or confusion of the expression of the substances in question. The temptation is strong to say that all there is in the corporeal substance is the mental, and that the passive and the active, the two constituents of the corporeal substance, are nothing but aspects of a single mindlike entity. But it would be wrong to give in to this temptation. Read in the context of Leibniz’s larger view here, a somewhat different and rather more complex picture emerges. This is what I would like to argue in some detail. Let me begin by noting that the term ‘expression’ doesn’t always involve the mentalistic notion of a perception for Leibniz, nor does the thesis about the lack of metaphysical causality always involve the distinctness or confusion of mind-like entities alone. In his letter to Arnauld of October 9, 1687, Leibniz writes: One thing expresses another (in my terminology) when there exists a constant and fixed relationship between what can be said of one and of the other. This is the way that a perspectival projection expresses its ground-plan. Expression is common to all forms, and it is a genus of which natural perception, animal sensation and intellectual knowledge are species. (G 2:112)

Leibniz goes on to explain how it is that in this precise sense, every Cartesian body expresses every other Cartesian body in the world of extended things: Now, this expression occurs everywhere, because every substance is in harmony with every other and undergoes some proportionate change which corresponds to the smallest change occurring in the whole universe. . . . I believe that M. Descartes himself would have agreed with this, for he would undoubtedly grant that because of the continuity and divisibility of all matter the smallest movement extends its effect over neighboring bodies and consequently from neighbour to neighbour ad infinitum, but proportionally decreased; thus our body must be affected in a way by the changes of all the others. (G 2:112)

Because there are no vacua, empty places in the Cartesian world, every event is transmitted to the entire rest of the world. In this way, even Cartesian bodies, without souls, without perceptions, can be thought of as expressing the entire world in which they exist. To the extent to which the effects are ‘‘proportionally decreased’’ as the distances become greater, perhaps we can talk here about more or less confused expression in Cartesian bodies (though Leibniz does not, at least not explicitly). The talk about mutual expression in the corporeal world does not, by itself, get us causal relations, though. While every body expresses every other body in the world in which it exists, and while we can get some handle on how we can talk about distinct and confused expression among extended things, it is difficult to see how exactly to understand Leibniz’s thesis that links distinct expression to cause and confused expression to effect with any concrete cases of body-body interaction.

102 Leibniz: Nature and Freedom

Now, Leibniz is quite comfortable about extending the thesis of the noncommunication of substances to the world of physics. For example, in the celebrated First Truths paper, now dated to 1689, Leibniz writes: Strictly speaking, one can say that no created substance exerts a metaphysical action or influx on any other thing. . . . What we call causes are only concurrent requisites, in metaphysical rigor. This is also illustrated by our experience of nature. For bodies really rebound from others through the force of their own elasticity, and not through the force of other things, even if another body is required in order for the elasticity (which arises from something intrinsic to the body itself) to be able to act. (A 6.4:1647/AG 33)

But though it is easy enough to understand how the noncommunication of substances doctrine can hold in the world of bodies, it is more difficult to understand the other part of his doctrine of causal relations. Though nothing exerts a metaphysical action on anything else, strictly speaking, certainly Leibniz wants to say that one body is the cause of the motion, and another the effect. But how is it that distinct expression is linked to cause, and confused expression to effect in bodies? Here, I think, Leibniz does slide into understanding the notion of expression in a mentalistic way. How are we to understand this strand of his thought? And does it force us to admit that during these years, Leibniz was, indeed, pressing an idealistic metaphysics? Discourse on Metaphysics 15, for example, seems to force us toward an idealistic reading of Leibniz’s thought at this time. But let me put that passage into its larger context. In an earlier draft of the Discourse on Metaphysics, the passage that was to become section 15 begins in a different way. In a passage found at the very end of section 14, crossed out and replaced by Discourse on Metaphysics 15, Leibniz gives his account of causal relations as follows: It is certain, above all, that when we desire some phenomenon, and that phenomenon happens at a given moment, and that happens regularly, we say that we have acted and that we are its cause, as when I will that which one calls moving my hand. Also, when it appears to me that at my will, something happens to what I call another substance, and that this happens by that means (as I judge by frequent experiences), even though the other substance didn’t will it, I judge that this substance is acted on [patit], as I admit of myself, when this happens to me following the will of another substance. . . . One can say briefly that with respect to the mind, our volitions and our judgments are actions, but that our perceptions or sensations are passions, and with respect to the body, we say that the change which happens is an action when it is the consequence of a preceding change, but otherwise it is a passion. In general, to give our terms a sense which reconciles metaphysics with practice, when several substances are affected by a single change (since, in fact, every change touches them all) one can say that that which through this passes immediately to a greater degree of perfection or continues in the same degree of perfection acts, but that which becomes by that immediately more limited, so that its expressions become more confused, is acted upon [patit]. (A 6.4:1551–1552)

It is a bit troubling, perhaps, that Leibniz deleted this passage; my own suspicion is that he decided that he had gone into more detail than he wanted to in this

Leibniz and Idealism

103

context, and chose, instead, the rather more economical (and frankly enigmatic) account of action and passion that he gave in the text of section 15, which begins with a later draft of the phrases that end this rejected fragment (‘‘But, without entering into a long discussion, in order to reconcile the language of metaphysics with practice . . .’’). In any case, this passage certainly does give us some insight into how Leibniz was thinking about causal relations. I want to begin unpacking it by looking at what Leibniz says here about action. The example Leibniz gives of an action is my moving my hand: when I have a volition, and the motion of my hand follows in a regular way, that constitutes an action. Clearly we are talking about a corporeal substance with a body here, a human body, my body as I perform a voluntary action. It is the feeling of acting, the feeling of having a volition that makes me the cause in this situation. This example is quite consistent with what Leibniz wrote a few years earlier in a short essay about motion, De Causa Motus, dated by the Akademie edition at summer 1678–winter 1680/1(?). In that essay Leibniz is considering the problem of the real subject of motion: when two bodies are in motion with respect to one another, which is really in motion and which is really at rest? There he wants to argue that real motion is, in a sense, independent of change of place! In the essay he gives an example in which a person is walking along a deck of a ship, while, at the same time, that same ship is moving with the same speed that the person is moving, but in the opposite direction. The person is at rest with respect to the shore, yet, at the same time, he feels the effort of walking, and thus is moving, says Leibniz. However, we should say that he is in motion: for, at any rate, that man will feel that he is becoming tired by walking, and he will be able to push or pull something by means of his motion. And thus, he acts. Therefore, since we attribute motion to that in which the cause of change of situation is found, that is [seu], to that which acts, we will, at any rate, say that the body e moves, even if the place it has in the world, and which is taken with respect to fixed points does not change. (A 6.4:2019)

Again, it is clear that we are talking about a corporeal substance, a soul and a body. It is the volition in the soul that makes the person the actor in this situation, but it is volition in an embodied soul. Leibniz’s examples only deal with human agents, but it is not hard to extend this account to other agents. Since, Leibniz claims, bodies are aggregates of corporeal substances, and all corporeal substances are thought of as animate beings with forms (souls), on analogy with human beings, they too could be conceived of as having something analogous to a volition when they act. Returning to the passage I quoted from the early draft of the Discourse on Metaphysics, I think that it is fair to infer that it is having this efficacious volition, the feeling of effort followed by a motion of the body, that constitutes ‘‘passing to a greater degree of perfection’’ for Leibniz. In that passage Leibniz doesn’t talk explicitly about distinct expression in this connection, but from the fact that he does talk about confused expression in connection with being acted upon, it is not unreasonable to infer that having an efficacious volition is what he takes to be a state of distinct expression. This is quite consistent with the way Leibniz expresses

104 Leibniz: Nature and Freedom

himself in the passage that appears in the contemporaneous letters to Foucher and to Arnauld I quoted earlier, where he says that the substance that acts ‘‘expresses more distinctly than the other the cause or reason of the changes.’’ Here we might understand the having of an efficacious volition as itself the distinct expression of the cause of the change. Leibniz seems to have something similar in mind in a passage from a note he wrote for a letter to Arnauld in November/December 1686: However, there is good reason to say that my volition is the cause of the motion of my arm, and that a dissolution of the continuity of the matter in my body is the cause of my pain, since the one expresses distinctly what the other expresses more confusedly, and one should attribute action to the substance whose expression is the more distinct. (G 2:71)

In the case of action, Leibniz’s conception of causality in terms of distinct expression and increase in perfection may require a soul (or something like a soul), capable of a volition (or something like a volition). But the examples he gives suggest that he has in mind here an embodied soul, a soul whose volitions are the causes of bodily motions. In this way there is nothing in Leibniz’s account of activity that pushes us toward idealism. But the account of matter, passive force, and passivity is somewhat more complicated. In what sense, then, is confused expression connected with passivity and being acted upon? Consider the case where I am pricked in the finger by a pin. When my body is acted upon we are dealing with a case in which someone or something else has the ‘‘volition’’ and I am not the cause of the state of my body. Here Leibniz often talks explicitly about a ‘‘confused expression.’’ But of what is it a confused expression? His position seems to be that I have a ‘‘confused expression’’ of the cause, to use the language of the Foucher/Arnauld passage, and that that confused expression is itself the sensation of pain. This is what he argues in a letter to Arnauld, October 9, 1687, where he is discussing what happens in the soul when a pin pricks us. He writes: Now, states of the soul are naturally and essentially expressions of the corresponding states of the world, and particularly of the bodies which then belong to them; therefore, since the prick forms part of the state of the body at moment B [i.e., following the pin prick], the representation or espression of the prick, which is pain, will also form part of the state of the soul at moment B. . . . (G 2:114; emphasis added)

Remember also the passage I quoted earlier from 1684–1685/6, where Leibniz implies that in the corporeal substance, it is the substantial form that expresses the world: ‘‘there is in every substantial form a certain knowledge [cognitio] that is an expression or representation of external things in a certain individual thing’’ (A 6.4:1508; emphasis added). When we are acted upon, this particular kind of confused expression, pain in the case of the pin prick, is the mark in the soul of our passivity, the passivity that we possess as corporeal substances. Understood in this way, Leibniz’s account of being acted upon in terms of confused expression does not seem to require any strong idealistic metaphysics, any more than his account of activity in terms of distinct expression did. Having pain

Leibniz and Idealism

105

would seem essentially to involve the body and its states. While one could perhaps give a purely idealist account of what is going on here, Leibniz himself doesn’t: it is the human being who is being acted upon, and the sensation is the mental expression. All the doctrine of causality requires is that bodies have souls and be capable of perceptions as well as actions; it does not require that everything be soul. Let us now return to that puzzling passage that started us out on these reflections: Substances have metaphysical matter or passive power to the extent that [quatenus] they express something confusedly, and active power to the extent that they express something distinctly. (A 6.4:1504)12

This can be read as an expression of idealism: for a substance to have matter is simply for it to have confused perceptions. But it need not be read in this way. Confused expression (perception) can be identified with matter. But confused expression could also be a consequence of the embodiment of a soul in an organic body; this passage could be read as saying not that confused expression is what constitutes matter, but that a substance or its form has confused expression because it has matter. Leibniz’s later monadological writings suggest the former more idealistic reading, but the examples he uses contemporaneously with this brief but pregnant passage suggest the latter. The pain I feel when I am pricked by a pin, when I am acted upon, is a confused expression in my mind, my soul, my substantial form, but in the texts that I have been examining, Leibniz makes no suggestion that this is an event that does or could happen without the body. It is the corporeal substance, soul and body that is acted upon, that suffers, but it is the mind that feels the pain. Insofar as a mind has confused expression, it must have matter: the confused expression, on this reading, is the mark of the embodiment of the soul. I don’t deny that what Leibniz wrote in the 1680s can be taken out of context and made consistent with what he wrote ten or fifteen or twenty years later in his more unabashedly monadological writings. But these passages can also be made consistent with what he wrote in writings of the same period, and it seems much more sensible to do that than to try to argue that these few passages are, in some way, glimpses of a hidden doctrine or prefigurings of what he will later think.

Leibniz Idealist? What conclusions can be drawn from my reading of these difficult texts? It has been striking to me that even though one can interpret Leibniz’s writings in the 1680s as consistent with his later idealism, there is nothing among the texts that forces us to read them in that way, no place where Leibniz explicitly says that ‘‘there is nothing in things but simple substances, and in them, perception and appetite,’’ as he will write to De Volder some years later. This led me in the past to the view that Leibniz was not an idealist during those years. My assumption was that Leibniz was some sort of realist during the 1680s and 1690s, and that he recognized something, some kind of matter, distinct from soul, mind, or substantial form.

106 Leibniz: Nature and Freedom

But it is just as difficult to find a clear and unambiguous statement of this kind of realism in these years as well. Leibniz does say to Fardella: ‘‘I do not say that the body is composed of souls, nor that body is constituted by an aggregate of souls, but that it is constituted by an aggregate of substances.’’ This seems to exclude a strongly idealistic reading of Leibniz’s views in this period.13 But, at the same time, what Leibniz says about matter in the technical sense in which it is contrasted with form or soul is not very clear, and does leave the reader wondering what exactly he had in mind. Perhaps we are misunderstanding something here. Perhaps the issue of idealism and its contraries simply was not on the horizon for Leibniz at this moment in his career. While he would later argue explicitly for idealism, at this moment, I suggest, his philosophical interests lay elsewhere. What were those interests? In the 1680s (and the 1690s too), I suggest, Leibniz’s interest was in refuting the broadly Cartesian account of body, and replacing it with his own. What he opposed was the idea that body is made up of extended stuff alone, passive, infinitely and arbitrarily divisible, continuous, and that there is nothing else in body. The replacement he proposed was that body is composed of substances, corporeal substances, matter and form: ‘‘the force of the argument consists in this, that body is not a substance, but substances or an aggregate of substances,’’ as he wrote in the Fardella memo. Sometimes, as in the Arnauld letters, he emphasized that these substances are thought of as animate creatures, little animals, like the ones that the microscopists were discovering every day. Sometimes he emphasized that unlike the Cartesian world, there is life everywhere in nature, and that there are an infinity of souls in the smallest bit of matter. Sometimes he explicitly related his view to that of the schoolmen, and argued that he was reviving the substantial forms of the Scholastic philosophy or even the distinction between natural and artificial basic to Aristotelian physics. Sometimes he took the analysis deeper, and argued that form and matter had to be understood in terms of active and passive forces, the forces that were central to the science of dynamics that he was in the process of inventing during these decades. This was the opposition that concerned him during these years, the opposition between the Cartesian world of extended substance and inert bulk and his own living world of soul and body, form and matter, active and passive force. As to the question of whether it’s all mind or soul, or whether there is something in the world that goes beyond, I don’t want to deny that he may have raised the question with himself from time to time. But I do want to argue that it was not his main concern, and that he did not have a settled position on the matter. I see no evidence that he subjected the notion of corporeal substance to the level of analysis that would have allowed him to address that question. It simply was not part of his program. Or, at least, so I would claim.

Notes Earlier (and shorter) versions of this essay were presented at the Seventh International Leibniz Congress in Berlin in September 2001, and at the Leibniz Society of North America, meeting with the Eastern Division of the APA in Atlanta in December 2001. I would like to thank audiences there for their helpful comments. But above all, I would like to thank Robert Sleigh. Some of my earliest discussions of

Leibniz and Idealism

107

these issues were with him during the year we were together in 1982–83, when he was at the Institute for Advanced Study working on what was to become his classic book, and I was visiting at Princeton University and teaching a seminar on Leibniz. I would also like to thank him more generally for his friendship and his philosophical wit and wisdom through the years. His kindness and enthusiasm for all things Leibnizian has sustained us all. 1. Leibniz to De Volder, January 19, 1706. 2. See Garber 1985. My reading has been opposed by two important books, Sleigh 1990, especially chaps. 5 and 6, and Adams 1994. 3. Translations of the correspondence with Arnauld are taken from LA. Since this translation is keyed to G, I will not reference it separately. 4. I follow here the earlier text as actually received by Arnauld. See LR 187. 5. Again, I follow the text in LR 186. 6. It should be remarked that this is probably Fardella’s rendering of what Leibniz said in conversation. 7. One nuance worth mentioning is Leibniz’s frequent hesitation about whether or not anything in our immediate experience, with the exception of us, indeed, whether or not anything at all but us constitutes genuine substances of this sort in the world. Sleigh (1990: chap. 5) emphasizes these passages to suggest that in this period, Leibniz was more of an idealist than I am inclined to make him. For a discussion of Sleigh’s position here, and my answer, see my review of his book (as well as of Wilson 1989) in Garber 1992b. 8. A search of G shows 133 texts in which the term appears. Of these one is before 1690 (Discourse on Metaphysics 35), which also appears in A, ser. 6, vol. 4; there are two texts in which it appears in the 1690s, a total of four occurrences in 1695–96 in connection with the Syste`me Nouveau (G 4:279; G 4:291–292). Other than that, there are a total of 130 texts in which it appears, the earliest being a letter to De Volder in 1700 (G 2:206). 9. Unless otherwise noted hereafter, the translations are my own. 10. This is the theme of a short piece written by Leibniz somewhere between 1683 and 1685: A 6.4:1466–1467. 11. ‘‘Substantiae habent materiam Metaphysicam seu potentiam passivam quatenus aliquid confuse exprimunt, activam quatenus distincte.’’ 12. See note 11. 13. Pace Adams’s (1994: 275) somewhat contorted reading of this passage. I discuss the text in more detail in my essay ‘‘Leibniz and Fardella: Body, Substance and Idealism,’’ forthcoming in the proceedings of the Florence Leibniz conference of November 2000.

5

Compossibility, Expression, Accommodation CATHERINE WILSON

n Leibniz and Arnauld, Robert Sleigh raised the following question: What is the

I relationship between the notions of compossibility, expression, and accommo-

dation in Leibniz’s writings in the period of the Discourse on Metaphysics and the associated Correspondence with Arnauld? Each of these two-term relations is regarded by Leibniz as obtaining universally, between all substances chosen pairwise. For all actual substances, in other words: 1. x is compossible with y 2. x expresses y 3. x accommodates to y

According to Sleigh, it was for Leibniz a necessary truth that all actual substances are compossible with one another, and the place of this theorem in his system is unproblematic. Clearly, the actual substances of our world cannot be incompossible, so the proposition is trivial. But, Sleigh goes on to say, the claim that all actual substances express one another, though treated by Leibniz as a necessary and even trivial truth, does not seem to be trivial. And the claim that all substances accommodate to one another does not even seem to be a necessary truth, and it is certainly not a trivial truth. It is unclear, he concludes, exactly what Leibniz meant by ‘‘accommodation’’ or, as he sometimes calls it, the ‘‘harmony’’ of two or more things, above and beyond their ‘‘expressing’’ one another.1 Even though Leibniz claims that universal harmony is a consequence of his theory of substance, it cannot be true that compossibility entails expression, or, for that matter, that compossibility or expression entail accommodation. Yet Leibniz sometimes sounds as though, given the nature of substance, accommodation follows. At one point, as Sleigh notes, Leibniz even says that accommodation, or ‘‘concomitance,’’ is a consequence of expression: The hypothesis of concomitance is a consequence of the hypothesis I have of substance. . . . The hypothesis of concomitance or of the accord of substances

108

Compossibility, Expression, Accommodation

109

among themselves follows from what I said that each individual substance envelops forever all the accidents that will occur to it, and expresses the entire universe in its own manner. (draft of a letter to Arnauld, December 8, 1686; LA 68–70, as quoted in Sleigh 1990: 170)

And in a passage quoted later, Leibniz seems to identify expression with a state of harmonious accommodation amongst substances. If the phenomena of different substances failed to harmonize, ‘‘there would be as many systems as substances’’ (Sleigh 1990: 178). But Sleigh denies that Leibniz could really have meant that concomitance was a logical consequence of universal expression, and if he did not think it to be a logical consequence of universal expression, Leibniz could not have thought it a logical consequence of compossibility either. And Sleigh calls Jacques Jalabert to task for the suggestion that ‘‘universal harmony’’ follows simply from God’s decision what to create; Sleigh argues that this is tantamount to accepting the claim that compossibility entails harmony (1990: 172).2 For his part, though, Jalabert was surely right to suggest that Leibniz wished to present experienced mirabilia such as the harmony of mind-body relations as consequences that can be derived from or be seen to flow from a consideration of the nature of substances—what they must be. Often, it is the intention of a metaphysician to derive, from a few apparently trivial, self-evident, or otherwise unquestionable truths, some significant, value-laden, action-orienting conclusions about our situation. (Consider Spinoza’s procedure in the Ethics, for example, in deriving a whole chain of specific ethical propositions from some seemingly self-evident truths about substance.) Acknowledging the problem of explaining the nature of the entailments between compossibility, expression, and accommodation, Sleigh expresses understandable frustration. ‘‘I have wrestled with these matters for some time and I have lost,’’ Sleigh writes. ‘‘My hope is that what follows will motivate someone to step forward and take up the challenge’’ (1990: 171). The challenge is irresistible. Compossibility, to begin with, is an all-or-nothing relation: two substances are either compossible or incompossible; they can’t be ‘‘barely compossible’’ or ‘‘extremely compossible.’’ But both expression and accommodation are relations admitting of more or less. Further, expression has a cognitive dimension that compossibility lacks—if A expresses B, it seems that we can learn something about B from examining A—and accommodation has a further aesthetic dimension that expression lacks. If A and B are harmonious, a spectator can derive pleasure from observing A and B. The instances of harmony and accommodation that are to be found in our world range from the existence of simple and elegant laws of nature, to the marvelous contrivance of the bodies of animals and the adaptations of creatures of all sizes to their various environments, to laws of the union of soul and body, and to the proportionality of ultimate rewards and punishments to deeds (Rutherford 1995a: especially 199). Further, the harmony of substances is, in Leibniz’s opinion, an excellence-making feature of our world, and reflection on it is supposed to be ethically motivating. The mere compossibility of all the substances within it cannot immediately be construed as an excellence-making feature. So, in ‘‘deriving’’ harmony from compossibility, there must be a process of what might be called ‘‘value-addition’’ that the commentator can bring to light,

110 Leibniz: Nature and Freedom

that the reader is disposed to accept as nonarbitrary. How is this value added in such a way that it can seem to a commentator like Jalabert that the intrinsic implications of an idea have merely been ‘‘drawn out,’’ and that nothing has in fact been added? The transition from compossibility to accommodation is mediated, as Sleigh suggests, by the intermediate notion of expression, which already ‘‘adds’’ some value to compossibility. But harmony, concomitance, or accommodation of things is more than the mutual expression of substances. Sleigh’s intuition that we have three distinct notions in play3 suggests that the metaphysical situation (not in Leibniz, but in reality) is as follows: Accommodation —? Expression —? Compossibility Compossibility —/? ? Expression —/? Accommodation

One of Jalabert’s points (I take it that Sleigh will go along with this) can be conceded: in our optimal world, all substances that are compossible also express each other and accommodate to each other. But if the metaphysical situation is as represented, there are some possible worlds, that is, worlds of compossible substances, in which there is a failure of one or both of expression or accommodation. The escalation, then, might be justified by Leibniz’s optimality-requirement. In that case, it will be possible to describe a world of compossible, but non–universally expressive, substances, and also to see what is inferior about that world with respect to a world in which expression holds. And it will be possible to describe a world of universally expressive, but nonaccommodating, substances, and also to see what is inferior about that world with respect to one in which both expression and accommodation hold. Indeed, worlds with compossible but non–universally expressive substances are possible, but are not even possible Leibniz-worlds; possible Leibniz-worlds must at least contain universally expressive substances, even if the substances are nonaccommodating. Now, the first claim implies that Leibniz-worlds are a subclass of possible worlds. Of course, anyone who believes that Leibniz’s conception of substance and his theory of space and time constitute a set of logically necessary truths will contest this and maintain that all possible worlds are in fact possible Leibniz-worlds. But we are not forced, as yet, to that admission. It is prima facie coherent to maintain that some possible worlds are composed of Leibnizian substances, others of material atoms, others of a single Spinozistic substance, others of two substances as reported by Descartes, others of Aristotelian matter-form amalgams, and so on, insofar as none of these ontologies has been demonstrated to be self-contradictory. What would a world of compossible substances that did not express one another look like? This condition is easily imagined. A world composed of material atoms would instantiate it, as just noted, and so would many Descartes-worlds. It does seem, though, that in all Leibniz-worlds, compossible substances do express one another. For, if compossible substances did not express each other, then, since substances have no intrinsic spatiotemporal relations, there would be no reason to say that they existed in the same world.

Compossibility, Expression, Accommodation

111

This leaves the second question: what would a world of universally expressive substances without accommodation look like? Sleigh believes that we can provide a kind of model for expression-without-accommodation. And he employs one of Leibniz’s own images, the image of the individually sequestered orchestra players, to do so. Leibniz says: This concomitance [accommodation] I uphold is like many different bands of musicians or choirs, playing their parts separately, and so placed that they cannot see or hear each other but can nevertheless harmonize, by each one following his own notes. (Leibniz to Arnauld, April 30, 1687; LA 95, as quoted in Sleigh 1990: 177)

Sleigh argues that the situation described models expression that is, in this case, contingently associated with harmony: Whether the separate bands succeeded in producing a euphony or a cacophony, there would be a certain ‘constant and fixed relation’ between the two based on certain ‘inner states’ of the two groups—the notes and silences produced at a given time. So one would express the other, whatever strange sounds were forthcoming, that is whether they succeeded in harmonizing [accommodating] or not. (Sleigh 1990: 177)

Thus, a situation can be described in which there is expression without accommodation, and something over and above the former is required for the latter. The pattern can fit substances as well. On Sleigh’s interpretation, as far as I understand it, the analogy works as follows: Notes played by Musician A express notes played by Musician B and Musician C and . . . : The experiences of Substance A express bodily events b and c and. . . . Notes played by Musician A harmonize with notes played by Musician B and Musician C and . . . : The experiences of Substance A harmonize with bodily events b and c and . . .

And Sleigh’s point, supporting his difference of meaning thesis, is that expression in both cases stops short of what is required for harmonization. If a harmony and not a cacophony is produced for a being-who-hears-through-walls, then Musicians A, B, C, and D must all express each other’s scores. But the converse does not hold. The sequence of notes and pauses played by Musician A can express whatever is played by Musicians B and C without the being-who-hears-through-walls hearing a harmony. For, even if they produce a cacophony, we can write out a table of correspondences from the notes played. The difficulty with this interpretation is that the players do not seem to express one another; rather, they express a common score that exists in multiple copies. If they did express one another, it is hard to see how they could produce a cacophony, and how they could in that case fail to accommodate. We can write out a table of correspondences that documents what note each is playing at a given moment. But merely being able to write out such a table of correspondences does not imply that the musicians are expressing each other’s scores, not just their own scores. Otherwise, any sequence could express any other sequence, and this is not what Leibniz has in mind. Some correlations will fall short of the standard required

112

Leibniz: Nature and Freedom

even of relatively poor expression. Suppose A is playing Chopin and B is simultaneously playing Beethoven. We could write out a correspondence table for each temporal interval, but it is just false that the scores express each other. Expression is a value-adding relation, relative to mere co-occurrence. It allows for degrees of perfection with reference to a standard of perfect expression, the type of expression that we can image most adequately by thinking of the relation between an object and its image in a flawless mirror. An equation expresses a curve, a color-experience expresses certain ‘‘shapes and motions’’ pertaining to the underlying physical state of the sensory organs (RB 403). And, when I perceive a jar of ink on my desk, it seems that my soul expresses the ink jar, and perhaps expresses God as well, or even instead, since, according to the Discourse, ‘‘although all substances express the whole universe, the other substances express the world rather than God, while minds express God rather than the world’’ (DM 36/AG 67).4 So the problem of differentiating expression from accommodation remains. The claim that expression is value-adding might seem problematic in light of the following well-known passage from the Discourse: [L]et us assume . . . that someone jots down a number of points at random on a piece of paper, as do those who practice the ridiculous art of geomancy. I maintain that it is possible to find a geometric line whose motion is constant and uniform, following a certain rule, such that this line passes through all the points in the same order in which the hand jotted them down. (DM 6/AG 39)

Clearly, the geometric equation expresses the physical array of points. And Leibniz infers from the possibility of writing such an equation that ‘‘[n]othing irregular occurs in the world.’’ That may be true. But he also claims: ‘‘We would not even be able to imagine such a thing.’’ This must be false if there are possible worlds in which substances do not express one another. One might infer from these remarks that as long as two sequences are mathematically compatible (we cannot map real numbers into integers or circles into squares), any sequence can be said to express any other sequence. Leibniz appears to claim in the geomancy passage that all phenomena are and must be orderly—that deep disorder is inconceivable because we can always fit some curve to the phenomena. The passage is troublesome, moreover, because it appears to contradict Leibniz’s main point in his criticism of Descartes’ laws of motion. Leibniz complained: ‘‘From the Cartesian rules no continuous line whatever can be derived for the results which correspond to the continuous line representing the variable data; on the contrary, a figure is produced which is most erratic and contrary to our criterion . . . or to the law of continuity’’ (G 4:384/L 403). Descartes’ collision rules translated into a ‘‘monstrous’’ (discontinuous) figure that could not correspond to the truth, represented by an ‘‘orderly’’ (continuous) figure (G 4:382/L 412). According to the geomancy passage, as long as we can write down a law—any kind of law—and we always can, the phenomena can be judged to be orderly, and it is not clear why Descartes’ curve should not be judged orderly in its own way. But perhaps what Leibniz was trying to say through his geomancy example can be put this way: Any empirical phenomenon we observe or could ever expect to observe in our world, no matter how chaotic or random it appears, can be described by rules. There must

Compossibility, Expression, Accommodation

113

be an equation that captures the location of spots on the leopard and stripes on the zebra, and that can be used to predict where a handful of stones dropped from the hand will fall. If the phenomena are complex, the laws describing them need not be simple, but the laws of motion—being somehow fundamental—cannot be complex. And perhaps the rebarbative equation that I write to generate the array of pebbles thrown by the geomancer is merely provisional. I regard the geomancer merely as the perpetrator of a vain artifice. I lay a piece of graph paper over his stones and write a function. But suppose I am a serious student of stone-scattering. I shall deepen my inquiry, and study the dynamics of stones projected from the hand with a certain force and striking the ground a certain distance away. I shall no doubt find that, as with the natural process that distributes spots on the leopard, there must be a beautiful and simple equation. But what is it for substances to express states of each other, if they are windowless and have only internal, perceptual, and volitional states? Because this question is seemingly unanswerable, the temptation to identify expression and accommodation is overwhelming. But it can be resisted if we take Leibnizian expression to be analogous to certain relations that can be described as occurring between two beings who are spatially sequestered as they are in Sleigh’s example, and who do not touch each other or touch any common objects. Leibnizian accommodation can then be taken to be analogous to certain relations that can be described between two beings who are in a common space and who touch each other and/or touch certain common objects. The ‘‘walls’’ separating the musicians signify the absence of a common spatiotemporal world, though, were musicians in our world to be separated by soundproof walls, each musician would still touch his floor, which would touch the walls of his chamber, which would touch the floor of the musician in the adjoining room, and so on, so that the sequestered musicians would be in mediated physical contact. To conceive the situation modeled, we need to interpret their partial isolation as complete isolation. Here, to illustrate, is a paradigmatic case of expression-without-physical-interaction in which we have to take just the partial as modeling the complete. Two identical twin acrobats are facing each other a few feet apart. When one raises her right arm, the other raises her left. When one takes a step backward with her right foot, the other takes a step backward with her left foot. We observers can see that they are expressing each other’s movements perfectly: call this ‘‘imitative expression.’’ And although physical signals are in fact passing back and forth between them, so that they are, from the physicist’s and the neurophysiologist’s point of view, interacting, we can agree to regard them as noninteracting. We could write a table that reads: A right arm up ¼ B left arm up A right foot backward ¼ B left foot backward But suppose now that when A raises her right arm, B does too. They no longer look like mirror images of one another; still, their expression seems to be imitative. Now suppose that when A raises her right arm, B raises her left foot, but does so one-half

114 Leibniz: Nature and Freedom

second later. When A raises his left arm, B raises his right foot one-half second later. If we were making up a table, we could write: A right arm up ¼ B left foot up þ :5 sec: A left arm up ¼ B right foot up þ :5 sec: This might be called ‘‘semiimitative expression.’’ Now, suppose that the first time A raises her right arm, B sings E-flat, but the second time, B sings F-sharp, the third time, E-flat . . . , and so on. Now we could write out a table that said: A right arm up 1 ¼ B E-flat A right arm up 2 ¼ B F-sharp A right arm up 3 ¼ B E-flat So long as B is conceived as following a rule that specifies that she respond to an action of A in a specific way, we may be quite willing to describe B’s movements as expressive of A’s, no matter how nonimitative they are. But suppose we have to begin at the other end. Someone presents us with a table of A and B correlated actions in which we can discern no pattern and can state no rule. Under what conditions might we be persuaded that B expresses A’s actions? Only if we can recognize A’s and B’s actions as variants of familiar expressive patterns in which it is possible to derive information about B from observing A (copying or mimicking, signaling with flags or gauge readings, translating notes or letters into sounds, employing an enigmatic code) or as (possibly failed) attempts to produce some variant of a familiar expressive pattern, is there any reason to describe the correlation as expression. When two musicians play individual sequences of notes that would produce a cacophony if heard by the beingwho-hears-through-walls, there is no reason to say that they are expressing the same score, or each other’s scores, in the absence of further information about what they are trying to do. The Chopin-Beethoven case earlier fails the test of expression since we assume that musicians playing Chopin are trying to play Chopin and that musicians playing Beethoven are trying to play Beethoven.5 Accommodation is also a scalar notion. Consider now some paradigmatic cases of accommodation pertaining to bodies in physical interaction. A and B are dancing together in a ballroom. They have clasped each other in the way prescribed for waltzing. When A moves her right foot forward, B moves his left foot back. When B heads off to the left, so does A. When one bends forward, the other bends backward. Sensing resistance, A refrains from breaking B’s back. They do not step on each other’s feet, or knock their heads together. Here is another example: Two workers on a job site are building a brick wall. A lifts the bucket of cement up to B, who pulls it up; B lowers the empty bucket to A, who shifts her weight to balance it. Each adjusts her body to respond to the motions of the other. And this harmonic behavior enables them to perform their tasks. Where A and B are in the same space and physically interacting, and where B’s response to A is an optimal response where human interests—having a good time at a dance,

Compossibility, Expression, Accommodation

115

getting the wall built, staying upright, not wrecking things and getting hurt—are served, we observers will perceive harmony and accommodation. Now imagine that our two dancers both try to move forward at the same time or that they try to move off in opposite directions. Or, as builder A lifts the bucket up to B, B pushes down on it, or as B lowers the empty bucket to A, it throws A off balance. To the eye of the observer, there is now comic disarray. The movements of A and B are inharmonious. Failure of accommodation of motions is the principle of slapstick. Stepping on the banana peel, I fail to accommodate my posture to the motion under my foot and fall flat on my back. Even slapstick has an order to it. It would be possible to write out a table of correlations between Larry, Moe, and Curly and the various objects in their physical environment including each other, as Larry, Moe, and Curly bang things up and get banged up in turn. At no point do Larry, Moe, and Curly or the objects in their environment violate any laws of nature; accommodation is value-added with respect to interaction. Not just any set of actions that can physically occur in the same space at the same time is harmonious, just as not every two piano pieces that can be played in the same interval express each other. So now back to Sleigh’s problem. I have rejected his claim that the mere possibility of writing a table that correlates certain notes and pauses of Musician A with notes and pauses of Musician B is sufficient for it to be true that A and B express each other’s parts. Must we nevertheless accept the claim that in any and every world in which Substance A and Substance B do express each other’s states, A and B also accommodate harmoniously to one another? Certainly the converse seems to be true, according to the following passage from the Monadology. This interconnection or accommodation of all created things to each other, and each to all the others, brings it about that each simple substance has relations that express all the others, and consequently, that each simple substance is a perpetual, living mirror of the universe. (Mon 56/AG 220)

But the passage seems to contain the solution to the problem. It is ‘‘simple substances’’ that express, where expression is to be understood with the help of the acrobat model, but ‘‘created things’’ that accommodate, where accommodation is to be understood with the help of the dancer or builder model. This hypothesis can be made to square with section 9 of the Discourse, which says: ‘‘All other substances in turn express [a given] substance and accommodate themselves to it . . . .’’ Evidently, Leibniz is talking here about our world, in which there is both expression of substances and accommodation of things. But we have succeeded in prizing apart the meanings of expression and accommodation by stating conditions under which expression may occur without accommodation and showing what value is added under accommodation. All that is needed for expression without accommodation to be possible is for it to be possible for there to be worlds other than ours in which substances fail to project into a harmoniously ordered, common spatiotemporal (but nevertheless phenomenal) world of ‘‘things.’’ For the phenomenal world of spatiotemporally related, interacting ‘‘things’’ is a projection from the noumenal world of nonspatial, nontemporal, windowless ‘‘substances.’’ (To say that level 2 is well-founded in virtue of the

116 Leibniz: Nature and Freedom

existence of level 1 is to say that level 1 projects to level 2 in the way analogous to the way that the ancient atomists and their seventeenth-century corpuscularian successors conceived the nearly qualityless atoms to project to a world of noisy, colorful, scented, flavored, living things.) Consider a possible world consisting only of Leibniz and Locke and the monads that composed their bodies. These ensembles would project into Leibniz and Locke—the embodied human beings—but not in such a way as to guarantee their accommodation. Since only they exist and, ex hypothesi, no other objects or people, they cannot meet or exchange letters. They cannot interact in a common space of things; they can only express one another.6 And that is why Leibniz claims obscurely that ‘‘the entire concept of space and time that we have is based on this harmony’’ (letter to Arnauld, October 9, 1687; LA 115, quoted in Sleigh 1990: 178). Leibniz explains accommodation as follows. Because ‘‘substances impede or limit each other,’’ one can say that ‘‘in this sense, they act upon one another and are required, so to speak, to accommodate themselves to one another’’ (DM 15/AG 48). But Leibniz and Locke would not seem in this condition to limit or impede one another, and so would not seem to act upon one another, or to accommodate to one another either. If this sketch is right, an interesting conclusion follows. There can be, not simply a possible world, but even a possible Leibniz-world in which universally expressive substances fail to project to a harmonious world. For, in the first place, the phenomenal, spatiotemporal world of interacting objects appears to be emergent, and to require some unspecified quantity—perhaps even an infinity—of individual substances. Just as two atoms, however real, cannot make there be phenomena for the human eye, perhaps tiny duplex worlds and other small worlds cannot produce the phenomena of spatiotemporal location, solidity, resistance, and motion. They cannot produce a world that physics can describe. Some critical mass of substances seems to be needed before we can have the sort of world in which entities can impede and limit one another and can be said to act upon one another. Further, even in possible Leibniz-worlds that are of a certain size, and in which there is expression between substances, there is not necessarily harmonious accommodation. Imagine a suboptimal but nevertheless possible world of sufficient size to allow the emergence of a spatiotemporal world. In this world, groundstory substances might express each other regularly and perfectly, but, in the phenomenal world of the upper story, persons, animals, plants, and inanimate objects—whatever ‘‘created things’’ exist—would fail to accommodate to one another, forming a weakly or strongly disharmonious rather than a harmonious world. In these worlds, some of what happens is like bad dancing and bumbling bricklaying, or a camping trip with Larry, Moe, and Curly. Perhaps, in such a world, the ‘‘laws’’ of nature can only be written down as monstrous curves, or keep changing, or are written with pens that often fail to express the thoughts or to move in accordance with the intentions of the writer. There might be neither pleasure nor production in such a world. But in the best of all possible worlds (ours), on Leibniz’s view, substances are strongly accommodating. The analogues of dancing and building come easily; the laws we can write down are wonderfully simple and elegant, and our fingers and pens mostly do what we want them to. In some

Compossibility, Expression, Accommodation

117

possible worlds that are inferior to our own, compossible substances express one another very inadequately. To summarize, at least some possible Leibniz-worlds are, like ours, two-story structures in which substances on the lower ‘‘foundational’’ story project into persons, plants, and animals, and material objects and their movements and changes on the upper ‘‘phenomenal’’ story. Expression can be construed as a symmetrical relation between first-story substances, accommodation as a symmetrical relation between second-story things. Substances project into all the foregoing sort of things, and things are well founded in substances. But doesn’t this account imply that it is contingent rather than necessary that the maximum compossible set of substances projects into this sort of visibly harmonious empirical world? This disagreeable conclusion can be escaped by noting that the claim ‘‘It is necessary that [The maximum compossible set of substances projects into a visibly harmonious world]’’ is shockingly ill formed. Necessary truths are true in all possible worlds, and the truth-value of ‘‘The maximum compossible set of substances projects into a visibly harmonious world’’ is arguably undefined for all worlds but ours. Is this what Leibniz meant to say? Well, one could hardly hope to find textual evidence for such an anachronistic claim, and that is the source of my reservations about having solved Sleigh’s interesting problem. But if there is a better solution to the paradoxes of expression and accommodation, a solution that can show how, under Leibnizian assumptions, accommodation or harmony flows from expression, which flows in turn from compossibility, value being added at each stage, it has not so far been presented. Consider again the image of an orchestra in which each musician plays her part in a soundproof room where she cannot hear the others. To the being-who-can-hearthrough-walls, the resulting sound, assuming each plays her part well, will be the same as if each played her part listening to the others. And so it is with creatures: each of us, considered as an individual substance, experiences his or her states and doings in causal isolation from all the rest. Nothing can flow between other creatures and us, such that we (transitively) see them. Nothing leaves my soul in virtue of which I can inflate my arm and leg muscles, or move my tongue, or touch an object outside of me. Each of us is a world apart. As Leibniz says in the famous letter to Des Bosses of May 26, 1712: One cannot prove from harmony that there is anything in bodies but phenomena. For from other considerations, it is clear that the harmony of phenomena in souls does not arise from the influx of bodies, but is pre-established. And this would be enough if there were only souls or monads, in which case all real extension (not to mention motion) would vanish, and this reality would be reduced to mere mutations of phenomena. (G 2:444/AG 201)

Yet there is a common world in which we not only see each other but touch each other, in which particles of light fall on our eyes arousing motions in our nerves and causing us to see different colors, and in which pulsations of animal spirits drive the movements of our bodies. Billiard balls smash into others, causing

118 Leibniz: Nature and Freedom

them to rebound. Caesar crosses the Rubicon, inducing Vercingetorix to marshal his forces. Leibniz insists that we can conceive ourselves both as musicians playing together and as musicians in soundproof rooms. We can hear through walls as inhabitants of the phenomenal world, but we also know that we really cannot hear each other and that we do not need to in order to play our parts.7 It is not simply the case that God—the one who can hear the orchestra playing together through soundproof walls—hears the harmony, but that the musicians themselves do. ‘‘I’’—and every other monadic individual—can be found in two places: I exist first in the foundational lower story of immobile world-apart substances without spatial location that project to the upper story, where objects, aggregates, and soulbody composites are engaged in the rough and tumble of physical interaction, some of them coming into being and passing away. In the lower story, I do not express external conditions, for there is nothing external to me, but instead I express the internal states of all the substances of my world. But I also exist in the upper story, for some substance projects to my empirical self. (Other substances project to parts of rocks, trees, vegetables, planets, dust particles, etc. or to other living organisms.) Where events in the upper story are concerned, to say that my mind expresses what happens in my body, as well as everywhere else in the universe, is to say that there is accommodation between conditions in some nerves, muscles, and sensory organs, such that they are part of my body, and the volitions and experiences of my empirical self. When one lower-story, ‘‘world-apart’’ individual—to employ Sleigh’s felicitous phrase (Sleigh 1990: 110, 137)—is said to express an internal state of another lowerstory, world-apart individual, the only way to understand this is as the claim that both world-apart individuals have the perceptions of the phenomenal common world they have in the top story. The characterization of the foundational world seems to be arrived at by considering what might be subtracted from our world, our experiences remaining the same, just as the characterization of the sequestered musicians is arrived at by subtracting their experiences of hearing one another. The foundational lower story seems to be the projected upper story minus space, time, causation, and the aggregations we call matter. So far, the analogy with the upper and lower story of the material atomists is preserved, for the characterization of their lower story too is arrived at by the subtraction of emergent properties. But the experiences of every soul or self on the foundational level seem to replicate its experiences in the phenomenal world, and this redundancy is hardly a positive thesis of Leibniz; it is merely an awkward implication of the interpretation just offered. Nevertheless, it is possible to understand why Leibniz thought as far into this scheme as he did. The claim that we are in some way like sequestered musicians is a response to a gap in our theoretical understanding. We don’t understand how one thing can make another move. We don’t understand how material atoms can produce experiences in minds. It seems utterly impossible. So—it seems—we must conclude that we only think we see this happening, and conceive that it does happen, when in fact no two things interact and nothing external produces internal experience. But one can’t be pushed to the conclusion that, if a process is not intelligible to the understanding it can’t occur, without being pulled by something else. And Leibniz

Compossibility, Expression, Accommodation

119

is evidently pulled by the notion that each substance in the foundational story can be a world unto itself, overheard and overseen by God alone. What makes this idea appealing is that it implies that, although we have to endure many things, we are not, as Spinoza claimed, assaulted by external forces that no finite creature can withstand for very long. Nothing is actually outside us; we are consequently invulnerable to predation on the part of others. Metaphysics can thus model for us a morally ideal condition. The image of the sequestrated musicians illustrates, finally, Leibniz’s claim that the harmony of things is objective, not subjective. Spinoza had claimed that harmony is produced by motions of the sense organs and the nerves (Ethics IApp/ CWS 445). ‘‘Men prefer order to confusion,’’ he says, ‘‘as if order were anything in nature more than a relation to our imagination’’ (Ethics IApp/CWS 444). Leibniz turns this argument on its head. If certain motions of the nerves produce the experience of harmony in us, this is because there is a concomitance between mind and body. That certain sequences of notes that we are able to play act on our ears and brains in the upper story to produce such ecstatic experiences is only one instance of the accommodation that distinguishes our world from inferior possible worlds. To Spinoza’s remark ‘‘Men have been so mad as to believe that God is pleased by harmony’’ (Ethics IApp/CWS 445), Leibniz’s response is that if God were not pleased by harmony, there would be no reason for this world to exist in preference to many other possible worlds composed of compossible substances but in which there is neither pleasure nor production.

Notes 1. A look at their historical precedents does not really help to answer the question whether Leibniz meant anything different by the two terms. His understanding of the term ‘expression’ seems to derive from Spinoza’s notion that expressive relations obtain both between ideas and the human bodies that have them, and between human ideas and God’s nature. The sources of Leibniz’s understanding of harmony are more difficult to trace. But in any case, the problem presented by the passages in question is conceptual, not historical. ‘‘Singular thoughts, or this or that thought, are modes that express God’s nature in a certain and determinate way’’ (Ethics IIP1/CWS 448). Compare: ‘‘Singular things are modes by which God’s attributes are expressed in a certain and determinate way, i.e., things that express, in a certain and determinate way, God’s power by which God is and acts’’ (Ethics IIIP6/CWS 498). Our ideas are said by Spinoza to ‘‘indicate or express a constitution of [our own] Body (or some part of it)’’ insofar as ‘‘its power of action, or force of existence, is increased or diminished, aided or resisted’’ (Ethics III, General Definition of the Affects/CWS 542). Leibniz’s uses of the term ‘harmony’ are very wide indeed. As Rutherford (1995a: 13) points out, he defines harmony as unity in variety or diversity compensated by identity, though his actual uses of the term are difficult to bring in under this rubric. 2. Sleigh quotes the following passage from Jalabert (1981: 87–88): ‘‘For Leibniz, the correspondence and the harmony [of nature] exist already at the level of the possibility of things, as they are conceived by the divine understanding. Each possible universe includes such a correspondence: phenomena that are not harmonious cannot constitute the same universe. So the universal harmony . . . is also pre-established in the sense that it is already constituted, from the possibility of things, that God contemplates prior to his decision to create.’’ 3. Sleigh 1990: 177. 4. However, the claim that humans express God rather than the world does not seem to be a consistent theme of Leibniz’s, and its appearance here may represent a transitory moment of enthusiasm. For a thorough survey of Leibniz’s uses of the term, see Kulstad 1977. Leibniz often offers asymmetrical

120 Leibniz: Nature and Freedom examples of expression. While the circle might seem to express the equation x2 þ y2 = 1, as well as vice versa, it is not clear that Leibniz would agree that the state of the sensory organs and the external world express the color experience. But it is not always the superior entity that expresses, for it is only humans who are said to express the divine nature. But this is perhaps in virtue of their standing higher in the scale of being than the nonminded expressers. Expression seems to imply a hierarchical relation of expresser to expressed; at the same time every substance in the universe expresses every other substance. This oscillation between distinction and generality can probably be demonstrated elsewhere in the structure of Leibniz’s metaphysics. 5. Wittgenstein’s answer to the question of when B is following a rule based on A’s actions was, I think, that the inner state of B is irrelevant. B expresses A’s actions if and only if there is some conceivable training process for B involving his observation of A that someone considers it functional or at least amusing to inculcate in B (Wittgenstein 1958: 90). 6. This raises an interesting question: if only God and I existed in a Leibniz-world (provided this is possible), would we necessarily accommodate to one another or only necessarily express one another? Sleigh claims that God is not part of any world (1990: 181). Indeed, Leibniz describes God somewhere as a ‘‘supramundane’’ intellect, implying that God is not an element of the world. But apparently, we do express God, and the things a substance expresses seem to be coextensive with things in its world, so perhaps the question whether God is an element of our world or not can be allowed to float for the purposes of the argument. 7. I borrow the idea of the two-storied house as a model for Leibniz’s metaphysics from Deleuze 1993. Deleuze conceives ‘‘downstairs’’ as a frothing mass of tiny unconscious creatures and ‘‘upstairs’’ as the habitation of intelligent self-conscious spirits; my assignment to the two floors is different. Donald Rutherford also allows two levels, one of which he describes as monadic, the other as panorganic: ‘‘They represent complementary ways of understanding the universe: one from the point of view of reality as it is in itself, a system of harmoniously related monads; the other from the point of view of the order determined by those monads’ expression of themselves as embodied creatures naturally subordinated to one another’’ (1995a: 230). I take this to be closer to Leibniz’s intentions, insofar as Leibniz appears to be substituting monads or immaterial atoms for material atoms as real causes of well-founded phenomena. See Wilson 1989: 180. But I prefer to say that the (noumenal) monads express one another’s states while (phenomenal) things, including the self of which I am aware and its body, harmonize, and that the monads project a world (as material atoms project a world of substances and qualities) as well as expressing the world they project.

6

Leibniz and Occasionalism NICHOLAS JOLLEY

hat is really at issue in Leibniz’s critique of Malebranche’s occasionalism?

W Ever since Arnauld, readers have found it difficult to answer this question. At

first sight the difficulty is surprising, for superficially the battle lines between the two philosophers seem to be clearly drawn: Malebranche holds that God is the one true cause, whereas Leibniz maintains that created substances are genuine causes as well. Problems arise as soon as we encounter Leibniz’s claims to the effect that occasionalism introduces perpetual miracles and that it appeals to God as a deus ex machina; for even casual acquaintance with Malebranche’s occasionalism reveals that it makes heavy use of the concept of law.1 Indeed, in the eyes of a critic such as Arnauld, the problem with occasionalism is not that it leaves too much room for miracles, but that it leaves too little.2 It may seem, then, that Leibniz’s critique of occasionalism is directed against a straw man, and that the debate as a whole is of little philosophical interest. It is true, as I will show, that some of Leibniz’s criticisms of occasionalism seem misguided, but such a negative verdict on the controversy would be a mistake, for the debate between the two philosophers centers on fundamental issues in metaphysics and the philosophy of science. Perhaps the best way of approaching the debate is to begin by considering the deep philosophical motivation for occasionalism. In a famous chapter of The Search after Truth entitled ‘‘The Most Dangerous Error in the Philosophy of the Ancients’’ (6.2.3), Malebranche explains that the belief in natural causal powers is a legacy of the despised Aristotelian-Scholastic tradition; such powers are in effect occult qualities which must be rejected by a philosophy founded upon clear and distinct ideas. Many years ago the French scholar Levy-Bruhl summed up this aspect of the philosophical motivation for occasionalism when he remarked that ‘‘as a worthy successor of Descartes, Malebranche replaces the confused scholastic notion of cause by the clear scientific notion of law’’ (quoted in McCracken 1983: 102). On this view Malebranche is the natural forerunner of Bertrand Russell, who, in a famous essay, argued that the notion of cause was a relic of an outdated conceptual scheme that had no place within the framework of modern science 121

122 Leibniz: Nature and Freedom

(Russell 1913). If Malebranche is a forerunner of Russell, Leibniz, by contrast, is a neo-Aristotelian who argues that the concept of natural causality must be retained; as he puts it in the New System, explanations must be ‘‘derived from the order of secondary causes’’ (G 4:483/L 457). What is at issue between the two philosophers is not just whether laws are ontologically grounded in causal powers, but whether laws constitute the rock bottom in scientific explanation. The logical relationship between these two issues might be brought out in the following way. It is clear that one can deny that laws are ontologically grounded in causal powers without holding that laws are explanatorily basic: one might believe that the latter role is indeed occupied by causal powers. More controversially, it seems consistent to hold that laws are grounded in causal powers while also holding that laws are explanatorily basic. For, as I will show hereafter, one might believe that causal powers are located at a metaphysical level below that with which science is concerned, and that they are thus irrelevant to the project of scientific explanation.

Leibniz and the ‘‘Disturbance’’ of Laws of Nature Recent commentators who have written about the debate between Leibniz and occasionalism have helpfully distinguished logically distinct strands in his critique. It has been observed that Leibniz’s objection that occasionalism introduces perpetual miracles is of wholly general application; it targets occasionalists’ claims about the relations between bodies no less than their proposed solution to the mind-body problem. But though this is Leibniz’s most celebrated objection to occasionalism, it is not of course the only one; he also offers an argument that is specifically aimed at the occasionalist stance on the relationship between mind and body (Woolhouse 1988: 176). According to this objection, the trouble with occasionalism is that its account of this relationship disturbs the laws of physics. For the purposes of this essay there is reason to begin by addressing this strand in Leibniz’s critique, since once the weaknesses of this argument are exposed we shall see that he will be forced to fall back on a second line of argument that brings into focus the real issue between Leibniz and occasionalism. In the Theodicy and other texts Leibniz objects that Malebranche’s occasionalist account of mind-body interaction represents no real improvement over Descartes’ interactionist account; it too involves a disturbance or derangement of the laws of nature (G 6:136). As Woolhouse suggests, Leibniz seems to have two distinct charges in mind (Woolhouse 1988: 176). The first is strictly physical: occasionalism, like Descartes’ interactionist position, is inconsistent with the principle of the conservation of momentum. When I raise my arm, the change in the momentum of my arm must be counterbalanced by a corresponding change in momentum elsewhere, but it obviously makes no sense to suppose that this change of momentum could be located in an immaterial substance (Woolhouse 1988: 175). The second objection is more general and more metaphysical: occasionalism, again like Cartesian interactionism, fails to respect the principle that the mental and the physical realms are alike causally closed; that is to say, mental events have exclusively mental

Leibniz and Occasionalism

123

causes, and physical events have exclusively physical causes. For Leibniz, this principle implies that it is never the case that a physical event has a mental event as either its real or occasional cause. By contrast, occasionalists are committed to claiming that at least some physical events have mental events as occasional causes, and that at least some mental events have physical events as occasional causes.3 Unfortunately for Leibniz, neither of these charges against occasionalism seems compelling. The occasionalists can meet the first objection by conceding that, by virtue of the principle of the conservation of momentum, any change of momentum must be counterbalanced; they must also concede that it is nonsensical to suppose that the locus of this counterbalancing could be an immaterial substance. But there is nothing in occasionalism that precludes them from saying that the change of momentum could be counterbalanced elsewhere in the physical system. That is to say, there is no reason why Malebranche should not offer exactly the same account as Leibniz himself of how, in the case of arm-raising, the principle of the conservation of momentum is satisfied. The more metaphysical objection, which turns on the general principle of the causal closure of the mental and physical realms, also runs into trouble. Robert Sleigh has argued that Leibniz is not well placed to defend his principle against occasionalism (Sleigh 1990: 166–167); in particular, with regard to the apparent action of mind on body, he is not in a position to say that, for any physical event, there is no sequence of causes, real or occasional, that has a mental event, such as an act of volition, as one of its members. For he himself tells Arnauld: ‘‘Nevertheless, one is quite right to say that my will is the cause of this movement of my arm . . . for the one expresses distinctly what the other expresses in a more confused way, and one must ascribe the action to the substance whose expression is the more distinct’’ (G 2:71/LA 87). If, however, Leibniz falls back on a weaker principle to the effect that, for any physical event, there is at least some sequence of causes, real or occasional, that has only physical events as its members, then again he is not well placed to refute occasionalism. For as Sleigh points out, there is nothing in occasionalism per se that precludes the recognition of some such quasi-causal history. Perhaps Malebranche in fact believed in the absence of such a quasi-causal sequence, but the problem would not be in his occasionalism but in his ignorance of physiology (Sleigh 1990: 167). This is a penetrating analysis of Leibniz’s position, but it is perhaps open to Leibniz to say that his remark to Arnauld should not be taken at face value. When Leibniz says that my will is the cause of the movement of my arm, he may be speaking with the vulgar but thinking with the learned; the concept of expression serves strictly as a replacement for the concept of causality rather than as an analysis of it. Thus Leibniz might say that he is still committed to the principle of the strict causal closure of the physical realm; by contrast, on his view occasionalism fails to satisfy this principle since it recognizes volitions as at least occasional causes of some physical movements (as in the case of arm-raisings). If Leibniz insists on this position, then it is open to the occasionalist to reply by emphasizing just how weak is the conception of an occasional cause. Indeed, Malebranche could say that talk of occasional causes commits him to no claims about mental or physical events as metaphysically real causes; it is thus incapable

124 Leibniz: Nature and Freedom

of constituting any sort of challenge to Leibniz’s principle of causal closure. Here it is worth remembering that Malebranche has rightly been seen as an ancestor of Hume, and the doctrine of occasional causes really amounts to no more than a recognition of what Hume was to call constant conjunction.4 To say that events of type A are the occasional causes of events of type B is to say that events of type A are regularly followed by events of type B. Thus Leibniz cannot deny that volition is an occasional cause of arm-raising without denying the existence of incontestable empirical regularities. It is true that Malebranche might wish to challenge Leibniz’s principle of the causal closure of the mental and physical realms. But his basis for doing so, I believe, is not that he is committed to the existence of at least quasicauses that straddle the two realms; it is rather that he wishes to insist that the whole category of natural causality is an outdated one that has no place in ontology or the philosophy of science. As Levy-Bruhl suggests, for Malebranche, the category calls for replacement by the more precise concept of law.

Miracles, Laws, and Natures To discover the heart of the debate between Leibniz and occasionalism one must turn to his notorious charge that occasionalism involves perpetual miracles; as I will show, it is here that we find the real controversy over the issue of whether laws or causal powers are explanatorily and ontologically basic. When Leibniz charged occasionalism with reintroducing perpetual miracles, Bayle expressed his puzzlement in a note to the article ‘‘Rorarius’’ in his famous Dictionary: The reason why this clever man finds the Cartesian system not to his taste seems to me to be based upon a false supposition; it cannot be said that the system of occasional causes, with its reciprocal dependence of body and soul, makes the actions of God into the miraculous interventions of a deus ex machina. For since God intervenes between them only according to general laws, in doing so he never acts extraordinarily. (WFPT 197)5

Today it is natural for us to share Bayle’s sense of puzzlement, for we are likely to agree with his implicit conception of a miracle as a suspension of a law of nature; moreover, as we have seen, even a superficial acquaintance with occasionalism reveals, as Bayle says, that its universe is governed by laws. Thus it seems that Bayle must be right to say that Leibniz wholly misses the mark in charging occasionalism with perpetual miracles. Yet such a reaction would reveal a misunderstanding of Leibniz, for according to his definition, he is justified. For Leibniz, a miracle is not to be defined in terms of a suspension of a law of nature; it is to be defined rather as an event that exceeds the causal powers of created things (G 4:520/WFPT 205). On this definition it is trivially true that, according to occasionalism, in the universe all events are miraculous, for if genuine causal powers are at issue, then creatures have no such causal powers (see Sleigh 1990: 162). Leibniz’s insistence that a miracle is correctly defined as an event that exceeds the powers of creatures suggests a possible strategy for replying to Bayle. Consistent with his definition Leibniz could have conceded that occasionalism recognizes the existence of laws of nature; he could nonetheless have insisted that, given his

Leibniz and Occasionalism

125

definition of a miracle, such a recognition has no bearing on the issue of perpetual miracles. Superficially it may seem that Leibniz is indeed tempted by this strategy, for he tells Bayle: Let us see, however, whether the system of occasional causes really doesn’t involve a perpetual miracle. Here it is said that it does not, because the system holds that God acts only according to general laws. I agree that he does, but on my view that isn’t enough to remove miracles. (G 4:520/WFPT 205)

But despite appearances, this is not the strategy Leibniz chooses to adopt. Leibniz does not deny that occasionalism recognizes the existence of laws or rules in the sense of divinely ordained regularities, but he does deny that such regularities constitute genuine laws of nature. With regard to the occasionalist laws of the union of body and soul, Leibniz writes, in some comments on Lamy: ‘‘But I do not at all agree that such a rule would be a law of nature, nor that the general laws of nature are purely arbitrary’’ (G 4:594). And, as Leibniz goes on to say, all miracles are exceptions to the laws of nature, even if they are not to be defined in these terms (G 4:594–595). Indeed, it seems that Leibniz would recognize an equivalence here: x is beyond the causal powers of creatures just in case x is an exception to a law of nature. Leibniz’s position, then, is not that the occasionalists recognize the existence of laws of nature, but that this is irrelevant to the issue of miracles; it is rather that the occasionalists do not, strictly speaking, recognize the existence of laws of nature at all. This has led some commentators to suggest that the real issue between Leibniz and occasionalism is the proper conception of a law of nature. Donald Rutherford, for instance, says that Leibniz wants to insist against Malebranche that laws of nature are laws of the natures of things (Rutherford 1993: 145). There is no doubt that Leibniz does wish to uphold such a claim, but it remains a question how it should be interpreted. When the claim is properly construed, I believe, we shall see that the fundamental issue between Leibniz and Malebranche is whether it is causes or laws that have ontological and explanatory primacy. It is Leibniz’s commitment to the primacy of causes that surfaces in his insistence that being an exception to a law of nature is only a derivative truth about miracles; it does not constitute their essence. One obvious difficulty in understanding Leibniz’s critique of occasionalism on this score lies in knowing what he means by ‘natures.’ Rutherford seems to assume that the natures of things are equivalent to their essences;6 on this basis he argues that, according to Leibniz, for a generalization to qualify as a law it is not sufficient that it should describe a regularity willed by God; ‘‘it must, in addition, be possible to conceive of the effects it describes as ‘explicable modifications’ of the natures of their subject’’ (1993: 145). According to Rutherford, from Leibniz’s standpoint the trouble with the occasionalist conception of a law is that it makes no provision for the intrinsic intelligibility of the created world (1993: 147). In this connection, and elsewhere, Rutherford speaks of Leibniz’s commitment to what he calls the Principle of Intelligibility. Rutherford’s assumption that the natures of things are equivalent to their essences is not an unreasonable one; such an equivalence does indeed seem to hold in the writings of Descartes and Spinoza, for example. Moreover, there are some passages that suggest that Leibniz himself subscribes to such an equivalence.

126 Leibniz: Nature and Freedom

In the Preface to the New Essays, for instance, Leibniz mounts a critique of Locke’s ‘‘thinking matter’’ hypothesis that is in many ways reminiscent of his case against occasionalism: But to explain myself more distinctly: it must be borne in mind above all that the modifications which can occur to a single subject naturally and without miracles must arise from limitations and variations of a real genus, i.e., of a constant and absolute inherent nature. For that is how philosophers distinguish the modes of an absolute being from that being itself; just as we know that size, shape, and motion are obviously limitations and variations of corporeal nature (for it is plain how a limited extension yields shapes, and that changes occurring in it are nothing but motion) . . . So within the order of nature (miracles apart) it is not at God’s arbitrary discretion to attach this or that quality haphazardly to substances. He will never give them any which are not natural to them, that is, which cannot arise from their nature as explicable modifications. (RB 66)

A striking feature of this passage is its strongly Cartesian flavor. For the Cartesians of course the essence or nature of matter is constituted by extension, its principal attribute, and any (nonmiraculous) property of matter is a modification or way of being of this attribute; thus being circular or revolving is supposed to be clearly a way of being extended. Now Leibniz of course does not subscribe to the Cartesian principle that the essence of matter is constituted by a principal attribute, extension; but he seems to be accepting the more general principle that any nonmiraculous property of matter must be an intelligible modification of its essence or nature. Thus for the purposes of arguing against Locke Leibniz seems to be helping himself to a Cartesian identification of nature and essence. Yet it is more characteristic of Leibniz, in his attacks on occasionalism, to seek to draw a distinction between the concepts of nature and essence. In the ‘‘Conversation between Philarethe and Ariste,’’ as Sleigh notes, Leibniz remarks: ‘‘One can distinguish three degrees among predicates: the essential, the natural, and what is simply accidental’’ (G 6:584/L 621; Sleigh 1990: 78). And in the antioccasionalist work De Ipsa Natura Leibniz clearly implies a distinction between the concepts of nature and essence, even if he does not explicitly state it: If, on the other hand, the law God decreed has in fact left some trace of itself impressed upon things—if things have been formed by the command in such a way that they are capable of fulfilling the meaning of the command—then it must be admitted that things have been given a certain ability, a form or force (such as we usually call a ‘nature’), from which the series of phenomena follows in accordance with the dictates of the original command. (G 4:507/WFPT 213)

Here the nature of a thing is equated with a force, and the force of a thing can hardly be identical with an essence; on this view natures are rather those causal powers or forces which the occasionalists are committed to denying. As Vailati remarks, a crucial difference between essences and natures is that, for Leibniz, God finds essences in his understanding, whereas through his will he is responsible for the natures of created things (Vailati 1997: 149). Thus when Leibniz says that laws must be grounded in the natures of things, he is claiming in these contexts that laws must be grounded in causal powers.

Leibniz and Occasionalism

127

Against occasionalism, then, Leibniz maintains that a law or rule is not truly a law of nature unless it is grounded in the nature, that is, the causal powers or forces of created substances. Although this is perhaps his central objection to occasionalism, Leibniz is not always careful in his arguments for this thesis; indeed, his arguments suffer from two weaknesses. Consider in the first place Leibniz’s argument that since they are not thus grounded in natures—in the Leibnizian sense—occasionalist laws of nature must be arbitrary: But I do not agree that such a rule would be a law of nature, nor that the general laws of nature are purely arbitrary. It is not an absolute necessity which has led God to establish them; however he has been led to them by some reason consistent with his supreme wisdom and by a certain conformity with the nature of things. (G 4:594)

In this passage Leibniz is making an entirely familiar point: his own position on the laws of nature occupies a middle ground between Spinozist necessitarianism on the one hand and Cartesian pure voluntarism on the other. But in seeking to show that occasionalists are committed to an objectionable Cartesian voluntarism Leibniz seems guilty of a non sequitur. From the fact that occasionalist laws of nature are not grounded in natures or causal powers, in the Leibnizian sense, it does not follow that they are thereby condemned to be purely arbitrary.7 To avoid this charge it is sufficient that there be reasons for God’s choice of the laws that are instantiated in our world; and Malebranche is insistent that this condition is satisfied. Indeed, Malebranche has a story to tell about God’s choice of the laws of nature that is similar in key respects to Leibniz’s own: God’s wisdom is guided by considerations of simplicity and fecundity.8 It may indeed be true, as Leibniz says, that God establishes the laws of nature both by some reason consistent with his supreme wisdom and by a certain agreement or conformity with the natures of things, but in order to rebut the charge of arbitrariness Malebranche is under no philosophical pressure to concede the second of these conjuncts. Thus Leibniz is wholly unsuccessful in his attempt to tar Malebranche’s occasionalism with the brush of Cartesian voluntarism. In the second place, Leibniz seems guilty of a curious equivocation in his critique of occasionalism. Consider Leibniz’s insistence that unless laws are grounded in natures, they will still be perpetual miracles: Thus it is not sufficient to avoid miracles that God makes a certain law, unless he gives creatures a nature capable of executing his orders. It is as if someone said that God has ordered the moon to describe freely in the air or in the ether a circle round [a` l’entour de] the globe of the earth without their being either an angel or intelligence which governs it, or solid globe which carries it, or vortex or liquid globe which transports it, or gravity, magnetism or any other mechanically explicable cause which prevents it from swerving away from the earth or moving off at a tangent to the circle. (G 4:595)

In this passage the thesis that laws must be grounded in natures is treated as equivalent to the thesis that bodies must have natures capable of executing God’s orders. But this way of explaining the thesis seems to trade on a confusion between the prescriptive and the descriptive senses of the term ‘law.’ When ‘law’ is taken in

128 Leibniz: Nature and Freedom

the prescriptive sense, Leibniz’s critique of the occasionalists has a certain plausibility. If I order you to run a mile in four minutes, my order will not have much point to it unless you possess the appropriate physique; in Leibnizian terms you must have a nature or set of causal powers that renders you capable of executing my order. Analogously, God will not give laws, in the prescriptive sense of the term, to his creatures unless he also gives them a nature capable of executing his order. Clearly, in the background to this discussion is the principle that ‘ought’ implies ‘can.’ When the prescriptive sense of ‘law’ is at issue, Leibniz’s thesis is plausible, but it is not at all clear that it is so compelling when the term is understood in the descriptive sense; and of course it is natural for us to say today that this is the only relevant sense. Certainly it is not difficult to see how Malebranche might reply to Leibniz’s strictures. As a good Cartesian Malebranche will of course concede that bodies have an essence, extension, and that this essence does not logically fix or determine the laws of physics in our world; consistent with their essence bodies might indeed describe circular paths instead of rectilinear ones. But it is open to Malebranche to add that there is no need for God to endow bodies with a nature, in the Leibnizian sense, over and above their essence. To bring about the observed behavior of bodies in our world it is sufficient that God will the laws of nature, for the fact that these laws are the volitions of an omnipotent being guarantees that bodies behave in accordance with these laws. From Malebranche’s perspective the Leibnizian model of God’s activity involves a needless duplication of effort on the part of the deity; not merely must he endow bodies with natures but he must further decree laws that are somehow in conformity with their natures. Leibniz’s second criticism of occasionalism thus seems to trade on an equivocation between the two senses of ‘law’ that we are in the habit of distinguishing sharply today. If Leibniz was indeed unclear about the distinction, he would have been in rather good company; as various scholars have shown, the habit of thinking of the laws of nature in prescriptive terms proved surprisingly tenacious among early modern philosophers. Boyle, for instance, remarks that ‘‘nothing but an intellectual being can be properly capable of receiving and acting by law’’ (quoted in C. Wilson 1987: 162). Somewhat similarly Clarke observes that ‘‘Matter [is] evidently not at all capable of any Laws or Powers whatsoever any more than it is capable of Intelligence,’’ and again, that ‘‘dull and lifeless Matter is utterly incapable of obeying any Law, or being indued with any Powers’’ (quoted in Vailati 1997: 141, 142). Clearly Boyle and Clarke are thinking of a law of nature as a rule or command that must be understood as a condition of being followed or ‘obeyed’ by any of God’s creatures. It is natural to observe that the two senses of ‘law’ are likely to be conflated when it is common ground among philosophers that laws are divine decrees. Even so, though the theological framework may encourage such a conflation, it does not of itself dictate it; it is still possible to distinguish clearly between a prescriptive divine decree, addressed to bodies, of the form: ‘‘Fall to the ground when unsupported,’’ and a descriptive one of the form: ‘‘Let it be the case that unsupported bodies fall to the ground.’’ Moreover, as various scholars have observed, no conceptual innovation on the part of early modern philosophers

Leibniz and Occasionalism

129

was required to make the distinction, for it had already been clearly drawn by the late Scholastic philosopher Suarez (Vailati 1997: 142; see C. Wilson 1987: 162). Despite the apparent parallels in the thought of Boyle and Clarke, it is difficult to believe that Leibniz was systematically confused about the status of the laws of nature. This is not just a matter of conventional piety toward a great philosopher. Apart from the fact that Suarez, with whose thought Leibniz was acquainted, had already drawn the distinction, there are reasons internal to Leibniz’s system for resisting such a view. As we have seen, there is evidence that Leibniz thought of himself as forging a middle path between the necessitarian conception of the laws of nature that he associated with Spinoza and the strictly voluntarist conception that he attributed to Descartes: the laws of nature are contingently true, but morally necessary in the sense that God’s choice of them is determined by his subscription to the Principle of the Best. Clearly such a position makes sense only on the assumption that laws of nature are universal propositions with a truth-value; if the laws of nature were simply imperatival in form, the question of their status as necessary or contingent would not arise. Perhaps, in the case of Leibniz, it is safe to say that he was generally clear about the distinction between the prescriptive and the descriptive senses of the term ‘law,’ even if for polemical purposes he was sometimes tempted to blur the distinction. Indeed, trading on the prescriptive sense may, on occasion, have served the purposes of an ad hominem argument, for as Vailati remarks, Clarke exploited the prescriptive conception of the laws of nature precisely in order to argue for a version of occasionalism (Vailati 1997: 141–142).

Occult Qualities Some of Leibniz’s arguments for the thesis that occasionalism involves perpetual miracles seem surprisingly flawed. And Leibniz may not seem to strengthen his case philosophically when, for polemical purposes, he further helps himself to the notion of occult qualities. Commenting on the supposition that God could order the moon to move in a circle without there being a mechanical explanation thereof, Leibniz concludes: ‘‘To deny that it would be a miracle is to return to absolutely inexplicable occult qualities which are decried today with so much reason’’ (G 4:595). Initially, what is perhaps most striking here is the irony of Leibniz’s invocation of occult qualities in a critique of occasionalism. For, as we have seen, it is plausible to suppose that the deep philosophical motivation for occasionalism is the desire to eliminate natural causality as the last occult quality; the successor concept to that of natural causality is the concept of law, and this concept is clear and distinct because laws can be expressed in mathematical terms. Yet here Leibniz objects that it is precisely the occasionalists who are guilty of reintroducing occult qualities by their elimination of natural causality. Such a move on Leibniz’s part may be rhetorically effective, but we may suspect that it is lacking in philosophical substance. In fact, such a judgment would be premature: Leibniz’s invocation of occult qualities in his critique of occasionalism proves to be surprisingly illuminating. Today when we think of occult qualities we are likely to call to mind the vis

130 Leibniz: Nature and Freedom

dormitiva explanations of the Scholastics, and such dispositional properties serve to clarify Leibniz’s case, at least by way of analogy. Consider a philosopher who subscribes to the following commitments. First, ultimate scientific explanations can be given in terms of dispositional properties; they are not merely placeholders for deeper explanations in terms of microstructural properties. Thus, when we say that the glass broke because of its fragility, no further explanation is called for or even possible. Second, as a matter of ontology, these dispositional properties are not grounded in anything more basic, such as the chemical structure of the glass. Such a combination of commitments is likely to strike many of us as ‘‘Scholastic’’ in a pejorative sense. If we now substitute laws for dispositions, and causal powers for nondispositional grounds, we are in a position to understand Leibniz’s hostility to occasionalism. For Leibniz sees the occasionalists as analogously holding that ultimate explanations can be given in terms of laws, and that these laws do not need to be grounded in natures or causal powers. And to a neo-Aristotelian, such a set of commitments seems as objectionable as the Scholastic position on dispositional properties is likely to do to us. Of course the case of dispositional properties serves as an analogy only; it is not literally an illustration of Leibniz’s point.For whereas we protest that dispositional properties need nondispositional grounds, Leibniz objects that laws need to be grounded in causal powers, and thus what plays the role of grounding, for Leibniz, is itself at least quasi-dispositional in character.9 But the modern critique of Scholastic appeals to dormitive powers and Leibniz’s critique of occasionalism have one important thing in common: they both turn on the point that there is something seriously missing in the position under attack; they both fail to satisfy the grounding requirement. And this is sufficient to enable us to see the point of Leibniz’s seemingly merely rhetorical claim that occasionalism involves the reintroduction of occult qualities.

Laws, Causal Powers, and the Theory of Monads One issue that has received little attention is how Leibniz’s critique of occasionalism is affected by the development of his system in the direction of the theory of monads that is characteristic of his later years. Fortunately, Leibniz’s correspondence with the Cartesian De Volder supplies much of the material necessary to answer such a question. In this exchange Leibniz is concerned not just to expose the inadequacies of the Cartesian conception of matter for physics, but also to explain the relationship between his science of dynamics and his theory of monads; indeed, Leibniz insists that the theory of physical forces is the gateway through which to pass to the true metaphysics (G 2:195/L 593). And though occasionalism is not the principal target of attack in this exchange, it is not forgotten either; in a number of places Leibniz insists that his theory of monads is the true alternative to occasionalism. Occasionalism may not be the chief issue in the exchange with De Volder, but there is no doubt about the implications of Leibniz’s position for that system; in explaining his theory of monads to De Volder Leibniz shows that he is in a sense moving even further away from occasionalist tenets. Leibniz continues to maintain

Leibniz and Occasionalism

131

his principle that laws must be grounded in the natures of things, but his position is complicated by the fact that he now recognizes two levels of grounding corresponding to two levels of forces. Against the Cartesians, including the occasionalists, Leibniz insists that the laws of motion force us to recognize the existence of dynamic properties in bodies by virtue of which these laws obtain: You very rightly observe, in agreement with my own opinion, that it is in opposition to the laws of power, cause, and effect, for a great body to be moved by a smaller one without penalty. But from this fact I prove that the body contains something dynamic by virtue of which the laws of power are observed. It therefore contains something besides extension and antitypy, for no such thing can be proved from these two alone. (G 2:184/L 520)

Such claims are of course all that Leibniz needs in order to distance himself from the occasionalists, but in fact he goes further; for he now insists that ‘‘the forces which arise from mass and velocity are derivative, and belong to phenomena’’ (G 2:251/L 530); what these physical forces derive from is of course the primitive active forces of monads or simple substances. Leibniz is at some pains to emphasize the explanatory autonomy of his science of dynamics; the physicist does not need to concern himself with the properties of monads: But in phenomena, or [seu] in the resultant aggregate, everything is explained mechanically, and so masses are understood to impel each other. In these phenomena it is necessary to consider only derivative forces. (G 2:250/L 529)

But, as Leibniz goes on to explain, though the physicist need not concern himself with such issues, the metaphysical truth of the matter is that these forces stand in need of ontological foundations. When De Volder ventured to suggest that there might be only such derivative forces, he received a dusty answer from Leibniz (G 2:251/L 530). Thus from the standpoint of his later metaphysics, the occasionalists are at two removes from the truth, as it were. Not merely do they fail to see that laws must be grounded in physical forces, but they also fail to see that these forces must in turn be grounded in the reality of monads. Although there is no doubt that this is Leibniz’s official later position, it is worth asking whether monadology does not have the resources for a more accommodating reply to occasionalism. Let us begin by considering one famous Leibnizian objection to occasionalism that we have so far ignored: the occasionalist conception of substance as being wholly devoid of genuine causal powers leads to Spinozism: This again shows that the doctrine of occasional causes which some defend can lead to dangerous consequences . . . though the consequences are doubtless not intended by its very learned defenders. Far from increasing the glory of God by removing the idol of nature, this doctrine seems, with Spinoza, to make God into the very nature itself of things, and to reduce created things to mere modifications of a single divine substance. For that which does not act, which has no active force, which is robbed of any distinguishing characteristic, and finally of all reason and ground of permanence, can in no way be a substance. (G 4:515/ WFPT 221)

132

Leibniz: Nature and Freedom

Thus Leibniz consistently subscribes to the principle that if x is not the bearer of genuine causal powers or forces, x is not a substance. Suppose, now, that Leibniz were also committed to the converse conditional: if x is not a substance, then x is not the bearer of genuine causal powers or forces. Let us call this the ‘‘Principle of Substantial Causality’’ (PSC). But in Leibniz’s later philosophy the only true substances are monads or simple substances; as Leibniz tells De Volder, ‘‘only simple things are true things, the rest are beings by aggregation, and therefore phenomena’’ (G 2:252/L 531). Now monads occupy the ground floor of Leibniz’s metaphysics, and are thus below the level of phenomena with which physics is concerned. Thus, if monads alone are true bearers of forces, then any appeal to such forces would be banished from the province of scientific explanations; such explanations would appeal rather to ‘‘abstractions,’’ as Leibniz terms them, such as the laws of physics (G 2:252/L 531). As metaphysicians we would indeed need to recognize that the realm of phenomena—the subject matter of physics—is grounded in the properties of monads, but as philosophers of science we could blithely ignore this fact; we would insist on the autonomy of scientific explanations with respect to their metaphysical foundations. Thus if Leibniz subscribed to PSC, he could still uphold his metaphysical objection to occasionalism, but as a philosopher of science he could accommodate the occasionalist insistence on the irrelevance of genuine causal powers and forces to the enterprise of science.10 The evidence of the exchange with De Volder strongly suggests that Leibniz does not subscribe to PSC; as we have already seen, he tells his correspondent that the derivative forces belong to phenomena that are not substances, but merely aggregates of substances (G 2:251/L 530). But though this is Leibniz’s official position, there are some passages—even ones bearing directly on occasionalism— that seem more sympathetic to PSC. In places Leibniz seems to short-circuit the discussion of physical, derivative forces, and move directly to a consideration of forces at the ground-floor metaphysical level. Consider the following passage: Last of all, you add: ‘‘Particulars act on each other and are subject to change with respect to actions. How this can be explained by substances which do not act upon each other is obscure to me.’’ This seems to be aimed at my opinion about the preestablished harmony between simple substances, which cannot act upon each other. Yet they do produce a change in themselves, and it is necessary for this to happen from your own point of view as well. For you acknowledged above that there is an internal basis for forces or actions, and so we must recognize an internal principle of change. And unless we do, there will be no natural principle of change and therefore no natural change. For if the principle of change were external to all and internal to none, there would be none at all, and we should have to turn back with the occasionalists to God as the only agent. It is therefore truly internal to all simple substances, since there is no reason why it should be in one rather than another, and it consists in the progress of the perceptions of each monad, the entire nature of things containing nothing besides. (G 2:271/L 538)

Here Leibniz argues directly for forces in monads on the basis that they satisfy the demand for a principle of change that is internal to things; there is no suggestion that this demand might be equally satisfied by recognizing a principle of change in bodies, that is, at the phenomenal level. Perhaps it may be said that in this passage

Leibniz and Occasionalism

133

Leibniz is responding directly to an objection to his theory of preestablished harmony, and that he does not mean to exclude the recognition of ‘‘principles of change’’ at the level of bodies or phenomena. But at least this passage suggests that the theory of monads contains the resources for confining causal powers or forces to the ground-floor metaphysical level; certainly there is nothing in such a restriction that would weaken the response to the metaphysical challenge of occasionalism. And if Leibniz were to respond in this vein, he could nonetheless accommodate the occasionalist insight that science is an investigation into laws, not causes. But officially at least Leibniz is not prepared to take this line. Leibniz’s debate with occasionalism has traditionally been seen as a rather quaint and outdated controversy over God’s relationship with the created world. It is true of course that the debate is often couched in theological terms: Leibniz famously complains that the occasionalists appeal to perpetual miracles and that they invoke God as a deus ex machina. But despite the theological language, the issues that divide Leibniz and Malebranche are anything but quaint or outdated; in fact they are surprisingly modern ones. For what is really in question is the nature of the scientific enterprise and its metaphysical foundations. In this debate, as we have seen, a partisan of a law-based conception of science is confronted by a partisan of a cause-based conception. There is of course a sense in which, by defending the latter position, Leibniz is the more conservative of the two thinkers, but to say this is certainly not to say that his position has been superseded. On the contrary, whether science is fundamentally about laws or about causes remains very much a live issue today.

Notes Earlier versions of this paper were read at the California Conference on Early Modern Philosophy, University of California, Irvine, July 2000, and at a conference entitled ‘‘Leibniz and his Correspondents,’’ Tulane University, March 2001; I am grateful to the commentators, Timothy Crockett and Justin Smith respectively, and to the editors of this volume for helpful criticisms. Translations other than those cited are my own. 1. Throughout this essay I restrict my discussion of occasionalism to Malebranche’s version of the doctrine. 2. Arnauld draws a distinction between perceptible miracles (miracles sensibles) and hidden miracles (miracles cachez) and claims that Malebranche’s doctrine of general volitions leaves no room for miracles of the latter kind: ‘‘On the supposition of this distinction between perceptible miracles and hidden miracles, one easily sees how it is an untenable paradox to claim that God does not act in the order of nature by particular volitions, but only by general volitions.’’ Antoine Arnauld, Reflexions philosophiques et theologiques sur le nouveau systeme de la nature et de la grace (Cologne, 1695), 1:196. Arnauld’s example of a ‘‘hidden miracle’’ is the case of a holy bishop who is gradually cured in response to the prayers of the faithful. 3. It may be objected that Malebranche cannot accept the principle of the causal closure of the physical realm since he is committed to the thesis that every physical event has God’s volition as its real cause, and God’s volitions are mental. The issue of divine real causality is a large topic that I cannot discuss here. However, it may be said that God is the cause of physical events in the minimal sense that he wills the initial conditions and the laws of nature that the events instantiate. 4. For a valuable account of Hume’s relationship to the philosophy of Malebranche, see McCracken 1983: 254–290.

134 Leibniz: Nature and Freedom 5. The original text is found in Dictionnaire historique et critique (Amsterdam, 1696–97), 1:967. 6. Rutherford (1993: 142) remarks that the Principle of Sufficient Reason requires that ‘‘within the ‘order of nature’ it is not enough simply that there be some reason for anything to happen as it does; in addition, there must be what Leibniz calls a ‘natural reason’: a reason that displays the effect in question as following in an intelligible manner from the nature or essence of some created being.’’ 7. See Rutherford (1993: 145, n. 17): ‘‘Malebranche would obviously contest this charge [that occasionalists are committed to the arbitrariness of the laws of nature].’’ 8. Although there are similarities between Leibniz and Malebranche in their conceptions of divine wisdom, there are also important differences; for example, as Rutherford says, unlike Malebranche Leibniz is concerned to uphold the conception of God as ‘‘an infinitely skillful craftsman, one who is disposed to create that world which in and of itself contains the greatest possible perfection’’ (1993: 158). 9. Donald Rutherford has pointed out to me that, for Leibniz, a causal force is not simply a power that would act if certain conditions were realized, but a power that does act unless it is prevented. See, for example, ‘‘On the Correction of Metaphysics and the Concept of Substance’’ (G 4:469/L 433): ‘‘Active force . . . contains a certain act or [sive] entelechy and is thus midway between the faculty of acting [facultatem agendi] and the act itself and involves a conatus. It is thus carried into action [operationem] by itself and needs no help but only the removal of an impediment.’’ Leibniz’s account here makes it appropriate to describe causal powers or forces on his view as quasi-dispositional. 10. In this spirit Leibniz need not take a stand on the issue of whether terms that appear to refer to physical forces refer to nothing at all or simply to nothing that is fully real.

7

Leibniz’s Two Realms JONATHAN BENNETT

Leibniz’s Commitment to Mechanism Leibniz insists that any bodily event can be explained purely in terms of ‘‘mechanism,’’ meaning impact mechanics. The latter’s laws are all quasi-causal (Sleigh’s label) rather than causal; they describe patterns among events that are embedded in the universal harmony, and do not imply that any body acts on any other. When declaring how things must go in physics, Leibniz does not often remind us that his topic is quasi-causation, not real transeunt causation; but that is always his view. Similarly, in those contexts he seldom reminds us that bodies are phenomenal rather than basically real; but in the mature years that is always his view too. In his claim for the power of mechanism, Leibniz is refusing to explain any particular events in terms of Aristotelian ‘‘forms,’’ for example, appealing to ‘‘the form of oak’’ to explain why an acorn develops as it does. In this he aligns himself with the Galilean revolution, which he sees has come to stay—‘‘the great light of our age,’’ he calls it. This may be less a philosophical opinion than a reading of where science is moving, and above all an upshot of seeing that after two millennia of virtual stasis science at last is moving (see PAB, G 7:343–344/AG 319).1 Leibniz’s support for ‘‘mechanism’’ may be encouraged also by his denial that any material event can be explained through a mental one. The Galilean revolution may encourage that denial, but not to the extent of requiring it. Descartes was a prime revolutionary who firmly rejected the Aristotelian paradigm, yet he held that human behavior requires some input from an immaterial mind. Leibniz dissents: ‘‘I attribute to mechanism everything which takes place in the bodies of plants and animals except their initial formation,’’ he writes at NE 2.9.11 (RB 139). ‘‘I do not approve of bringing in the soul when plant and animal phenomena have to be explained in detail’’ (NE 2.9.11/RB 139). This holds for humans too. Leibniz’s faith in mechanism’s extent draws strength from his conviction that the value of this world consists partly in how simple the quasi-causal principles are that govern it. Because, for him, mechanistic principles are unbeatably 135

136 Leibniz: Nature and Freedom

‘‘simple,’’ his adherence to them pays tribute to God, ‘‘the reason for the greatness or power in the mechanism of the universe as now constituted’’ (UO, G 7:305/AG 152). My focus here is on Leibniz’s view that all the events with which physics (including biology) has to do can be explained in terms of efficient quasi-causes. To explain any material event E, he holds, one needs to invoke only preceding events that quasi-made E occur; not to later ones that E was for. Yet he also gives teleological concepts a place in the truth about the world. How can this be? That is my chief question, and I shall evaluate Leibniz’s answers to it.

Three Roles for Teleology Even if we completed a perfect total science, Leibniz holds, we ought still to ask: Why are these the basic principles of physics? His answer is teleological; the principles obtain because God wants them to do so: These laws of motion . . . cannot be explained by the mere consideration of efficient causes, or of matter. For I have found that we have to bring in final causes, and that these laws do not depend on the principle of necessity . . . but on the principle of compatibility, the choices of wisdom, that is. (PNG 11/WFPT 263)

This serves ‘‘to clear the mechanical philosophy of the impiety with which it is charged’’ (DM 23/WFPT 75). It gives Leibniz—he says elsewhere—a basis on which to protest against ‘‘the Spinozist view . . . which dismisses the search for final causes and explains everything through brute necessity’’ (RB 73; see also NI, G 4:505/ WFPT 211, and DM 10/WFPT 61–62). That first answer to my chief question raises no problem: a world wholly governed by mechanism could, obviously, result from the purposes of its creator. This, however, is philosophically uninteresting in Leibniz’s hands, as in those of most Christian philosophers. They typically attribute activities to God without considering how he performs them, perhaps assuming that God needs no ways or means—no how—for doing anything, so that they can attribute purposes to him without inquiring into teleology as such. Furthermore, the thesis about divine teleology does not imply that there is any interplay between teleology and mechanism. Yet Leibniz often commits himself to there being such an interplay; that creates the problems that I want to address; and Leibniz’s first answer to my question brings no grist to that mill. Second, teleological notions belong in biology, Leibniz holds. I have quoted him saying that mechanism covers all of biology ‘‘except the initial formation’’ of plants and animals. Given that an organism exists, its life and procreation can be mechanistically explained, he seems to say, but the fact of its existence—the fact of those particles’ being interrelated like that—can be explained only through God’s purposes. I do not know how serious he was in this retreat from his usual doctrine that mechanism can explain every material event. It grates against several things in his philosophy; and I have found it only once in his writings, whereas the contrary view that mechanism covers everything is ubiquitous. For example, he hopes to reconcile ‘‘those who hope to explain mechanically the formation of the first tissue

Leibniz’s Two Realms

137

of an animal and the whole machinery of its parts, with those who account for this same structure using final causes,’’ and says that ‘‘both ways are good’’ (DM 22/ WFPT 75). Anyway, insofar as Leibniz admits teleology in the ‘‘except the initial formation’’ manner, he again provides me with nothing to discuss. In postulating a final-cause explanation for an event that cannot be explained in terms of efficient quasi-causes, Leibniz does not imply that the two sorts of explanation intersect in a manner that we should inquire into. Final causes more often enter his biology through the concept of function. He writes, against Descartes, that biological structures give us evidence about God’s purposes: ‘‘The reasons for what was created by an understanding are the final causes or plans of the understanding that made them. These are apparent in their use and function, which is why considering the use parts have is so helpful in anatomy’’ (G 4:299/AG 242). The core of the concept of biological function was made clear in Wright (1976); critics have picked away around the edges, but the main idea is clearly correct. To say that the function of (C) the contraction of subcutaneous capillaries when the skin cools is to produce (R) a reduction in heat-loss is to say: 1. C leads to R. 2. It is because C leads to R that the organism is so structured that C occurs.

In most cases, clause 2 is true for evolutionary reasons: the structures producing C have been selected for because they lead to R. There is nothing properly teleological about this; see Bennett (1976: 78–80) for a defense of that. However, Leibniz would have to support the concept of function differently. For him, as we have just seen, clause 2 is true because of God’s purposes (see for instance DM 22/WFPT 74–75), and of course purposes do bring in genuine teleology. But in this context as in others, Leibniz is not philosophically curious about the concept of teleology as applied to God; nor does he carefully explore the question of how these purposes of God make themselves felt in a material world wholly governed by mechanism. Third, Leibniz deploys notions that he takes to be teleological in the thick of doing inorganic physics, and not merely in reflecting on why its basic principles are true. Although ‘‘the [mechanistic] way of efficient causes is deeper and in some sense more immediate and a priori,’’2 nevertheless ‘‘[t]he way of final causes is . . . frequently of use in discovering important and useful truths which it would take a long time to find by the other, more physical route’’ (DM 22/WFPT 75; see also SD, GM 6:243/WFPT 163–164). Although he says ‘‘frequently,’’ Leibniz seems to have only one example of this; it is the only one he cites, and he cites it often. It is Snell’s law, which relates the angles of incidence and of refraction whenever light passes between two translucent mediums. Leibniz takes this to imply that the light always follows the ‘‘easiest,’’ or anyway the most determinate, way to pass from a given point in one medium to a given point in the other (TA, G 7:273/L 479). I shall go along with this, on the strength less of his obscure explanation than that of Feynman, Leighton, and Sands (1963: chap. 26.4).

138 Leibniz: Nature and Freedom

Snell offered his law in an essay that does not survive and that Leibniz did not see. The law was first published by Descartes, who sought to derive it mechanistically from an opaque me´lange of points, principles, and models. Leibniz published a supposed demonstration that the law could be reached by asking what he took to be a teleological or final-cause question, namely ‘‘What path for a light ray would be the easiest to get from A to B?’’ He conjectured that Snell did reach it in that way, and that Descartes had not discovered it for himself but merely took it from Snell. Of Descartes’ explanation of the law he writes: It is extremely forced and too hard to understand. . . . It is an afterthought adjusted somehow to the conclusion, and not discovered by the method he gives. So we may well believe that we should not have had this beautiful discovery so soon without the method of final causes. (TA, G 7:274/L 480; see also DM 22/WFPT 75)

This does not repeat the old point that patterns in nature bring in teleology through the thought that God purposefully instituted them. Rather, Leibniz is now saying that some of the patterns are teleological, so that one might be helped to discover them by asking teleological questions. Granted that Leibniz’s easiest-path formulation accurately expresses Snell’s law, why call it teleological? Most writers in this area take Leibniz’s word for it, but he and they are wrong. Of course we can talk about God’s wanting the light to follow the easiest path; but so we can about any physical law. ‘‘Newton’s gravitational law reflects God’s desire that the relative acceleration between two otherwise isolated bodies should vary inversely with the square of the distance between them.’’ No one who had worked on teleological concepts per se would count Snell’s law as teleological. If it is so, then Leibniz ought to explain how events in a mechanistically patterned material world can be revealed by asking teleological questions. He does not try to.

Teleology within the Individual Monad Fourth, Leibniz claims to build teleology into his ground-floor metaphysic of monads: There are only monads, and in them there is only perception and appetite. The word ‘appetite’ implies a claim to teleology, which is reinforced by much that Leibniz writes, including: ‘‘The laws of appetites [are] the laws of the final causes of good and evil’’ (PNG 3/WFPT 259). The basic account of appetite is straightforward. Whereas a perception is a short-lived intrinsic state of the monad, an appetite is the monad’s endogenous change from one state to another: ‘‘The action of the internal principle which brings about change, or the passage from one perception to another, can be called appetition’’ (Mon 15/WFPT 269). In the case of humans, Leibniz writes to Remond, ‘‘appetite is called will’’ (G 3:622), which repeats the teleological claim. Momentary states of a monad correlate with the rest of the universe in such a way as to carry information about it; and that thesis, which lies deep in Leibniz’s metaphysic, gives color to calling such states ‘‘perceptions.’’ Nothing comparable

Leibniz’s Two Realms

139

supports designating as ‘‘appetition’’ a monad’s change of state. Such changes are caused from within the monad; and that is a point they have in common with voluntary action, which we ordinarily think of as endogenous rather than forced from without. But that is a thin likeness. Unlike our ordinary voluntary/involuntary concept, this one draws no line through the monad’s life, because according to Leibniz every monadic change is caused from within. Furthermore, an account of the will should provide for attempts that fail, as when one tries and fails to meditate on God or to stop thinking about Istanbul. Leibniz offers to provide for that, through the notion of tendency. Sometimes he says that appetitions are tendencies (PNG 2/WFPT 259), but all he needs here is his more usual view that an appetition involves or manifests a tendency: The action of the internal principle which brings about the change or passage from one perception to another can be called appetition; it is true that the appetite cannot always completely reach the whole perception toward which it tends, but it always obtains some part of it, and reaches new perceptions. (Mon 15/WFPT 269)

So an appetite may tend toward something without reaching it. The only clear examples I can find of unrealized tendencies involve a thing’s tending to become F but not doing so because of some fact about its environment; but the tendencies of Leibniz’s monads cannot be like that, because they are perfectly self-contained, never acted on from outside. Even if he can overcome that difficulty, Leibniz’s account of monadic appetite does not justify his redescriptions of it in such teleological terms as ‘final causes’ and ‘ends and means.’ His fundamental theory says only that the individual monad runs through its history in accordance with laws given to it by God, laws that govern the efficient causality of its unfolding. I agree with Robert Adams: The language of ‘appetite’ and ‘ends and means’ may be somewhat misleading. It suggests the pursuit of a desired future state of affairs, but the action of a Leibnizian substantial form is more like what is sometimes called ‘acting on principle.’ In Leibniz’s view, the ‘‘internal principle’’ governing ‘‘the passage [of a substance] from one perception to another’’ is not based on the desirability of the later perception in itself, but rather on the following of certain laws of nature. (Adams 1994: 318)

There seems to be no followable route from Leibniz’s basic metaphysic to the notion of doing something for the sake of an end, let alone the more fully teleological notion of doing something because one thinks it will lead to a certain end. Writing informally about the human condition, Leibniz makes free with intentions and purposes; but when his metaphysic is close at hand he has little to say about teleology except for the blank assertion that final causes reign within the monad because of appetition. The only passage I can find that puts a little flesh on those bones is this strange one: since the nature of a simple substance consists of perception and appetite, it is clear that there is in each soul a series of appetites and perceptions, through which it is led from the end to the means, from the perception of one object to the perception of another. (C 14/MP 175)

140 Leibniz: Nature and Freedom

The monad, we are told here, is led ‘‘from the end to the means,’’ a fine ad media. That seems to imply that appetition involves genuine teleology, in which ends, or anyway thoughts of them, help to explain means. If that were Leibniz’s point, something would be seriously wrong. In his metaphysic the move from one perception to the next is a matter of efficient causation; if it were also a move from end to means, that would subject teleology to Spinoza’s jibe that it absurdly ‘‘reverses the order of nature,’’ treating effects as causes. Of course Spinoza was wrong about that: in treating ends (or thoughts about them) as explanatory, teleology does not treat effects as causes; it does not imply that when I stretch out to pluck the apple from the tree, the plucking causes the stretching. Yet that is what Leibniz commits himself to if he holds that the efficient-causal move from one perception to the next involves going ‘‘from the end to the means.’’ I hope that it was a slip of the pen, although ‘‘through which it is led from the means to the end’’ would hardly have been better. Adams’s criticism and mine rest on our seeing Leibniz as a thoroughly bottomup philosopher, not a partly top-down one. Leibniz’s theory-building proceeds in a strictly upward direction: he aims to develop his metaphysic in an austere way, allowing himself the concepts of logic and mathematics, and those of substance, quality, time, correlation and little else, not availing himself of concepts from folk psychology with no warrant except their familiarity. For him, I contend, our psychological concepts are to be clarified and justified by being developed out of the parsimonious initial stock of concepts. A partly top-down project would allow him to import into his monadic metaphysic concepts taken unanalyzed from armchair psychology. Then he might explain appetition partly through the idea that such changes have something volitional or purposive about them—with those concepts being taken on board, unexplained, on the strength of their quotidian familiarity to us. That would get him out of trouble that he otherwise has, but at a high price in philosophical interest. Leibniz’s actual bottom-up project is potentially illuminating because when we see a great philosopher try to make bricks without straw, we can learn something about the right way to make bricks.

Two Realms Let us now join Leibniz in supposing that he has established teleology in the individual monad, and examine what structure he erects on that foundation. According to Leibniz, an organism is an infinitely complex machine associated in a special way with a single monad, its mind. Although he sometimes writes as though he were a realist about the material machine, his real view is that it is the appearance to us of an aggregate of monads. Its mind, on the other hand, can only be thought of as a single monad, and thus ultimately real in Leibniz’s metaphysic. So when he contemplates an animal, Leibniz is attending at once to phenomenal matter and to a real monad. He holds that the former should be intellectually handled purely in terms of efficient quasi-causes or mechanism, the latter partly in terms of final causes or teleology:

Leibniz’s Two Realms

141

Souls act according to the laws of final causes, through appetition, ends and means. Bodies act according to the laws of efficient causes, or of motions. And the two realms [re`gnes], that of efficient causes and that of final causes, are in mutual harmony. (Mon 79/WFPT 279)

This two-realms doctrine, asserting the harmony of the mechanistic and teleological orders, is my topic in this and the next three sections. The final/efficient duality of realms risks being muddled with two other dualities. Each muddle is hinted at in a passage where Leibniz writes of the two realms: These realms are governed each by its own law, with no confusion between them, and the cause of perception and appetite is no more to be sought in the modes of extension than is the cause of nutrition and other organic functions to be sought in the forms or souls. (CT, G 4:391/L 409–410)

Of course perception and appetite, being properties of monads, could not be causally explained in terms of ‘‘modes of extension,’’ that is, of phenomena rather than basic reality. So they lie outside the range of ‘‘mechanism’’ when this is understood as the physics of bodily collisions. But the duality we are tracking is that of (1) efficient causality and final causality,

not that of (2) impact mechanics and something else.

The monadic status of perception and appetite does automatically put them on the right in contrast (2), but not in (1), which is my topic. On the face of it, as Adams pointed out, the causality within the monad belongs on the left of (1). The other possible muddle is the confusion of (1) with the duality of (3) necessity and freedom.

There are hints of that in the passage from which I have been quoting. Just before the bit last quoted, Leibniz writes: ‘‘Nature has, as it were, a dominion within a dominion, a double realm [regnum] of reason and necessity—that is, of forms and of material particles’’ (CT, G 4:391/L 409–410). The phrase ‘reason and necessity’ hints at (3): Leibniz does sometimes connect reason with freedom, and the link between ‘necessity’ and (3) is obvious. Less obvious, but perhaps more potent, are the implications of the phrase ‘‘a dominion within a dominion’’ (imperium in imperio). Spinoza had used this very phrase to characterize a supposed error: ‘‘They seem to conceive man in nature as imperium in imperio. For they believe that man disturbs rather than follows the order of nature’’ (Ethics IIIPreface/CWS 491). Now, Spinoza was defending (3) determinism against radical freedom, rather than (1) mechanism against teleology; but I don’t doubt that Leibniz steered for this collision, and his doing so suggests that he was at risk of collating (1) with (3).

142 Leibniz: Nature and Freedom

Are these two dualisms different? Well, on most reasonable understandings of what freedom is, it involves teleology. But the two are not equivalent: we have a clear enough idea of a creature that acts in accordance with what it believes will achieve certain goals, but which is utterly in thrall to its goals, having nothing that we could recognize as freedom. It is confusing, to say the least, for the realm of teleology to be labeled as the realm of freedom. Furthermore, for Leibniz as for many others, the freedom question is tangled with issues about the scope of deterministic causation, the latter being pointed to by his reference to reason and ‘‘necessity.’’ When ‘‘freedom’’ is looked at in this way, teleology is not necessary for it. Leibniz speaks of God (and sometimes even of humans, alas) as acting from sources that ‘‘incline but do not necessitate.’’ This offers freedom of a kind, but through a weakened, nondeterministic, efficient causality; it has nothing to do with teleology.

Hidden Harmony between the Two Realms Accepting for discussion’s sake that each monad has a teleological aspect, let us now ask: Does that mental teleology relate to the material organism in such a way that it too can be characterized in teleological terms? If Leibniz said no, he would be drawing a picture in which mental teleology does not spread to the material organism: the mind’s wants and goals have no bearing on the physical behavior of the organism. Either the mind cares only about itself, or it cares about the body too, and is nearly always frustrated. This repellent scene is not Leibniz’s. An ensouled animal (which he here calls a ‘‘corporeal substance’’) belongs to both realms, he says. Their laws are very different, he continues, but God ‘‘brings it about . . . that two very different series in the same corporeal substance respond to each other and perfectly harmonize with each other’’ (CT, G 4:391/L 410). This harmony is between teleological events in the mind and mechanistic ones in the body. Leibniz is plainly paving the way for the mind’s appetitions to have something to do with what happens in the body; presumably wanting to provide a metaphysical underlay for common beliefs about how goals and purposes are exhibited in behavior. How, then, are the two series harmonized? With what specific relation do we cash in ‘‘have something to do with’’? Leibniz gives one answer when he relates appetitive events to bodily ones through his general view that all events in one’s mind closely mirror events in one’s body. He usually states this in terms of perceptions or sensations: It is thoroughly reasonable that the effect should correspond to the cause; and how could one ever be sure that it does not, since we have no distinct knowledge either of the sensation of blue (for instance) or of the motions which produce it? It is true that pain does not resemble the movement of a pin; but it might thoroughly resemble the motions which the pin causes in our body, and might represent them in the soul; and I have not the least doubt that it does. (RB 131–132)

So far as this concerns color sensations, it has turned out to be right (Hardin 1988: 133). To speculate that it holds for all sensations of whatever kind, however, is bold indeed.

Leibniz’s Two Realms

143

Leibniz goes further still. He applies the thesis to our conative as well as to our cognitive and sensory states, and to regular accompaniments that are not even quasi-causes; and that yields an answer of a sort to my question about how mind ‘‘harmonizes’’ with body. Here is the clearest statement of it that I have found: There is an infinity of shapes and motions, present and past, which play a part in the efficient cause of my present writing; and there is an infinity of tiny inclinations and dispositions of my soul, present and past, which play a part in its final cause. (Mon 36/WFPT 272–273)3

This conjectures that the flurry of submicroscopic material events leading to my hand’s going up is mirrored by a dense cloud of mostly unconscious conative items—pros and cons, inclinations of my soul—whose resultant is that I choose to raise my hand. That is too wildly speculative to be a soberly interesting suggestion for how the two realms might harmonize. Anyway, it does not yield the kind of harmony that I am inquiring after. What Leibniz ought to want, and what his rhetoric seems to promise, is a metaphysic that underpins our ordinary common-sense attributions of goals, wants, and so on to human and other animals. The teleology that we are supposing to reside in the animal’s ‘‘dominant’’ monad ought to harmonize not with the neural goings-on in my body but with my observable behavior. That is the need that Leibniz should try to meet.

Teleological Patterns in Behavior Anything like a credible solution to this problem must avail itself of the idea of a teleological pattern in an animal’s behavior. As a first approximation, we can say that that is a pattern that makes the behavior at least prima facie a candidate for being brought under teleological concepts. Then the thesis of two harmonized realms will say that the real teleology that (we are supposing) is present in the appetitions of an animal’s mind or dominant monad matches the teleological patterns in its behavior. In a nutshell: its behavior will prima facie support the attribution to it of food as a goal when and only when its mind is the subject of a desire for food, an intention to get food, or the like. I have not found Leibniz writing explicitly about teleological patterns in animal behavior; but they have to be part of any story that captures our ordinary beliefs and assumptions about the goals and purposes of animals, including ourselves. Teleological patterns would not be needed if we settled for point-blank teleology. This involves the simplest possible relation of wants to movements: when an animal’s body undergoes a turn of the head or an opening of the mouth, pointblank teleology says that the animal wanted or intended to make that very movement. End of story. That is too flat and featureless to be of any use. One mark of its degeneracy is that although it would provide ‘‘explanations’’ of a sort—‘‘Why did he jump?’’ ‘‘Because he wanted to’’—it would not yield even the beginnings of a hint of a tentative prediction.

144 Leibniz: Nature and Freedom

A teleology worth having must be not point-blank but long-distance, providing for the idea that the animal moves thus in order to bring about some later state of affairs. For point-blank teleology, we have only to look to the given animal movement; it is automatic and easy. Long-distance teleology, however, requires us to pair a movement with some subsequent possible state of affairs; and to select the latter we need rules, criteria, principles, general theory. These will serve to delineate the teleological patterns in the behavior. In sketching these I shall borrow from previous publications of mine, but I do not claim ownership. The central ideas inform the work of most competent workers in this field today, though I have laid them out more explicitly than most (Bennett 1976). The needed principles rely on two ideas: that purposive behavior ordinarily reflects character-traits, purposes that the animal often has; and that a fair amount of purposive behavior succeeds. These two jointly yield the crucial notion of a teleological pattern of the form: For any action-kind A of which Animal is capable, whenever it is so situated that performing an A will lead to G, Animal performs the A.4

If Animal is so placed that digging would bring it food, it digs; if swimming would bring it food, it swims; if jumping would bring it food, it jumps. That prima facie qualifies Animal to count as a food-seeker. For a given animal there is a teleological pattern for each relevant value of G. For this to be realistic, the generalizations defining the patterns must be weakened, to allow for nonperformance because of competing goals, physical incapacity, satiety, and so on. Also, provision must be made for failures. These are ignored by the formula I have offered, which says that Animal does A whenever doing A would lead to G. The remedy for this also repairs another flaw in the account as so far given. The remedy is to change the teleological patterns from ‘‘. . . whenever performing an A will lead to G . . .’’ to ‘‘. . . whenever Animal thinks that performing an A will lead to G . . .’’ That allows for failure resulting from error about means and ends. It also heads off the complaint that it would be miraculous if Animal’s behavior were to depend on how it is situated rather than on how it thinks it is situated. (Leibniz would agree; see NI, G 4:505/WFPT 211.) Other fine-tuning is also needed, but this is not the place to go into all the details. That is what Leibniz needs if he is to round out his account of human teleology. In the ‘‘realm’’ of the organic machine there will be events that fall under teleological patterns of something like the kind I have sketched; in the ‘‘realm’’ of the dominant monad there will be corresponding appetitions. I have not found Leibniz explicitly laying out anything like the concept of teleological pattern that I have sketched. But there is no obstacle to his having it, and it is at least hinted at in things he writes, such as: ‘‘Volition is the effort or endeavour (conatus) to move towards what one finds good and away from what one finds bad, the endeavour arising immediately out of one’s awareness of those things’’ (RB 172). This, however, needs work. A teleological scheme in which the central idea is

Leibniz’s Two Realms

145

that Animal tries to do what it finds good is close to vacuous; it needs to be replaced, for each individual animal, by something that says what the animal ‘‘finds good’’; which is what my teleological patterns do.

Harmonizing Teleology with Mechanism Given teleological patterns in an animal’s behavior, and wanting to use them in explanations, what are we to do about the mechanistic explanations that we think are always appropriate? This is the problem of how to harmonize the teleological and mechanistic realms without letting the former carve off for itself any of the territory that would otherwise belong to the latter. It confronts all of us who think that teleological explanations can be legitimate although every event can be explained mechanistically. The best and probably only solution for it has informed the work of various philosophers for several decades, most notably that of Daniel Dennett, though I may have done most to make it explicit (in Bennett 1976: 72–78). Its crux is the idea that a teleological pattern, although each instance of it can be explained mechanistically, may not correspond to a single mechanistic explanation. For any value of A within the repertoire of this thermostat, whenever its doing A would bring the room’s temperature nearer to 68 degrees, it does A; yet we do not credit it with having a 68-degree room as a goal. What justifies our reluctance? The obvious answer, ‘‘It is an artifact, and so not alive, and so does not have goals,’’ goes too fast and throws no light. A better answer says that we do not attribute a thermal goal to the thermostat because the entire range of its behavior falling under the teleological pattern admits of a single mechanistic explanation: the gap between its temperature and 68 degrees causes a strip of metal to expand or contract, that opens or closes a switch, and that controls the flow of electricity which controls the temperature. That covers the entire teleological pattern. Because the teleological account of the thermostat’s pattern of behavior is matched by a single mechanistic account, the latter should prevail and we should reject the former as having no honest work to do (Taylor 1964: 29). Now consider an animal with a large repertoire of food-seeking behaviors, triggered by a range of different sensory clues: various sounds, smells, and sights lead it to dig, swim, run, climb, and so on. Each episode can be explained mechanistically in terms of how sensory inputs stimulate nerves, and how neural outputs make muscles contract; but the episodes employ different mechanisms; no unitary mechanism covers the whole range of them. So the teleological explanation in terms of food-seeking provides a grouping of episodes—a way of understanding the animal—that no mechanistic explanation delivers, and our concerns and interests may make that of value to us. Complete knowledge of the whole mechanism for each episode, even if we had it, would not exhibit the pattern that the teleological explanation yields; so the latter earns its keep by doing work that mechanism cannot do. For more about this, see Bennett 1976: 72–78; Dennett 1989. I cannot see any obstacle to Leibniz’s having these thoughts, and accepting them. Indeed, a unitary teleological pattern arising out of a swirling cloud of

146 Leibniz: Nature and Freedom

different mechanistic ones should strike him as a fine example of the kind of ‘‘harmony’’ that God in his greatness can achieve. But I have no evidence of Leibniz’s thinking any of this. The little he says on the harmonizing of the two realms is strikingly perfunctory, and is always based on the assumption that the teleological realm is mental and the mechanistic physical. This is typical: ‘‘It is not only the voluntary inner acts of our minds which follow from this conatus, but outer ones as well, i.e. voluntary movements of our bodies, thanks to the union of body and soul which I have explained elsewhere’’ (RB 173). How did we get into this situation? Leibniz was offering to harmonize mental teleology with the material behavior of animals (including humans); I said that for that we needed to know what sort of physical behavior is involved; and in developing that story I seem to have provided a sufficient basis for teleological concepts in animal behavior considered on its own, without bringing mental appetitions in at all. I did bring in what the animal believes; but the account could be developed in such a way as to extrude even that mental element; and in any case that is (in Leibniz’s terms) an appeal to the monad’s perceptions but not to its appetitions. What has happened? I cannot draw any interesting moral from this situation. It seems to me that Leibniz simply did not think hard about the essentially explanatory nature of teleological concepts. He had presumably some notion of the kind of behavior that could plausibly be handled teleologically—behavior falling into teleological patterns— and thought of instances of such patterns as merely accompanied by corresponding appetitions in the dominant monad. That is a plausible enough position. Someone who looked at a thorough account of animal teleology of the sort I have been sketching, and saw that it could draw the needed distinctions, yield the wanted predictions and explanations, and so on, might still think that something is missing. ‘‘For an animal’s movement to be explained by its having a certain goal,’’ he might say, ‘‘there must be an internally felt wanting or intending. The story you have told lacks the experienced heart of teleology, which resides in the mind and not in behavioral patterns.’’ We can see Leibniz’s two-realms story as an attempt to give that objector what he wants. It is a prima facie reasonable thing to want to do; but we should bear in mind that while it aims to supply teleology with its felt, experienced, inward aspect, it contributes nothing to its workings.

Teleology as Explanatory I now confess to a major gap in the teleological story that I have told; and I need to investigate whether monadic appetitions could fill it. As I have just implied, teleological concepts are nothing unless they are explanatory: identifying an animal’s goals is essentially a matter of discovering certain explanations for some of its behavior. Attempts to identify goals in a spirit of mere description, with explanation perhaps coming later, have always come to grief. (For some details see Bennett 1976: 42–46.) So far, I have provided nothing to secure that the patterns are explanatory rather than coincidental. Given that Animal conforms to a mouse-catching pattern, we shan’t attribute to it the goal of catching mice unless we think that on most

Leibniz’s Two Realms

147

occasions when it does what it thinks will give it a mouse it does so because it has that thought. The mere existence of that pattern is not enough; there must be something about it that makes it explanatory. (The account I gave in Bennett 1976 did not meet this requirement. I remedied the omission in Bennett 1991a, b.) One might think that monadic appetitions can enter at this point, playing a working role: ‘‘Leibniz can say that what makes it legitimate to explain the animal’s movements in terms of a mouse-catching goal is the existence at the same time of a toward-mouse-catching appetition in its dominant monad.’’ It is not clear how that would work. For the pattern to have explanatory force, one would have thought, is for it to be more than merely coincidental; and it is not clear how that is headed off by the existence of a certain mental event each time the pattern is instantiate. It might be different if the appetition functioned as the cause of the behavior; then the pattern would drop out of the picture, for the behavior would be causally explained purely by the occurrence of the appetition. But Leibniz will not say that, for it involves transeunt causation, which he firmly and deeply denies; and it involves giving a material event an explanation that supplants any mechanistic one, which he comprehensively condemns. What does solve the present problem has nothing to do with mentality. What is to be ruled out is the possibility that it is a sheer coincidence that this animal houses a multitude of mechanisms that jointly make it a mouse-getter. For the behavior caused by those mechanisms to be appropriately explained in terms of mouse-seeking, it must be the case that the animal contains that package of mechanisms because it has the overall effect of making the animal a mouse-getter. That could be the case for any of three reasons. (1) God’s purpose in building that package of mechanisms into the animal was to make it a mouse-getter. (2) The animal contains that package of mechanisms because they were evolutionarily selected for; their effect of making their owners mouse-getters was selected for. (3) The individual animal has assembled that package within itself through a process of learning what clues indicate mice and what behavior leads to catching them. I believe that (2) and (3) cover the ground. Leibniz, having not heard of (2) and being a theist, could and presumably would think that (1) and (3) between them exhaust the cases. That, so far as I can see, would not set up any special strains within his overall philosophy; but would still leave appetites idle.

Locke’s Theories of Motivation Leibniz apparently offers to provide for the conative side of the human condition by appetitions. I have argued that their teleological nature is dubious at best; the shape of their fit to bodily behavior is seriously underdescribed; and they are given no proper explanatory role. These deficits show up when in the New Essays Leibniz writes about the conative, motivational side of the human condition, doing so more fully and freely (I think) than anywhere else in his work. The prompt for this is the treatment of motivation in Locke’s Essay 2.21. Leibniz’s comments on this material are lively, energetic, and supremely intelligent; they are also friendly and collaborative, as

148

Leibniz: Nature and Freedom

befits his judgment that Locke here ‘‘makes especially evident the depth and penetration of his mind’’ (RB 164). Furthermore, they tap into Leibniz’s benevolence and decency: his interest in issues about motivation is energized by their bearing on people’s moral well-being (see RB 191). Leibniz is here focusing sharply, intensely, and with moral concern on the voluntary behavior of humans, putting himself under pressure to do the best he can to get straight about this matter. We shall find that these heartfelt pages draw no sustenance from his metaphysical doctrines—of monadic teleology and of the ‘‘two realms’’—and indeed that the pressure of what he wants to say drives him ever further away from those doctrines. He employs the key term ‘appetition,’ but the deeper he gets into his topic the less teleological appetitions become. Eventually, they have nothing to do with forethought, purposes, intentions, or the like; Leibniz explicitly divorces them from all that, which he hands over to a different faculty— reason. Properly to present all this, I must set the scene. Locke first held that a person always acts for what he thinks to be his own good, so that self-harming behavior always involves ignorance or error about consequences.5 He rightly came to doubt this, and offered a second theory that expands on the first by postulating a mediating link in the causal chain from valuejudgments to volitions: Good and evil, present and absent, ’tis true, work on the mind. But that which immediately determines the will from time to time in every voluntary action is the uneasiness of desire fixed on some absent good. (Essay 2.21.33)

To the thesis that A judgment about comparative good causes an act of the will, Locke now adds the further detail that It does so through an intermediate item, a state of uneasiness. Uneasiness acts reliably on the will, he holds, but the power of value-judgments to cause it is an unsteady affair. Someone who is sure that fornicating will bring him some transient pleasure yet cost him an extra decade in Purgatory may nevertheless proceed with his sexual adventure because the thought of that decade makes him less uneasy than does the thought of remaining chaste. Thus, akrasia does not have to be explained in terms of error about consequences. Locke regards his uneasiness theory as virtually standing to reason: ‘‘When a man is perfectly content with the state he is in, which is when he is perfectly without any uneasiness, what industry, what action, what will is there left but to continue in it?’’ (2.21.34). My trying to bring about some state of affairs S is intelligible (he argues) only if I am dissatisfied with the present non-S state of the world. My awareness that the nonobtaining of S is unsatisfactory to me is my uneasiness—my sense of something wrong—and in acting I attempt to cure it by making S obtain. This was a mistake on Locke’s part. It is simply not true that voluntary action is intelligible only as an attempt to improve on the present. Purposive action involves comparing a possible future not with the present but with other possible futures. (I have not been convinced by the defense of Locke and Leibniz against this criticism by Vailati 1990: 214–216.) This error of Locke’s is repeated, more subtly but quite definitely, in Leibniz’s commentary on him. His account of human activity is permeated by the Lockean idea that voluntary behavior is always remedial, always directed toward putting us

Leibniz’s Two Realms

149

at our ease. We shall see this helping him to move away from real long-term teleology toward a mere itch-scratching that is not clearly teleological.

Leibniz’s Response to Locke’s Theories Though he thinks that Locke dismisses his first theory too briskly, Leibniz agrees with the second: something like uneasiness mediates between value judgments and the will. But there is a difference. Where Locke thinks of episodes of uneasiness as fully present to consciousness, Leibniz says that they may be barely noticeable or even lie below the threshold of awareness. I shall use ‘‘disquiet’’ for the phenomenon Leibniz is talking about; he calls it inquie´tude, which Coste had used to render Locke’s ‘‘uneasiness,’’ but Leibniz says that that was a bad translation. Leibnizian disquiet, because it can be subliminal, plays a very different role from the one that Locke assigns to uneasiness. In the preceding chapter, before Locke has come to his theories of motivation, Leibniz ushers in disquiet, which he describes as ‘‘imperceptible little urges which keep us constantly in suspense.’’ (The implication that episodes of disquiet are all ‘‘imperceptible’’ is not his usual view about them.) He further characterizes the ‘‘little urges’’ or ‘‘disquiets’’ thus: These impulses are like so many little springs trying to unwind and so driving our machine along. . . . That is why we are never indifferent, even when we appear to be most so, as for instance over whether to turn left or right at the end of a lane. For the choice that we make arises from these insensible stimuli, which, mingled with the actions of objects and of our bodily interiors, make us find one direction of movement more comfortable than the other. (RB 166)

This aims to unify the entire range of human behavior—from your deliberated career choice right down to your crossing your legs without thinking about it. Here as elsewhere Leibniz gets long mileage out of his view that some mental states lie below the threshold of awareness. But there is trouble. In the next chapter Leibniz will openly identify the episodes of disquiet, the ‘‘impulses,’’ with appetitions (RB 190); those are features of monads, so that ‘‘these insensible stimuli’’ should belong to the mentalistic part of the story he is telling. Yet he speaks of them as ‘‘mingling’’ with bodily events: at the end of the lane I turn left because of insensible mental stimuli mingled with a slight breeze blowing that way and a faint pain in one knee. I cannot see what sort of ‘‘mingling’’ this could be, consistent with Leibniz’s metaphysic. A worse trouble, alluded to earlier, is that the picture Leibniz is drawing here does not seem teleological; it is too much like mere itch-scratching. That is even more true of what he writes next: In German, the word for the balance of a clock is Unruhe—which also means disquiet; and one can take that for a model of how it is in our bodies, which can never be perfectly at their ease. For if one’s body were at ease, some new [event] would at once alter the balance and compel those parts of the body to exert some tiny effort to get back into the best state possible; with the result that there is a perpetual conflict which makes up, so to speak, the disquiet of our clock. (RB 166)

150 Leibniz: Nature and Freedom

This charming passage has nothing teleological about it. But then it is explicitly about ‘‘the body’’; so one might think that Leibniz means here to be writing about only one of the two realms. But not so. In the following chapter, NE 2.21, Leibniz openly identifies disquiets with appetitions, and says that they ‘‘stimulate the will’’ (RB 189). All of this, then, falls within the ambit of the ‘‘two realms’’ doctrine, and ought to be anchored in Leibniz’s basic metaphysic of teleology. However, it is not. That metaphysic says that each monad falls within a realm of final causes, but all Leibniz is offering us here are spurs, springs, imbalances—efficient causes. Here is the climax: These appetitions, whether small or large . . . are truly the first steps that nature makes us take; not so much towards happiness as towards joy, since in them one looks only to the present, but experience and reason teach us to govern and moderate them so that they can lead us to happiness. . . . Appetitions are like a stone’s endeavour to follow the shortest but not always the best route to the centre of the earth; it cannot foresee that it will collide with rocks on which it will shatter, whereas it would have got closer to its goal if it had had the wit and the means to swerve aside. In the same way, by rushing straight at a present pleasure we sometimes fall into the abyss of misery. That is why reason opposes appetition with images of greater goods or evils to come, and with a firm policy and practice of thinking before acting and then standing by whatever is found to be best. . . . (RB 189)

Although twenty pages earlier Leibniz seemed to make disquiet drive the whole range of human motivation, he now assigns the ‘‘spurs’’ and appetitions to the stupid, obsessed, thoughtless part of our conative nature. In fact, he has been implicitly doing this all through, but now it comes to the surface because the spurs—now openly called ‘‘appetitions’’—are contrasted with another element in the human condition, namely ‘‘reason.’’ Whereas in appetition ‘‘one looks only to the present,’’ reason brings a concern for what is ‘‘to come’’: reason is forwardlooking, and able to control appetitions, redirecting them and perhaps lessening their force. The teleological load in Leibniz’s account of the human condition, we now find, is carried purely by ‘‘reason.’’ Let us see whether Leibniz has prepared reason for this task, and what task it is. I give these a section apiece.

The Place of Reason in Leibniz’s Metaphysic Whereas all monads have appetite, in Leibniz’s scheme of things, only high-grade ones have reason. Those high-grade monads are marked off by having more distinct perceptions than do the rest: more of their perceptions are distinct, or their perceptions are on average more distinct, or . . . It seems that Leibniz never fine-tunes this doctrine, but always leaves it vague: ‘‘Monads . . . are limited and differentiated by their level of distinct perception’’ (Mon 60). Nor does he explain in fundamental terms what the distinctness of a perception consists in. The best attempt I know to fill this gap for him is Brandom (1981), and even that is open to question (see Bennett 2001: sec. 125). Having noted these two lacunae, let us press on: let us take it that Leibniz selects a subset of monads in terms of something called distinctness of perceptions, and see what he does with that.

Leibniz’s Two Realms

151

He thinks it is the source of three other differences. High-grade monads differ from the rest, he maintains, in having (1) memory, (2) (more) awareness of their own states,6 and (3) reason. (1) Leibniz holds that every monad at every moment contains traces from which its whole past could in principle be read off; and to have a memory, properly so-called, is to be reflectively aware of such a trace. The ‘‘insensible perceptions,’’ of which one is not aware, merely ‘‘preserve the seeds of memory’’ (RB 239–240). So (1) memory is drawn in by (2) reflective awareness, and does not need separate discussion. (2) Leibniz sometimes ties our awareness to distinctness, writing that we are unaware of many of our perceptions because they are not distinct enough. He links this theme with his further thesis that perceptions may escape our notice or reflective awareness because they are ‘‘too minute’’ (RB 164), are ‘‘not forceful enough’’ (RB 54), go by too fast (see RB 403), or are too uniform: There is in us an infinity of perceptions, unaccompanied by awareness or reflection; that is, of alterations in the soul itself, of which we are unaware because these impressions are either too minute and too numerous, or else too unvarying, so that they are not sufficiently distinctive on their own. (RB 53)

So we have Leibniz maintaining that we are aware of our perceptions to the extent that they are large or forceful, conspicuously varied, and fairly slow-changing. I do not think that these three features define ‘distinct’ for him, but they accompany distinctness; perceptions that lack them Leibniz often calls ‘‘confused.’’ (3) When he writes about reason as a faculty, Leibniz often calls it ‘‘intellect’’: ‘‘We should undertake the analysis of them . . . by reason in so far as they can be made more capable of being treated by the intellect’’ (RB 120); ‘‘These depend upon intellectual truths, grounded in reason’’ (RB 444). Also, he assigns to ‘‘intellect’’ matters such as the exploration of necessary truths, which he also assigns to ‘‘reason,’’ as in his reference to ‘‘[I]ntellectual ideas, from which necessary truths arise’’ (RB 81). Now consider this: We are aware of many things, within ourselves and around us, which we do not understand; and we understand them when we have distinct ideas of them accompanied by the power to reflect and to derive necessary truths from those ideas . . . . ‘Understanding’ in my sense is what in Latin is called intellectus, and the exercise of this faculty is called ‘intellection,’ which is a distinct perception combined with a faculty of reflection, which the beasts do not have. Any perception which is combined with this faculty is a thought, and I do not allow thought to beasts any more than I do understanding. So one can say that intellection occurs when the thought is distinct. (RB 173)

Any intellectual activity (reasoning) involves thoughts, and Leibniz says that to have a thought is just to have a distinct perception and a reflective awareness of it.7 Analogously, to do carpentry one needs sharp tools and a good grip on them. Of course there is more to carpentry than that; having firmly grasped a fine chisel one should then do something with it; and Leibniz presumably also held that for reasoning or intellection one must operate in certain ways on the distinct perceptions/ideas of which one is reflectively aware. But he does not say anything

152

Leibniz: Nature and Freedom

about what those activities are—or, anyway, he does not relate them to the terms of his basic metaphysic. I cannot find him explaining why it is that the monads that meet some standard of distinctness of perceptions are all and only the ones that have reason.

Reason as Controller of Desire What role does Leibniz assign to reason in the conative life of a thoughtful person? Even if we cannot get from him a satisfactory rooting of the concept of reason in his basic metaphysic, we can still ask this question about how he situates it in his picture of the human condition. Such answers as I can find come mostly from the New Essays; I have found little in other works. It may be symptomatic that there is no entry for ‘‘reason’’ in the indexes of AG, MP, Mates (1986), Wilson (1989), Ross (1984), Jolley (1995), or Sleigh (1990). Adams (1994) and L have only four between them, none relevant to my present topic. The best I can offer are four ideas, of which two are little more than tentative suggestions. (1) Discussing ways of using some desires to block others, Leibniz writes that without a certain technique one could not resist certain desires ‘‘in any free and voluntary way in which reason could play a part’’ (RB 195). When he discusses how we can do this, offering ‘‘methods and stratagems’’ for maintaining conative hygiene, he presents only one clear way in which ‘‘reason could play a part’’ in it. One of the stratagems—the ‘‘best of all’’—is to ‘‘become accustomed to proceeding methodically and sticking to sequences of thoughts for which reason, rather than chance (i.e., insensible and fortuitous impressions), provides the thread’’ (RB 196). This brings in the contrast between ‘‘empirics’’ and disciplined scientists, which Leibniz often aligns with what separates experience and memory from ‘‘reason’’ (Mon 28; PNG 5). He is apparently saying that our desires and thus our behavior will be better if our habits of thought are rooted in well-tested scientific theories rather than in random anecdotes. He may be right; but this invocation of ‘‘reason’’ does not take us far; and it contributes nothing to my theme of reason’s role in the teleological aspect of the human condition. (2) Sometimes Leibniz takes reason to be a controller: the conative side of a mature and thoughtful person involves appetite governed by reason. I have a suggestion about how that might be. Associating reason with reflective self-awareness, and thus with mental episodes that are not defeatingly small, fast, or lacking in variety, Leibniz may think that my bringing reason to bear on my appetitions involves their standing out from their mental surroundings, so that I can become aware that I have them and aware of what they are. ‘‘We think of many things all at once, but pay heed only to the thoughts that stand out most distinctly’’ (RB 113). He also holds that something of which we are not aware can be brought to our awareness retrospectively: When we are not alerted, so to speak, to pay heed to certain of our own present perceptions, we allow them to slip by unconsidered and even unnoticed. But if someone alerts us to them straight away, and makes us take note, for instance, of

Leibniz’s Two Realms

153

some noise which we have just heard, then we remember it and are aware of just having had some sense of it. (RB 53–54)

This could be applied to appetitions as well, in the manner of some current styles of psychotherapy, in which control of one’s drives is sought partly through becoming sharply aware of them. Leibniz was an acute observer of the human psyche, and it would be in character for him to have some such thing as that in mind. He wrote this to Bayle: The soul does many things without knowing how it does them—when it does them by means of confused perceptions and unconscious inclinations or appetitions, of which there are always an extremely large number, so that it is impossible for the soul to be conscious of them, or to distinguish them clearly. (G 4:550/ WFPT 238)

Leibniz elaborates this as it applies to perceptions. When we enjoy a musical sound, he writes, we are responding to some facts of which we are not distinctly aware about the rate of vibration in the instrument; and when physics and music theory enable us to know the facts, ‘‘reasoning has to come to our aid’’ (G 4:551/ WFPT 238). He does not return to appetitions in this context; but I guess that he thinks reasoning can come to our aid with those too. That, though, would not help much with my theme of teleology. It gives reason a role in positioning you to make better, more informed decisions about which of your drives and desires to go with, and which to discourage. This contribution of reason is over before the teleological work begins.

Reason as a Source of Moral Guidance (3) Leibniz frequently says that reason is an instrument for the discovery of truths about morals and value. ‘‘To speak against reason is to speak against the truth, for reason is a chain of truths. This is to speak against oneself, and against one’s own good, since the principal use of reason consists in knowing the good.’’ (RB 199). ‘‘Justice, taken quite generally, is nothing other than goodness in conformity with reason’’ (PNG 9/WFPT 262–263). I have not found him explaining how reason serves in moral discovery, but I shall not press that point here. Moral knowledge obviously has a place in the life of one’s desires. In the interplay among my various desires and inclinations, my knowledge or beliefs about what would be good or right makes a difference. So that is another way in which reason can enter the conative picture. Writing in one place about what we need if we are ‘‘finally to gain control both of our passions and of our insensible inclinations, or disquiets’’ (RB 188), Leibniz depicts reason not as controlling but as guiding, not holding the tiller but displaying the compass. He writes that we should acquire ‘‘the custom of acting in conformity with reason,’’ and so ‘‘we must make this rule for ourselves once and for all: wait until you have the findings of reason and from then on follow them.’’ That could give reason a working role in our lives; but once again it leaves the teleological aspect dangling. A person may decide to act thus rather than so, taking

154 Leibniz: Nature and Freedom

into account everything she knows about what that will involve—pleasure, misery, social success or failure, moral success or failure. Reason-as-moral-compass supplies her with important material to take into account, but it does not enter into the deciding, the teleological procedure in which someone acts because of a thought about a possible future. (4) If moral knowledge or belief is not merely a cognitive matter but has something conative built into it, the picture changes. On that view, accepting a moral proposition is not (only) believing or knowing something but essentially involves endorsing an imperative addressed to oneself. That is a conative state: to accept a moral proposition is to be inclined to behave thus rather than so. If reason can give me the moral ‘‘knowledge’’ that essentially includes such a disposition, then it is after all woven tightly into the conative side of my nature. Leibniz sometimes wrote as though he took that view of moral principles. He may be doing so in a passage I have already quoted: ‘‘In these ‘appetitions’ . . . one looks only to the present; but experience and reason teach us to govern and moderate them so that they can lead us to happiness’’ (RB 189). If reason teaches us to govern our appetitions, then it is a source not merely of moral knowledge but of moral injunctions. However, the matter is not clear, because the French could mean that reason teaches us how to govern our appetitions, and that would fall under option (3) rather than (4). Other passages, though, unambiguously point to (4) reason as inherently conative: ‘‘Unless appetite is directed by reason, it endeavours after present pleasure rather than that lasting pleasure which is called happiness’’ (RB 199–200). If reason can ‘‘direct’’ appetite, that is presumably because the moral propositions that reason lets us discover have imperative force. There is further evidence in something that I chopped off the end of a previous quotation: ‘‘. . . the principal use of reason consists in knowing the good and pursuing it.’’ I cannot understand this except as implying that moral knowledge (or belief ) involves conative attitudes. The same thing is implied when Leibniz writes of the ‘‘inclinations which reason gives us’’ (RB 194). I do not quarrel with the thesis that any moral proposition incorporates an imperative, so that accepting it involves adopting a conative attitude. (For sophisticated developments of this overall position see Gibbard 1990; Hare 1952, 1963, 1981.) But most of those who have adopted a position of this kind have associated it with a certain subjectivism about the nature of morality, which has led them to find its human roots in feelings rather than in reason. They have indeed been pushed that way by their inability to devise a credible epistemology for morality if it is taken to be objective and factual. Someone who holds as Leibniz does that reason can discover objective moral truths, and who also holds (as I am conjecturing that Leibniz also does) that accepting such a truth involves a conative element, has a lot of explaining to do. I have not found that Leibniz does any of it. R. M. Hare, uniquely among moral philosophers who accord to moral propositions a kind of subjectivity and hold that accepting such a proposition is like accepting an imperative, maintains that morality can be established a priori. His argument for this is a vivid tour de force that seems not to have converted anyone. It would be astonishing, to say the least, if Leibniz ever entertained anything like it.

Leibniz’s Two Realms

155

Notes 1. In addition to those common to this book, the following abbreviations are used to refer to Leibniz’s works. CT: Critical Thoughts on the General Part of Descartes’s Principles (G 4:354–392/L 383– 412); NI: On Nature Itself (G 4:504–516/WFPT 209–222); PAB: A Physicist against Barbarism (G 7:337– 344/AG 312–320); SD: Specimen of Dynamics (GM 6:243–254/WFPT 153–179); TA: Tentamen Anagogicum (G 7:270–279/L 477–485); UO: On the Ultimate Origination of Things (G 7:302–308/AG 149–155). 2. For Leibniz (sometimes) and his predecessors, an a priori explanation of P explains why it is the case that P, whereas an a posteriori one merely provides reason for thinking that P is true. 3. Something similar appears at RB 56. Leibniz there seems to assign the conative role to perceptions, but I suppose he means changes of perception, i.e. appetitions. 4. I owe the idea of such a generalization to Taylor 1964. My most recent development of it is in Bennett 1991b. 5. This theory occupies eleven sections that run across the bottoms of pp. 248–273 in the Nidditch edition. The theory with which Locke replaced it occupies Essay 2.21.28–59. 6. ‘Awareness’ is the correct translation for Leibniz’s ‘apperception.’ To use ‘apperception’ as an English word is historically insensitive and linguistically absurd. For details see Bennett 2001: sec. 113. 7. He speaks of distinct ‘‘perceptions’’ and of distinct ‘‘ideas’’; and, although I agree with Brandom (1981: 453–459) that often Leibniz took these to be different concepts, he usually runs them together when, as here, he is trying to separate reason or intellect from the other capacities of the mind.

8

Leibniz on Spontaneity DONALD RUTHERFORD

hilosophers have long held that it is a necessary condition for acting freely that

P an agent act spontaneously, or that she determine herself to act, as opposed to

being determined by external causes. Only if an action can be attributed to an agent’s will as its cause can the action be judged to have been performed freely. The condition of spontaneity, however, can be understood in a variety of ways corresponding to the different kinds of constraint that may be thought to restrict the operation of an agent’s will. Most weakly, spontaneity may require only the absence of physical constraint. An agent can be judged to have acted of her own will, provided she has not been determined to act by external physical causes, for example, she has not been pushed or drugged. More restrictively, spontaneity may demand also the absence of psychological constraint. Traditionally, this has been associated with cases of coercion, where an agent is forced to act against her will because of fear induced by a threat. But the idea of psychological constraint can be generalized to include the force of any passion (grief, anger, lust) that disrupts an agent’s ability to exercise her will. For Spinoza, the absence of such constraint is both a necessary and sufficient condition for freedom, although he also believes that the possession of this freedom is consistent with the determination of an agent’s actions by prior causes.1 The last claim is challenged by Kant, who argues that the freedom required for morality presupposes that an agent’s will possess an ‘‘absolute spontaneity,’’ defined as the absence of all causal constraint. This strongest form of spontaneity consists in the will’s independence from all determination by prior causes, whether those causes be external physical events or internal mental representations.2 On any weaker construal, Kant writes, the consciousness of spontaneity, if taken as freedom, would be a ‘‘mere delusion’’ (Ak 5:101). The preceding range of views provides a starting point for understanding Leibniz’s account of spontaneity. Leibniz defends spontaneity as a necessary condition for freedom, and defines it generally in terms of the absence of constraint.3 At the same time, he extends the scope of spontaneity in a way that many have found 156

Leibniz on Spontaneity

157

puzzling. According to Leibniz, every substance is the sole cause of all of its own states; thus, every change that occurs in it occurs spontaneously. On the face of it, this may seem close to Kant’s view that any meaningful notion of spontaneity must be absolute: complete independence from the causal order of nature. In fact, however, the positions of the two philosophers are sharply opposed. Although Leibniz believes that any substance is causally isolated from every other substance, he insists that each of a substance’s states is causally determined by prior states of it. Thus spontaneity, for Leibniz, does not imply the absence of causal determination.4 This is highlighted in his famous image of the monad as a ‘‘spiritual automaton’’ (T 52), which Kant scornfully dismisses as leaving us with nothing more than the ‘‘freedom of a turnspit’’ (Ak 5:97). The division between Leibniz and Kant on the issue of spontaneity hinges on difficult questions about the sort of freedom that morality demands and about what can plausibly be claimed on behalf of the will’s freedom. I do not intend to enter into these disputes here. My concern rather is with a problem internal to Leibniz’s account of spontaneity and its relation to his conception of freedom. The problem is one that arises because of Leibniz’s intention both to uphold central features of traditional treatments of spontaneity—particularly spontaneity as absence of physical constraint—and to extend the scope of spontaneous action to include every change in the state of a substance. The former commitment implies that there is a salient distinction between cases in which an agent acts spontaneously, and hence may be judged to have acted freely, and cases in which an agent cannot be judged to have acted freely because the condition of spontaneity has not been met. Yet, if every change in a substance occurs spontaneously, then it seems that such a distinction cannot be drawn. Hence Leibniz fails to offer a satisfactory account of spontaneity’s relation to freedom. In this essay I reconstruct Leibniz’s theory of spontaneity in a way that makes clear his ability to respond to this objection. The crucial step involves distinguishing what I call ‘‘monadic spontaneity’’ and ‘‘agent spontaneity’’—the latter of which requires the absence of physical constraint. A second objective is to understand why Leibniz attaches such importance to the notion of monadic spontaneity in the analysis of freedom. Although the thesis of monadic spontaneity appears at first glance to rule out the possibility of constraint, I argue that the thesis in fact supports what Leibniz takes to be the only adequate explanation of this phenomenon. In the next section I sketch the distinction between Leibniz’s two notions of spontaneity and the different roles they play in his theory of action. Thereafter I examine the two notions of spontaneity in greater detail, showing finally how they are brought together by Leibniz in a single theory capable of accounting for both the will’s freedom and its constraint by external causes.

Two Notions of Spontaneity One of Leibniz’s best known explications of the concept of freedom appears in Theodicy 288:

158 Leibniz: Nature and Freedom Freedom, such as is required by the schools of theology, consists in intelligence, which includes a distinct knowledge of the object of deliberation; in spontaneity, in virtue of which we determine ourselves; and in contingency, that is, in the exclusion of logical or metaphysical necessity. Intelligence is, as it were, the soul of freedom, and the rest is as its body and basis. The free substance determines itself by itself, and this according to the motive of the good perceived by the understanding, which inclines it without necessitating it; and all the conditions are contained in these few words. (G 6:288)

Leibniz isolates three conditions as individually necessary and jointly sufficient for freedom: intelligence, spontaneity, and contingency. He labels the first condition the ‘‘soul of freedom.’’ Intelligence is the power of mind by which rational beings are distinguished from other substances, and rational minds alone possess freedom. Further, the free acts of such minds are limited to those in which choice is directed by a distinct knowledge of the object of deliberation. The other two conditions, spontaneity and contingency, Leibniz describes as the ‘‘body and basis’’ (le corps et la base) of freedom. The latter requirement, specified as ‘‘the exclusion of logical or metaphysical necessity,’’ is an obvious condition on freedom. In addition, Leibniz insists that an action is free only if it is one in which ‘‘we determine ourselves.’’ How precisely to understand this condition is left unclear in the passage. If, as Spinoza holds, to determine oneself to act is to be ‘‘led by reason’’ (CWS 584), then one might question whether spontaneity warrants inclusion as a separate condition on freedom. If one determines oneself to act only if one acts rationally, then one acts spontaneously only if one’s choices are guided by intelligence. However, this is not Leibniz’s position. The final sentence of the quoted passage suggests, instead, that he sees himself as offering three independent conditions on freedom: ‘‘The free substance determines itself by itself [spontaneity], and this according to the motive of the good perceived by the understanding [intelligence], which inclines it without necessitating it [contingency].’’ Although a free substance ‘‘determines itself by itself . . . according to the motive of the good perceived by the understanding,’’ there is no implication that this is the only way a substance can determine itself or act spontaneously. This is confirmed in dramatic fashion two sections later in the Theodicy, when Leibniz claims that, strictly speaking, every action of a substance, whether or not that substance is a rational mind, occurs spontaneously: [W]hen it is a question of explaining oneself precisely, I maintain that our spontaneity suffers no exception and that external things have no physical influence on us, speaking with philosophical rigor. In order to better understand this point, we must realize that a genuine spontaneity is common to us and all simple substances, and that in the intelligent or free substance this becomes a dominion over its actions. . . . (T 290–291/G 6:289)

In the case of substances endowed with intelligence, there is a distinction to be drawn between free and unfree acts; however, this distinction is not drawn on the basis of the spontaneity of those acts, for, ‘‘speaking with philosophical rigor,’’ every change in every substance is a consequence of its spontaneous action.5

Leibniz on Spontaneity

159

This is no trivial commitment on Leibniz’s part. It is the metaphysical bedrock of his system of preestablished harmony. Given this, one may wonder whether Leibniz has anything interesting to say about spontaneity as it relates to the understanding of freedom. In Theodicy 288, Leibniz groups spontaneity and contingency together as the ‘‘body and basis’’ of freedom. This suggests conditions that are presupposed by freedom but are not themselves the ground of the distinction between free and unfree acts; that role is reserved for intelligence, the ‘‘soul’’ of freedom. Contingency clearly qualifies as such a condition. If it were logically or metaphysically necessary that an agent act in a certain way, then she could not be said to have acted freely. Likewise, if an agent does not act spontaneously—if she does not determine herself to act but instead is determined to act by external causes—then she cannot be said to have acted freely. At this point, however, Leibniz makes what seems an unwarranted leap to the doctrine of monadic spontaneity: the view that any substance is the sole cause of all its own states. In doing so, he appears to overlook a narrower notion of spontaneity—agent spontaneity—that is integral to the understanding of freedom. Spontaneity in this more restricted sense is intended to mark the contrast between changes that we produce as agents and changes that occur as a result of external causes acting on us. Given the thesis of monadic spontaneity, however, it seems that we cannot even draw this distinction, for, according to that account, all of a substance’s active and passive states are produced by its own actions.6 Many of Leibniz’s statements about spontaneity leave him open to this criticism. In invoking the doctrine of monadic spontaneity, he seems to have opted for the wrong notion of spontaneity, one that is not directly implicated in the understanding of freedom. The relevant condition on freedom is one that defines it in terms of the absence of external causes that could have preempted the operation of the will, not one that renders constraint by external causes impossible. A further line of argument nonetheless leads Leibniz to conclude that monadic spontaneity is the ‘‘body and basis’’ of freedom. As he sees it, our ability to draw a distinction between constrained and unconstrained action presupposes an answer to a fundamental metaphysical problem. To explain what it means to be constrained by external causes requires an account of how the soul’s volitions can be resisted by its body and by the actions of external bodies on it. Similarly, the explanation of unconstrained action requires an account of how the soul’s volitions can be effective in altering the states of its body and through it the states of other bodies. In Leibniz’s view, no previous philosopher has offered a satisfactory explanation of these phenomena, and no explanation can be given, provided that we take them to involve a real causal interaction between substances. According to Leibniz, the distinction between constrained and unconstrained action can only be rendered intelligible in terms of the hypothesis of preestablished harmony, which itself rests on the doctrine of monadic spontaneity. Thus, although agent spontaneity may be a requirement of freedom, the possibility of agent spontaneity presupposes the truth of monadic spontaneity. Therefore, monadic spontaneity is properly the ‘‘body and basis’’ of freedom. The hypothesis of preestablished harmony affirms the causal independence of soul and body: ‘‘everything happens in the soul as if there were no body, just

160 Leibniz: Nature and Freedom

as . . . everything happens in the body as if there were no soul’’ (G 4:560/WF 113). The foundation of this hypothesis is the doctrine of monadic spontaneity, which maintains that any substantial form, or ‘‘entelechy,’’ is causally sufficient for the production of all its own states.7 On the basis of this principle, Leibniz extends the scope of a soul’s spontaneous action to encompass all of the changes that occur in it, including states in which it represents itself as acted on by physical causes: [E]veryone who accepts immaterial indivisible substances attributes to them a simultaneous multitude of perceptions, and a spontaneity in their reasonings and their voluntary acts. I am therefore only extending that spontaneity to their confused and involuntary thoughts, and showing that their nature is to contain relations with everything that is external. (G 4:564/WF 118)

As much as its voluntary acts, states in which the soul represents itself as constrained by physical causes come about through its own action, independent of any action on it by external things. What we take to be instances of the interaction of soul and body are, from the perspective of the soul, explained entirely in terms of the determination of the soul’s representational states by prior states of it.8 Strictly speaking, then, there is ‘‘no constraint in substances except in external appearances’’ (G 4:558/WF 111). Yet just because constraint is not grounded in facts about real causal relations between substances, it does not follow that it can or should be ignored. The significance of this point goes beyond Leibniz’s stated goal of preserving ordinary ways of speaking.9 The distinction between constrained and unconstrained action is indispensable for our conception of ourselves and other creatures as agents who possess a will. Not being God, finite creatures represent their agency as effected through a body, which can resist their will, either directly or through the action of other bodies on it. The category of constraint expresses our conception of such hindrances or failures of agency, just as the category of unconstrained or spontaneous action expresses our conception of ourselves as acting of our own will. As Leibniz draws the distinction in the New Essays: [W]e can take to be its ‘action,’ and attribute to the substance itself, any change through which it comes closer to its own perfection; and can take to be its ‘passion,’ and attribute to it an outside cause (though not an immediate one), any change in which the reverse happens; because the change can be explained in an intelligible way by reference to the substance itself in the former case and by reference to outer things in the latter. (NE 2.21.72/RB 211)

Regardless of how the soul’s states are determined causally, the concepts of agency and constraint (or, more generally, action and passion), remain basic to how we explain what we do and what happens to us. Leibniz sees it as a strength of his theory that, in the strictest sense, the soul’s actions are not constrained by any external causes. Nevertheless, saying just this leaves us without a way of conceptualizing what we take ourselves to be capable of doing as agents. It is this notion of agent spontaneity that is presupposed by the analysis of the will’s freedom, and for this reason the contrast between it and constrained action is indispensable to us.10 Leibniz’s considered view, therefore, is that the notion of spontaneity must be understood in two distinct senses, corresponding to what I have called ‘‘monadic

Leibniz on Spontaneity

161

spontaneity’’ and ‘‘agent spontaneity.’’ The doctrine of monadic spontaneity articulates a basic metaphysical truth: any substance is self-determining in the production of all its own states. At the same time, Leibniz allows that, among these states, we can distinguish those in which a substance operates as an agent promoting change in the world, relative to its own perspective and that of other substances (including God), and those in which it operates as the passive recipient of the effects of external causes, relative to its own perspective and that of other substances (including God). Both of Leibniz’s notions of spontaneity raise significant interpretative issues. As Leibniz understands it, monadic spontaneity is defined by a kind of causation in which states of a monad determine, without necessitating, subsequent states of it. In the next section, I present an account of the dynamics of such change that aims to demonstrate how monadic spontaneity is supportive of both the contingency of change and the effectiveness of individual monadic states in producing change. In subsequent sections I explore the resources Leibniz has for developing a conception of agent spontaneity within the framework defined by monadic spontaneity.

Monadic Spontaneity and the Contingency of Change Leibniz relies on a general model of change that he formulates early in his career and retains as an essential component of his theory of monads. The model makes two significant assumptions: (1)

Within nature, all real change is modal change, i.e., change in the states or modifications of substances. Once created, substances endure through all change and can be destroyed only by God. (2) Any nonmiraculous change in a substance’s states can be explained in terms of the action of that substance, without reference to any other created substance.11

Leibniz has various things to say on behalf of these assumptions, but here I propose to take them as entry points to his discussion of change. The second assumption introduces the idea of the causal isolation of substances, which forms the backbone of the theory of preestablished harmony. Leibniz’s account of change is intended to explain how this demand can be met. His critique of intersubstantial causality— in particular, physical influx—is meant to show that there is no other coherent explanation of modal change. Change, according to one of Leibniz’s definitions, is ‘‘an aggregate of two contradictory states’’ (A 6.4:556), that is, two states of the same substance that exclude each other. In another study from the 1680s, he writes: ‘‘A state of a thing exists if there is any true contingent proposition whose subject is the thing. Change is an aggregate of contradictory states’’ (A 6.4:569). By itself, however, this definition appears inadequate, since it includes no notion of one state following from another. Accordingly, Leibniz amends the definition to read: ‘‘Change is an aggregate of two contradictory states, from one of which the other follows immediately; by reasoning, I understand that it follows immediately, i.e., such that no other like state intervenes’’ (A 6.4:569).

162 Leibniz: Nature and Freedom

Change thus involves both a pair of contradictory states and an ordering relation that defines one state as the successor of the other. The idea of one state following from another is subject to further analysis in other texts from the period. The result is a general account of the succession of states in a substance: Of two contradictory states of the same thing, that one is prior in time, which is prior by nature, or which involves the ground [rationem] of the other, or, what is equivalent, which is understood more easily [facilius intelligitur]. Just as in a clock, in order to understand perfectly the present state of the gears, it is required that we understand the ground of this state, which is contained in the preceding state, and so on. And the same is true in any other series of things, for there is always a definite connection, though it is not always necessary. (A 6.4:563)

As part of his explanation of change, Leibniz proposes an analysis of the temporal ordering of states in terms of their causal ordering, which he casts as an ordering on the basis of ‘‘grounds.’’ Of two contradictory states, one is temporally prior to another only if the former is ‘‘prior by nature’’ to, or contains the ground of, the latter. Leibniz interprets this grounding relation as both an ontological and an explanatory relation: the grounding state contains what is sufficient to bring about and to render intelligible the existence of the state which it grounds. In this sense, the grounding state is a ‘‘predeterminant’’ of the grounded state.12 As Leibniz’s example makes clear, the ordering relation does not have to be one in which a given state logically entails its successor. The relation may be only hypothetically necessary, or necessary relative to a certain ‘‘mode of producing’’ (A 6.4:872). Thus, the present state of the clock’s gears does not logically entail a unique subsequent state but does so only relative to certain laws of motion. This point is crucial to Leibniz’s defense of the contingency of change within the created world. The preceding analysis remains at a high level of abstraction. It is brought down to earth when Leibniz specifies the nature of the ground responsible for the succession of a substance’s states. Another version of his definition of change reads: ‘‘change [transitio], or variation, itself . . . is nothing but a complex of two states which are immediate and opposite to each other, together with a force or [seu] reason for the change, which reason itself is a quality’’ (C 9/MP 134). Here the ground or reason for change is identified as a force, which is described as a quality, or ‘‘intrinsic accidental denomination,’’ of substance (C 9/MP 134). At this point Leibniz’s account of change is united with his theory of substance. According to Leibniz, every substance is necessarily active, or a source of change. On its weaker reading, this thesis implies that whatever is a substance is not purely passive but has within it a force or power by which it acts to bring about changes in its own states. The stronger version of the thesis—monadic spontaneity—adds the claim not only that a substance is causally active but that the force it contains is sufficient to determine each of its successive states. Thus, monadic spontaneity implies that in all of its changes a substance is self-determining: it has the particular states it does by virtue of acting on itself and not by virtue of any other created substance acting on it.13 Several distinct lines of argument converge on this thesis. In writings from the 1680s, Leibniz stresses the ‘‘completeness’’ of the nature of any substance, the fact

Leibniz on Spontaneity

163

(defended by considerations of individuation and divine foreknowledge) that part of what it is to be an individual substance is to possess a nature that includes the ground of everything that will ever be true of that substance. Coupled with the proposition that any substance is by nature active, or a source of change, this implies that any substance is causally self-sufficient: capable of bringing about all of its own states, as if only it and God existed (DM 14). Later, in the Monadology, Leibniz presents the thesis differently. He writes that he will ‘‘take for granted that every created being, and consequently the created monad as well, is subject to change, and even that this change is continual in each thing’’ (Mon 10). However, in a substance that is truly simple or without parts, it is impossible to explain how such change can occur through the influence of other created beings. A monad has no parts that can be added to or subtracted from; and it has ‘‘no windows through which something can enter or leave’’ (Mon 7). It follows that a monad’s ‘‘natural changes’’ can come only from ‘‘an internal principle, since no external cause can influence it internally’’ (Mon 11). Thus, he concludes, ‘‘one can call simple substances or created monads ‘entelechies’. . . . They have a self-sufficiency (autarkeia) that makes them the sources of their internal actions, and, so to speak, incorporeal automata’’ (Mon 18). There is no reason to think that these accounts are fundamentally at odds; however, if our aim is to understand a substance’s capacity for change, we should focus on the idea that a substance by nature is endowed with a force sufficient to bring about changes in its own states. One of the most important points Leibniz makes about this force is that it is a conative power, akin to appetite or desire. At the most basic level, change comes about in a substance as a result of an ‘‘endeavor’’ or ‘‘striving’’ for future states. In the New Essays, Leibniz describes the soul as ‘‘an entelechy—i.e., a primary or substantial endeavor . . . accompanied by perception’’ (RB 170). In more technical writings, he refers to this same principle as ‘‘primitive active force.’’ Distinctive about this entelechy, or primitive force, is that it is not simply the capacity or potential to act but a force from which action spontaneously follows. Active force, he writes, ‘‘involves an endeavor [conatus] or striving [tendentia] towards actions, so that, unless something impedes it, action results’’ (G 4:395/AG 252; see G 4:469–470/L 433).14 Leibniz is less explicit than he might be about the dynamics of modal change.15 One unresolved issue is whether the force relevant to explaining change in a substance is the ‘‘primary or substantial endeavor’’ definitive of the substance’s nature, or the varying moments of endeavor that characterize its successive states. For purposes of discussion, call the former the ‘‘primitive force theory,’’ the latter, the ‘‘derivative force theory.’’16 The primitive force theory explains spontaneity by crediting substance with a nature that is both productive and law-like. The states of a substance are determined to evolve in a fixed pattern, by virtue of the fact that the substance’s nature, or primitive force, is expressed in accordance with a determinate principle of order—its ‘‘law of the series.’’17 From this, monadic spontaneity follows directly. By virtue of its nature, any substance is, as it were, an ‘‘incorporeal automaton’’ that, once created, runs through a succession of prearranged states. The primitive force theory cites the fact that a substance’s activity is expressed in accordance with a ‘‘law of the series’’ as an explanation of the changes in its

164 Leibniz: Nature and Freedom

states. Against this, the derivative force theory holds that changes in a substance’s states can only be explained adequately in terms of a property of those states themselves: their inherent tendency to give way to new states. In late works, Leibniz refers to these tendencies as a monad’s ‘‘appetitions’’: a monad, in itself and at a moment, can be distinguished from another only by its internal qualities and actions, which can be nothing but its perceptions (that is, the representation of the composite, or what is external, in the simple) and its appetitions (that is, its tendencies to go from one perception to another) which are the principles of change. (G 6:598/AG 207)18

An analogy may help to motivate this alternative account of change. Imagine a line of dominoes that has been arranged to fall in a particular order once the first member of the series has been pushed against the second. The dominoes have been ordered such that the resulting cascade realizes a certain pattern. Suppose we now ask why domino n þ 1 falls after domino n. One answer would be, simply, that this is how they have been arranged to fall: the setup has been designed such that this event happens as part of a larger pattern. Yet this might be found insufficient as an answer to the question of why domino n þ 1 falls as it does, or alternatively why it fails to fall. To answer this question, we need to know exactly how n þ 1 is struck by n, and how this action does, or does not, lead to the intended outcome. So, the proponent of the derivative force theory concludes, if we want to explain the change in a substance’s states—the fact that state n is succeeded by state n þ 1—we need to refer to more than the substance’s law of the series. We require, in addition, an explanation of why the appetitive force associated with n is sufficient to bring about n þ 1, as opposed to any other state. The primitive force theory makes one point that is uncontroversial. Ontologically, primitive force is prior to derivative force, a monad’s entelechy prior to its appetitions. If there were no substantial principle of action, there would no instances of acting in this way or that—any more than there would be shapes if there were no extension to be shaped.19 With respect to the explanation of modal change, however, there is reason to question the adequacy of the primitive force theory. Recall that, for Leibniz, change is defined in terms of a pair of mutually exclusive contingent states, and a reason or force by which those states are connected as prior and posterior. On his account, the reason for the transition from one state to another is identified as a property specific to the prior state (Leibniz refers to it as ‘‘an intrinsic accidental denomination’’) that explains why that state is succeeded by its actual successor and not by some other state. A defender of the primitive force theory might argue that such an explanation is precisely what the law of the series delivers: state n þ 1 succeeds state n because that is how the law determines it to be. But such a response is unsatisfactory. In the first place, it is unclear what it means to say that a substance’s law of the series determines the changes in its states. In speaking of a ‘law,’ we usually mean something propositional, an item of knowledge. Taken in this sense, a law’s determination of a substance’s states cannot be a case of causal determination. At most, it can be construed as a type of entailment: given the law of the series, it is true that S’s being in state n is followed by S’s being in state n þ 1. Appeal to the law of the series in this

Leibniz on Spontaneity

165

sense fails as an explanation of the dynamics of change. The law of the series expresses the global ordering of a substance’s states, but it does not explain how, in a given substance, there is a transition from state n to state n þ 1. Of course, Leibniz thinks of the law of the series as more than a formula or proposition. He identifies it with a real substantial power, or entelechy. But we are, I suggest, no better off if we follow Leibniz in interpreting the law of the series in this way. For it is characteristic of such a power qua law that it is perpetual and unchanging; it is that which ‘‘contains’’ all the particular changing states of the substance. Although such an unchanging ground is presupposed by change, it does not explain change: the fact that S’s being in state n is followed by its being in state n þ 1. Consequently, the primitive force theorist either is left with no explanation of change, or is pushed back to an explanation in terms of derivative force, saying something like: the law of the series determines that state n þ 1 succeeds state n, because n is a modification of the primitive force expressed in the law of the series, and n is causally sufficient for the occurrence of n þ 1.20 Comparing theories in this way, we can surmise that rather than being rivals, the primitive and derivative force theories are better understood as complementary components of a single theory of monadic spontaneity.21 Rather than offering competing answers to the same question, the two theories provide mutually supportive answers to different questions—what we might distinguish as the ‘‘why’’ and ‘‘how’’ of change. The primitive force theory provides an answer to the question of why there is change at all in a monad. As the term implies, derivative forces are only causally efficacious to the extent that substance itself is. It is the primitive force, entelechy, or substantial power, that is the ultimate ground of change in the created world. Yet, insofar as we are concerned to understand the how of change, or to explain the changes that occur in the states of substances, there are good textual and philosophical reasons for taking Leibniz to hold that such changes are explained in terms of the determination of one state by another, where the relevant causal factor is the derivative force, or appetition, associated with the prior state. Somewhat more speculatively, we may suppose that to the extent that a substance, or its nature, is properly conceived of as a cause, it is so only in the sense of possessing a determinable power that can be expressed in a variety of determinate ways. A substance’s primitive active force acts in a determinate manner productive of change only insofar as it is modified by some derivative force or appetition. The preceding account shows clearly why Leibniz is entitled to think of modal change as contingent. From the point of view of dynamics, no logical, metaphysical, or conceptual necessity determines that a monad should pass from state n to state n þ 1. Such a transition occurs only if n is endowed with a force that is sufficient, relative to the appropriate dynamical law, to bring about state n þ 1.22 In this respect, the changes that occur within the monad are on a par with the physical changes that occur within a system of bodies. Such a system undergoes a transition from some dynamical state d to state d + 1 only if d is characterized by forces that are sufficient, relative to the appropriate dynamical laws, to bring about d þ 1. No one would suggest in this case that d + 1 is logically necessitated by d, and there is no more reason to suppose that this is so in the case of the transition between two

166 Leibniz: Nature and Freedom

states of a monad. Thus, the causal determination that underlies monadic spontaneity is fully consistent with the contingency of modal change.

Teleology and Monadic Spontaneity Leibniz conceives of the monad as subject to continual change: monadic states are always giving way to new states, as a consequence of the forces inherent in them. This would seem to imply that monadic states are genuinely efficacious: they (or their properties) are causes that make things happen. Yet what kind of causes precisely? Are they, for example, properly characterized as efficient causes? Intuitively, it seems, the answer should be yes. The appetitions associated with particular monadic states are productive of new states. Hence they are efficient causes of those states. While this may seem an obvious conclusion, it is in tension with some of the things Leibniz says about soul-like substances. In Monadology 79, he writes: ‘‘Souls act according to the laws of final causes, through appetitions, ends, and means. Bodies act according to the laws of efficient causes or of motions’’ (G 6:620/AG 223). Here, as in other texts, Leibniz appears to limit efficient causation to the physical interactions of bodies, and to identify changes in the soul, or monad, with the operation of a kind of final causation. As suggested earlier, Leibniz conceives of monadic activity on analogy with appetite or desire. According to his metaphysics, every monad is by nature active (an entelechy or source of change), and its activity is inherently conative (the endeavor toward a future state). What must be added to this is that monadic activity is also inherently teleological, that is, an endeavor toward future states based on a conception of their goodness or the goodness represented in them. As Leibniz writes in the Principles of Nature and Grace, ‘‘perceptions in the monad arise from one another by the laws of appetites, or of the final causes of good and evil’’ (G 6:599/AG 207). Recognition of the role played by teleology in the operation of the monad adds an important dimension to Leibniz’s account of spontaneity. Yet to the extent that Leibniz himself opposes causation within the monadic realm to the efficient causation operative in the physical realm, it also raises a problem for his account. If the appetitive forces associated with monadic states do not function as efficient causes, how are we to understand the capacity of those states to lead spontaneously to new states? The problem in this case is largely terminological. While maintaining that souls, or monads, ‘‘act according to the laws of final causes,’’ Leibniz is clear that appetitive forces are genuinely efficacious in bringing about changes in a monad’s states.23 His point, therefore, is not that monadic states do not act as efficient causes of subsequent states but that those actions cannot be explained by appeal to the laws of efficient causation that govern the actions of bodies. Rather, changes within the monad are produced by forces that act according to the ‘‘laws of final causes.’’ Saying just this, however, leaves unexplained the relation between appetitive forces and the laws of final causation by which they operate. Suppose s1 and s2 are successive states of a monad, and the reason or ground of the change from s1 to s2 is the appetitive force associated with s1. Suppose further that appetitive force,

Leibniz on Spontaneity

167

in general, is an endeavor toward some good, and that this endeavor is stronger in proportion to the magnitude of that good. On the face of it, the efficacy of this force can be understood in either of two ways: in terms of the intrinsic goodness of the state toward which it tends (s2), or in terms of the goodness of s2 as represented by s1. If the force operates according to what I shall call the ‘‘law of natural teleology,’’ then the effectiveness of the force in bringing about a transition from s1 to s2 is explained by the fact that s2 is objectively the best state for the monad to be in next. If, on the other hand, the force operates according to the ‘‘law of desire teleology,’’ then its effectiveness in bringing about a transition from s1 to s2 is explained by the fact that s1 represents s2 as the state of greatest goodness relative to it. If appetition operates according to the law of natural teleology, then changes within the monad are determined in accordance with an external standard of goodness. Given any state of the monad, there is a determinate answer to the question of what its next state should be relative to the rest of the universe, and the appetitive force associated with the prior state will be just that force sufficient to bring about the appropriate successor state. If appetition operates according to the law of desire teleology, on the other hand, this conclusion does not follow. The appetitive force associated with any monadic state will depend upon what that state represents as goods attainable in future states. In this case, there is no assumption that the good for the sake of which the monad acts is the greatest good, relative to it or the universe. The forces that determine changes in its states are a function of what appear to it as goods, and in the case of any created monad such appearances can be mistaken. Leibniz draws on both these notions of teleology in explaining the changes that occur in a monad’s states. Sometimes he discusses monadic change in terms of the realization of an objectively optimal state of affairs, for example, universal harmony among the states of soul-like substances; at other times he conceives of this change as the product of a monad’s striving for what it represents as the greatest good. The question is how these two notions of teleology relate to each other. One possibility is that Leibniz posits a division of labor between them, based on the phenomenon to be explained. In the domain of psychology, where we aim to explain how appetite and volition contribute to action, the law of desire teleology is most appropriate. Here we are interested in how rational and nonrational creatures come to act, given their particular representations of the good—representations that may be inaccurate. By contrast, if we are concerned to explain the general fact of monadic change, then our attention should be focused on the law of natural teleology, since it is Leibniz’s view, developed in the doctrine of preestablished harmony, that any change in a monad is regulated in such a way as to bring about an optimal outcome. The generality of the latter requirement suggests that it is a mistake to think of a division of labor between these two forms of teleological explanation. Instead, Leibniz’s position must be that any case of a monad acting according to the law of desire teleology is also a case of its acting according to the law of natural teleology, for the latter is a completely general form of explanation that applies to all changes in the monad. But have we any reason to think that the converse also holds: that any case of a monad acting according to the law of natural teleology also must be

168 Leibniz: Nature and Freedom

a case of its acting according to the law of desire teleology? If we are sensitive to the difference between the notions of appetition and desire, I believe, the answer is no. Although every change in every monad can be explained in terms of appetitive forces operating according to the law of natural teleology, not every such change is an example of the monad acting according to the law of desire teleology, that is, of its acting to bring about the state that it represents as containing the greatest good. Distinguishing cases in this way, I shall argue, is the best way of understanding Leibniz’s notion of agent spontaneity, and of why it deserves to be treated as an independent condition on freedom.

Agent Spontaneity In late works such as the Theodicy, Leibniz defends a general theory of the will, according to which appetite and desire operate in the same manner in all created beings.24 Without making any claim for completeness, we can summarize the theory in five propositions: (1) There is no absolute indifference of the will (T 35, 46, 302–303; T k16). (2) There is no appetite or desire except as determined by the representation of some good (T 33, 45). (3) The motivational force of any appetite or desire is proportional to the apparent goodness of its object (T 287). (4) When the soul acts (or chooses) it does so in accordance with its strongest motive; that is, the desire for the greatest apparent good prevails (T k13). (5) Under the influence of competing motives, the soul infallibly chooses the greatest apparent good, but it is not necessitated to do so, for it is determined to act by motives that ‘‘incline without necessitating’’ (T 45, 310; T k14).25

Leibniz relates the spontaneity of the soul’s actions most directly to propositions (2) and (4). According to the former, the soul’s natural endeavor is rendered determinate as the desire for certain goods. This could be understood simply as a matter of the soul’s reacting to the impressions made on it by external objects. This was Hobbes’s view.26 Leibniz, however, interprets desire as arising not in response to objects themselves but to representations of objects as goods: Although an active substance is determined only by itself, it does not follow that it is not moved by objects: for it is the representation of the object within it which contributes towards the determination. Now the representation does not come from without, and consequently there is complete spontaneity. Objects do not act upon intelligent substances as efficient and physical causes, but as final and moral causes. . . . When we say that an intelligent substance is moved by the goodness of its object, we do not assert that this object is necessarily a being existing outside the substance, and it is enough for us that it be conceivable: for its representation acts in the substance, or rather the substance acts upon itself, insofar as it is disposed and influenced by this representation. (T k20–21/G 6:421–423/H 427–428)

Leibniz on Spontaneity

169

According to Leibniz, desire can be understood as the result either of the represented object acting on the soul as a final cause, or of the soul’s acting on itself, ‘‘insofar as it is disposed and influenced by this representation.’’ Because the soul’s desires are determined by representations of objects, and representations are internal to the soul, spontaneity is preserved up to this point. Leibniz’s second main claim about spontaneity, articulated in (4), is that when the soul acts (or chooses), and thereby undergoes a change of state, it does so as determined by its strongest desire. The rationale for this claim is conveyed in an image Leibniz employs in his fifth letter to Clarke: ‘‘reasons in the mind of a wise being, and motives in any mind whatsoever, do that which corresponds to the effect produced by weights in a balance’’ (G 7:389/L 696). In any soul in which there are two competing desires that cannot both be satisfied, the strongest desire—that for the greatest apparent good—prevails, just as the heaviest weight tips the balance. In Theodicy 325, Leibniz develops this account using a different physical analogy, which illustrates why, of a plurality of simultaneous endeavors toward different goods, only one determines the soul to act: As very often there are several courses to choose from, one might, instead of the balance, compare the soul with a force which puts forth effort on various sides simultaneously, but which acts only at the spot it finds easiest or where there is least resistance. For instance, air if it is compressed too firmly in a glass vessel will break it in order to escape. It exerts force on every part, but finally pounces upon the weakest. Thus do the inclinations of the soul extend over all the goods that present themselves: they are antecedent acts of will; but the consequent will, which is their result, is determined in the direction of that which is most influential [ce qui touche le plus]. (G 6:309/H 322)

Leibniz defends spontaneity at this second stage, then, by appeal to a natural operation of the soul, whereby the strongest desire—that for the greatest apparent good—determines the will by which the soul acts. This does not mean that reason always prevails over appetite, or that the greatest good as such is always chosen. Leibniz claims only that the will acts in pursuit of the greatest apparent good, which tracks the good more closely in proportion to the distinctness of the soul’s perceptions. Leibniz’s theory of the will offers the basis for an explanation of agent spontaneity. According to Leibniz, the state of the soul at any time is characterized by a complex array of perceptions, at least some of which represent objects as goods. Associated with the latter perceptions are desires for those goods. Competition among the soul’s desires terminates in its acting in pursuit of the greatest apparent good, and this act in turn produces a spontaneous change in its state. Consistent with the notion of agent spontaneity, changes of this sort are distinguished by the fact that the soul determines itself to act on the basis of its representations of goods. Thus, we can say that a soul exhibits agent spontaneity, if its desire for some future good is effective in bringing about a change in its state, or, equivalently, if the change in its state can be explained by appeal to the law of desire teleology.27

170 Leibniz: Nature and Freedom

The Unified Theory The preceding account of agent spontaneity falls short of a general explanation of the changes that occur in a monad’s states. Leibniz’s theory of the will depicts the competition among a soul’s desires as culminating in an act, whereby the strongest desire determines a motion toward the greatest apparent good. Such an account, however, can make no definitive claim about the subsequent perceptual state of the soul, for that depends not only on the soul’s effort to acquire a certain good but, in addition, on the resistance it confronts, or the factors that oppose the exercise of its will. Obviously, when we act to satisfy our desires, we sometimes fail to do so. Arguably, such cases may still qualify as instances of agent spontaneity, since whether or not we achieve the goods we desire, we nonetheless act to achieve them. In other cases, though, this condition fails to be met. Often the soul is affected in ways that seem to directly contradict the supposition that changes in its states result from a desire for the good. An example of this sort was proposed by Pierre Bayle as part of his critique of Leibniz’s New System.28 A dog is happily eating and taking pleasure in its food. A man comes up from behind and without warning strikes the dog on the back with a stick. The dog’s pleasure immediately gives way to pain. As Bayle saw it, based on the doctrine of preestablished harmony, Leibniz is committed to there being some appetite that determines the dog’s soul to pass from a state of pleasure to a state of pain, without it being influenced by anything external to it. But this cannot be true, both because there is no appetite for the unpleasant and because the dog had no way of knowing that it was about to be struck. From the perspective of the dog’s present state, its subsequent state of pain is neither apparent nor a good. Bayle interpreted this case as a counterexample to the doctrine of preestablished harmony. Yet this assumes that Leibniz believed that every change in the state of the dog’s soul can be explained as the consequence of an appetite or desire for what it represents as a good. This assumption is by no means obvious, and in my view it should be rejected. The lesson of Bayle’s dog, as I read it, is that if appetition is the internal force, or endeavor, that explains the transition of the dog’s soul from a state of pleasure to a state of pain, then that endeavor cannot operate in the same way as volitional states such as desire. Appetition cannot be understood as functioning according to the law of desire teleology, for there is no conceivable scenario under which the dog would have represented its subsequent state of pain as a good relative to its present state of pleasure. Against this it might be objected that Leibniz’s own philosophy contains the resources needed to explain how in Bayle’s example the change in the dog’s soul can be understood as the product of a desire for its subsequent painful state. Although the dog does not consciously desire to be in such a state, it might be drawn to such a state by unconscious desires that confusedly represent the state as a state of pleasure. Leibniz is well known for ascribing an important psychological role to petites perceptions that represent objects unavailable to consciousness. Among these are the objects of future perceptual states, since, in Leibniz’s view, every monad contains ‘‘marks and traces’’ of its entire history. To the extent that the

Leibniz on Spontaneity

171

soul contains such representations, which are admittedly extremely obscure and confused, one might infer that it also includes unconscious desires for the objects represented in those future states. Thus, the objection concludes, there is nothing amiss in thinking of the change in the dog’s soul as coming about in a way that is consistent with the law of desire teleology. I take the objection to concede the point that whenever a monad acts for an end that it represents as desirable, or good, this end is perceived as a source of pleasure. As Leibniz puts it, ‘‘in substances which are capable of pleasure and pain every action is a move towards pleasure, every passion a move towards pain’’ (NE 2.21.72/ RB 210).29 Thus, if the change in the dog’s soul can be understood as occurring in accordance with the law of desire teleology, the dog either must mistakenly perceive its future pain as a sensation of pleasure, or it must perceive the painful sensation as part of a larger state of affairs that promises a proportionally greater pleasure. But neither of these descriptions is plausible. There is nothing in the sensation of pain as such that could be misconstrued as a sensation of pleasure, so it is inconceivable that the dog could have misperceived that sensation as in any way desirable.30 Accordingly, the objector must hold that what the dog’s soul represents as desirable is the complex state of affairs of which its painful sensation is a part; that is, the dog must obscurely represent the subsequent state of the universe as being so desirable that it offsets the pain that it itself will experience as part of that state. Yet this, too, is implausible. The objection trades on the intuition that the dog’s representation of what is good, or desirable, includes in a confused way the desirability of the world as a whole. However, if the perception of something as desirable is linked to its perception as pleasurable, there is no reason to think that this is the case. Since the dog’s soul lacks the power of reason by which it might form representations of the goodness of things in general, its perceptions of the good as good are limited to the circumstances affecting its own body. And these circumstances are ones that it perceives as painful. Consequently, accepting that the dog’s soul does move from a state of pleasure to a state of pain, the case must be interpreted as one in which the law of desire teleology is violated and agent spontaneity fails.31 For Leibniz, this admission is consistent with the truth of monadic spontaneity. Although the case is one in which, relative to its point of view, the dog is passive rather than active in the determination of the change in its state, any such change is the product of appetitive forces internal to the soul. When the dog’s soul moves from a state of pleasure to a state of pain, it does not function as a spontaneous agent, acting to bring about what it represents as the greatest good, but it does undergo a spontaneous change of state, brought about by the appetitive forces operating within it. In general, every change in the dog’s perceptual state can be ascribed to the operation of appetitive forces, but only some such changes are ones in which the dog’s soul acts spontaneously in pursuit of the greatest apparent good. To draw this distinction in a compelling way requires separating the roles played in Leibniz’s philosophy by, on the one hand, volitional states such as desire and, on the other, the broader class of appetitions.32 Volitional states, which operate according to the law of desire teleology, reflect the standpoint of an individual agent. For any such agent, desire is endeavor for an object represented as a good. Accordingly, the range of an agent’s desires is constrained by the soul’s power

172 Leibniz: Nature and Freedom

of representation. Regardless of the intrinsic goodness of an object, if it is not represented by the soul as good, there can be no desire for it. Assuming such desires are present, they may or may not be effective in producing a change in the soul’s state. If they are effective, and if the change in the soul’s state can be explained in terms of an agent’s acting for the sake of the greatest apparent good, then that agent acts spontaneously, or exhibits agent spontaneity. Appetition operates in a fundamentally different way, which reflects the soul’s representation of a complex physical world of bodies as opposed to its own ends. There is, of course, a close relationship between these two sets of representations. Included within my representation of the physical world is a representation of my body, which defines the locus of my agency and for the sake of which I desire certain goods. Furthermore, insofar as I act, I am able to bring about changes in my body’s relation to other bodies. Clearly, though, my ability to affect the physical world in this way is limited. Almost all of what I perceive to be changes in the physical world come about independently of my will, or even in opposition to my will. If Leibniz’s account is correct, however, the appetitions of my soul are responsible for bringing about everything I perceive of the physical world, including representations of distant and microscopic events of which I am entirely unaware and which are in no way apparent goods for me. What a monad’s appetitions track, therefore, is not a representation of the world from the standpoint of agency—a world of apparent goods— but a representation of the world of physical events governed by mechanical laws. Such a distinction forms the basis of Leibniz’s own response to Bayle’s counterexample: The crux of M. Bayle’s objection here is that we have no spontaneous inclination towards what we find unpleasant. I make a distinction: I admit this when we know that something will displease us, but in this case the dog does not know. We must also distinguish between the spontaneous and the voluntary. The principle of change is in the dog, the disposition of its soul moves imperceptibly towards giving it pain: but this is without its knowing, and without its wanting it. The representation of the present state of the universe in the dog’s soul produces in it the representation of the subsequent state of the same universe, just as in the things represented the preceding state actually produces the subsequent state of the world. In a soul, the representation of causes are the causes of the representations of effects. And since this subsequent state of the world includes the blow on the dog’s body, the representation of that subsequent state in its soul includes the pain which corresponds to that blow. (G 4:532–533/WF 78; Leibniz’s emphasis)33

According to Leibniz, the dog’s soul can pass spontaneously from a state of pleasure to a state of pain, because the sequence of its perceptions is determined not by its representation of desirable ends, but by its representation of the causal history of the universe. At any moment, the dog’s soul contains a representation of the complete state of the physical world (though most of this is unconscious), and consequently it contains representations of the causes of subsequent states of the world. The causes that bring about future physical states are infinitely diverse; hence, so too are the tendencies of the soul’s perceptions to change, for in the soul, the ‘‘representations of causes are the causes of the representations of effects.’’ Leibniz elaborates on this picture in Theodicy 403:

Leibniz on Spontaneity

173

The operation of spiritual automata, that is of souls, is not mechanical, but it contains in the highest degree all that is beautiful in mechanism. The movements which are developed in bodies are concentrated in the soul by representation as in an ideal world, which expresses the laws of the actual world and their consequences, but with this difference from the perfect ideal world which is in God, that most of the perceptions in the other substances are only confused. For it is plain that every simple substance embraces the whole universe in its confused perceptions or sensations, and that the succession of these perceptions is regulated by the particular nature of this substance, but in a manner which always expresses all of universal nature; and every present perception leads to a new perception, just as every movement that it represents leads to another movement. (G 6:356–357/H 365)

When Leibniz says that monadic perceptions tend of their own accord toward new perceptions, what he means, most fundamentally, is that they tend toward the representation of a new state of the physical universe. According to the thesis of monadic spontaneity, the changes that occur in a monad’s perceptual states are explained in terms of the operation of appetitive forces internal to those states. However, the action of these forces, in turn, is explained in terms of their effectiveness in advancing the monad through successive representations of the complete state of the physical universe. In an important sense, therefore, Leibniz gives explanatory priority to a notion of lawful change in the physical world, for it is in terms of the latter that the effectiveness of appetitive forces is understood.34 Earlier, I suggested that appetitions operate according to the law of natural teleology: they act so as to bring about what is objectively the best state for the monad to be in next. How is this to be reconciled with the explanatory priority of the physical? What, if anything, makes a subsequent state of the physical world good, or desirable? From the standpoint of the agent or organism—the soul defined by the physical point of view of its body—the future state of the world may not be desirable at all. Witness Bayle’s dog. From the point of view of God’s will, on the other hand, it is always desirable, for the future state of the world is the best way of continuing the physical universe, the best state for the universe to be in following its present state. As it is this order that governs the development of a monad’s perceptions, its appetitions collectively tend toward the representation of a physically optimal state of affairs—a state whose goodness is defined in terms of considerations of lawfulness and order. On the reading defended here, there is no general account of a monad’s appetitions as endeavors toward what it represents as goods, that is, desirable ends or objects. Rather, explanatory priority is given to the lawful evolution of the physical universe, and monadic perceptions tend toward representations of whatever the next best state of the universe is. Leibniz’s theory of agent spontaneity is developed against this backdrop. Agent spontaneity presupposes the perspective of an agent, acting to achieve what it represents as goods. The perspective of an agent is defined by its powers of representation, which determine its relation to the physical world and the range of its goods. On Leibniz’s account, human agency is understood to have both a rational and a sensory component: there are intelligible goods represented by reason and bodily goods represented by the senses. Desire is the power by which an agent endeavors to achieve the things it represents as goods; and to the extent that desire

174

Leibniz: Nature and Freedom

is effective in promoting the attainment of those goods (and its own perfection), an agent acts spontaneously. Yet no finite agent can escape completely the limits of its perspective. Consequently, no finite agent realizes the condition of perfect spontaneity. Whatever its degree of activity, it is always to some degree passive, or responsive to the state of the physical world. The importance of appetition for Leibniz is that it explains the source of this passivity: the changes that occur in a soul’s representations in spite of its best efforts to change the world. To the extent that a soul exhibits agent spontaneity, changes in its perceptual states can be explained in terms of its effort to acquire what it represents as goods. Such cases, however, constitute a relatively small proportion of the changes that occur in most monads’ perceptual states. For the most part, such changes occur without any effort of will or even in opposition to the will, when the world is represented as resisting the soul’s efforts to change it. In such cases, the soul’s passivity, or submission to external causes, is accounted for by the operation of appetition, which spontaneously leads to new representations of the physical world, independently of what the soul represents as its own volitions, or efforts of will. For Leibniz, this complex interplay of activity and passivity is worked out within the confines of preestablished harmony, according to which there occurs no real interaction between soul and world. At the deepest level, there is only a correlation between two series of perceptions, each evolving according to its own laws. How we characterize perceptual change is a subtle matter for Leibniz. Leibniz ascribes all of the changes that occur in a soul’s perceptual states to the operation of appetitive forces, which act according to the ‘‘laws of final causes’’ (G 6:620/AG 223). On the account presented here, however, it is necessary to distinguish among these endeavors, based on whether they function merely as appetitions, leading the monad toward new representations of the physical world, or as desires for apparent goods. In the first case, change is explained in terms of the law of natural teleology, whereby the soul endeavors to achieve what is (from the perspective of God) the next best state of the universe. In the second case, change is explained in terms of the law of desire teleology, according to which the soul endeavors to achieve what it represents as the greatest good. Noteworthy about this distinction is that it is drawn entirely in terms of the content of a monad’s perceptions, and that in the first case the relevant content is simply the successive states of the physical world.35 Given this, Leibniz is able to retain the commonsense distinction between the soul’s activity and passivity: its acting on and being acted on by an external world. In terms of their content, only volitional states involve the soul’s acting for the sake of apparent goods; hence they alone correlate with the notion of agent spontaneity.

Spontaneity and Freedom In a 1704 letter to Lady Masham, Leibniz writes: ‘‘freedom requires that we act with spontaneity and with choice; my system augments our spontaneity and does not diminish our choice’’ (G 3:364).36 Leibniz’s reason for claiming that his system augments our spontaneity is that on his account—that of Theodicy 288—every

Leibniz on Spontaneity

175

change in a substance’s states occurs spontaneously, as the result of forces internal to it. Leibniz saw an important relation between this thesis of monadic spontaneity and the preservation of freedom, which he associated with a substance’s independence from the effects of external causes. This point is highlighted in Theodicy 300–301: [T]he establishment of this system [of preestablished harmony] demonstrates beyond a doubt that in the course of nature each substance is the sole cause of all its actions, and that it is free of all physical influences from every other substance, save the customary cooperation of God. And this system shows that our spontaneity is real, and not only apparent. . . . The spontaneity of our actions can therefore no longer be questioned; and Aristotle has defined it well, saying that an action is spontaneous when its source is in him who acts. . . . Thus it is that our actions and our will depend entirely upon us. (G 6:295–296/H 309–310)

In linking spontaneity to an agent’s power of self-determination, Leibniz allies himself with a philosophical tradition stretching back to Aristotle. According to this tradition, an agent acts spontaneously when it is the source or principle of its own action—a self-mover—as opposed to being constrained to act by external causes. Yet Leibniz fails to establish in a sufficiently clear way the connection between this idea and the thesis of monadic spontaneity. It is worth recalling that monadic spontaneity does not enter Leibniz’s system through an analysis of the problem of freedom; rather, the thesis is a consequence of answers Leibniz gives to a set of fundamental metaphysical questions concerning change, causation, unity, and identity. Only after answers to these questions have been given does he attempt to link spontaneity, so conceived, to the issue of freedom.37 When Leibniz explicitly addresses the problem of freedom the relevance of monadic spontaneity is far from obvious. In Theodicy 34, there is the following characteristic airing of the issues: I am of the opinion that our will is exempt not only from constraint but also from necessity. Aristotle has already observed that there are two things in freedom, namely, spontaneity and choice, and therein lies our mastery over our actions. When we act freely we are not being forced, as would happen if we were pushed on to a precipice and thrown from top to bottom; and we are not prevented from having the mind free when we deliberate, as would happen if we were given a draught to deprive us of discernment. There is contingency in a thousand actions of nature; but when there is no judgment in him who acts there is no freedom. (G 6:122/H 143)

Here Leibniz appears to locate three potential threats to freedom. The first is an incapacity of our judgment, or a lack of freedom of deliberation. This impinges directly on our power of choice, or our ‘‘mastery over our actions.’’ The other two threats, which are not clearly distinguished, are related to the condition of spontaneity. One of these is the ever-present specter of necessity: spontaneity (and hence freedom) presupposes the contingency of our actions. The other is the classic problem of physical constraint—as Leibniz puts it, that ‘‘we are not being forced, as would happen if we were pushed on to a precipice and thrown from top

176 Leibniz: Nature and Freedom

to bottom.’’ In neither of the latter two cases does monadic spontaneity respond directly to the underlying worry. With respect to the first problem, it is essential for Leibniz to be able to argue that monadic spontaneity is consistent with the contingency of our actions. This I have claimed he is able to do, and consequently the threat of metaphysical necessity can be repelled. At the same time, monadic spontaneity is not entailed by the solution to this problem. Leibniz’s solution, the doctrine of hypothetical necessity, is consistent with a variety of causal theories, including ones in which at least some of substance’s states are caused by the actions of external substances. In the second case, the challenge is to distinguish ordinary instances of unconstrained action from special circumstances in which the exercise of our will is constrained or preempted by external causes. Monadic spontaneity addresses this concern by maintaining that we are never acted on by external causes; everything that happens to us is the product of actions internal to our soul. But this has the effect of rendering the phenomenon of constraint nonexistent, contrary to our intuition that there is a significant distinction between constrained and unconstrained action. Once again, we confront the criticism that in generalizing the notion of spontaneity to cover any change in a soul-like substance, Leibniz severs spontaneity’s relation to freedom. I have argued that Leibniz has the resources to construct a more promising account of spontaneity—agent spontaneity—that meets this objection. The answer, in a nutshell, is found in a sentence from Theodicy 64: ‘‘All that happens to the soul depends upon it, but it does not always depend upon its will; that would be too much’’ (G 6:137). In order to make sense of the idea of agent spontaneity, it is necessary to preserve the distinction between a substance’s active and passive states: those in which it acts to bring about changes in its own states and those in which such changes are ascribed to external causes. I have shown that Leibniz has a way of preserving this distinction within the framework of preestablished harmony, which itself rests on the doctrine of monadic spontaneity. Monadic spontaneity thus serves as the metaphysical foundation for Leibniz’s treatment of freedom, but it is not the notion he requires in order to define the concept of freedom. This is made clear in his most careful treatment of the topic in the New Essays.38 Given the presence of a will that acts for the sake of the greatest apparent good, Leibniz distinguishes two principal kinds of freedom and two associated forms of constraint. The first is the ‘‘freedom to do,’’ or ‘‘the power to do what one wills.’’ Agent spontaneity is sufficient for the possession of this freedom, provided that an agent has the use of those things customarily in her power, above all, the free use of her body. Such freedom is blocked by physical constraint, or the action of external causes that preempt the spontaneous operation of her will. The second kind of freedom is the ‘‘freedom to will’’ or ‘‘the power to will as one should.’’39 For Leibniz, this is freedom in the morally relevant sense. In reasoning about this freedom, he writes: ‘‘the question is not whether a man can do what he wills to do but whether his will itself is sufficiently independent’’ (NE 2.22.21/RB 181). For one’s will to be free in this sense, it is necessary that one exercise a will, or that one act for the sake of the greatest apparent good. However, it is not necessary that one have the means to render one’s will effective. The critical factor instead is the form of one’s will, namely, whether it is governed by intelligence, or a

Leibniz on Spontaneity

177

distinct knowledge of the good. In rational beings, such freedom is blocked by moral (i.e., psychological) constraint, or the ‘‘bondage of the will’’: the inner constraint of the passions, which oppose the determination of choice by intelligence. To the extent that nonrational beings act to satisfy desires, they exhibit agent spontaneity; they act for the sake of the greatest apparent good. Lacking intelligence, however, such beings are incapable of freedom in the morally relevant sense. To attain ‘‘a mastery over one’s actions,’’ it is necessary to possess intelligence, on the basis of which one is able to deliberate and choose. In creatures with the capacity for choice, freedom in this sense is limited by the degree to which one’s ends are set by confused thoughts or passions rather than a distinct knowledge of the good: ‘‘we may say that we are immune from bondage in so far as we act with a distinct knowledge, but that we are the slaves of passions in so far as our perceptions are confused’’ (G 6:288/H 303). Nevertheless, whether slaves of our passions or masters of ourselves, we act spontaneously, provided that we act for the sake of the greatest apparent good.

Notes My thanks to Jan Cover, Michael Futch, Matthew Kisner, Wayne Martin, Sam Rickless, and Eric Watkins for their comments on previous versions of this essay. 1. According to Spinoza, freedom is the state in which an agent ‘‘complies with no one’s wishes but his own’’ (CSW 584). This occurs only if he is the ‘‘adequate cause’’ of changes in him or outside him—which he is only if he is led by reason alone and not by the passions. See Ethics IIIP3, IVP66S, and IVP68D. 2. Critique of Pure Reason, A 445/B 473, A 553/B 561; Critique of Practical Reason (Ak 5:96–97). 3. In an undated text, Leibniz defines ‘freedom’ as ‘‘the spontaneity of an intelligent being,’’ and ‘spontaneity’ as ‘‘contingency without constraint [coactione], that is, the spontaneous is that which is neither necessary nor constrained. . . . That is constrained whose principle is external’’ (G 7:108). Later in the same piece he writes, ‘‘The more anything is determined by itself, the more spontaneous or (if it is intelligent) free it is; the more it is determined from elsewhere, the more it is subject [servit] or constrained [coactum]’’ (G 7:109). 4. In a 1711 letter to Des Bosses, Leibniz writes: ‘‘I certainly maintain that a power of determining oneself without any cause, or without any ground of determination, implies a contradiction . . . but from this the metaphysical necessity of all effects does not follow. For it suffices that the cause or reason not necessitate metaphysically, although it is metaphysically necessary that there be some such cause’’ (G 2:420). 5. See T 65, 400; NE 2.21.72/RB 210; G 3:403/AG 195. 6. This objection is noted by both Murray and Greenberg in their essays in this volume. 7. According to Leibniz, ‘‘there is no real spontaneity in a mass. . . . [S]trictly speaking, what is spontaneous is the entelechy’’ (G 4:558/WF 111–112). Any soul-like substance, or monad, has an entelechy, form, or ‘‘primitive force.’’ This leaves the question of how to account for the apparent spontaneity exhibited by bodies. One answer is that some of these bodies (animated organic bodies) are substances in their own right, whose unity and activity are ascribed to their substantial forms. Another answer is that a reductive analysis of body terminates in an account of it as consisting solely of monads. See C 13–14/ MP 175; G 7:502. 8. Leibniz explains the harmony between soul and body in terms of the relation of ‘‘ideal causation’’: ‘‘[O]ne may give a true and philosophic sense to the mutual dependence which we suppose between the soul and the body. It is that the one of these two substances depends upon the other ideally, in so far as the reason of that which is done in the one can be furnished by that which is in the other. This had already happened when God ordered beforehand the harmony that there would be between them’’ (T 66/G 6:138/H 158–159).

178 Leibniz: Nature and Freedom 9. See T 65, 290. 10. In the New System, Leibniz claims that preestablished harmony ‘‘has this great advantage, that instead of saying that we are free only in appearance and in a way sufficient for practical purposes . . . we are determined only in appearance, and that, in rigorously metaphysical language, we have a perfect independence relative to the influence of every other creature’’ (G 4:485/AG 144). I have argued that this independence is inadequate for freedom in the morally relevant sense. For that agent spontaneity is required. But agent spontaneity, no less than physical constraint, is defined relative to a soul’s representational states. So does that mean that Leibniz is committed to the position that we are free ‘‘only in appearance’’? As will become clear, the answer is no: a soul is ‘‘physically constrained’’ by what it represents as the actions of external things on it; however, its spontaneous acts are effects of its own power (and not merely representations of its power). 11. The doctrine of divine concurrence, to which Leibniz assents, raises problems that I cannot tackle here. For an attempt to do so, see Sleigh 1990: 183–185. 12. There is room for only a brief sketch of the steps Leibniz takes in reaching this conclusion. Among his primary definitions is that of one thing being a sufficient condition, or inferens, of another: ‘‘If with A supposed to exist, it follows that B exists (although not at the same time), A will be an inferens, B, an illatum’’ (A 6.4:563). Conversely, any inferens is said to be conditioned by that for which it is a sufficient condition: ‘‘If with B supposed not to exist, it follows that A does not exist, B will be a condition, A, a conditioned’’ (A 6.4:563). To these he adds the notion of one thing’s being prior by nature to another: ‘‘If a condition is prior by nature to a conditioned, then it will be said to be a requisite [requisitum], and the conditioned will be called a requirens’’ (A 6.4:563). On the basis of these definitions, Leibniz offers the following account of one’s things being a full cause, or predeterminant, of another: ‘‘A full cause is an inferens prior by nature to an illatum, or that which involves all the requisites sufficient for it (i.e. that from which all the remaining requisites follow). Here it should be noted that whatever involves all the requisities of some means of producing also involves all the requisites of the thing to be produced. . . . From this it is evident that it is not necessary that a cause and an effect exist at the same time, nor that they not exist at the same time. Moreover, just as a full cause is specified with respect to a requisite, or a condition prior by nature, so a determinant is specified with respect to some condition. For that is a determinant which involves all the conditions of that which is determined, and consequently a predeterminant, or a determinant prior by nature, is a full cause. It is obvious, too, that an inferens and a determinant are exactly the same; but an inferens prior by nature can be said to be a predeterminant, or a full cause’’ (A 6.4:563–564). The definitional studies in which these ideas are worked out are discussed in greater detail in Rutherford 1995a: chap. 5 and Futch 2001. 13. The latter claim may seem to go beyond the former, since one can imagine cases of concurrent causation (or of causal overdetermination) where it is true both that a substance acts on itself to produce changes in its states and that those changes are the effects of the actions of other substances on it. Leibniz offers independent arguments against the latter possibility that he appeals to in defending the thesis of monadic spontaneity. See the passages from the Monadology cited in the next paragraph of the main text. 14. Earlier I say that force involves an endeavor or striving for future states; here Leibniz says, an endeavor or striving towards actions. Since ‘action’ and ‘passion’ are defined as species of change, i.e., an aggregate of contradictory states (A 6.4:556; C 9/MP 134), an endeavor toward action is an endeavor toward a state incompatible with the substance’s present state. 15. Compare Sleigh: ‘‘But the texts are curiously vague about the character of the relation between a state of a substance and its predecessor, in virtue of which, according to Leibniz, the former is a consequence of the latter’’ (Sleigh, Chappell, and Della Rocca 1998: 1263). 16. Cover and O’Leary-Hawthorne 1999: chap. 6 defend a version of the former; Kulstad 1990 a version of the latter. See also Bobro and Clatterbaugh 1996. 17. Evidence of this idea can be found in texts as early as 1676. See Rutherford 1995a: 148–154. 18. I interpret the content of a monad’s perceptual state and the tendency of that state to change (its ‘‘appetition’’) as properties of the same state. Compare the following passage from a 1706 letter to Christian Wolff: ‘‘In truth, whatever is in the soul may be understood generally to be reduced to two things: the soul’s harmonious expression, in accordance with its body, of the present state of external things, and the endeavor [tendentia] to a new expression that represents the endeavor of bodies (or

Leibniz on Spontaneity

179

external things) to a future state, in a word, perception and tending toward perception [perceptionem et percepturitionem]’’ (GLW 56). 19. See, e.g., T 396. 20. This conclusion is supportive of Sleigh’s interpretation of the doctrine of spontaneity as it figures in the Arnauld correspondence: ‘‘Throughout the texts from our period it is clear that the series of states, constituting the history of an individual substance, must be generated by a relation of causality, applied to its initial state’’ (1990: 129). 21. A similar suggestion is advanced by Jolley (1998: 611, n. 14). 22. In addition to the law of the series that encompasses the complete history of any monad’s states, Leibniz appeals to general laws that govern transitions between these states. Consistent with the analysis presented here, one state follows from another relative to a certain ‘‘mode of producing,’’ articulated in the relevant general law. And Leibniz insists that such laws themselves are contingent (DM 13; NE 2.21.13/ RB 178–179). On this point, see Kulstad 1990: 139–142, and note 25 below. 23. ‘‘The action of the internal principle which brings about the change or passage from one perception to another can be called appetition’’ (Mon 15/AG 215). 24. I use ‘will’ in the sense of what Leibniz calls ‘volition’: ‘‘the effort or endeavour (conatus) to move towards what one finds good and away from what one finds bad, the endeavour arising immediately out of one’s awareness of those things’’ (NE 2.21.5/RB 172). As Kulstad notes (1990: 147), Leibniz is not entirely consistent in this usage, since he also characterizes volitions in terms of our awareness of the endeavor itself as opposed to its object: ‘‘desires and endeavours of which we are aware are often called ‘volitions’ . . . whether or not they prevail and take effect’’ (NE 2.21.39/RB 192). 25. The contingency of acts of will (including acts of free will) is defended by Leibniz in the same way as the contingency of modal change in general. Such acts follow from causes that are sufficient, relative to the appropriate laws, to determine them. This is just what it means to say that the soul’s motives ‘‘incline without necessitating’’: ‘‘the event has nothing in it to render it necessary and to suggest that no other thing might have happened in its stead. And as for the connection between causes and effects, it only inclined, without necessitating the free agency, as I have just explained; thus it does not produce even a hypothetical necessity, save in conjunction with something from outside, namely, this very maxim, that the prevailing inclination always triumphs’’ (T 53; G 6:131–132/H 152). See NE 2.21.8/RB 175; 2.21.13/RB 178–179. I believe this account is consistent with Adams’s reading, defended in this volume, of Leibniz’s doctrine of moral necessity; for a dissenting view, see Murray’s essay in this volume. 26. Leviathan, chap. 6. 27. Following Leibniz, we can also describe such changes as ‘‘endeavors towards perfection’’ (RB 210). ‘‘Goodness or striving [appetitio] for the good is the perfection of the will’’ (T cd18/G 6:441), which leads any created substance towards a state of greater perfection. 28. Dictionnaire historique et critique, article ‘‘Rorarius,’’ Note H. The passage is translated in WF 73–74. 29. See also T 289: ‘‘In truth, we will only that which pleases us’’ (G 6:289/H 303). Against this, there is one passage in the New Essays where Leibniz characterizes desire more broadly, in a way that does not presuppose an awareness of ‘‘actual pleasure or suffering’’ (2.21.39/RB 192). 30. Here the objector may reply that what the dog misperceives is not the sensation of pain but the bodily effects of being struck. This might mean either that the dog correctly (but obscurely) perceives the damage done to its body but incorrectly perceives it as associated with a sensation of pleasure, or that it misperceives the effects of being struck and therefore misperceives them as pleasurable. The first alternative is ruled out by Leibniz’s psychological theory. Although one may have a mistaken belief that such-and-such bodily state is correlated with such-and-such mental state, there is no way to prize apart one’s perception of a bodily state and one’s affective response to that perception. They are aspects of the same mental state. On the second alternative, we are meant to imagine that the dog’s perceptions misrepresent the future state in which it is struck as a beneficial, pleasure-producing state as opposed to a harmful, pain-producing state. Since there is nothing special about the case in question, this would require supposing a systematic error in the dog’s perceptual apparatus, such that at the level of petites perceptions harmful bodily effects were routinely represented as beneficial ones. This is surely too high a price to pay.

180 Leibniz: Nature and Freedom 31. A different answer could be given in the case of a rational agent. Since such an agent has the capacity to form an idea of the goodness of things in general, based on a conception of their metaphysical or moral perfection, and since pleasure is just ‘‘the perception of perfection,’’ it is conceivable that such an agent might (consciously or unconsciously) desire a state in which the pain associated with the condition of its body was outweighed by the pleasure it took in the perfection of the universe. This, I claim, is inconceivable in the case of an agent, such as a dog, that lacks reason. 32. Leibniz himself highlights this distinction: ‘‘There are other efforts, arising from insensible perceptions, which we are not aware of; I prefer to call these ‘appetitions’ rather than volitions, for one describes as ‘voluntary’ only actions one can be aware of and can reflect upon when they arise from some consideration of good and bad’’ (NE 2.21.5/RB 173). 33. In notes on the same article as it appears in the second edition of Bayle’s Dictionnaire, Leibniz uses the example of a man stung by an insect while eating jam (G 4:546–547/WF 103). 34. Of course, strictly speaking, there is no physical universe over and above the system of phenomena represented in monadic perceptions. 35. In comments on Lamy’s De la Connoissance de soi-meˆme, Leibniz explicitly frames the distinction as one between the order of efficient causes and the order of final causes: ‘‘perceptions which express the laws of motion are linked together just like those laws, which they express according to the order of efficient causes. But the order of voluntary perceptions, which is that of final causes, is in conformity with the nature of the will’’ (G 4:580/WF 155). 36. A similar comment appears in a letter to Jaquelot from the same year (G 3:471). 37. See, e.g., the conclusion of the New System (G 4:485/AG 144). 38. At NE 2.21.8, beginning ‘‘The term ‘freedom’ is highly ambiguous’’ (RB 175). On the forms of constraint, see NE 2.21.13/RB 179. 39. Leibniz also characterizes it as ‘‘freedom of mind’’ or ‘‘self-mastery.’’ See NE 2.21.8/RB 175; T 289/G 6:288–289.

9

Moral Necessity ROBERT MERRIHEW ADAMS

eibniz frequently describes free decisions as ‘‘morally necessary’’ but not

L ‘‘metaphysically necessary.’’ This way of speaking is characteristic of Leibniz’s

Theodicy; more generally, it is especially frequent in writings from the last ten years of his life. That is particularly true of the historically pivotal application of the concept to God. Gaston Grua, who knew the texts as well as anyone in the twentieth century has known them, said that to his knowledge, it was only in 1707 that Leibniz began to speak of actions of God as ‘‘morally necessary’’ (Grua 1953: 234–235). In those last years, however, God’s choices join those of the virtuous sage as paradigms of the morally necessary for Leibniz; ‘‘it must be said that God, that the sage, is carried to the best by a moral necessity’’ (T 132). But such claims do not stop there; strange as it may seem, the category of moral necessity is extended in some passages to all voluntary choices, even sinful ones.1 Leibniz’s Jesuit friend Des Bosses got him to agree in correspondence in 1711, the year after the Theodicy was published, that he should not say ‘‘that there was a moral necessity of sinning in Adam or in anyone else,’’ and that ‘‘it is better never to acknowledge a moral necessity except to [do] good.’’2 This private concession has limited credibility, however; Leibniz linked it with the comment that he would prefer his words in general to be interpreted in such a way ‘‘that nothing bad sounding follows’’ (G 2:420). And in the last months of his life we find Leibniz speaking to Clarke of ‘‘a necessity which is moral, whereby [not only] a wise being chooses the best, [but] every mind follows the strongest inclination’’ (LC 5.4). What metaphysical necessity is for Leibniz we know pretty well. It is the type of necessity on which most ink has been spilled by his interpreters—an absolute necessity whose mark, for Leibniz, is that the negation of what is metaphysically necessary implies a contradiction. It is harder to say what he means by moral necessity, and that is the main subject of this essay. It is one that seems to me to have some intrinsic philosophical interest, whether or not the idea of moral necessity will go as far as we might like it to as interpretation of Leibniz. 181

182 Leibniz: Nature and Freedom

Leibniz found the expression ‘moral necessity’ used extensively in the jurisprudence of his time (as documented in Grua 1953: 222–226). In jurisprudence the expression typically meant obligation; it expressed the strongest modality in deontic logic. In previous writing on this subject I have interpreted ‘moral necessity’ as having that meaning in Leibniz. That was also Samuel Clarke’s interpretation; he took a moral necessity of doing good to mean no more than ‘‘that a good Being, continuing to be Good, cannot do Evil,’’ and complained, with some plausibility, that a necessity of that sort was not relevant to a dispute about freedom and determinism (G 7:423). Leibniz gives definitions that leave no doubt that he sometimes uses ‘moral necessity’ in this deontic, jurisprudential sense (A 6.1:465; Grua 608). Michael Murray (1996: 47–48) has argued, however, that Leibniz also uses the expression in another, metaphysically richer way, and in particular that he does so in the Theodicy; and I am persuaded that Murray is right about that. In this richer sense, to say that a decision is morally necessary is not merely to say that it ought to be made; it is to say something by way of explaining how and why it is made. Murray points out that the terminology of moral necessity had been used in such an explanatory sense by several Catholic theologians, mostly Spanish Jesuits, in the seventeenth century, beginning with Diego Ruiz de Montoya (1562–1632) and Diego Granado (1571–1632), and that Leibniz acknowledged acquaintance with the work of some of them.3 Ruiz and Granado developed a theory according to which a free choice—for instance, a choice of a perceived good—is neither metaphysically nor physically necessary, but is morally necessary, in view of the agent’s inclinations and perceptions of value. That this is an important part of the background to Leibniz’s thought about free will is highly plausible. Murray, and Sven Knebel (1991a, 1992), have been bringing to light a current of seventeenthcentury, mainly Jesuit, philosophical theology that is very interesting and deserves to be better known than it has been. It was certainly known to Leibniz, and provided precedents for some aspects of his thought, including his ‘‘optimism.’’ Nonetheless, I think we cannot rely very heavily on the Jesuits for the content of our interpretation of Leibniz’s views on freedom, for several reasons. (1) My most general reason for this caution is that Leibniz seems to me a remarkably learned writer, but not a very scholarly one. He is generally eager to present himself as agreeing with other writers, but seems to care much less whether he agrees with them in precisely the sense intended by them. (2) He often cites by memory, and his appreciative mention of these Jesuit authors tends to be accompanied by disclaimers of current or extensive knowledge of their writings (G 2:362, 366, 436, 450). (3) There are possible disagreements to be considered between Leibniz and his Jesuit precursors. In particular, we may wonder whether Leibniz agrees with Ruiz and Granado in opposing moral necessity to physical necessity, or only to metaphysical necessity—a point discussed by Murray to which I shall soon return. (4) There are also significant differences in the development of the idea of moral necessity by different seventeenth-century theologians, as Knebel and Murray make clear, so that even if we suppose Leibniz to be agreeing with views of theirs, there will remain a question as to which version of the views he has adopted. For these reasons, and because I think I see how to do it in a way that may be interesting,

Moral Necessity

183

I will try simply to fit to the texts of Leibniz some speculations about moral necessity.

Moral and Physical Necessity It is certainly not only to metaphysical necessity that moral necessity is opposed in Leibniz’s thought. The competition here is among answers to questions of the form ‘‘Why is it certain, or determined, that p?’’ The ‘‘machinery,’’ so to speak, of determination and explanation invoked by the competing conceptions is quite different: formal or logical implication and contradiction in the case of metaphysical necessity; ‘‘final causes,’’ as Leibniz calls them, in the case of moral necessity. But these are not the only relevant alternatives; moral necessity is opposed to any valuefree mechanism determining choice. One passage in which Leibniz appears to oppose moral necessity to physical necessity, or at any rate to physical causation, is particularly illuminating. Arguing against Thomas Hobbes, Leibniz writes: M. Hobbes is no more willing to admit talk of a moral necessity, because in effect everything happens by physical causes. But there is reason, nonetheless, to make a big distinction between the necessity that obliges the wise to do good, which is called moral, and which has a place even in relation to God, and that blind necessity by which Epicurus, Strato, Spinoza, and perhaps M. Hobbes have believed that things exist without intelligence and without choice, and consequently without God, who in effect would not be needed according to them, since according to that necessity everything would exist by its own essence, as necessarily as two plus three must make five. And that necessity is absolute, because everything it carries with it must happen, whatever one does; whereas what happens by a hypothetical necessity happens in consequence of the supposition that this or that has been foreseen or resolved, or done in advance; and moral necessity carries an obligation of reason, which always has its effect in the wise. This species of necessity is happy and desirable, when one is carried by good reasons to act as one does; but necessity blind and absolute would overturn piety and morality. (T h3)

Moral necessity is opposed in this passage both to ‘‘physical causes’’ and to an ‘‘absolute necessity’’ of which mathematical necessity is an example, though Leibniz is not at pains here to distinguish these alternatives to moral necessity. What is most important to Leibniz, I believe, about the types of necessity opposed here is that they are all blind necessity, necessity that operates ‘‘without intelligence and without choice.’’ Moral necessity is not blind; in its operation ‘‘one is carried by good reasons to act as one does.’’ Another important mention of ‘‘blind necessity’’ in the Theodicy comes in a discussion of Spinoza, who, according to Leibniz, ‘‘appears to have explicitly taught a blind necessity, having denied to the author of things understanding and will, and imagining that good and perfection relate only to us and not to him’’ (T 173). Particularly significant in this comment is the clear indication that ‘‘good and perfection’’ will play an essential part in any pattern of determination and explanation that avoids blind necessity. Value has explanatory force in the Leibnizian

184

Leibniz: Nature and Freedom

scheme of things. The best of all possible worlds is chosen and actualized because it is the best, and Leibniz even implies that God exists because God is the most perfect possible being (see Adams 1994: 164–76). This is connected, I believe, with the idea of final causes, whose soundness and importance Leibniz fervently and persistently advocated. In final causation, as conceived by Leibniz, something is done, or occurs, because of the value, or apparent value, of an end that is an object of choice or appetition.4 Final causes are presented by Leibniz, in a central metaphysical context, as opposed to causes that we might call physical. It will be important to remember that our sense of ‘physical’ is narrower than its usual sense in Leibniz’s lifetime. Our sense may be close to what Leibniz meant in speaking about ‘‘physical causes’’ in the passage about Hobbes that I quoted; but he usually speaks of ‘‘bodies’’ where we might speak of the ‘‘physical,’’ and it is efficient causes and the realm of bodies that are opposed to final causes in a particularly interesting section of his final letter to Clarke: All the natural forces of bodies are subject to mechanical laws, and all the natural forces of minds are subject to moral laws. The former follow the order of efficient causes, and the latter follow the order of final causes. The former operate without liberty, like a watch; the latter are exercised with liberty. . . . (LC 5.124)5

Though moral necessity is not mentioned in this section, this is the same letter in which Leibniz says it is by moral necessity that ‘‘every mind follows the strongest inclination’’ (LC 5.4). We may surely infer that ‘‘moral laws’’ and ‘‘the order of final causes’’ are the way in which Leibniz thinks moral necessity works. The contrast between efficient and final causes is one of Leibniz’s ways of characterizing one of the most fundamental distinctions in his system, that between monads and bodies. In section 79 of the Monadology he says: Souls act according to the laws of final causes by appetitions, ends, and means. Bodies act according to the laws of efficient causes or of motions. And the two realms, that of efficient causes and that of final causes, are harmonious with each other.

In this text ‘‘the laws of efficient causes or of motions’’ are surely the laws of Leibniz’s mechanistic physics. And while only ‘‘souls’’ are mentioned here on the other side, Leibniz’s system of preestablished harmony surely requires that all monads or simple substances be included in the ‘‘realm of final causes’’—the more so inasmuch as the internal causal structure of all monads is supposed to work by ‘‘appetitions,’’ which we must certainly understand teleologically. I do not know of a text in which Leibniz explicitly applies the concept of moral necessity or moral laws to the actions of subrational monads; perhaps he would not and should not. But the difference between moral necessity and whatever necessity determines the operations of humbler monads can hardly be one of the most fundamental differences in his system; for the appetitions of the humblest monads are conceived by analogy with conscious human purposiveness, and even in our own case ‘‘an infinity’’ of factors of which we are not conscious is apt to be involved in the final cause of our actions (Mon 36). The idea on which I think it may be most interesting to focus in trying to understand what Leibniz says about moral necessity and final causation is this: that

Moral Necessity

185

the value or apparent value of an end, or some analogue of apparent value, is sufficient of itself to determine the action of a substance. God chooses the best just because it is the best, other minds choose the apparent best just because it is the apparent best, and even the humblest monad does something analogous. In relation to minds, it would be claimed that the ‘‘moral laws’’ are strong enough to determine choice infallibly in advance. Sure as the moral laws are to be followed, however, there would be no logical incoherence in a mind’s failing to follow them; the choice is in that sense not metaphysically necessary. And there is no value-free mechanism that makes the mind obey the moral laws; the choice is in that sense not physically necessary. The boldest form of this thesis would be one in which it is denied that facts of value can be reduced to anything metaphysically more fundamental, but I will not venture here into the complex topic of Leibniz’s views about the metaphysical structure of facts of value. An account in which the value or apparent value of an end is ‘‘sufficient of itself’’ to determine the action of a substance is one in which there is no further mechanism to explain how final causation works.6 In any system or pattern of explanation it seems there will be a bottom layer, at which, at least in practice, explanation comes to an end. In typically modern patterns of efficient causal explanation, the most general laws of nature are allowed to be part of such a bottom layer. I take Leibniz to have supposed that ‘because it seemed to s to be the best thing to do’ can function in that way as part of a bottom layer of explanation and causation—a rock bottom ‘‘mechanism’’ of final causation, so to speak. I don’t think he thought there must be a further how here, a how simple substances choose what seems to them best.7 It will hardly escape notice that the interpretation I have just sketched would make Leibniz in one way a precursor of Kant’s theory of freedom. On Kant’s view, freedom is a sort of causality, and as such requires a law. Its law is the moral law. To be free is to able to do something just because the moral law requires it. Kant’s view is less deterministic than Leibniz’s, however: Kant does not think the moral law determines choice infallibly; we can and do violate it. Kant may seem less optimistic than Leibniz at this point, and in some sense he is certainly a less optimistic philosopher; but we must bear in mind that the ‘‘moral laws’’ we all follow infallibly according to Leibniz are much less ideal than the Kantian moral law, as they are tailored to our very imperfect apprehension of the good. It is time to focus on the deterministic aspect of Leibniz’s thought. In denying that free actions are necessitated, he generally goes out of his way to say that it does not follow that they are not determined. Indeed, as I have indicated, I think the concept of moral necessity is supposed to play a part in explaining how free actions are determined. I have suggested that they are supposed to be determined by the value or apparent value, or something analogous to the apparent value, of an end; and presumably the real value of an end plays at least a part in determining its apparent value. But values can hardly work alone in determining an action. If values can determine the action of a substance, that is because the substance is sensitive to value, has the capacity to perceive and appreciate value to some degree, and the capacity to pursue an end because of its perceived value, or an analogue thereof—the capacity for value-oriented appetition.8 These capacities will be part

186 Leibniz: Nature and Freedom

of the nature of the substance. Thus it seems that the action of the substance— even the free action of a mind—will be determined jointly by its nature and the values of the ends for which it could have appetition. At this point in the argument we have returned to the relation between moral and physical necessity. Only here it seems not to be a relation of opposition. Rather, moral necessity seems likely to be subsumed under physical necessity. To see why that is so, we must remember that in Leibniz’s time, ‘physical’ does not just mean ‘corporeal.’ ‘Physical’ comes from the Greek j
usiB which means nature, and physical necessity is generally necessity flowing from the natures of things.9 But then, if it is determined by the natures of minds, even without metaphysical necessity, that they always choose the apparent best, is that not physically necessary, in a broad sense? That the actions of substances are predetermined by their natures—that is, by their primitive forces—is unquestionably a thesis of Leibniz’s philosophy; the preestablished harmony depends on it. There is still the question whether that predetermination allows of an interesting opposition between moral necessity and other sorts of necessity. We have already seen how moral necessity is supposed to be opposed to ‘‘blind’’ necessity, and that seems to me relatively unproblematic. The opposition between moral and metaphysical necessity seems to require Leibniz to deny that free actions follow logically, by the principle of noncontradiction, from the natures of substances. That may be more problematic for him; I’ll come back to that in the third section herein. The question before us at the moment is whether there is yet another sort of necessity, to be called ‘‘physical’’ necessity, by which free actions do not follow from the natures of substances. Michael Murray suggests that ‘‘Leibniz did not think physical or causal necessity compatible with freedom.’’10 I am uneasy about the use of the term ‘causal necessity’ here. I have not noticed Leibniz using it (though I could of course have overlooked it), and it is not obvious with what we should identify causal necessity in a Leibnizian context. More than one candidate is available. Leibniz’s explanation of morally necessary free actions is explicitly in terms of causes—final causes. That explanation is opposed to an explanation in terms of ‘‘efficient causes’’; but what is meant by the latter seems to be quite specifically and narrowly a mechanical explanation. Leibniz certainly does not mean to deny, but rather to affirm, that free actions are caused—or at any rate, done—by agents, substances. And he holds that they are predetermined by prior states of those substances, as required by the preestablished harmony. It is not surprising, then, that in an earlier essay Leibniz spoke of efficient causes in a broader sense, contending that ‘‘even final causes can be referred to efficient causes, that is, when the agent is intelligent, for then it is moved by thought’’ (Grua 28). Even in his fifth letter to Clarke, Leibniz describes the operation of moral necessity in terms that I think most of us would call strongly causal: It is true that Reasons in the mind of a Wise being, and Motives in any mind whatsoever, do that which answers to the effect produced by weights in a balance. The author [Clarke] objects that this notion leads to necessity and fatality. (LC 5.3)

Moral Necessity

187

It is in the course of distinguishing types of necessity in response to this objection that Leibniz makes the statement I have quoted already that by moral necessity ‘‘a wise being chooses the best, and every mind follows the strongest inclination’’ (LC 5.4). This is not an isolated instance; Leibniz liked mechanical analogies for the will (RB 193; T 22; G 7:304/L 488). Volition is like mechanical causation in that the effect is determined with certainty, and by prior facts; they are unlike, of course, in that volition does not have a value-free mechanism. They are unlike further, as Leibniz claims a bit later in the letter, in that the choosing mind is active in a way that a balance is not. ‘‘Properly speaking, motives do not act upon the mind, as weights do upon a balance; but ’tis rather the mind that acts by virtue of the motives, which are its11 dispositions to act’’ (LC 5.15). But the fact that the action of the motives is the mind’s own action does not show that the motives determine less surely than the weights in a balance; on the contrary, it shows that ‘‘if the mind should prefer a weak inclination to a strong one, it would act against itself, and otherwise than it is disposed to act’’ (LC 5.15). If in one passage that we have seen Leibniz opposes ‘‘physical causes’’ to ‘‘moral necessity,’’ elsewhere he brings them into positive connection, and even subsumes physical necessity under moral necessity. At the outset of the Theodicy he tells us that ‘‘physical necessity is founded on moral necessity, that is on the choice of the wise, worthy of his wisdom; and that the one as well as the other must be distinguished from geometrical necessity’’ (T pd2). The reason for this dependence is that the laws of nature, seen here as the source of physical necessity, depend on God’s free choice,12 which is produced by moral necessity but not by metaphysical necessity, which is here called ‘‘geometrical.’’ And undoubtedly God’s choice of the best is the single most important and most foundational case of final causation in the Leibnizian scheme of things. Leibniz’s system also offers us, however, another and less theological reason for thinking that final causation and moral necessity are fundamental to all natural causation. As I have noted earlier, the Leibnizian simple substances, the monads, act by appetition, by a sort of final causality, and by moral necessity or an analogue thereof. But it is the simple substances, and they alone, that are the ultimately real agents in the Leibnizian universe. The bodies, the ‘‘realm of efficient causes,’’ are only ‘‘well-founded phenomena.’’ Their motions, governed by mechanical laws of efficient causation, exist only as harmonious appearances to monads. The ultimately real events that ‘‘found’’ such reality as the motions possess are changes in monads and are governed by moral or quasi-moral laws of final causation. It is only the processes of final causation that are metaphysically fundamental and that determine what motions shall appear to occur in accordance with the laws of mechanical or efficient causation. That the processes of final causation dictate an appearance of motion in accordance with mechanical laws is due, I take it, to the great beauty and value of mechanical order. From these considerations I believe we must conclude that the Leibnizian system does not include a metaphysically fundamental type of causal necessity that can be opposed to moral necessity. Still it is true that Leibnizian moral necessity operates in a very different way from what most philosophers today would think of

188 Leibniz: Nature and Freedom

as causal necessity. It operates by final causes, and excludes determination by any value-free mechanism. Other questions can be raised about the relation of moral necessity to laws of nature. I believe the Leibnizian system will allow us to ascribe to free decisions at least the following independence from laws of nature: no free action can be predicted, precisely and with certainty, from law-like generalizations and other facts, by any calculation that can be performed by finite minds (see DM 7 and 17; C 21–22/MP 102). Of course, that same sort of independence can be ascribed to some motions of bodies. In particular, and most obviously on Leibnizian assumptions, it can be ascribed to those motions that are the corporeal, phenomenal counterpart of free decisions. These factors seem to me to distinguish Leibnizian moral necessity less fundamentally than the value-laden teleological character of its operation.

Moral and Metaphysical Necessity I want now to take a closer look at the relation between moral necessity and metaphysical necessity. As we have seen, Leibniz says that free decisions are morally but not metaphysically necessary. I have argued (Adams 1994: chap. 1) that Leibniz has two main accounts of how any truths can be metaphysically contingent, given certain constraints operative in his system that might be thought to force him to regard all truths as necessary. I believe that Leibniz’s conception of moral necessity presupposes both of these accounts, and thus is not an alternative to them, but at most a supplement to them—though we may consider interpreting it in a way that puts it in tension with some of his views about the basis of metaphysical necessity. In one of Leibniz’s theories of metaphysical contingency, facts are allowed to be contingent if they have alternatives that are possible in their own nature, completely free of internal contradiction or incoherence, whether or not a being possessing the perfect nature that God possesses could coherently choose to actualize them. It is this theory of contingency that appears to be invoked in Leibniz’s explication of moral necessity in his fifth letter to Clarke, when he says that choice of the good is not absolutely necessary. For when God (for instance) chooses the best, what he does not choose, and is inferior in perfection, is nevertheless possible. . . . For God chooses among possibles, that is, among several alternatives, none of which implies a contradiction. (LC 5.8)

In this account of free choice, the centerpiece of the argument that ‘‘good, either true or apparent . . . inclines without necessitating’’ (LC 5.8) is an appeal to the internal coherence of alternatives for choice, rather than a claim about the nature of the chooser. Here I shall set this line of argument aside, however, in order to focus on Leibniz’s other main theory of contingency, the one that starts with his thesis that in every true affirmative proposition the concept of the predicate is contained in some way in that of the subject, and proceeds to distinguish necessary truths, as

Moral Necessity

189

those that can therefore be reduced to identities in a finite series of steps of analysis, from contingent truths, whose analysis will yield no such result though it be continued to infinity. We should not expect Leibniz to exempt facts of free choice from his conceptual containment thesis about truth, since he regularly says that it applies to contingent as well as necessary truths (see Adams 1994: 42–43). We should ask rather how predicates of free choice are included in the complete concepts of choosers, and whether there is a manner of such containment that is consonant with moral necessity. It may be helpful in this inquiry to consider first the question how existence is contained in the concepts of those creatures that actually exist. The answers Leibniz proposes to this question13 do not treat existence as a fundamental or primitive constituent of the definitive concepts of finite things. The only being that exists by its own nature alone is God. The existence of other things that actually exist does not follow from their own concepts alone but only in conjunction with considerations about God. The existence of an actually existing creature, we may say, is contained in its individual concept, not directly and immediately, even in an infinite analysis, but only in the sense that the concept contains reasons that are sufficient (together with facts about God) to determine and explain the existence of the creature. And the most significant fact about these reasons, from the point of view of a theory of moral necessity, is that Leibniz clearly assumes that the reasons for existence to be found in the concepts of creatures will turn on the value of the creatures or, more precisely, on the comparative values of the possible worlds of which they are parts. They are reasons for the existence of the creatures because they are reasons for God to choose those creatures. And God’s choice is guided, with moral necessity, by the values of the alternative possible worlds; God chooses the best. In this pivotal example we see a way in which the conceptual containment thesis and the concept of moral necessity seem to mesh very well. The conceptual containment thesis admits more than one pattern of determination and explanation of predicates by the concept of the subject. One such pattern, as we have just seen, involves the same sort of value-laden teleological structure that is involved in moral necessitation. If the existence of a finite substance can follow from its concept in this way, so, surely, can the choices a substance makes follow from a concept that has among its constituents, in the first instance, not the choices themselves, but valueladen reasons from which the choices follow by moral necessity. And this necessity, we may add, will not be a metaphysical necessity because no finite analysis would suffice to derive the choices from the reasons that are primitive constituents of the concept. With this fusion of moral necessity and the infinite analysis conception of metaphysical contingency, a Leibnizian theory of moral necessity could reach its conclusion. And so perhaps it should; but I’d like to play a bit with the thought of another step in this direction, a further mitigation of Leibniz’s determinism, though I doubt that it can work for him. The further step would be a strengthening of the claim that when an agent chooses freely, there would be no formal or logical incoherence in the agent’s choosing otherwise. If Leibniz relies only on the infinite analysis theory to shield free choice from metaphysical necessity, then the claim just mentioned means

190 Leibniz: Nature and Freedom

only that there would be no finitely provable incoherence in the agent’s doing otherwise. One might want more than that, for there may also be broadly formal incoherence that is not finitely provable. Suppose, for the sake of argument, that Goldbach’s conjecture is true but not provable. Every even number is the sum of two primes, but there is no proof of that mathematical fact. Then the hypothesis of a number that is even but not the sum of two primes is arithmetically incoherent in a way, though it is not provably incoherent. There are clearly points in Leibniz’s philosophy at which he demands a broadly formal coherence that is more than freedom from provable incoherence. Possible worlds, for example, must have the more extensive, broadly formal, but not merely proof-theoretical coherence. If there is, for example, no possible world that is as good as the actual world without the horrors of the Thirty Years’ War, that is surely not because there is a finitely provable dependence of great goods on those evils. Let us ask, then, whether a free agent’s choosing otherwise must be free, not only from finitely provable incoherence, but from any sort of broadly formal incoherence. It seems that an affirmative answer to this question would yield a less deterministic version of Leibniz’s philosophy. For if the agent’s choosing otherwise would be formally incoherent, though not provably so, that would seem intuitively to be a metaphysically strong sort of necessitation, even if Leibniz chooses not to classify it as ‘‘metaphysical necessity.’’ But if choosing otherwise would be formally coherent in every way, then any ‘‘moral necessity’’ that lies behind the choice would seem to have much less metaphysical bite. But in that case, how can the particular choice that is made be determined and perfectly certain in advance? Thoroughgoing indeterminists would be happy to say that it is not determined with certainty in advance, but Leibniz is certainly not indeterminist enough to accept that conclusion. Can he appeal at this point to the Principle of Sufficient Reason? It is determined and certain in advance that the agent will choose the apparent best, because the apparent bestness of the choice is a sufficient reason for it, and there is no sufficient reason for any alternative choice. On this view the Principle of Sufficient Reason would indeed be ‘‘the principle of contingent truths.’’ It would have to be strongly independent of the Principle of Noncontradiction; otherwise there would be at least a broadly formal incoherence, after all, in something happening without a sufficient reason. Indeed it seems that in order to play this role, the Principle of Sufficient Reason itself would have to be metaphysically contingent in the strong sense that its falsity would not involve even a broadly formal incoherence. But then how could the Principle of Sufficient Reason be strong enough to ground the determination of our choices, with certainty, in advance? There is in some sense supposed to be no real chance in the Leibnizian universe that any rational agent will choose other than the apparent best. What could be the status of a principle that keeps that from being a real chance? It must be synthetic, on this hypothesis, in the sense that, as I have said, its falsity would not involve even a broadly formal incoherence. And of course, nothing that is not analytic—indeed, finitely analytic—will count as ‘‘metaphysically necessary’’ for Leibniz. Some philosophers, however, have thought there are synthetic necessary truths—necessary,

Moral Necessity

191

obviously, not in Leibniz’s sense, but in some other way that seems metaphysical enough and is often called ‘‘metaphysical.’’ Could the Principle of Sufficient Reason have for Leibniz something like the status that other philosophers have conceived as synthetic necessary truth? Leibniz will not call that status ‘‘metaphysical necessity,’’ but is that more than a verbal point? This suggestion faces objections on two fronts. On the one hand, we may doubt that it does much to mitigate Leibniz’s determinism. If an agent’s doing otherwise would violate a principle that is something like a synthetic necessary truth, a principle that God knows will never be violated, does that leave us able to say that the agent ‘‘could have done otherwise’’ in a sense much stronger than that in which we could say it if doing otherwise would violate a principle that is broadly analytic though not finitely provable? That is a difficult question. The opposite objection regards only the interpretation of Leibniz. It is not easy to determine what Leibniz believed about the necessity or contingency of the Principle of Sufficient Reason, and its relation to the Principle of Noncontradiction.14 It is clear, however, that in some of his logical writings he committed himself to the view that all truths are at least broadly analytic—that all facts are determined, at bottom, by purely formal considerations. This is, I think, an implausible view—so implausible that Leibniz, like virtually every other philosopher, says a number of things not easily reconcilable with it; and many of his ideas, fortunately, are detachable from it. Nonetheless, it appears to be a view that he held and thought important; it seems to be presupposed in his infinite analysis theory of contingency. And it obviously implies that the Principle of Sufficient Reason, and all facts about free choice, are broadly analytic, and their negations incoherent, even if not provably so. Similar things can be said about the other line along which Leibniz might try to explain the infallible predetermination of free choices without appeal to even broadly analytic truth—namely, in terms of the nature of the choosing agent. In an early essay he wrote that ‘‘even moral causes are natural, for they are taken from the nature of the mind’’ (Grua 28).15 Early and late he writes that ‘‘God wills the best by his own nature’’ and that ‘‘it follows from the nature of God that he prefers the most perfect.’’16 And I have already argued that if values determine choices, in a Leibnizian context, the nature of the chooser must participate in the determination. Certainly Leibniz thought that particular free choices do not follow from a mind’s nature by a finite formal demonstration, and are thus not metaphysically necessitated in his strict (and arguably artificial) sense. May he also have supposed that the connection between the mind’s nature and its choices is not even broadly analytic?—that the hypothesis of a mind with the same nature choosing otherwise would not be attended with even the sort of incoherence that may be thought to attend a mathematical hypothesis that is false but not provably false? And may he at the same time have supposed that the connection between the mind’s nature and its free choices is nonetheless strong enough for the nature to determine the choices with certainty in advance? This thought faces a pair of objections similar to those that attended the alternative account in terms of the Principle of Sufficient Reason. On the one hand, if the nature of a mind infallibly determines its choices, that seems to imply a strong

192 Leibniz: Nature and Freedom

determinism, hardly weaker than what is normally thought of as causal determinism or causal necessitation of the choices (though not, of course, determination by a value-free mechanism). We may wonder how much this mitigates Leibniz’s determinism. It is not clear to me whether it would be more mitigating than determination by a Principle of Sufficient Reason that would have something like the status of a synthetic necessary truth. The opposite objection attacks the claimed nonanalyticity of the connection between the mind and its choices. Even outside the Leibnizian context, when it is said that some fact follows from the nature of a thing, it is often assumed that the connection is at least broadly analytic—perhaps because of the relation between the concepts of nature and definition in the philosophical tradition. For Leibniz, of course, to the extent that he is committed to a theory that grounds all truth on formal syntactical relationships, the following of particular choices from the natures of the agents must be at least broadly analytic, as all truths must be. And his statement, which I have quoted, from his fifth letter to Clarke, that ‘‘if the mind should prefer a weak inclination to a strong one, it would act against itself, and otherwise than it is disposed to act’’ (LC 5.15), seems to suggest pretty specifically that there would be something incoherent in a mind’s acting otherwise than according to its strongest preexisting inclination. I conclude that while things Leibniz says about moral necessity fit into very interesting views about how free choices are determined and explained, they are unlikely to lead us to a much softer reading of his determinism.

Notes This essay was presented to the Pacific Division of the American Philosophical Association in April 2000, and to the philosophy department of Purdue University in September 2000; I first developed some of the main ideas in a seminar on Leibniz at Yale University. I am indebted to many for helpful discussion—in particular to Jan Cover, Martin Curd, Patricia Curd, Sukjae Lee, and especially to Michael Murray, who presented extensive written comments on the paper at the APA. 1. Most explicitly in T a3; apparently also in T 282. Both passages are explicitly targets of Des Bosses’s objection (G 2:421). 2. G 2:418–420; see also G 2:421–424. The concession is noted in Grua 1953: 237. 3. Murray 1995: 106–107; see Murray 1996: 37–43. 4. In Leibniz’s conception of final causation, unlike Aristotle’s, the valued end must be an object of perception and appetition to the subject or agent. Arguably, also, the good of the whole universe, as distinct from that of the agent, plays a larger role in Leibnizian than in Aristotelian final causation. I am indebted to discussion with Patricia Curd on these points. 5. See G 7:451/L 472, from 1696. 6. I am indebted to Jan Cover for pressing me on the issue addressed in this paragraph. 7. In Leibniz’s system this may apply most cleanly to God’s choosing; I am abstracting here from the role that God’s ‘‘concurrence’’ may play in explaining the choices of finite substances. 8. Such capacities could of course figure in indeterministic as well as deterministic theories of final causation. 9. When Granado, for instance, says that when ‘‘God wills the best, even if it is not necessary physically, but rather absolutely free . . . still it is morally necessary’’ (quoted in Knebel 1991a: 5, n. 11), he is presumably not making the irrelevant (because platitudinous) exclusion of corporeal necessity from God, but rather allowing that God’s choice of the best does not follow necessarily from the divine nature.

Moral Necessity

193

10. Murray 1995: 97; see Murray 1995: 95, and Murray 1996: 49. 11. Here I follow Clarke’s translation, as reprinted by Alexander, and G 7:392, which reads ses, as the sense of the passage seems to demand. The critical edition, Robinet 1957, has des. The reading I have chosen implies nothing that is not clearly implied in the passage as a whole. 12. See RB 178–179. 13. See the texts cited in Adams 1994: 43 (including n. 65 on that page). 14. See Parkinson 1965: 62–69. 15. Dated 1683–86 by watermark, according to A.6.4:2323. 16. Grua 289 (probably from the early 1680s) and 393 (from 1698).

10

Spontaneity and Freedom in Leibniz M I C H A E L J. M U R R A Y

n the history of metaphysics philosophers have disagreed, sometimes with vigor,

I over whether or not an agent can be free and responsible in choosing, while also

being determined to so choose. In contemporary metaphysics those who think an agent can be free and responsible while so determined we call ‘‘compatibilists’’ and those who do not we call ‘‘libertarians.’’ Historically, those who are committed to libertarianism are usually so committed for at least one of two reasons. First, some are convinced that the very idea of an agent acting freely and responsibly is incoherent when sufficient conditions for the choice obtain, whether internal or external to the agent. If, it is claimed, the choice of the agent can be traced back to states of affairs that are sufficient for the choice, the choice is simply a consequence of those conditions, and thus not an act of the agent himself. As a result, such choices are neither free nor something for which the agent is morally responsible. Others, however, are committed to libertarianism for theological reasons. The intuition here is that if the agent is determined to do what she does by external forces, then the agent is not ultimately responsible for her actions and, what is worse, God is. As a result, libertarians of this sort aim to preserve the notion of free and responsible action in the world as a way of preventing culpability from working its way back to God. Libertarian freedom provides the needed buffer. Likewise, those who are committed to compatibilism are so committed for one of two reasons. First, some are convinced that the very idea of an agent acting freely and responsibly is incoherent unless the agent is the ultimate source of the action. However, the compatibilist thinks that ultimacy requires (merely?) that cognitive and/or conative states internal to the agent be the proximate cause of the act. Further, the compatibilist thinks that this is a sufficient condition for an act being free. As a result the compatibilist might agree, and find it uninteresting, that the internal states that are the proximate cause of the act are themselves, in turn, caused by other states, perhaps even other states external to the agent. 194

Spontaneity and Freedom 195

Other compatibilists, more rare among metaphysicians these days, are motivated by theological considerations. The Reformed theological tradition is noted, or perhaps notorious, for endorsing compatibilism on such grounds. This tradition regards God’s unfettered exercise of providence over nature to be a centerpiece of proper theology. Thus, Reformed theologians typically defend the claim that all events in the created order happen as a result of unconditional divine decrees. In order for God to exercise robust control over creation, all events in the created order must find their ultimate cause in God. Thus Francis Turretin, a typical seventeenth-century Reformed theologian, affirms that the divine decrees must be held to necessitate each event, offering five arguments for the claim: (1) All things were decreed of God by an eternal and unchangeable counsel; hence they cannot but take place in the appointed time. Otherwise the counsel of God would be changed, which the Scriptures declare to be impossible. . . . (2) Scripture predicts such necessity . . . ‘‘The Son of Man goeth, as it was determined’’ (Luke 22:22). . . . (3) The most fortuitous and casual things are said to happen necessarily. . . . (4) As all things are foreseen by an infallible foreknowledge, so they must necessarily happen infallibly. (5) They are certainly predicted as future so that the word of God cannot fail, nor can the Scripture be broken. Therefore, they must happen necessarily. (Turretin 1992: vol. 1, 320)

Still, in spite of the affirmation of the necessitation of all events by unconditional divine decree, Reformed thinkers still believed there to be room for creaturely freedom. Thus, Turretin continues: Although in relation to the first cause, all things are said to be necessary, yet taken according to themselves, certain things can be free, contingent, and fortuitous, because each thing is and may be judged according to proximate and particular causes. . . . Hence we might rightly say, ‘‘Adam sinned freely and necessarily’’: the latter with respect to the decree and the futurition of the thing; the former with respect to his will and as to the mode. For no matter what the necessity of the decree, still Adam sinned voluntarily and consequently most freely. (Turretin 1992: vol. 1, 321)

Yet while these libertarians and compatibilists come down on opposite sides of the question of whether or not freedom and determination are compatible, they all agree that free and responsible acting requires that the agent be the locus of explanation in some central sense. And thus each camp is, in its own way, committed to the claim that when it comes to free action, the explanation for the act properly terminates in the agent in some important sense. This internal, selfinitiating character of free acts is what both compatibilists and libertarians mean when they talk of ‘‘spontaneity.’’ Careful attention to the way in which figures in the history of metaphysics use the word ‘spontaneous’ (and its antonyms, such as ‘violent’ and ‘coerced’) is useful in getting to the heart of their respective positions on freedom, because it allows one to discern just what kinds of determination each figure takes to be consistent or inconsistent with freedom. For each different sort of determination that presents a

196 Leibniz: Nature and Freedom

threat to freedom, there is a corresponding variety of ‘‘spontaneity.’’ But what sorts of determinism were seen to be threatening to freedom in the seventeenth century, and what sorts of spontaneity were touted as necessary conditions for free action? In his Institutio, the Reformed theologian Turretin provides a list of six sorts of determination or necessity that were regarded by various figures as inconsistent with spontaneity, and thus with freedom. They are as follows: First, the necessity of coercion arising from an external agent (he who is compelled, contributing nothing). Second, physical and brute necessity occurring in inanimates and brutes who act from blind impulse of nature or a brute instinct and innate appetite, without, however, any light of reason . . . and without any choice. The third necessity is the necessity of the creature’s dependence on God . . . (to wit, the government of providence: [1] in the antecedent decree; [2] in the subsequent execution). This necessity is called hypothetical both of infallibility (with respect to prescience) and of immutability (with respect to the decree and actual concourse). Fourth, rational necessity of determination to one thing by a judgment of the practical intellect (which the will cannot resist). Fifth, moral necessity or slavery arising from good or bad habits and the presentation of objects to their faculties. . . . Hence it will happen that the will (free in itself) is so determined either to good or to evil that it cannot but act either well or badly. . . . Sixth, the necessity of the existence of the thing or of the event, in virtue of which, when a thing is, it cannot but be. (Turretin 1992: vol. 1, 661–662)

Advocates of the most extreme forms of libertarianism in the period would require that free actions are not determined in any of the first five respects. Leibniz, no friend of libertarianism, does not want to go so far. But Leibniz sees more of these sources of determination as threats to freedom than are traditionally acknowledged. In this essay I will argue that Leibniz takes freedom-preserving spontaneity to be inconsistent with the first three types of necessity described here. In addition, I will argue that Leibniz has similar difficulties with necessity of the fourth and fifth sorts when the necessity is understood to be metaphysical or physical necessity. To that end, I begin, in the second section, with an examination of a variety of spontaneity often endorsed by Leibniz, which clearly will not serve his deeper metaphysical aims. This sort of freedom is the sort Leibniz attributes to substances in virtue of his basic metaphysical commitments, which include the claim that substances contain within themselves the complete ratio for all of their properties. In the third section I then go on to show that Leibniz rejects the first two sorts of necessitation, physical necessity and coercion. In the fourth section, I show that Leibniz is willing to accept the claim that the will is necessitated by practical deliberation and/or moral habits, as long as the sense of necessity is carefully qualified. Finally, in the fifth section I argue that Leibniz rejects the third sort of necessity, that is, divine necessitation, by arguing that such necessitation, if present, would both undermine creaturely freedom and render God culpable for creaturely sin. From this I will conclude that Leibniz has many more anticompatibilist sympathies than is commonly recognized.

Spontaneity and Freedom

197

Freedom and Spontaneity-for-Free Throughout his philosophical career Leibniz affirms and reaffirms his commitment to the claim that spontaneity is a necessary condition for freedom. The oftcited definition of freedom from the Theodicy is typical: I have shown that freedom, according to the definition required in the schools of theology, consists in intelligence, which involves a clear knowledge of the object of deliberation, in spontaneity, whereby we determine ourselves, and in contingency, that is, in the exclusion of logical or metaphysical necessity. Intelligence is, as it were, the soul of freedom, and the rest is as its body and foundation. The free substance is self-determining and that according to the motive of good perceived by the understanding, which inclines it without necessitating it: and all the conditions of freedom are comprised in these few words. (T 288/H 303)

And shortly thereafter he affirms the standard doctrine of spontaneity as follows: The spontaneity of our actions can therefore no longer be questioned; and Aristotle has defined it well, saying that an action is spontaneous when its source is in him who acts. . . . Thus it is that our actions and our wills depend entirely upon us. (T 301/H 309–10)

Clearly, Leibniz thinks there is something about his account which guarantees that we act with spontaneity. What is it? The answer can be discerned from a variety of texts, of which the following are typical: [T]rue spontaneity is common to us and all simple substances, and . . . in the intelligent or free substance this becomes a mastery over its actions. That cannot be better explained than by the System of Pre-established Harmony, which I indeed propounded some years ago. There I pointed out that by nature every simple substance has perception, and that its individuality consists in the perpetual law which brings about the sequence of perceptions that are assigned to it, springing naturally from one another. . . . Whence it follows that the soul has in itself a perfect spontaneity, so that it depends only upon God and itself in its action. (T 291/H 304)1

In short, Leibniz thinks that his doctrine of preestablished harmony secures the spontaneity of substances ‘‘for free.’’ Seen in a certain way, Leibniz’s defense of spontaneity is not surprising. After all, the preestablished harmony doctrine does entail that the complete explanation for any state of a substance is contained in the prior state of that substance. And thus, Leibnizian substances seem to have all the spontaneity it is possible to have. Given the preestablished harmony, spontaneity seems to be secured for free. On the other hand, securing spontaneity-for-free, as Leibniz so often does when writing on freedom, seems in many respects to undercut his own view. The reason for this is that it simply does not serve to nail down the distinction between free and unfree action that spontaneity was supposed to secure. Rather, spontaneity is invoked, by all parties, as a way of excluding from the domain of free

198 Leibniz: Nature and Freedom

action both (i) those acts that are directly caused by external agents, and (ii) coerced acts. On the former we find Bishop Bramhall, as one representative example, claiming: For the clearer understanding of these things, and to know what spontaneity is, let us consult awhile with the Schools about the distinct order of voluntary or involuntary actions. Some acts proceed wholly from an extrinsical cause; as the throwing of a stone upwards, a rape, or the drawing of a Christian by plain force to the idol’s temple; these are called violent acts. Secondly, some proceed from an intrinsical cause, but without any manner of knowledge of the end, as the falling of a stone downwards; these are called natural acts. . . . So then the formal reason of liberty is election. The necessary requisite to election is deliberation. Deliberation implyeth the actual use of reason. But deliberation and election cannot possibly subsist with an extrinsical predetermination to one. How should a man deliberate or choose which way to go, who knows that all ways are shut against him and made impossible to him, but only one? This is the genuine sense of these words voluntary and spontaneous in this question. (Hobbes 1839: vol. 5, 83–84)

And later, with respect to coercion, Bramhall continues: All this proceeds from the sensitive passion of fear, which is a perturbation arising from the expectation of some imminent evil. . . . Fear is commonly of one, deliberation of more than one; fear is of those things which are not in our power, deliberation of those things which are in our power; fear ariseth many times out of natural antipathies, but in these disconveniences of nature deliberation hath no place at all. In a word, fear is an enemy to deliberation, and betrayeth the succours of the soul. If the horse did deliberate, he should consult with reason, whether it were more expedient for him to go that way or not; he would represent to himself all the dangers both of going and staying, and compare the one with the other, and elect that which is less evil; he should consider whether it were not better to endure a little hazard, than ungratefully and dishonestly to fail in his duty towards his master, who did breed him and doth feed him. This the horse doth not; neither is it possible for him to do it. (Hobbes 1839: vol. 5, 86–87)

If we are to take Leibniz at his word in the Theodicy and numerous other texts where he endorses the spontaneity-for-free doctrine, we will have to conclude that Leibniz holds that the Christian being dragged into the idol temple, and the person who is coerced by fear, are acting spontaneously. This would be a very high price to pay, and one surely Leibniz did not want to pay. In the context of discussing freedom, at any rate, Leibniz was going to have to deliver some way of making the distinction that does not simply fall out of the preestablished harmony. Thus, readers of Leibniz must dispense with the sort of confident bluster one gets when he says, for example, that: There is no coercion in voluntary action, because, even though numerous representations of external things might exist in our minds, our voluntary actions are always spontaneous, because such acts have their causes within the agent. The theory of the harmony between body and mind pre-established by God from the beginning explains this more clearly than has been possible until now. (G 6:455)

Spontaneity and Freedom

199

Freedom from External Causes Although Leibniz frequently insists that spontaneity-for-free gives him the spontaneity requisite for genuine freedom, we have seen that he cannot consistently maintain such a view. And in fact, in numerous other texts he backs down from this bold claim, and offers an account of spontaneity and external determination that moves him closer to a traditional notion of spontaneity. Leibniz thinks that there are two types of external quasi-causes that must be absent for spontaneous action to occur. First, and most obviously, it must be the case that the action of the agent is not physically quasi-caused. In the Theodicy Leibniz remarks: When we act freely we are not being forced, as would happen if we were pushed on to a precipice and thrown from top to bottom; and we are not prevented from having a free mind when we deliberate, as would happen if we were given a potion to deprive us of discernment. (T 34/H 143)

Second, Leibniz holds, in keeping with tradition, that free acts must be free of coercion. In the following, he describes two sorts of ‘‘constraint’’ that are incompatible with freedom, the second of which is coercion: As for ‘constraint,’ it is useful to distinguish two sorts: physical, as when a man is imprisoned against his will or thrown off a precipice; and moral, as for example the fear of a greater evil. (NE 2.21.13/RB 179)

In both of these cases, the acts are not spontaneous because they can be attributed to an external source or quasi-cause. Since Leibniz is committed to the claim that each state of a substance can be fully explained in terms of its own law of the series alone, he is going to have to provide some account to distinguish those states of the substance that are genuinely spontaneous from those that are rightly ascribed to external quasi-causes. In various texts Leibniz appears to provide three related but distinct accounts of how such discriminations might be made. According to the first account, spontaneous states consist of clear perceptions in the monad, while confused perceptions are not spontaneous. In a published response to Bayle, Leibniz employs this distinction: Everyone who accepts immaterial indivisible substances attributes to them a simultaneous multitude of perceptions, and a spontaneity in their reasonings and their voluntary acts. I am therefore only extending that spontaneity to their confused and involuntary thoughts, and showing that their nature is to contain everything that is external. . . . In another sense, however, it is reasonable to call those things which consist in confused thoughts, through which there is involuntariness and incomprehension, perturbations (as the Ancients did) or passions. And this is what in ordinary speech we not unreasonably attribute to the conflict of the body with the mind, since our confused thoughts represent the body or the flesh, and constitute our imperfection. (G 4:564–565/WF 118)2

Here Leibniz simply identifies the spontaneous states of a substance with its clear perceptions.

200 Leibniz: Nature and Freedom

In other texts, Leibniz offers a second account. On this second account, spontaneous states are those that result when an agent moves from having less clear to more clear perceptions, and thus from lesser to greater overall perfection: As I have already said, anything which occurs in what is strictly a substance must be a case of ‘action’ in the metaphysically rigorous sense of something which occurs in the substance spontaneously, arising out of its own depths; for no created substance can have an influence upon any other, so that everything comes to a substance from itself (though ultimately from God). But if we take ‘action’ to be an endeavour towards perfection, and ‘passion’ to be the opposite, then genuine substances are active only when their perceptions (for I grant perceptions to all of them) are becoming better developed and more distinct, just as they are passive only when their perceptions are becoming more confused. . . . From that point of view a body can be said to ‘act’ when there is spontaneity in its change, and to ‘undergo passively’ when it is pushed or blocked by another body; just as with the true action or passion of a true substance, we can take to be its ‘action,’ and attribute to the substance itself, any change through which it comes closer to its own perfection; and can take to be its ‘passion,’ and attribute to an outside cause . . . any change in which the reverse happens. (NE 2.21.72/RB 210–211)3

Here, whether or not a state is spontaneous depends on facts about the immediately preceding state of the substance. If the transition from state to state is one in which the substance moves from lesser to greater perfection, then the succeeding state is spontaneous, and if not, it is externally quasi-caused. There is what appears to be a still further account in other texts. On this account, spontaneity depends on the explanatory relationship between states of distinct substances. Specifically, if state S1 of substance B1 is a clear perception, and S1 provides us with an explanation for the existence or coming to be of a state S2 of some distinct substance B2, then B1 is active with respect to S1 and B2 is passive with respect to S2. [W]hat is active in certain respects is passive from another point of view; active in so far as what we distinctly know in it serves to explain what takes place in another, and passive in so far as the explanation of what takes place in it is to be found in that which is distinctly known in another. (Mon 52/AG 219–220)

In virtue of what, one might wonder, could the state of one substance serve to ground the explanations of the state of another? The answer, found earlier in the text, is that in harmonizing substances, God brings it about that B2 exhibits S2 in order to make its coexistence with B1 harmonious. My interest here is not in arguing for the ultimate consistency of these accounts or for their relative merits taken severally. What is important is that we see that Leibniz was not content to rest with spontaneity-for-free. Such spontaneity could not distinguish free acts from physically determined or coerced acts— something Leibniz saw a need to do, and for good reason.

Freedom from Psychological Determinism Earlier we saw that some figures in the seventeenth century were keen to preserve another sort of spontaneity in freedom. This is the sort of spontaneity that preserves

Spontaneity and Freedom 201

causal independence of volitions from the deliverances of the practical intellect. In other words, this variety of spontaneity aimed to exclude psychological determinism or, more specifically, volitions that are causally necessitated by psychological antecedents. Two questions arise concerning this sort of spontaneity. First, why did various thinkers want to endorse spontaneity of this sort? Or, more specifically, what sort of problem was psychological determinism of volition via practical intellect seen to present? Second, what reasons are there to think that Leibniz wanted to endorse spontaneity of this sort? This second question is especially significant since one might think there is good reason to suppose that psychological determinism would not trouble Leibniz at all. There are at least two reasons for this. First, he seems to accept intersubstantial quasi-causal determinism at the phenomenal level of description. Second, he also seems to accept intrasubstantial causal determinism between adjoining temporal states of a substance at the ground floor metaphysical level.4 Thus, if Leibniz wanted to reject psychological determinism as I claim, one wants to know what could motivate such a rejection. Surely, Leibniz was not motivated by the libertarian intuitions outlined earlier, according to which freedom is incompatible with sufficient conditions. Nonetheless, there are passages where Leibniz indicates that efficient causal sufficient conditions are not compatible with choice. Here are two: Mr. Hobbes refuses to listen to anything about a moral necessity either, on the ground that everything really happens through physical causes. But there is nonetheless reason to draw an important distinction between the necessity which constrains the wise to do good, and which is called moral, even existing in relation to God, and that blind necessity. . . . (T h3/G 6:390)5 All the natural forces of bodies are subject to mechanical laws, and all the natural forces of mind are subject to moral laws. The former follow the order of efficient causes, and the latter follow the order of final causes. The former operate without liberty, like a watch, the latter are exercised with liberty. (LC 5.124/G 7:419; emphasis added)

What distinction did Leibniz mean to make in those passages where he endorses ‘‘moral necessity’’ and ‘‘moral laws’’ in choice but not ‘‘physical necessity’’ or ‘‘mechanical laws’’? Elsewhere I have argued that Leibniz’s use of the phrase ‘moral necessity’ in the context of the psychology of choice was meant to signify an alliance with one well-established tradition in faculty psychology (Murray 1996). This connection is important, in part because one of the distinctive features of this tradition is that it endorses sufficient reasons for choice while denying that psychological antecedents physically necessitate choices. Why think Leibniz appropriated this tradition? I turn to that question presently. Those in the faculty psychology tradition agreed that free action was to be explained ultimately in terms of the activity of intellect and will. Without going into details, those in this tradition held that the intellect carries on the task of deliberating about alternative courses of action and issuing practical judgments, while the will chooses one course of action judged to be good. But even the most

202 Leibniz: Nature and Freedom

compatibilist-friendly participants in this tradition would firmly reject the claim that practical judgments could physically necessitate the will (with one exception: the case in which the course of action under consideration is loving God ‘‘understood in his essence’’). Courses of action judged to be good, they argued, could ‘‘incline’’ the will but cannot necessitate it. All of them took this to entail that consideration of any such course of action failed physically and metaphysically to necessitate choice.6 There is a dizzying variety of variations on this basic theme. Some—Scotus and Molina for example—held that even after all the evidence was in from practical deliberation, the will maintained an utter indifference, remaining free to choose the object judged best, some lesser object, or no object at all.7 Others found nothing unsettling about the will infallibly choosing the single course of action judged best, as long as metaphysical and physical contingency were maintained (for example, see Bellarmino 1873: vol. 5, 590–591).8 Of course, exactly how the ‘infallibility’ was to be understood was far from clear. At the end of the sixteenth century there were roughly two classes of views held by those within the scholastic faculty psychology tradition: intellectualists and voluntarists.9 Intellectualists were wedded to the metaphysical impossibility of ‘‘self-moved movers.’’ As a result, they held that the will could only form a volition when it was moved to do so by some, we might say, sufficient reason—in this case, a practical judgment that such-and-such course of action should be chosen here and now. Critics of this view argued that it involved a vicious intellectual determinism since it appeared to make the will determined by the practical intellect. This, they argued, would undercut the freedom of the will since the operation of the intellect, even in practical deliberation, was regarded as determined. Some tried to thwart this objection by arguing that the will exercises control over the process of practical deliberation, rendering the activity of the intellect free, albeit in a derivative sense. However, critics argued that this view falls prey to an equally vicious infinite regress, since the intellectualists claimed that each act of will in turn required a judgment of the intellect to move it. Voluntarists held that the aim of practical deliberation was rather to determine a slate of acceptable possible courses of action, from which the will was then able to select one. Critics contended, among other things, that the view was metaphysically impossible since it required that the will be reduced from potency to act by itself, making it a ‘‘self-moved mover.’’ The voluntarist view is not strictly relevant here since Leibniz makes it no part of his own intellectual heritage. On the other hand, the developments within the intellectualist tradition are critical. Any plausible defense of intellectualism would have to show how one could avoid determinism on the one hand, and the vicious infinite regress on the other. How might this be done? One way perhaps is to press hard on the ‘‘inclination without necessitation’’ claim. One could then argue that if the intellect judges a certain course of action to be best, such a judgment might not suffice to physically determine the choice of the will, and yet the will might find all other courses of action correspondingly less attractive, and thus would never in fact choose these other courses.

Spontaneity and Freedom

203

Is this view coherent? That is, is it open to the intellectualist (i) to affirm that the will does infallibly follow the last practical judgment but also (ii) to deny that the last practical judgment causally determines the choice of the will? One notable school of Spanish Jesuits, the ‘‘moral necessitarians,’’ thought so. The two originators of this third main variant, Diego Ruiz de Montoya and Diego Granado, argued that this pair of positions could be maintained, and viewed this as an alternative to voluntarism and the prevailing intellectualist model. They claimed that while the choice of the will follows infallibly given the last practical judgment, this relationship fell short of physical necessitation, and they described the modality governing the relationship between last practical judgment and choice as having necessitas moralis, moral necessity.10 The moral necessitarian view itself spawned a host of variants in the seventeenth century. A central point of contention among critics and defenders was whether or not moral necessity was in fact distinguishable from physical necessity. Moral necessitarians argued that it was, and expended a good deal of effort, especially after 1650, trying to tease out a distinction. Thus we find numerous passages like these: Thus, a subject has a metaphysical necessity to act when . . . if it failed to happen, two contradictories would be given, which is certainly repugnant. Something is physically necessary, however, when it could not fail to happen naturally and without a miracle, even if it could happen miraculously. Thus, finally, something is morally necessary when, by way of inclination, that which usually, or always, or almost always is accustomed to occur, cannot fail to happen, even if it can fail absolutely or in light of a law of nature. (Izquierdo 1670: vol. 2, tract. 10, disp. 32, q. 3, 454) In things and in objects there exist many necessities concerning existence or essence. The first is called Moral, and this occurs when a thing occurs always or almost always in the same manner. But this necessity does not exclude Physical and Metaphysical contingency: not Physical because the thing is able to happen otherwise Physically; not Metaphysically because of the fact that if some other thing would happen, a contradiction does not follow. The second necessity is Physical, which excludes Moral contingency because out of the nature of the thing it exists as it does, since it cannot naturally happen otherwise since the thing does not have the Physical power to the opposite effect; for example, when fire is applied to a combustible patient and it does not burn, it is a miracle, and thus it cannot happen otherwise naturally. But this necessity does not exclude Metaphysical contingency because if combustion did not occur a contradiction would not follow. The third necessity is Metaphysical, which excludes all contingency, Moral and Physical, because it is impossible that a contradiction ever occur, and the Physical power for this to occur cannot be given, and this is the highest of all necessities. (de Pen˜afiel 1663: 522a)

Early advocates of moral necessity were unwilling to regard the will as morally necessitated in every act of choice. However, as defenders began to see ways of pressing the view into theological service (explaining foreknowledge, providence, and election), some began to argue that every volition was morally necessitated. This view seems to have been the majority view, for example, among Franciscans

204 Leibniz: Nature and Freedom

at the Sorbonne during the 1670s, the period when coincidentally Leibniz was in Paris. The following derives from a definitive defense of the position from the Parisian Franciscan Jeronimo de Sousa in 1680: You ask . . . whether the will, whenever it acts, always acts in a morally necessitated way. . . . I respond affirmatively, for this necessarily agrees with the will; just as it is necessary that the will have the good for its motive object when it operates, for this good both lures and moves, and also predetermines and necessitates it. (Sousa 1680: 29)11

The affinities between this moral necessitarian view and Leibniz’s own views on the psychology of freedom are striking. Compare the following text, for example: From this it can be understood what is that ‘indifference’ which accompanies freedom. Just as contingency is opposed to metaphysical necessity, so indifference excludes not only metaphysical but also physical necessity. It is in a way [quodammodo] a matter of physical necessity that God should do everything in the best possible way. . . . It is also a matter of physical necessity that those confirmed in the good . . . should act in accordance with virtue. . . . Again, it is a matter of physical necessity that something heavy tends downwards. . . . But it is not a matter of physical necessity that men should choose something in this life, however specious and apparent a good may be; though there is sometimes a very strong presumption to that effect. It indeed may never be possible for there to be an absolute metaphysical indifference, such that the mind is in exactly the same state with respect to each contradictory, and that anything should be in a state of equilibrium with, so to speak, its whole nature. . . . Yet the mind has this much physical indifference, that it is not even subject to physical necessity, far less metaphysical; that is, no universal reason or law of nature is assignable from which any creature, no matter how perfect and well-informed about the state of the mind, can infer with certainty what the mind will choose—at any rate naturally, without the extraordinary concourse of God. (C 21–22/MP 101–102)12

One might think that Leibniz could not have been a partisan of moral necessity since he sometimes denies one of the moral necessitarians’ central claims, namely that the will does or must infallibly follow the last practical judgment of the intellect. For example, in one place Leibniz remarks, ‘‘we do not always follow the latest judgment of the practical understanding when we resolve to will’’ (T 51/H 151). Careful examination makes it clear, however, that Leibniz’s motives for denying this maxim are not libertarian. Instead the denial springs from his account of the role of passions in choice. According to most variants of the intellectualist tradition, passions affect choice by influencing the way in which deliberation itself takes place. We can see this at work in Aquinas’s account of akrasia. On his view, passions affect choice by changing the premises that the intellect employs in deducing the last practical judgment. I know that sweet things are good to taste and I know that high-fat foods are bad for me. So, as I examine the piece of chocolate cake, my practical intellect must come to regard it either as a sweet thing (to be enjoyed) or as an unhealthy (to be avoided). I might well have some understanding of the fact that the thing before me is both. But when it comes to the termination of practical deliberation, only

Spontaneity and Freedom

205

one of these renderings becomes salient. The role of passions for Aquinas is to influence which of the two judgments will become effective (ST 1a2ae, q. 10, a. 3). Leibniz’s own account of the role of passions has to be somewhat more direct since, for him, passions are themselves appetitions resulting from unconscious or confused perceptions or apprehensions. In keeping with the fundamental metaphysical claim that all properties of monads are either perceptions or appetitions, Leibniz counts passions among the latter. As such, passions compete with the conatus induced by practical judgment. Thus, when we see Leibniz holding that the last practical judgment of the intellect is not always effective in moving the will to choice, it is not because he is departing from the moral necessitarian tradition in thinking that the will can choose something judged less good. Instead, Leibniz is merely admitting that, in addition to the appetites generated by the conscious process of practical deliberation, confused perceptions unconsciously generate competing unconscious appetites. As a result, although choice is not determined by the deliverances of practical reason, it is determined by the deliverances of perception tout court including adequate ideas, which figure into practical deliberation, and confused imperceptible ideas, which give rise to competing passions (G 3:401/AG 194). Thus, the full text of the passage cited earlier reads as follows: Besides, we do not always follow the latest judgment of the practical understanding when we resolve to will; but we always follow, in our willing, the result of all the inclinations that come from the direction both of reasons and passions, and this often happens without an express judgment of the understanding. (T 51/H 151)

Of course, even when the practical intellect does issue in a final judgment, the inclinations induced by this judgment can still be overridden by desires induced by inadequate ideas. As a result, only those rational agents who are free from passions altogether or who have their passions thoroughly harnessed are capable of choosing in accord with the deliverances of practical reason alone: I think it is only God’s will that always follows the judgments of the understanding: all creatures are subject to some passions, or to perceptions at least, that are not composed entirely of what I call adequate ideas. . . . As for us, in addition to judgments of the understanding, of which we have an express knowledge, there are mingled therewith confused perceptions of the senses, and these beget passions and even imperceptible inclinations, of which we are not always aware. These movements often thwart the judgment of the practical understanding. (T 310/H 313–314)13

Thus, while Leibniz does not hold that willing is subject simply to the deliverances of the practical intellect, he does hold that willing is the result of a ‘‘vector sum’’ of desires for those things that are apprehended as good, whether by way of adequate or inadequate ideas. I have argued that Leibniz deploys moral necessity in much the way that the Jesuit moral necessitarian tradition did. Does this mean that Leibniz, in fact, appropriated this tradition for himself? It is worth noting that Leibniz explicitly indicates his agreement with this tradition in the correspondence with Des Bosses.

206 Leibniz: Nature and Freedom

It is also notable that some early critics of the Theodicy criticized it because (they claimed) it appropriated this Jesuit tradition. Furthermore, defenders of Leibniz replied to these critics by defending this Jesuit tradition.14 Others have expressed doubts about the similarity between Jesuit and Leibnizian moral necessitation. Those who have commented on the matter and its relevance for the issue at hand have said that Leibniz means to distinguish moral necessity from: (1) Only metaphysical necessity (Spinozistic necessity) (2) ‘‘Blind necessitation’’ (3) Both metaphysical and physical necessitation.

Robert Sleigh (Sleigh, Chappell, and Della Rocca 1998: 1264–1269) says (1), Robert Adams (2),15 while I have argued for (3). By ‘‘blind necessitation’’ Adams means necessitation by factors that are, as he says, ‘‘value-free.’’ Adams points us toward a passage in which Leibniz explains at least one problem he has with Spinozist necessitarianism.16 In the Theodicy Leibniz mentions that Spinoza ‘‘appears to have explicitly taught a blind necessity, having denied to the author of things understanding and will, and imagining that good and perfection relate only to us and not to him’’ (T 173/H 234). Adams claims that Leibniz here objects not so much to the necessitarianism, but to the blindness of the necessity, that is, that the necessitation takes place without reference to ‘‘good and perfection.’’ According to Adams, Leibniz rejects metaphysical and physical necessity in free choice, human and divine, not because they are necessitating, but because they do not necessitate by way of ‘‘good and perfection.’’17 Moral necessity, then, is necessitation-via-value. I find Adams’s interpretation interesting and plausible. It is not clear to me what sorts of textual evidence might come to the surface that would suffice for deciding between his interpretation and my own. We both agree that freedom and physical necessity are incompatible but for different reasons. Resolution will require that we turn to the question of what it is about physical and metaphysical necessity that is problematic for Leibniz. I think the resources in the faculty psychology texts will not help us here. However, I think that certain theological texts give us good reason for thinking, at least, that it is the necessity, in addition to the blindness, that is troubling for Leibniz. I will return to this point in the next section. What of Sleigh’s claim that, for Leibniz, even physical necessity is no threat to freedom? There are two sorts of evidence in Sleigh’s favor. First, Leibniz sometimes says that metaphysical necessity is the only threat to freedom: ‘‘It will be shown that absolute necessity, which is also called logical and metaphysical . . . and which alone is to be feared, does not exist in free actions’’ (T, Preface/G 6:37).18 Of course, such arguments must be handled with care. Perhaps Leibniz simply spoke incautiously. More likely, Leibniz here mentions only those threats that are relevant or salient in context. As even Sleigh admits, there were few defenders of efficient causal determinism in this period. Necessitarians like Spinoza represented the most salient threat for Leibniz. Perhaps efficient causal compatibilists were simply not worth resisting here.19

Spontaneity and Freedom

207

Second, in at least one place Leibniz writes as if moral and physical necessity are modally equivalent: [F]or it must be admitted that when one thing follows from another in the contingent realm, the kind of determining that is involved is not the same as when one thing follows from another in the realm of the necessary. Geometrical and metaphysical ‘followings’ necessitate, but physical and moral ones incline without necessitating. (NE 2.21.13/RB 178)

This is a complex text. But what is important here is that Leibniz seems to assert that physical and moral ‘‘followings’’ are modally equivalent. Whatever holds for one seems to hold for the other. Thus, they are not distinct modalities after all, as they are for the Jesuit moral necessitarians.20 It is hard to know how to weigh this text against others cited earlier. Leibniz sometimes makes incompatible claims and we are left to judge on systematic grounds how to best understand his intentions, all things considered. It seems to me the preponderance of texts weighs against the Sleigh interpretation in this case. Nonetheless, I think theological grounds, to which I will turn next, are the deciding element. Let me add that it seems to me that many feel obliged to equate moral and physical necessity because of our own contemporary suspicions about final causality. Moral necessity is, as I have characterized it, intrinsically teleological. And since, some seem to suppose, final causes must ultimately be cashed out in terms of efficient causes, they must be similarly reducible for Leibniz. But contemporary work in the metaphysics of freedom might perhaps persuade us to be more accommodating. Recent libertarians take final causes to play an important role in the explanation of action, a role not reducible to talk about efficient causes.21 These libertarians regard free actions to be entirely causally undetermined. Still, they contend, a complete explanation of free action will make reference to the fact that the agent chooses with the aim of satisfying one of the agent’s desires. In this way, having desires is a necessary condition of an agent acting freely. And we might even imagine that in some case such desires are sufficient as well. For nothing seems to prevent its being the case that I am at some point positioned in such a way that I desire only a single course of action, including refraining from acting altogether. In such a case we might say that though an action is uncaused, nonetheless, the fact that there is only one course of action (including refraining) that is desirable might necessitate that this course of action be chosen.22 I suspect this is the sort of thing that the moral necessitarian Sousa had in mind in his own account. Given that the will cannot help but incline most strongly toward that course of action judged best, it acts in such a way that the action brought forth is aimed at the satisfaction of that desire. Of course, whether or not one is willing to admit this sort of account will depend on exactly what account of causation one adopts. If causation is to be cashed out simply in terms of the ability to support certain sorts of counterfactuals, this account will not fare well. But if we think of causation as instead a real relation between states of affairs or events, as Leibniz surely would when thinking of causal relations between temporally adjacent states of a substance, then an account like

208 Leibniz: Nature and Freedom

this seems to preserve the integrity of final causes that are not ultimately reducible to efficient causes. So, perhaps this is the sort of account we should favor when thinking about Leibniz. It is fair to say that many of his intellectual predecessors held such a view of final causes in this context. And it seems to fit well with his own talk about moral necessity as distinct from physical necessity and metaphysical necessity. This gives us a way of conceiving of what Leibniz is up to when it comes to the faculty psychology of freedom, and this is the place where moral necessity is primarily put to work. What did Leibniz hope to gain from such spontaneity? On the one hand, this is hard to answer because I know of no text in which Leibniz directly speaks to the question. On the other hand, such an absence should lead us to believe that he hoped to secure here what everyone hoped to secure by denying intellectual determinism, namely, the sort of independence that allows the will to count as genuinely active in the act of election. From the thirteenth century onward, critics of intellectual determinism argued that it made the will passive and thus not possibly free.23 There is no reason to think Leibniz’s own account would have demanded less.

Freedom from Divine Necessitation Unlike his Reformed counterparts, Leibniz also held that spontaneity required independence from divine determination. Here the motivation was as much to preserve the will’s freedom, as it was to preserve God from culpability for human sin: For who would call into doubt that the mind thinks and wills, that we elicit in ourselves many thoughts and volitions, and that there is a spontaneity that belongs to us? If this were called into doubt, then not only would human liberty be denied and the cause of evil things be thrust into God, but it would also fly in the face of the testimony of our innermost experience and consciousness, testimony by which we ourselves sense that the things my opponents have transferred to God, without even a pretense of reason, are ours. (G 4:510/AG 161)

There were many respects in which Leibniz believed free human action must maintain independence from divine determination. In various texts, Leibniz considers the threat to freedom proposed by God’s activities of creating, conserving, and concurring in human activity. Here I will focus on one controversial part of Leibniz’s worries. The worry can be stated in brief as follows. God, when considering the possible creature Adam, say, knows that Adam will choose to eat the forbidden fruit. Given this, how can it be the case that God wills that Adam be actual, without thereby coming to be the cause of Adam’s sin? If God were the cause of Adam’s sin, this would undermine Adam’s freedom and make God the ‘‘author of sin.’’ Leibniz argued that God avoids being the author of sin by not causing but merely permitting the sin of Adam. But how is Leibniz going to make this distinction between causing and permitting? As I will show, he does so by arguing that the creature has a variety of spontaneity sufficient to make it author of its own acts.

Spontaneity and Freedom 209

One of Leibniz’s earliest attempts to grapple with the independence of human free actions from God occurs in the Confessio Philosophi, where he attempts to deflect responsibility for evil from God by arguing that God does not cause, but merely permits, evil. Early in the dialogue he provides an account of permission that can be summarized as follows. P permits E if and only if: (1) (2) (3)

P fails to will that E P fails to will that not-E P brings it about that state of affairs S obtains by willing that S obtains (4) If S obtains then E obtains (5) P knows that (4) (6) P believes that the good entailed by S’s obtaining outweighs the evil entailed by E’s obtaining.24

Although Leibniz appears content with this view in the Confessio, concerns with the view as described seem to arise in a series of texts from the mid-1680s. The most critical one is De Libertate, Fato, Gratia Dei et Connexis.25 Among other things, Leibniz is, in this piece, looking to solve a knotty theological problem in which he had a keen interest, namely, the truthmaker for propositions of the form: If person S were in circumstances C, then S would freely do action A.

In the discussion concerning divine providence there were two widely endorsed Scholastic views on the truthmakers for such propositions, called conditional future contingents (CFCs), in the late sixteenth and early seventeenth centuries. The first view was primarily defended by Dominicans. The second view was defended primarily by Jesuits. For the Dominicans, CFCs have as their truthmaker a divine decree. It is, on their view, up to God whether or not Peter denies or performs some other act in circumstances C. As a result, the truth value of CFCs is said to be determined ‘‘postvolitionally.’’ The Jesuits held, on the contrary, that the truth values of CFCs must be determined prevolitionally (i.e., independently of any act of the divine will). The reason for this is the obvious one: if it is up to God what I choose under particular circumstances, then, they argued, the choice could not truly be free. Further, in those cases where what I choose is evil, God, it seems, is the direct cause of my performing the evil act since it was due to his decree that I chose to sin in those circumstances. Surprisingly, even though Leibniz is aware that there is an issue concerning what grounds the truth of CFCs, and he is aware of the competing solutions, it is not something that he considers in any detail in his early career.26 However, when we come to the period of 1685 through 1687, there is a flurry of activity on Leibniz’s part concerning this issue. In De Libertate, Fato, Gratia Dei et Connexis, Leibniz deals with a wide range of topics concerning human freedom and divine providence. After a brief discussion of the problem of foreknowledge, Leibniz turns to the more complex problem described earlier, namely, the truthmaker for CFCs. He introduces the problem by citing the stock example in discussions of this topic, an example

210 Leibniz: Nature and Freedom

springing from the biblical narrative in 1 Samuel 23. In the passage, David is hiding from King Saul in the city of Keilah. David fears that Saul will besiege the city and that the residents of the city will fail to protect him. Thus, David, wondering whether or not he should flee, inquires of God concerning whether the residents of Keilah will, in fact, turn him over if Saul takes the city. God tells him that they will, and David then flees. Obviously, the knowledge that God has revealed to David is not an item of simple divine foreknowledge, and for obvious reasons: the events described by David do not occur in the actual future. The question here is what the Keilites would do, were some circumstances to arise that do not in fact arise. In other words, God is herein displaying that he has knowledge of conditional future contingents in a case where the antecedent of the conditional is not satisfied. The question then is, in virtue of what does God know what the Keilites will do, or, alternatively, in virtue of what is it the case that, (7) If the Keilites were in circumstances C (where Saul takes the city), they would deliver David to Saul.

Leibniz then undertakes a defense of the Dominican view that the truthmaker for (7) is the divine will. In the course of his defense, Leibniz reiterates the claim that God is not the author of sin because rather than willing the sin, God merely permits it. Since Leibniz is defending the postvolitional, Dominican view, it is not at all surprising that the series of criticisms he raises are the stock objections of the prevolitionalist Jesuits. The first two criticisms he raises pose the natural questions: if CFCs have a decree of the divine will as their truthmaker, then it seems that human freedom is precluded and, in light of this, it appears that sins ought to be ascribed to God and not to the creature. In response to the second criticism Leibniz sets forth a view much like the one we find in the Confessio: God is not the author of sin because he merely permits rather than wills the sin, and further such permitting is excusable since the evil permitted results in a greater good. Thus, it cannot be said that God wills sin since, on this view, ‘‘properly speaking, God does not decree that Peter sin.’’ However, Leibniz was clearly dissatisfied with the view he lays out in this paragraph since he subsequently struck the entire paragraph and replaced it with a paragraph that goes in a quite different direction. What did Leibniz find unsatisfactory about the postvolitional view? Recall that in the Confessio Leibniz not only tells us that God’s relationship to sin in the world is one of mere permission, but he goes on to give us a detailed account of the nature of permission. In that account, the second condition was: (2)

P fails to will not-E.

This second condition is necessary since, if God were to will that E not occur, E would not occur. So, to make this account of permission work, Leibniz must be able to hold that with respect to a token evil, Peter’s denial of Christ, that: (30 )

It is not the case that God wills that Peter denies.

Spontaneity and Freedom

211

But in light of the account we have earlier, it looks like Leibniz is committed to the following claims: (8) (9) (10) (11)

God wills that if Peter is in C, then Peter denies. God wills that Peter exist. God wills that C obtain. God wills that Peter be in C.

It is hard to see, however, how Leibniz can coherently hold (8), (11), and (30 ) since it is reasonable to assume (8) and (11) entail the denial of (30 ). But what is it that Leibniz should surrender here? If he surrenders (30 ), then he must also surrender his account of permission since surrendering (30 ) would violate the first condition of his account of permission (i.e., that God does not will E where E is Peter’s denial). Surrendering (11) appears to carry a price too high to pay since (11) seems to be required by any orthodox account of divine creation. If Leibniz wants to give up (8), then he will have to surrender the claim that the divine will is the truthmaker for CFCs. This might not seem to be a very high price to pay. After all, the view that something other than a divine decree acts as truthmaker for CFCs was a view widely held by Jesuits, as I noted earlier. But the Jesuit view also required holding a libertarian view of freedom that Leibniz could not accept. Thus, giving up (8) would leave Leibniz in need of finding something else that could act as the truthmaker of CFCs. If the truth of the key conditional (12) If Peter is in C, then Peter denies.

is not grounded in the divine will, what are the options? Perhaps Leibniz could hold it to be a necessary truth. This would solve the problem of directly implicating God in Peter’s sin. Peter sins because Peter must. It is essential to Peter that he sin. However, Leibniz utterly rejects this move. Whatever the truthmaker for (8), Leibniz is committed to the claim that it is not the divine will, and it is not that Peter sins in C necessarily.27 Recall that if, as I claim, Leibniz is giving up the postvolitional view while attempting to retain the Principle of Sufficient Reason, he is now forced to look for something that can act as the truthmaker for CFCs. In another text from the same period, Leibniz begins to construct such an account by developing a view of the relationship between God and the creaturely free act. He then employs this view to answer the key question concerning the truthmaker for CFCs: Does God determine what the creature would do under specified circumstances, and if so, in what respect? To this end he says: It must be replied that the first cause [God] is determined by the second cause [the creature] taken ideally, i.e., the idea of the second perceived in the divine intellect determines the will of the first one, and the choosing of the second taken actually is determined by the first, or everything takes its being from him. (Grua 386)

In other words, we have an account quite akin to the prevolitional account. God does not, by an act of will, determine what the creature would do under given circumstances. But God, seeing what the creature would do, decides which

212 Leibniz: Nature and Freedom

possible creature to create and in virtue of that creative decree, they come into being. But now the question becomes, what is the truthmaker for the CFC, if it is not the divine decree? In other words, one might ask, when we abstract away considerations of the divine contributions toward the creaturely act, what is left that acts as the ‘‘sufficient reason’’ for the truth of the CFC? One of the issues that Leibniz was clearly concerned about in this period is one that likewise occupied other prevolitionalists, namely, exactly what can be abstracted away from the creature’s circumstances when God is considering what the creature will freely do? Or, to put it another way, what elements of C (the circumstances) are necessary for the CFCs having a determinate truth value? For example, it would appear that God cannot abstract away from C considerations of all divine causal contribution at the time of choice. Among God’s causal contributions would be his activity of sustaining the creature in existence; but if necessary conditions for creaturely existence are not preserved, neither is that creature’s existence or any action whatever—leaving the corresponding CFC without a determinate truth value. In addition, since orthodox metaphysicians held almost without exception that no causal power of a creature could be exercised without some (concurring) act of the divine will, it appears that determining what the creature would do in given circumstances would require consideration of whatever was necessary for the act to occur, and thus to consider not only God’s sustaining causal contribution, but some measure of God’s concurring contribution as well.28 In the next two paragraphs, Leibniz takes up just this question when he asks: ‘‘It is sought what it is in the human will that God attends to such that he chooses to concur with one action rather than another’’ (Grua 386–387). He then proposes his answer to the sufficient reason question, taking up a position that, to my knowledge, he holds for the remainder of his career: But when rightly considered we must hold that the certainty of God arises from the fact that He foresees that the mind will assent. It can be asked, whence he has foreseen this connection? Entirely from the consideration of minds as much as of graces, he knows both that which is the future state of mind and what grace will superadd to it. . . . Because it is an impossible condition that a creature operate without divine concurrence, it is impossible that God foresee what the creature, per se, would do by the power of free will alone. So God can only foresee that to which the creature is inclined. And so the matter is reduced to the doctrine of inclination but not necessitation. (Grua 387–388; emphasis added)

The answer, then, is that it is the inclination of the creature prior to choice that acts as (a) the truthmaker for the proposition, and further (b) that on which God bases the nature of his concurrence with the free act. This view, I believe, was suggested to Leibniz by a work he was reading at this time, and that he cites in another connection slightly later in the same essay. The work is Louis of Dole’s Concerning the Manner of the Concursus of God and Creatures. In the work, Dole describes a view very much like the one mentioned by Leibniz. Dole writes: God explores through the dictates of his middle knowledge the way in which the created will is bending itself by way of its own liberty on such and such an

Spontaneity and Freedom

213

occasion and at such and such a moment in time so that as a result of this prevision and foreknowledge, God determines himself and makes a resolve such that he will manifest his concursus in this way rather than that at this particular moment in time, and in such a manner that his concursus is toward willing rather than nilling and toward willing this rather than that object. (Dola 1634: part 2, chap. 1, 107)

And Leibniz comments favorably on this view.29 Thus, God first foresees how it is that the creature inclines, and from this knows how he will concur with the act of the creature. These together suffice to give God knowledge of the truth of the relevant CFC and thus, as Leibniz says in the passage I quoted earlier: ‘‘And so the matter is reduced to the doctrine of inclination but not necessitation’’ (Grua 388). We find here the coming together of the two species of spontaneity. What makes the free action genuinely the creature’s own, is the fact that it has an active will, which is not psychologically determined, and which is not necessitated by divine action on creatures. I urged in the fourth section that theological consideration seemed relevant, if not decisive, when considering Leibniz’s case against the consistency of physical necessity and freedom. The reason should now be clear. Leibniz expends a great deal of effort making the case that causal determination of creaturely acts (by God in this case) is sufficient to deflect moral responsibility away from the creature (and toward the ultimate cause, in this case God). If Leibniz were satisfied with the compatibility of freedom and causal determinism, such worries would simply never arise. And yet they do. This is, it seems to me, very powerful evidence of anticompatibilist sympathies in Leibniz’s thought. What is more, whatever case can be pressed against divine determinism can be equally forcefully made against garden-variety efficient causal determinism in choice. To block this, one would need a buffer analogous to the one Leibniz installs between God and creatures. Jesuit moral necessity provides just such a buffer.

Conclusion In the introduction I noted that compatibilists and libertarians have appealed to distinct pairs of reasons or guiding intuitions in defending their views on freedom. In both cases, one member of the pair is strictly metaphysical, while the other is theological. In light of the varieties of spontaneity that Leibniz aims to defend, we can now ask: where do his overall sympathies lie here? The answer is, as is typical for Leibniz, a mixed one. Surely, Leibniz has no sympathy at all with the first libertarian intuition, that is, that freedom is incompatible with sufficient conditions for choice. Rather, his sympathies here run in parallel with the compatibilist intuition that free actions require that the act be generated by certain internal states of the agent. In Leibniz’s case, these internal states are deliverances of practical reason that are in turn a species of clear perception. Still, while endorsing this intuition, Leibniz also seems to want to argue that the act generated by practical reason is not, strictly speaking, efficient causally determined by it.

214 Leibniz: Nature and Freedom

On the theological side, however, Leibniz comes down squarely on the libertarian side. Seventeenth-century theologians were well aware of the implications of straightforward compatibilism for the problem of evil. If God wills that certain events occur by means of deterministic processes that he implements, God is the author of those events. And thus, if the sinful acts of creatures are causally determined, God is the author of sin. Leibniz distances himself from this objection by first arguing that the actions of free creatures under specified circumstances are not under the control of the divine will. Thus, transitions between states within freely willing rational substances are not under immediate divine control. Rather, the transition from state to state in such substances is fixed by morally necessitated final causality. Since this sort of relation between intellect and will, and correspondingly, between successive states of the substance, is sufficient to secure spontaneity (or so Leibniz believes), and since it is on such grounds that God governs his own concurrence with the actions of substances, divine causal involvement in creaturely free acts neither destroys freedom nor implicates God in sin. As a result, the texts indicate a Leibniz who is not as sympathetic with fullblown compatibilism as he is sometimes held to be. Does mitigated Leibnizian compatibilism succeed in doing what Leibniz proposes for it? The answer to this question depends on whether or not tolerably good sense can be made of Leibniz’s appropriation of moral necessity. If, as some of Leibniz’s contemporaries claimed, there is no logical space for a modality that is weaker than physical necessity, then physical and moral necessity will collapse, as will Leibniz’s view. But if moral necessity as described is defensible, the view merits closer scrutiny.

Notes 1. See also T 59, 65, 290, 302, k20 (respectively G 6:135, 138, 289, 296, 421–422); also G 6:455/S 137; G 3:364; G 4:483–484/AG 143, and elsewhere. 2. This account also seems to be in play at G 7:109; G 2:69; G 3:402/AG 195; NE 2.21.12; G 3:70/ WF 131. The reader should take note of the fact that I have altered the translation of Woolhouse and Francks at a key point. They render ‘‘Mais on a raison dans un autre sens d’appeller perturbations. . . . ce qui consiste dans les pense´es confuses, ou` il y a de l’involontaire de et l’inconnu’’ as ‘‘In another sense, however, it is reasonable to call those things which consist in confused thoughts, and in which there is involuntariness and incomprehension, perturbations.’’ 3. This account also seems to be in play in the Arnauld correspondence, G 2:47. See also G 6: 138–139. 4. Leibniz sometimes even appears willing to talk about deliverances of the practical intellect causing volitions. See, e.g., NE 2.21.13. 5. See also G 6:300. I should note that some might charge that I have unfairly truncated the quotation. I deny the charge. Leibniz later in the passage goes on to deny metaphysically necessitating conditions for choice. But in this first part it is clear that physically necessitating conditions are in view. 6. See, for example, Aquinas, ST 1a2ae, q. 10, a. 1, ad2um, and De Malo, q. 16, a. 7, ad18um. 7. See Scotus 1950: 2, q. 39, a. 2, n. 5, and Molina 1953: disp. 2, n. 9. 8. For more details on these various views, see Murray 1996: 31–37. 9. For more detailed renderings of these views one can consult Murray 1996: 27–37. 10. One might wonder how this view differs substantially from the earlier intellectualist view. The answer is that it is rather more of a bold development. The intellectualists endorsed the sufficiency of practical judgment for choice, and also denied that practical judgment could necessitate the will. But there was no unified strategy for making sense of this nonefficient-causal sufficiency. As mentioned

Spontaneity and Freedom

215

earlier, the most widely endorsed strategy was to hold that practical judgments were not causally sufficient in an efficient sense since the will was a proximate cause of the intellect forming the last judgment. This was unsatisfying for reasons already described. There were other minority strategies attempted as well. But this represents the first explicit attempt to address the intellectualist worries by forging an attenuated sense of necessitation. For a more detailed treatment of this movement see Murray 1996: 37–43 and Knebel 1991a: 3–24. 11. Note that Sousa makes it clear that the necessity in play here is strictly necessitas moralis. 12. The claims that Leibniz make concerning God and the beatified do not undercut the interpretation offered here. Note that Leibniz only likens (quodammodo) God’s choosing to cases of physical necessitation. As for the blessed, the standard view was that since they were in the presence of God, and thus could see God ‘‘in his essence,’’ they were physically necessitated in choosing, since ‘‘loving God as seen in his essence’’ is the one practical activity that necessitates the will in this way. 13. See also NE 2.21.13; T 51/G 6:130; Grua 253, 272, 276, 301. 14. For a defense of these claims see Murray 1996: 51–53. 15. Adams, ‘‘Moral Necessity,’’ in this volume. 16. T 173/G 6:217. 17. Adams, ‘‘Moral Necessity,’’ this volume. 18. Sleigh also points to T 44, 302, 367 (respectively, G 6:127, 296, 333); G 3:401; and Grua 480–481. 19. Though, as we saw earlier, Leibniz does, in an appendix to the same work, distance himself from Hobbesian compatibilism. In addition, there are notable cases in which Leibniz has the opportunity to endorse this form of compatibilism and does not. For example, Francois Lamy challenges Leibniz with the remark that the sort of full-blown causal determinism that he takes Leibniz to endorse is not compatible with freedom (WF 147, from the original De la Connoissance de soi-meˆme, second edition, 1699). Leibniz replies by denying that the sort of connection between states of a substance is causally determined in the way Lamy claims (WF 154–155, 167). 20. Many understand Leibniz in this way, e.g., Grua 1953: 234 and Vailati 1997: Introduction. 21. See for example: O’Connor 1995: 173–200 and Ginet 1995: 69–94. 22. Peter van Inwagen (1995) defends such a possibility. 23. William de la Mere is widely known for directing the Franciscan attack on Thomism on this point in the fourteenth century. Voluntarist critics of the intellectualist view sustained the charge at least through the sixteenth century. For more on this one can consult Kent 1995: 106. 24. This summary derives from the portion of the text at A 6.3:129–131. 25. Grua 306–322 and A 6.4:1595–1612. This piece carries a date of 1686–87. Other important pieces treating the topic in the period are: Aus Ludovicus a Dola, De Modo Conjunctionis Concursuum Dei et Creaturum (A 6.4:1789–1792), and De Libertate et Gratia (Grua 384–388/A 6.4:1455–1459). 26. It is discussed briefly on a few occasions, however. For example, see C 26–27. Leibniz says things that suggest that he holds the view I claim he held prior to 1685. Here he seems to adopt the postvolitional view concerning CFCs when he says, ‘‘God knows future absolute things because He knows what He decreed; and future conditionals because He knows what He would have decreed’’ (A 6.1:545–546; 1670). Leibniz seems to lean in the opposite direction (toward the prevolitional view). But here Leibniz only makes a brief mention of the prevolitional view without giving any endorsement of it. 27. The first portion of this section was previously published in Murray 2000: 81–84. Cover and Hawthorne (2000b) have replied to the claim I make there by arguing that regarding (12) as a necessary truth would not commit Leibniz to the necessity of the consequent. Why then does Leibniz strike the postvolitionalist language in the first draft? Cover and Hawthorne contend that what Leibniz found troublesome was merely (8). Closure under willed material implication does seem to invite the judgment that God is culpable for human sin. However, (12) can be necessary and known true by God, without inviting the same judgment since willing under known material implication is (one might hold) not similarly closed. Thus, Leibniz’s denial of the necessity of the consequent does not settle the modal status of the conditional. The position Cover and Hawthorne defend merits a longer reply than I can provide here. To assess the plausibility of their reply one needs to consider just what it is in virtue of which the conditional is necessarily true. Here are a couple of sources of the necessity that might suggest themselves: (a) the nature of complete concepts (if, for example, Leibniz were a superessentialist), and (b) facts about

216 Leibniz: Nature and Freedom which world is best and the necessity of God’s willing to actualize the best world. Both of these would raise obvious problems. But then what does ground the necessity? I think that ultimately there are no grounds to be found that would do the work Leibniz needs done, and that this explains why he does not (here) avail himself of this answer. Still, it is fair to note that the sort of strategy Cover and Hawthorne ascribe to Leibniz looks very much like the ‘‘hypothetical necessity’’ strategy endorsed in other texts in this period. 28. The most detailed contemporary treatment of the discussion of this issue among sixteenth- and seventeenth-century figures is found in Knebel 1991b: 262–294. 29. A 6.4:1791.

11

Leibniz Against Molinism Freedom, Indifference, and the Nature of the Will SEAN GREENBERG

he problem of free will is standardly taken to arise from a tension between 1 In approaching early modern accounts of freedom from this interpretive standpoint, it is assumed, somewhat uncritically, that early modern philosophers conceived of the problem of freedom in the same way that we do. I believe that this assumption is mistaken. Early modern approaches to the problem of freedom were cast in terms of conceptions of the mind, and, in particular, of conceptions of the will. Considerations about the nature of the mind feature far less prominently in contemporary discussions of freedom, and consideration of the will is altogether absent, because the will no longer figures in our conception of the mind. Peter van Inwagen has drawn attention to this change, remarking that

T human freedom and causal determinism.

the ascription of free will to an agent . . . does not imply that the agent has a ‘faculty’ called ‘the will.’ It was not always so. Once upon a time, to say that X ‘had free will’ was to imply that X had something called a ‘will’ and that this will was not only unimpeded by external circumstances . . . but that X’s internal constitution left him ‘free’ to ‘will’ in various ways. . . . A tradition, however, is a changeable thing, and the classical tradition has abandoned these implications of the words ‘free will.’ (Van Inwagen 1995: 220)

Van Inwagen’s remark is provocative: someone of a historical bent might well wonder when and why the ‘‘classical tradition’’ abandoned its commitment to the faculty of the will, and whether this abandonment had any implications for how we conceive of the problem of freedom. These questions merit extended discussion; I will not be able to give it to them here. I do, however, want to take some steps toward opening up these questions, which have been raised in some earlier histories, but have nevertheless not yet received the attention that I believe they deserve.2 I propose to illustrate the value 217

218 Leibniz: Nature and Freedom

of considering the problem of freedom from the perspective of the nature of the mind (and will) by showing how this approach offers a fresh perspective on Leibniz’s criticisms of the Molinist conception of freedom.3 The Molinist account of freedom, so called after the Spanish Jesuit philosopher and theologian Luis de Molina, was characterized in the following formulation well known to early modern philosophers: a man is free, given that with all the requisites for action having been posited, he is able to act or not act. On this account, agents are said to be ‘‘indifferent’’ with respect to their choices, because they determine themselves to choose as they do on the basis of perceptions, desires, and so on, the ‘‘requisites for action’’: in other words, the requisites for action do not determine the agent’s choice. Molina originally developed this doctrine in order to account for the relation between human agency and divine grace as part of the Counter-Reformation attempt to counter Lutheran and Calvinist doctrines of the servitude of man.4 Yet a wide range of early modern thinkers, including Samuel Pufendorf, John Bramhall, and Samuel Clarke, adapted versions of this account of freedom to their own, different, philosophical problems.5 Leibniz’s criticisms of the Molinist conception of freedom, like the criticisms of his contemporaries Thomas Hobbes, Ralph Cudworth, and John Locke, center on the notion of indifference central to the Molinist definition of freedom, and he pursues a line of criticism akin to theirs. These philosophers all claim that the indifference required for Molinist freedom implies that an agent might be able to choose for no reason at all.6 On the face of it, such criticisms do not touch the Molinist position, because the Molinists do not actually claim that a free agent chooses for no reason at all. Although commentators have tried to sharpen Leibniz’s objections to Molinism, to my mind, these attempts have not gone far enough. In this essay, I try to go further by tracing Leibniz’s criticisms of Molinism to the divergences between his conception of mind and that of the Molinists. I lay the foundation of my discussion by sketching the Molinist account of freedom, focusing on Francisco Sua´rez’s formulation of the position in Disputation 19 of the Metaphysical Disputations (MD).7 I then turn to Leibniz’s criticisms of the Molinist account of freedom, arguing that the target of these criticisms is the conception of the will that underlies that account of freedom. Leibniz objects to the Molinist conception of freedom as he does because there is no place for the Molinist will in the Leibnizian mind.8 Indeed, I believe that Leibniz not only rejects the Molinist will, but that he wants to account for freedom without recourse to the will. This last claim may strike some readers as implausible, given that Leibniz uses the term ‘will’ in his writings and seems to want to give an account of what it is to will freely. Yet we should not make too much of the fact that Leibniz uses this term: on account of his oft-stated aim to reconcile various opposing philosophical positions, Leibniz often appropriates the language of those positions. The question is whether there is a functional analogue of the Molinist will in the Leibnizian mind, and what role that functional analogue plays in Leibniz’s conception of freedom. I argue that the Leibnizian will has no distinct functional role to play in Leibniz’s conception of freedom, and conclude by assessing the significance and success of Leibniz’s attempt to account for human freedom without recourse to the will.

Leibniz Against Molinism 219

Sua´rez on Freedom and Indifference Sua´rez’s account of human freedom forms a part of his discussion of efficient causality, and arises in the context of his attempt to explain the conditions under which efficient causes act.9 In particular, Sua´rez seeks to explicate the difference between ‘‘free’’ and ‘‘natural’’ efficient causes—that is, the distinction between rational agents, such as humans, angels, and God, and all other created things. According to Sua´rez, natural causes may only act if they are determined to one thing, whereas free causes may act even if they are not determined—that is, if they are indifferent—with respect to any particular thing. This difference is reflected in the definitions that Sua´rez gives of these types of cause. A natural cause is one such that with all the requisites for acting having been posited, it must act (MD 19.1.1); a free cause is such that with all the requisites for acting having been posited, it can either act or not act (MD 19.4.9).10 The following example derived from Freddoso (1986) may help to clarify Sua´rez’s distinction between natural and free causes. Provided that all conditions are normal and barring miracles, if one turns the knob on one’s stove, the burner will light. The stove will fail to light only if not all the conditions are normal—for example, if the gas is not on, or the pilot light is out—in other words, if there is some ‘‘requisite’’ of the stove’s lighting that is not in place in the actual situation. Given that a gas burner is the kind of thing that will produce only one effect, provided that all the conditions that are required for it to produce that effect are in place, it is not indifferent with respect to that effect, and may be said to be naturally necessitated. In contrast, if a human being sees a tasty piece of cake, and desires it, she is not determined to eat the cake, but must choose to eat it. Although the sight of a tasty piece of cake is a necessary condition for the agent choosing to eat the cake, it is not a sufficient condition, for the agent must also decide that the cake is indeed something that she wants. Sua´rez proposes two definitions of causation in order to emphasize the fact that the actions of natural causes are entirely determined by the ‘‘requisites for action,’’ whereas free causes are naturally indifferent to their actions, and must therefore determine themselves to action (MD 19.2.18–22). The heart of Sua´rez’s discussion of freedom is devoted to explaining how free causes may determine themselves, given that it is in virtue of this power of selfdetermination that free causes differ from natural causes (MD 19.2.18). According to Sua´rez, a free faculty’s indifference is preserved in a satisfactory way primarily and precisely through the faculty’s relation to the act or to the absence of the act. This is usually called freedom with respect to exercise, since it is by virtue of this freedom that the very exercise of the act is indifferent. (MD 19.4.9)

In accordance with the Aristotelian tradition, Sua´rez maintains that human beings are free—they enjoy freedom of exercise—because they have both intellects and wills (MD 19.5, 21). Sua´rez’s specific problem, then, is to determine whether the intellect or the will has freedom of exercise. In Sua´rez’s terminology, the question is that of determining which faculty is ‘‘formally’’ free.

220 Leibniz: Nature and Freedom

Sua´rez maintains that a faculty may have freedom of exercise only if it is intrinsically voluntary, that is, if the very action of the faculty is voluntary (MD 19.5.17). Intrinsic voluntariness is here contrasted with extrinsic voluntariness, which arises from the voluntariness of the faculty that brings about the action. For example, walking is extrinsically voluntary because it arises from one’s decision to move one’s limbs, and it is voluntary only because the decision to move one’s limbs is voluntary (MD 19.5.17). According to Sua´rez, only the action of the appetitive faculty—the will—is intrinsically voluntary, and only the will may determine itself (MD 19.5.18). Consequently, it is in virtue of the nature of the will that a free agent ‘‘may act or not act, with all the requisites for action having been posited.’’ Because the indifference attributed to the free agent in the Molinist definition of freedom is secured by the will, the will lies at the heart of Molinist freedom.

The Chimera of Molinism Leibniz criticized the Molinist conception of freedom throughout his philosophical career (A 6.3:132–133; A 6.4:1408–1409; A 6.6:196–197). In the Conversation with Steno, one of his earliest writings on freedom, Leibniz maintained that ‘‘the definition of freedom—that it is the power of acting or of not acting, with all the requisites for acting having been posited . . . is an impossible chimera. . . . This concept of freedom was unknown in antiquity. . . . It was announced by the later Scholastics’’ (A 6.4:1380). Tellingly, Leibniz did not originally attribute this innovation so generally to ‘‘the later Scholastics’’: as noted in Sleigh, Chappell, and Della Rocca (1998: 1258–1259), Leibniz originally wrote: ‘‘This concept of freedom was unknown in antiquity. . . . It was announced by the Molinists.’’11 In the Theodicy, Leibniz continues to pursue this line of criticism against the Molinists. He urges: [O]ne must not imagine that our freedom consists in an indetermination or an indifference of equilibrium; as if one had to be inclined equally to the side of the yes or no, or to different choices, if there is more than one to make. This equilibrium is in any case impossible . . . this equilibrium is also absolutely contrary to experience. (T 35)

In the remarks on King’s Essay on the Origin of Evil appended to the Theodicy, Leibniz notes that King, following the later Scholastics, claims that ‘‘true liberty depends on an indifference of equilibrium, vague, entire, and absolute, so that there is no reason to determine oneself, prior to that determination, neither in the agent who chooses, nor in the object’’ (G 6:401/H 406). Leibniz consistently objects to the assumption of the Molinist account of freedom that in free choice there must exist an ‘‘indifference of equilibrium.’’ According to Leibniz, the Molinist free agent is like Buridan’s apocryphal ass, placed between two equally appetizing piles of hay, with no reason to choose either, although, in the end, the agent (unlike the ass) is somehow able to make a choice.12 Ex hypothesi, the agent is able to choose for no reason at all. According to Leibniz, the choice that the Molinist account allows may be made in this situation

Leibniz Against Molinism

221

is an ‘‘impossible chimera’’ that violates the principle of sufficient reason. Moreover, Leibniz claims that the choice situation that seems to be presupposed in the example is also impossible, because it requires that the universe could be divided into identical parts, thereby violating the principle of the identity of indiscernibles (T 49). Because the conditions that the Molinist definition of freedom places on free choice can never be met, the Molinist account of freedom is itself an ‘‘impossible chimera.’’ These various criticisms of Molinism form a somewhat heterogeneous collection, although they are all centered around the principle of sufficient reason. In order to uncover the heart of Leibniz’s objections to Molinism, I want first to consider the resources available to Sua´rez to respond to Leibniz’s explicit objections to Molinism. I argue that Sua´rez has the resources both to respond to these objections and to a more general objection to Molinism based on the principle of sufficient reason itself. I believe, however, that a deeper objection to Molinism based on the nature of the mind is implicit in Leibniz’s various appeals to the principle of sufficient reason, and I go on to show how attention to this objection draws out the differences between Leibniz’s position and that of the Molinists. Sua´rez agrees with Leibniz that the Buridan’s ass scenario could not actually occur (MD 19.1.9). Moreover, he agrees with Leibniz that if an ass could be placed in the imagined situation—that is, if an ass were placed between two equally appetizing piles of hay—it would not be able to make a choice. Because the ass could not be determined to any one thing, all the requisites for its action would not be in place, and so it would not act at all, because natural causes are such that they cannot act unless all the requisites for acting are in place. Sua´rez explains that the sentient appetitive faculty can operate only in conjunction with a sufficient proposal of and movement by the object. But this is lacking in the present case, since neither of the objects in question is apprehended without qualification as something to be loved and pursued, as is required in order for the appetitive faculty’s effect to follow necessarily. Moreover, the reason why that sort of apprehension or judgment is not had in this case is that the objects mutually impede one another as long as they produce equal movement and resistance within their mode of causing. (MD 19.1.9)

Because the ass has only a sensitive appetite that must be determined to one thing if the ass is to act, the ass would be unable to act in the imagined situation. If, per impossibile, an agent were placed in this situation, she would not be in the same sorry state as an ass, because an agent has a will and may determine herself to one thing rather than to another. However, in determining herself in this way, an agent does not choose for no reason, as Leibniz charges: rather, the reason for which the agent chooses does not determine her to that one thing alone, and consequently she is indifferent (in Sua´rez’s sense of ‘indifferent’) with respect to that thing. While there is a reason why the agent chooses as she does, that reason alone is not sufficient to explain why she chooses as she does: one must also make reference to the agent’s choice itself in order to account for why that choice was made. Leibniz would claim that the Molinist has not gone far enough here, because an insufficient reason is no better than no reason at all. According to Leibniz,

222 Leibniz: Nature and Freedom

a sufficient reason explains why some event happened rather than another, and determines that event to take place (T 44). But the choice attributed to the agent by the Molinists seems not to admit of such an explanation. Most commentators take this implication of the Molinist conception of freedom to be the underlying target of Leibniz’s invocations of the principle of sufficient reason against the Molinists, and they seem to believe that this criticism hits the mark (Davidson 1996: 104–105; Vailati 1997: 92–101). Sleigh gives especially good expression to this line of interpretation, writing: Leibniz characterized a sufficient cause or reason for the obtaining of some state of affairs a as a total set of requisites for the obtaining of a. The principle of sufficient reason requires that, if a obtains, there is a sufficient cause or reason why it obtains. Leibniz also held that if a sufficient cause or reason for some state of affairs a obtained, then a obtains. Put these items together with Molina’s characterization of freedom and it is clear why Leibniz regarded it as an impossible chimera. (Sleigh et al. 1998: 1259)

Sleigh’s reconstruction of Leibniz’s criticism of Molinism does seem plausibly to capture the motivation behind Leibniz’s invocation of the principle of sufficient reason in his criticisms of Molinism, but I do not believe that this criticism would tell against the Molinists. Leibniz’s criticism of Molinism rests on his definition of a sufficient reason as the ‘‘total set of requisites for the obtaining of some state of affairs.’’ Given this definition, the Molinist account of freedom does indeed seem to imply that the ‘‘total set of requisites’’ does not produce some particular state of affairs. Yet Sua´rez anticipates this objection. He distinguishes two kinds of requisites for action: antecedent requisites are what is required if the will is to determine itself; concomitant requisites include the determination of the will (MD 19.4.10). In claiming that the Molinist account of freedom implies that a choice may arise indifferently from the ‘‘total set of requisites,’’ Leibniz seems to take the Molinist definition of freedom to apply to concomitant requisites and therefore does not seem to distinguish between the conditions that must be in place for the will to determine itself and the action that includes the determination of the will. Sua´rez articulates the significance of this distinction: For just as the action is required for acting, so too whatever is included in the action can be said to be required. Now the action is required as that whereby the faculty is formally determined and as that whereby it is constituted as actually acting; and for this reason the action cannot be included among those conditions in the presence of which the faculty has to be indifferent with respect to acting or not acting, since this involves an obvious contradiction. Therefore, neither should one include among those required things anything at all that pertains to the intrinsic nature of the action or is essentially included in the action. For the same argument that applies to the action applies to these things as well. (MD 19.4.12)13

The Molinist definition of freedom—a man is free, given that with all the requisites having been posited, he is able to act or not to act—is meant only to capture the conditions that must be in place if the will is to determine itself freely. Thus it is not meant to apply to the ‘‘total requisites of action’’ as Leibniz believes.

Leibniz Against Molinism

223

Leibniz’s objection based on the principle of sufficient reason seems to miss the Molinist position, because he neglects the crucial role that the will plays in the Molinist account of freedom that accounts for the agent’s ‘‘indifference’’ with respect to her choice. Nevertheless, I believe that further consideration of Leibniz’s objections to Molinism based on the principle of sufficient reason can uncover a deeper aspect of Leibniz’s criticisms of Molinism that reflects some sensitivity to the significance of the will in Molinist freedom. Leibniz maintains that it needs to be asked ‘‘how it was possible that some determination should finally arise, for which there was no source: for to say with Molina that it is the privilege of the free cause is to say nothing, it is to give it the privilege of being chimerical’’ (T 48; see G 2:420). Leibniz’s remark here is directed against the Molinist claim that an agent, a free cause, may determine herself, although this determination does not arise from any preceding circumstances but from the will itself. In this respect, the self-determination essential to the Molinist account of freedom does seem to violate the principle of sufficient reason. The Jesuit Molinists’ Scholastic opponents, the Dominicans, had standardly made just this sort of criticism, charging that the Molinists made the will a ‘self-moved mover’ that determines itself for no reason, and Murray (1996: 44) has suggested that Leibniz is here launching a similar objection to the Molinist position. In a later section of the Theodicy, Leibniz sharpens this criticism. He explains: To wish that a determination comes from a complete and absolutely indeterminate indifference is to wish that it comes naturally from nothing. It is supposed that God does not give this determination: it therefore has its source neither in the soul, nor in the body, nor in the circumstances, because everything is supposed absolutely indeterminate; and that here nevertheless it appears and exists, without any preparation, without any thing that disposes it to that action, without an angel or even God able to know or to show anyone how it exists. And not only does it arise from nothing, it arises on its own. (T 320; emphasis added)

The Molinist does believe that choices arise from the soul, in particular, from the will, and also believes that these choices do take place naturally, because the will is naturally indifferent to particular things, and therefore equally naturally has the power to determine itself (MD 19.8.17–18). But Leibniz has a different conception of the nature of the mind from the Molinists, and indeed from the Scholastics more generally, which I think underwrites his criticism of Molinism and to which I now want to turn. In the late formulation of his metaphysics, the Monadology, Leibniz explains that monads consist of perceptions and appetitions (Mon 15–16).14 In the Principles of Nature and of Grace 2, Leibniz offers a similar formulation, explaining that one monad by itself and at a single moment cannot be distinguished from another except by its internal qualities and actions, and these can only be its perceptions— that is to say, the representations of the compound, or of that which is without, in the simple—and its appetitions, that is to say, its tendencies from one perception to another, which are the principles of change. (G 6:598/L 636)

224

Leibniz: Nature and Freedom

In these works, Leibniz seems to suggest that two principles are required in order to specify the nature of the mind: perceptions, or representations, and appetitions, the ‘‘principles of change’’ that determine the transition from one perception to another. In other late texts, however, Leibniz suggests that perceptions are more fundamental to the mind than appetitions, and that there is merely a theoretical distinction between perception and appetition.15 In a letter to Christian Wolff, Leibniz explains: In truth, whatever is in the soul may be understood generally to be reduced to two things: the soul’s harmonious expression, in accordance with the body, of the present state of external things, and the endeavor to a new expression that represents the endeavor of bodies (or external things) to a future state, in a word, perception and tending towards perception. (GLW 56)16

Although there may be a theoretical or conceptual distinction between appetitions and perceptions, these remarks suggest that there is no functional distinction between them. This point is relevant to understanding Leibniz’s conception of the will. According to Leibniz, intelligent substances—the only free substances—do not merely have appetitions, they may also be conscious of them, and this consciousness of appetitions constitutes the human will. This idea may be found in a variety of Leibniz’s mature philosophical works. In a fragment written in the early 1680s, he defines the will as ‘‘the tendency to acting of which we are conscious’’ (A 6.4:1444). He elaborates on this definition in the New Essays, explaining that ‘‘volition is the effort or endeavour (conatus) to move towards what one finds good and away from what one finds bad, the endeavour arising immediately out of one’s awareness of those things’’ (A 6.6:172). The concluding clause of the preceding passage reflects the same reductionism with respect to appetitions as the passage from the letter to Wolff. Indeed, in several passages, Leibniz suggests that—to borrow an expression from Murray17—choices are the vector sum of the appetitions that arise out of various perceptions. For example, in the New Essays, Leibniz explains that ‘‘various perceptions and inclinations combine to produce a complete volition: it is the result of the conflict amongst them’’ (A 6.6:192), and he remarks that ‘‘everything which . . . impinges on us weighs in on the balance and contributes to determining a resultant direction, almost as in mechanics’’ (A 6.6:193). Because the Leibnizian will is not functionally distinct from perceptions, it lacks the functional independence from perceptions (from the deliverances of the intellect) of the Molinist will, and therefore cannot play the functional role that the Molinists attributed to the will. This difference between Leibniz’s conception of the will and that of the Molinists comes out especially clearly in their discussions of the proper objects of the faculties of the mind. In the Metaphysical Disputations, Sua´rez tries to illustrate the difference between the intellect and the will by contrasting the relation between the intellect and its proper object, the true, and that between the will and its proper object, the good. Sua´rez explains:

Leibniz Against Molinism

225

The formal object of the intellect is truth, and truth and falsity cannot exist in the same object and so the intellect—both of itself and as regards the merit of its object—is always determined to one effect with respect to the species of its act. Hence, if in some cases it is not adequately determined, this is only because the object is not sufficiently proposed or sufficiently apparent, and not because the intellect itself has an internal power or control over its own act. By contrast, the object of the will is the good, and one and the same object can be simultaneously both good and evil—that is, both fitting and unfitting in relation to different things or under different aspects. And so even if the object is perfectly proposed or cognized, the appetitive faculty [the will] can be indifferent . . . with respect to pursuing or rejecting such an object. (MD 19.5.16)

The crucial point for my purposes is the contrast that Sua´rez draws here between the intellect and the will. The intellect is naturally necessitated to assent to what is true and dissent from what is false, and because it is determined by its objects, it cannot determine itself. By contrast, the will is not naturally necessitated to any particular good or any particular end, and therefore the will must determine itself to some particular good or end. Sua´rez’s point is that the nature of the will is such that it must determine itself, while the intellect cannot do so. In the Theodicy, Leibniz comments on the conception of the relation between the intellect and the will common to the Aristotelian tradition. And with respect to the parallel between the relation of the intellect to the true and the will to the good, it must be noted that a clear and distinct perception of a truth actually contains within itself the affirmation of this truth and therefore the intellect is necessitated by it. But whatever perception one has of the good, the effort of acting after this judgment, which to my mind is of the essence of the will, is distinguished from this perception. Thus, because time is required to effect this effort, it can be suspended and even changed by a new perception. (T 311)

Like Sua´rez, Leibniz here distinguishes between the functioning of the intellect and of the will, and locates the difference between these faculties in their relation to their proper objects. According to Leibniz, the difference between the intellect and the will consists in the fact that the determination of the will—the effort or tendency that derives from the perception of the object—may only be diverted by other tendencies, arising from other objects, before it attains its end. Although according to Leibniz there is a difference between the intellect and the will that reflects the nature of those faculties, the Leibnizian will has no self-determining power, and is intrinsically as determined by the good as the intellect is by the true. Consequently, the nature of the Leibnizian mind rules out the real use of the will by the agent required by the Molinist account of freedom. Leibniz’s criticism of Molinism thus cuts more deeply than that of the Dominicans, to which it seems to bear some similarity, because the Dominicans shared the Jesuit Molinists’ conception of the faculties of the mind and simply disagreed with them about the nature of the interaction of these faculties. Leibniz, by contrast, rejects this conception of the mind altogether.18 Indeed, I think that Leibniz’s repeated appeals to the principle of sufficient reason against the Molinists may be seen as reflecting the differences between his conception of the mind

226 Leibniz: Nature and Freedom

and that of the Molinists. Because the Molinist account of freedom appeals to a notion of the will that does not fit in the Leibnizian mind, Leibniz believes that the Molinist account of freedom does not have a sufficient foundation in the nature of the mind. From Leibniz’s perspective, what is chimerical about Molinist freedom is the Molinist will.

Leibnizian Freedom and the Leibnizian Mind Leibniz’s and Sua´rez’s accounts of freedom reflect the differences in their conceptions of the mind and will. Some consideration of the definition of freedom that Leibniz proposes in the Theodicy will help to illustrate this claim. According to Leibniz, freedom, such as is required in the theological schools, consists in intelligence, which includes a distinct knowledge of the object of deliberation; in spontaneity, with which we determine ourselves; and in contingency, that is to say in the exclusion of metaphysical or logical necessity. Intelligence is like the soul of freedom and what remains is like the body or the foundation. (T 288)

Leibniz and Sua´rez agree that intelligence, spontaneity, and contingency are severally necessary and jointly sufficient conditions of human freedom, but they disagree about how these conditions are interrelated and how they are to be explicated. According to Sua´rez, it is because intelligent beings have wills and may determine themselves—in virtue of the fact that they are spontaneous, in the terminology of the Theodicy—that their actions are contingent and not naturally necessitated. Because the Leibnizian will has no self-determining power, and is not spontaneous in Sua´rez’s sense, Leibniz cannot account for contingency by appeal to the will, and so the will must have a much less central place in Leibniz’s account of freedom than it does in that of Sua´rez. Leibniz and Sua´rez consequently characterize human freedom in quite different terms. According to Sua´rez, the will is the ‘‘form’’ of freedom, while Leibniz terms intelligence, or reason, the ‘‘soul’’ of human freedom. Insofar as Sua´rez, like other Scholastics, believed that the soul was the form of a thing, Leibniz’s terminology could be taken to imply that he takes intelligence to be the form of human freedom. In other words, the implication is that intelligence plays the fundamental role in Leibniz’s account of freedom that Sua´rez assigns to the will. But is there any role for the will in Leibniz’s account of freedom? In order to answer this question, we need to consider Leibniz’s conception of spontaneity. According to Leibniz, a substance is spontaneous and may be said to determine itself just in case its states are not determined from without, and in Leibniz’s metaphysics, all substances are self-determining in this sense, because their states are determined by an internal principle of action. Leibniz explains: [E]ach simple substance naturally has perceptions, and its individuality consists in the perpetual law that determines the succession of perceptions that are assigned to it, and which arise naturally from one another. . . . Thus it follows that the soul

Leibniz Against Molinism

227

has thus in itself a perfect spontaneity, insofar as it only depends upon God and itself in its actions. (T 291)

The internal principle of action that accounts for the spontaneity of substances is represented as the will in conscious substances, so it does after all seem that there is a role for the will in Leibniz’s conception of freedom, although just not the role attributed to the will by the Molinists. But this sort of spontaneity cannot contribute in any meaningful way to Leibnizian freedom.19 This spontaneity is common to all substances—animals and human beings—so it does not explain freedom, which is the prerogative of intelligent beings. In this context, Leibniz’s clarificatory remark about spontaneity in the Theodicy is telling: ‘‘a perfect spontaneity is common to both us and animals, and in the intelligent or free substance it becomes control over one’s actions’’ (T 291; see DM 32; Grua 362). Spontaneity and the will (as the representation of the internal principle that accounts for spontaneity) do not after all play a role in Leibniz’s account of freedom; it is only because human beings are intelligent that they are responsible for their choices and may be said to determine themselves. Leibniz’s conception of the mind thus leads him to locate freedom in intelligence alone.20 In the New Essays, Leibniz redefines the problem of freedom in terms that do not make any functional appeal to the will. Leibniz distinguishes two senses in which one may speak of freedom. One is when one opposes it to the imperfection or slavery of the mind, which is a compulsion or constraint . . . like that which comes from the passions; the other sense is to be found when one opposes freedom and necessity. Employing the first sense the Stoics said that only the wise person is free. . . . It is in this way that God alone is perfectly free, and that created minds are free only to the extent that they are above passion; and this kind of freedom pertains strictly to our understanding. But the freedom of mind that is opposed to necessity concerns the bare will. . . . This is what is called free choice [le franc arbitre]; it consists in the view that strongest reasons or impressions that the understanding presents to the will . . . do not confer upon it an absolute or (so to speak) metaphysical necessity. (A 6.6:175)

The first sense of freedom, freedom of understanding, consists in rationality. According to Leibniz, substances may be more or less free in virtue of the extent to which their actions reflect ‘‘clear and distinct’’ instead of ‘‘confused and obscure’’ perceptions, intellectual perceptions rather than passions. The second sense of freedom—contingency—is what Leibniz believes that philosophers have been traditionally concerned with in discussions of freedom of the will, and is what Sua´rez sought to preserve by appeal to the self-determining power of the will. Although Leibniz does employ the term ‘will’ in his formulation of this second sense of freedom, his use of it carries no more ontological commitment than contemporary references to the will or volitions in discussions of freedom, for Leibniz makes no functional appeal to the will in his account of contingency. According to Leibniz, contingency simply consists in the absence of metaphysical necessity: that is, in the fact that an agent’s choices are not themselves metaphysically necessary.21 To my mind, freedom of the understanding is at the heart of Leibniz’s conception of freedom. The minimal absence of metaphysical or logical necessity

228 Leibniz: Nature and Freedom

required for contingency certainly does not suffice by itself to capture the nature of freedom, especially given the fact that Leibniz believed that all natural events, including the movements of physical objects, are contingent in this sense (A 6.6:176). By contrast, it is in virtue of having intellects that substances are capable of freedom at all, and the centrality of freedom of understanding mirrors Leibniz’s later formulation that intelligence is the ‘‘soul’’ of human freedom. This account of freedom, however, opens Leibniz to the objection that his account of freedom does not yield any genuine freedom. After all, Leibniz’s conception of the mind seems to rob human beings of the power to initiate their own actions, and his metaphysical reduction of choices to the sum of various tendencies in the mind seems to transform the mind into a mechanism. Kant gave perhaps the most famous formulation of this objection, remarking that Leibniz’s account of freedom yields nothing more than ‘‘the freedom of a turnspit, which when once wound up also carries out its motions of itself’’ (Ak 5:97). In a similar Kantian vein, J. B. Schneewind maintains that Leibniz’s account of free will ‘‘still leaves us with questions about freedom and the understanding of ourselves as agents rather than as mere loci of more or less confused perceptions’’ (Schneewind 1998: 242). Sua´rez would probably have objected to Leibniz’s conception of freedom along similar lines. I believe that Leibniz should not have been much troubled by this line of objection. The Kantian criticism is motivated by the conviction that freedom requires active self-determination by the will in order to explain why human beings may merit reward or punishment for their actions. Sua´rez, too, is concerned with such issues. He notes that ‘‘everyone believes that those who act badly deserve punishment by the very fact that acting in such a way has been placed within their own power and will’’ (MD 19.2.16). But Leibniz does not posit such a close connection between freedom and reward and punishment. In the Theodicy, Leibniz even flirts with a quasi-utilitarian justification of reward and punishment, which he claims could be justified even if all human actions were metaphysically necessitated (T 67–75).22 To my mind, one way of understanding this flirtation is as an attempt on Leibniz’s part to distinguish issues of reward and punishment (questions having to do with ethics and justice) from issues of freedom. In making this distinction, Leibniz separates himself from a deep-rooted tradition that conceives of freedom in terms of accountability—in terms of whether an agent is justified in being rewarded or punished for her actions—a tradition to which both Kant and Sua´rez belong. Leibniz may be seen instead as beginning to articulate a conception of freedom as attributability, according to which freedom simply consists in an agent’s capacity for rational self-government.23 The difference between these two conceptions of freedom is that freedom as accountability conceives of freedom as bound up with active self-determination, and therefore is quite congenial to a conception of the will, whereas freedom as attributability conceives of self-determination as rational determination rather than as an exercise of active power. Because freedom as attributability does not require the real use of the will, there is as it were a preestablished harmony between Leibniz’s conception of freedom and his conception of the mind.

Leibniz Against Molinism 229

Closing I have argued that Leibniz’s objections to Molinist freedom reflect his conception of the nature of the mind, and suggested that consideration of Leibniz’s conception of the mind helps illuminate what is distinctive about his conception of freedom. Leibniz breaks with the Scholastic tradition in reconfiguring the nature of the will and its place in accounts of freedom, and attention to Leibniz’s response to this tradition clarifies the nature of his break from it. I believe that further consideration of the role of the mind (and will) in early modern accounts of freedom will illuminate what is distinctive about those accounts. Today it is commonly believed that early modern philosophers transformed the problem of freedom by embarking on a line of enquiry that leads to the modern problem of free will framed in terms of a tension between determinism and human freedom.24 This received view obscures aspects of early modern approaches to the problem of freedom that do not cohere easily with our own contemporary conception of that problem. Historicizing the problem of freedom by tracking the evolution of particular concepts—such as the will—central to early modern formulations of the problem will combat such myopia and enable us to achieve a deeper, more contextually situated understanding of the early modern problem of freedom. A richer historical understanding of the early modern problem of freedom may also help us better to understand our own problem of freedom. By coming to understand the nature and source of the problems that drove early modern discussions of freedom and the conceptual resources drawn upon in these discussions, we may come better to appreciate the nature of the problem space that structures contemporary approaches to freedom, as we come to see the differences between our problem of freedom and the early modern problem. In this way a more historically sensitive approach to the problem of freedom in early modern philosophy may well yield a deeper philosophical understanding of the problem of freedom, both in its early modern and contemporary manifestations.

Notes An earlier version of this essay was presented at the 2001 New England Seminar in Early Modern Philosophy: thanks to the members of the audience for their discussion. Thanks also to Nate Bowditch, Paul Lodge, Matt McAdam, J. B. Schneewind, Alison Simmons, Kurt Smith, Carol Voeller, and the editors of this book, Jan Cover and Don Rutherford, for helpful comments and suggestions. I would like to register a special debt to Bob Sleigh, without whose encouragement and support I would never have written on Leibniz. 1. Both contemporary philosophers and historians of early modern philosophy tend to see the problem in this way. For example, in the introduction to an anthology of contemporary articles on free will, Gary Watson explains that ‘‘free will is problematic for many philosophers because of its controversial relation to determinism’’ (Watson 1982: 2). Although the recent survey of early modern approaches to this topic, Sleigh, Chappell, and Della Rocca (1998), is part of the section of The Cambridge History of Seventeenth-Century Philosophy entitled ‘‘Will, Action, and Moral Philosophy,’’ the article’s focus is reflected in its title: ‘‘Determinism and Human Freedom.’’ There is relatively little attention to the will in the body of the article.

230 Leibniz: Nature and Freedom 2. Alexander (1898) treats the concept of the will from the Pre-Socratics to Lotze; he undertakes relatively little in-depth analysis, but nevertheless provides a good general overview of the terrain. Bourke 1964 outlines but does not work out in detail the nature of changes in conceptions of the will. Bourke remarks that ‘‘one way of making this investigation would be to write a history of philosophy, taking will and volition as a central interest, and treating the various thinkers in schools or in chronological order. I must confess that this is what I had in mind originally. . . . In effect, I have assembled such a chronological account, in preliminary research notes, and it is a welter of materials which requires further organization’’ (8). Bourke’s ‘‘preliminary research notes’’ are at times illuminating, but his account is essentially a catalogue of the various meanings that the term ‘will’ has had over the years. Arendt 1978 considers the implications of different accounts of the will for conceptions of freedom, with special attention to Augustine, Aquinas, Scotus, Nietzsche, and Kant. Although Arendt’s observations are brilliant, her focus is not particularly historical, and her investigation is structured by the aim of clearing space for a phenomenology of political action. 3. Murray 1996 also advocates examining Leibniz’s account of freedom from the ‘‘axis of faculty psychology’’—by considering the role that the intellect and the will play in Leibniz’s account of freedom—although he comes to quite different conclusions on the basis of this investigation. 4. In particular, it was devised to provide an interpretation of session 6, canon 4, of the Council of Trent: ‘‘If someone were to say that man’s free decision [liberum arbitrium] moved and excited by God, by assenting to God’s call and excitement, in no way cooperates towards disposing and preparing itself in order to obtain the grace of justification and that it cannot dissent, if it wants [neque posse dissentire, si velit], but that like something inanimate it does nothing whatever and is merely passive, let him be anathema.’’ For Molina’s account of freedom, see Molina 1953: 1.2.3. 5. See Pufendorf 1934: bk. 1, chap. 4; Bramhall, A Vindication of Liberty (in Hobbes and Bramhall 1999); Clarke 1998. 6. In ‘‘Of Liberty and Necessity’’ Hobbes addresses the Molinist definition of freedom: ‘‘I hold that the ordinary definition of a free agent [the Molinist definition], namely that a free agent is that which, when all things are present which are needful to produce the effect, can nevertheless not produce it, implies a contradiction and is nonsense; being as much as to say that the cause may be sufficient, that is necessary, and yet the effect shall not follow’’ (Hobbes and Bramhall 1999: 39). Cudworth claims that ‘‘if the . . . will do not necessarily follow the last dictate of practical judgement of the necessary understanding, but still remains indifferent and doth fortuitously determine itself either in compliance with the same or otherwise, then will liberty of will be mere irrationality, and madness itself acting or determining all human actions’’ (1996: 169). Locke charges that ‘‘a perfect Indifferency in the Mind, not determinable by its last judgment of Good and Evil, that is thought to attend Choice, would be so far from being an advantage and excellency of any intellectual Nature, that it would be as great an imperfection, as the want of Indifferency to act, or not to act, till determined by the Will, would be an imperfection on the other side’’ (Essay 2.21.48). Locke later adds, in words reminiscent of Cudworth: ‘‘Is it worth the Name of Freedom to be at liberty to play the Fool, and draw Shame and Misery upon a Man’s self? If to break loose from the conduct of Reason, and to want that restraint of Examination and Judgment, which keeps us from chusing or doing the worse, be Liberty, true liberty, mad Men and fools are the only Freemen: But yet, I think, no Body would chuse to be mad for the sake of such Liberty, but he that is mad already’’ (Essay 2.21.50). 7. I take Sua´rez rather than Molina as my starting point here because Sua´rez’s account of freedom is subtler and more detailed than Molina’s account, although it does follow the framework laid down by Molina. Moreover, Leibniz may well have known Sua´rez’s work better than Molina’s. Although he does refer to Molina in the Theodicy, those references may be derived as much from the text of Pierre Bayle to which he was responding as from his own reading. Leibniz was certainly acquainted with Sua´rez’s work: his master’s thesis on the principle of individuation was largely a response to Sua´rez’s discussion of that topic. In the New Essays, he pays Sua´rez the following compliment: ‘‘But to be fair to the deeper Scholastics, such as Sua´rez . . . it should be acknowledged that their works sometimes contain substantial discussions—for instance of the continuum, of the infinite, of contingency, of the reality of what is abstract, of the principle of individuation, of the origin of forms and a vacuum among forms, of the soul and its faculties, of God’s communion with created things, etc., and even, in moral philosophy, of the nature of the will and the principles of justice. In short, it must be admitted that there is still gold in that

Leibniz Against Molinism

231

dross’’ (A 6.6:431; emphasis added). Leibniz himself thought Sua´rez’s work deserved to be taken seriously, and this provides another prima facie reason for considering it in this context. Nevertheless, I do not mean to attribute to Leibniz a deep or detailed understanding of the Metaphysical Disputations. In this respect, it is worth bearing in mind Sleigh’s cautionary remark about the relation between early modern philosophers and their predecessors: ‘‘It is entirely salutory that historians of early modern philosophy are paying increasing heed to the late mediaeval and Renaissance background of their subject. But it is important to take into account exactly what the philosopher under investigation knew of that background. It is worth remembering that we are not studying historians of philosophy; for the most part we are studying philosophers, whose familiarity with the texts of those whose names they bandied about may be surprisingly meagre’’ (Sleigh et al. 1998: 1275, n. 170). In what follows, I present only as much of Sua´rez’s account of human freedom as is necessary to focus the target of Leibniz’s criticisms of Molinism. 8. J. B. Schneewind recognizes this point, remarking that ‘‘Leibniz does not call upon a separate faculty of the will in order to account for freedom of choice’’ (1998: 243). Yet Schneewind does not, to my mind at least, explain clearly why Leibniz does not call upon a faculty of the will in order to explain freedom, or consider the implications of his explanation. 9. Murray 1996: 32–36 similarly surveys Sua´rez’s account of freedom. 10. It might be objected that these definitions do not adequately distinguish free causes from natural causes, because free causes, like natural causes, also may only act if ‘‘all the requisites for acting are in place,’’ with the difference consisting in the fact that one of the requisites of the action of the free cause is the agent’s choice. However, this is a misunderstanding. The ‘‘requisites for action’’ in question here should be understood to be external or antecedent conditions for action, conditions that apply both to natural and free causes. Sua´rez’s point is that these antecedent conditions do not determine an agent to act in the way that they determine a natural cause to act. Sua´rez explains: ‘‘when it is said that what is free is such that it is able to act and able not to act, given that all things required for acting having been posited, this should be understood as referring to the things that are required antecedently, and not to the others. . . . This very point is suggested . . . when it is claimed that a free faculty must be disposed and proximately prepared for its work in such a way that, given that disposition, it is able to act or not to act. For the preparation in question requires all those things that are required in the first act for operating’’ (MD 19.4.10). Leibniz himself seems not to recognize this point, and in fact raises an objection to Molinist freedom akin to the one considered in this note. I return to this point in the third section. 11. In A Treatise of Freewill, Ralph Cudworth makes a similar criticism, noting that ‘‘as for this scholastic definition of freewill, viz. that it is, after all things put [a variant translation of ‘with all things having been posited’ (omnis requisitis positis) in the Molinist definition], besides the volition itself, an indifferency of not doing or of doing this or that. This is an upstart thing, which the ancient peripatetics, as Alexander and others, were unacquainted with’’ (1996: 170). Both Leibniz and Cudworth were quite historically sensitive philosophers, so it is not surprising that they would both note that the Molinist definition of freedom was somewhat of an innovation in historical terms. One aspect of Cudworth’s criticism is worth remarking, because it differs from that of Leibniz. Leibniz tends to focus his criticisms of Molinism on the notion of freedom of exercise, which I suggested in the second section was central to the Molinist conception of freedom. In the quoted passage, however, Cudworth conflates freedom of exercise—the freedom to will or not to will some act—with freedom of specification, the freedom to will this or to will that object, to specify the object of one’s will. Although Sua´rez takes freedom of exercise and freedom of specification to be related, he is careful to distinguish them (MD 19.5.18). 12. The charge that the Molinist free agent is like Buridan’s ass is a later formulation of Leibniz’s basic criticism of Molinism that derives from Leibniz’s encounter with Bayle’s article on Buridan in the Dictionnaire Historique et Critique. See G 3:402. 13. Hobbes also raised an objection akin to that of Leibniz against his Molinist opponent, Bramhall, who responded to Hobbes in words reminiscent of Sua´rez. See Hobbes and Bramhall 1999: 38, 62. 14. Relatively little scholarly attention has been paid to Leibniz’s conception of appetition: one of the few exceptions is Kulstad 1990. To some extent, this is due to the fact that Leibniz himself does not devote the attention to appetition that he does to perception. 15. Rutherford 1995b: 138 similarly points out that the distinction is merely ‘‘conceptual.’’

232 Leibniz: Nature and Freedom 16. Thanks to Don Rutherford for drawing my attention to this passage. See also Leibniz’s response to Bayle at G 4:562, cited in Rutherford 1995b: 138–139. 17. ‘‘Spontaneity and Freedom in Leibniz,’’ this volume. 18. Ezio Vailati has noted this line of argument as well. His remarks, which arise in the course of his dicussion of Leibniz’s criticisms of Clarke’s conception of freedom, are worth quoting at some length. According to Vailati’s Leibniz, ‘‘acceptance of liberty of indifference ultimately destroys the very possibility of a science of the mind. . . . The point can be expanded by considering Leibniz’s critique of the notion of ’mere will’ which Clarke uses to explain choice in a state of equilibrium. Leibniz claimed that in addition to being unworthy of God, a mere will could not exist, since its notion is incompatible with the definition of will. . . . Leibniz did not elaborate, but presumably what he had in mind was the sort of definition of volition that one can see in the New Essays: ‘volition is the effort or endeavour (conatus) towards what one finds good and away from what one finds bad, the endeavour arising immediately out of one’s awareness of things’ [NE 2.21.5]. . . . Ultimately, if one takes the view that science deals with laws, as Leibniz saw, Clarke’s view leads to the conclusion that no science of volition, and hence of human action, can be given’’ (Vailati 1997: 95–96). I think that Vailati’s reading mislocates Leibniz’s objection to the Molinist account of freedom. Leibniz does not generally think of the natural world—the mind or the body—in terms of laws, but rather in terms of natures. (See Jolley, ‘‘Leibniz and Occasionalism,’’ this volume, for a consideration of the implications of this strand of Leibniz’s philosophy in the context of his debate with Malebranche.) Leibniz’s criticism of the Molinists turns on the idea that the nature of the Leibnizian mind does not leave any room for the Molinist will. The Molinist conception of the will undermines the possibility of a science of the mind because it postulates a faculty that does not fit within the (Leibnizian) mind. 19. Here I agree with Murray, ‘‘Spontaneity and Freedom in Leibniz’’ (this volume), although we understand the implications of this point very differently. 20. Commentators have begun to highlight the importance of intelligence to Leibniz’s conception of human freedom. See, among others, Phemister 1991 and Davidson 1998. 21. This claim will certainly strike some readers as controversial; it deserves more discussion in another context. I will briefly give my reasons for this claim. Throughout Leibniz’s philosophical career, from the Confessio Philosophi to the Discourse on Metaphysics to the Theodicy (A 6.3:1293; DM 4; G 6:31–32, 37/H 55–57, 61), Leibniz was concerned to show the fallaciousness of fatalism, which he believed had disastrous consequences for morality and religion. Leibniz believed that fatalism rested on ‘‘confused notions that had been formed concerning freedom, necessity, and destiny’’ (G 6:29/H 53). The confusion that led to fatalism was the belief that all actions were metaphysically necessitated, and therefore Leibniz sought to articulate a clearer notion of necessity that would help to turn back the threat of fatalism and restore faith in morality and religion. Consequently, it seems to me that throughout his work Leibniz is most concerned to show how his system could avoid the metaphysical necessity that he took to be a great threat to human freedom. 22. In this stretch of text Leibniz does admit that there might not be any justification for retributive punishment in a world in which human actions were determined by metaphysical necessity (T 73). Some readers might therefore take Leibniz to agree with Kant and Sua´rez here about the nature of the freedom required for genuine moral responsibility. However, Leibniz elsewhere suggests that he was not particularly interested in justifying retributive punishment. For example, Leibniz remarks that ‘‘judges do not seek the reasons which have disposed a man to have a bad will, but only stop to consider the extent to which this particular will is bad’’ (DM 30; see T 264). Moreover, I believe that even the kind of retributive punishment that Leibniz believed may have been appropriate did not require the kind of selfdetermination required by Kant or Sua´rez. According to Leibniz, the only threat to retributive punishment is metaphysical necessity, and in order to turn back the threat of metaphysical necessity Leibniz does not need an account of freedom along the lines of that of Kant or Sua´rez. 23. This terminology is derived from Watson 1996: 228–229. 24. Don Garrett gives especially clear expression to this widely held view: ‘‘The Scientific Revolution demanded a rethinking of the relation between human beings and nature. The success of mechanistic explanation led to conceptions of the causal relation that take events, rather than substances, as the most fundamental relata. . . . Event-centered conceptions of causation led, in turn, to more precise formulations of the doctrine of causal determinism within natural science, while the

Leibniz Against Molinism

233

explanatory successes of natural science inevitably raised questions about the applicability of its methods to human affairs. As a result, philosophers were forced to address anew the relation between natural events and human actions, the nature of human responsibility for actions, and the justice of divine rewards and punishments in a world of causal determination’’ (Garrett 1997: 118). Schneewind 1998 has already mounted a sharp challenge to a central assumption that Garrett makes in this passage—the assumption that natural science drove philosophical enquiry in the early modern period, including enquiry into ethics. I believe that investigation into the place of conceptions of the mind and will in early modern accounts of freedom may lead us further to rethink the traditional conception of the early modern problem of freedom to which Garrett gives voice.

12

Video Meliora Proboque, Deteriora Sequor Leibniz on the Intellectual Source of Sin J A C K D. D A V I D S O N

eciding to help Jason, thereby betraying her father and his kingdom, Ovid’s

D Medea says: ‘‘Video meliora proboque, deteriora sequor’’—I see the better

way and approve it, but I follow the worse way.1 Leibniz knew this quotation well; it appears, for example, in the very early Confessio Philosophi (1672/3) and thirty-seven years later in the Theodicy (1710). Ovid’s Medea’s violence is not explained by the fact that she is a barbarian witch; rather, her lament is moving precisely because it poetically expresses our common experience of knowing the wrong action is wrong, and doing it anyway (or knowing the right action, and failing to do it). Such experience seems presupposed by Saint Paul when he writes: ‘‘For what I want to do I do not do, but what I hate to do I do’’ (Romans 7:15). Socrates, as represented by Plato and Aristotle, denied this natural view, maintaining instead that no one ever knowingly does wrong.2 Let us call views like Socrates’, that no one ever willingly does wrong, intellectualism; according to the intellectualist, all volitional error is preceded by cognitive error. For the intellectualist, knowledge is sufficient for virtue, and so supports the following implication: for any goods x and y, we cannot both believe that x is better than y, and choose y. According to this view, akrasia, or weak-willed behavior, is conceptually impossible. The denial of intellectualism, anti-intellectualism, denies that knowledge is sufficient for virtue, and so supports the following implication: for any goods x and y, we can both believe that x is better than y, and do y. Anti-intellectualism did not wait long for a champion. Finding Socrates’ model of soul too simplistic to account for what he took to be obvious cases of psychological conflict, Plato presents the story of Leontius, who, walking outside the city walls, encounters some corpses lying in the execution field: 234

The Intellectual Source of Sin

235

He had an appetite to look at them but at the same time he was disgusted and turned away. For a time he struggled with himself and covered his face, but, finally, overpowered by the appetite, he pushed his eyes wide open and rushed toward the corpses, saying ‘‘Look for yourselves, you evil wretches, take your fill of the beautiful sight.’’ (Republic 439e–440)3

Unlike Socrates, Plato acknowledges akratic behavior, and uses it as evidence for his tripartite model of the soul. This essay examines Leibniz’s account of why we sin, concentrating in particular on the question of intellectualism. His views on practical reason and its occasional breakdown are fascinating in their own right, offering an explanation of a phenomenon many of us find both existentially and theoretically vexing.

The Seventeenth-Century Background For the medievals and for most early modern philosophers, choice was understood as the result of two faculties, the intellect and the will. The proper end of the intellect was truth, its role to assess the value of the choices available to an agent. The proper end of the will was the good, its role to incline and choose that possibility judged best by the intellect. Seventeenth-century philosophers disagreed, however, over the precise relation between the intellect and will. The nature of this relation was among the central issues at the heart of the intramural and extramural disagreements between Jesuits and Dominicans, and among the Lutheran, Reformed, and Remonstrant (or Arminian) confessions. Those who gave priority to the will often claimed that an action is free only when the will is indifferent. Philip van Limborch, the head professor of the Remonstrant seminary in Amsterdam and a staunch defender of the Arminian understanding of freedom, wrote to John Locke that indifference, the sine qua non of freedom, ‘‘is that energy of spirit by which, when all requisites for acting are present, it can act or not act’’ (Locke 1982: vol. 7, 367).4 Included in the present requisites are God’s general concurrence and the judgment of the intellect. According to the Jesuits and Remonstrants, an agent need not will or choose that which is judged best by his intellect. According to van Limborch, ‘‘the will is the master of our actions, and governs them according to its choice’’ (Locke 1982: vol. 7, 368). According to this libertarian version of anti-intellectualism, freedom rests in an undetermined act of the will, independent of the intellect’s evaluation and determination. Call this view voluntarism. In his Loci communes the influential Lutheran theologian Philip Melanchthon gives his version of the view: Knowledge serves the will (voluntas), and thus one calls the will . . . joined with the knowledge or with the understanding of the intellect by a new name, ‘free will’ (arbitrium). For the will . . . in man corresponds to the place of a despot in a republic. Just as the senate is subject to the despot, so is knowledge to the will . . . with the consequence that although knowledge gives good warning, yet the will . . . casts knowledge out and is borne along by its own affection. . . . Freedom cannot rightly be said to belong to the knowing part; rather it is subject to the will . . . and is driven hither and yon. (Melanchthon 1969: 23–24)

236 Leibniz: Nature and Freedom

Advocates of voluntarism acknowledge what many would call an obvious fact of our lives—that we often (or at least sometimes) do wrong knowingly—whereas intellectualism must explain it away. Second, advocates claim voluntarism makes us properly responsible for our sins, which in turn makes divine judgment just and morality itself possible. By contrast, intellectualism entails, in the words of William Molyneux, ‘‘that a Man shall be Damn’d, because he understands no better than he does’’ (Locke 1982: vol. 4, 601). Third, proponents of voluntarism argue that any other view violates orthodox teaching on soteriology. Scripture writes that God loves the whole world (1 John 3:16, 1 John 2:2). Voluntarists understand this to mean that those not elected freely reject God, not that the damned are lost through a failure to understand. For Christian philosophers concerned with exonerating God from various versions of the problem of evil, voluntarism of a libertarian stripe is an important component of the powerful free will theodicy or defense of God’s justice. Just as those who give primacy to the will claim that an action is free only when the will is indifferent, so those who give primacy to the intellect claim that an action is free only when it is rational. We find this view in Francis Turretin, the most important Reformed theologian of the second half of the seventeenth century and an eloquent advocate of intellectualism. In his influential Institutio Theologicae Elencticae he writes that the rational agent ‘‘cannot be rational without on that very account being free; nor can he be deprived of liberty without being despoiled of reason’’ (Turretin 1992: t. 10, q. 3, a. 11).5 Since the will is created by God to be ‘‘a rational appetite,’’ it must follow ‘‘the last judgment of the practical intellect’’ (t. 10, q. 2, a. 7), otherwise we would be seeking our own misery, which is absurd (t. 10, q. 13, a. 8). This entails that ‘‘every sin arises from ignorance since the will cannot seek evil as evil, but [only] under the idea of good’’ (t. 9, q. 13, a. 5). Turretin says that Medea’s famous lament is true only if read as reflecting a judgment preceding ‘‘the last judgment,’’ for the latter ‘‘can never oppose the decided and last judgment’’ (t. 10, q. 2, a. 15). Sin is ‘‘error in the mind’’ (t. 9, q. 13, a. 5). Turretin also illustrates the natural affinity between intellectualism and compatibilism, the latter being the view that cognitive and volitional systems can be determined and yet free. Rejecting the Molinist definition of freedom embraced by the Arminians (and present-day libertarians), Turretin proposed instead: ‘‘[I]t suffices [for freedom] that there be spontaneity and willingness depending upon a judgment of the reason’’ (t. 10, q. 3, a. 15; emphasis added). Far from destroying freedom, the determination of the will by the intellect ‘‘makes it perfect’’ (t. 10, q. 4, a. 12). Since indifference is not to be found in God who is ‘‘good most freely . . . but necessarily and immutably’’ (t. 10, q. 3, a. 5) and since the essence of freedom must be the same on earth as in heaven, ‘‘the Jesuits, the Socinians and Remonstrants’’ are in deadly error in their efforts to ‘‘bring back (either openly or secretly and by burrowing) Pelagianism and semi-Pelagianism and to place the idol of free will in the citadel’’ (t. 10, q. 1, a. 1). Advocates of intellectualism can interpret the biblical teaching that we are made in God’s image in a robust and straightforward fashion, for just as God’s will invariably follows the judgments of his intellect, so do ours. Second, the Christian intellectualist has an account of freedom that makes God the most free being, not, as libertarianism entails, the least free being. (For if God is essentially

The Intellectual Source of Sin

237

omniscient and morally perfect, there is no possibility for undetermined action of the sort the libertarian requires for freedom.) Third, intellectualism is attractive to theologians who stress the necessity of God’s grace against the Pelagian heresy. In Turretin’s austere words: It is of great importance, therefore, that the disciples of true and genuine grace should oppose themselves strenuously to these deadly errors and so build up the misery of man and the necessity of grace that the entire cause of destruction should be ascribed to man and the whole glory of salvation to God alone. (1992: t. 10, q. 1, a. 1)

Locating Leibniz Even a cursory examination of texts throughout Leibniz’s career shows that he takes the intellectualist side in this ancient debate. In the early (1672/3) Confessio Philosophi, an imaginary dialogue between a philosopher and a theologian, Leibniz takes up Medea’s words. The philosopher—Leibniz’s spokesperson—first cautions that the famous words, if not properly understood, are absurd. His gloss is that Medea did realize the injustice of the deed, but that the pleasure of revenge appeared to her a greater good ‘‘than the deed was evil’’ (A 6.3:134). The philosopher concludes: Therefore, from this one will not be able to prove that sometimes something is chosen which, in the opinion of the chooser, is worse, all things considered. Who believes the opposite, subverts all the principles of morality, and cannot even say what willing is. (A 6.3:135)6

Liberty ‘‘consists in the use of reason,’’ which is the ‘‘true root of liberty’’ (A 6.3:135). After rejecting the Molinist conception of freedom, the philosopher adds: If those passions are absent it is as impossible for us to err in thinking, to sin in willing, as it is for an attentive mind, with eyes open, not deprived by some fault, not to see, in a medium adequately lighted, a colored object in its proper distance and magnitude. (A 6.3:135)

To which, the theologian replies, ‘‘So all sin arises from ignorance,’’ and the philosopher agrees. Roughly ten years later (in the early 1680s) in some notes on Bellarmine, Leibniz writes: ‘‘It seems this can be said: it is certain and infallible that the mind determines itself to what appears to be the greatest good. No contrary example can be given where error would not have preceded sin’’ (Grua 298).7 Through the 1680s Leibniz is a card-carrying intellectualist. Leibniz retains his criticism of voluntarism and its elevation of the will’s power in his writings from the 1690s and beyond. In his ‘‘Comments’’ on Descartes’ Principles from 1692, Leibniz remarks on Descartes’ reply to the epistemological problem of evil. Of Descartes’ unyoking of the will from the scope of reason’s evaluation, Leibniz writes: ‘‘I do not admit that errors are more dependent upon the will than on the intellect. . . . The source of all errors is precisely the same in its own

238 Leibniz: Nature and Freedom

way as the reason for errors which is observed in arithmetical calculation’’ (G 4:361/ L 387–388). The Theodicy discussion of Ovid’s Medea appears to mirror that of the Confessio: But free will tends toward good, and if it meets with evil it is by accident, for the reason that this evil is concealed beneath the good and masked, as it were. These words which Ovid ascribes to Medea, Video meliora proboque, Deteriora sequor, imply that the morally good is mastered by the agreeably good which makes more impression on souls when they are disturbed by the passions. (T 154/H 220)8

Since unlike the notorious ‘‘Private Miracle’’ passage from De Necessitate et Contingentia (C19/MP 100–101), the main points made in the foregoing texts are repeated throughout Leibniz’s career in a variety of passages, this brief diachronic survey suggests Leibniz subscribes to intellectualism into and throughout his mature philosophy.9 This ascription locates Leibniz on one side of a rough taxonomy, but many of the most important details of his philosophical psychology remain unexplored. Is the intellect prior to the will (to put the question in Thomistic terms)? Must the will be a rational appetite? If Leibniz answers affirmatively, does the will do any real work, or is it merely a vestigial remnant of a Scholastic theory retained for the appearance of orthodoxy? If all sin is the result of ignorance, how are we culpable? Can the passions unseat reason, or does passion achieve its disruption only indirectly? Since it certainly seems to be an all too familiar fact that we sometimes act contrary to what we know to be right, what motivates Leibniz to adapt and defend so implausible a theory?

Leibniz’s Moral Psychology Since Leibniz’s psychology supervenes, at least in his mature philosophy, on his theory of monads, and since some of the same terms do double duty, it is with the latter that I begin. Spirits or rational minds are those monads, Leibniz’s term for basic substances, that are distinguished from the souls of nonhuman animals by being self-conscious and capable of knowing eternal and necessary truths. Minds, like all substances for Leibniz, are necessarily active. All minds act spontaneously, that is, initial states aside, each state is caused by some preceding internal state, with no intersubstantial causation between any created substances. Since bodies supervene on minds, all activity at the most fundamental metaphysical level is mental. Each substance differs in all its intrinsic, nonrelational qualities. While monads are ontologically simple, they are representationally complex. Leibniz describes the basic activity of monads as a complex endeavor (conatus) or appetition, which distinguishes the soul’s state at any moment and is the principle which drives it from one perception to another (Mon 15). Appetition is a teleological endeavor by which a soul seeks a more perfect state. For spirits or minds, this perfection is epistemological insofar as the soul seeks a more distinct understanding of the universe (Mon 60), and moral insofar as the soul seeks the good (PNG 3). This foundational theory of agency is extremely abstract and metaphysical, and bears little resemblance to the phenomenology of our lives as

The Intellectual Source of Sin 239

embodied beings interacting in a natural and social world. Leibniz frames his more detailed account of mental activity in terms of scholastic faculty psychology familiar to all and accepted by many of his early modern audience. Leibniz ‘‘psychologizes’’ his spare, metaphysical account of monadic activity by adopting the language and general framework of Thomas’s theory of human action. In this model, primacy is given the faculties of will and intellect. Each is understood teleologically and so has its proper object: the end of the will is the good, the end of the intellect is the true. Freedom depends on the proper functioning of each. According to Thomas, reason’s work is prior since it must first apprehend the good as true, ‘‘for the will cannot desire a good that is not previously apprehended by reason’’ (ST 1a2ae, q. 19, a. 13). In his comments on Aristotle’s discussion of akrasia, Thomas writes: [T]he opinion of Socrates was that knowledge can never be overcome by passion. Therefore he held every virtue to be a kind of knowledge, and every sin a kind of ignorance. In this he was somewhat right, because, since the object of the will is a good or an apparent good, it is never moved to an evil, unless that which is not good appear good in some respect to the reason; so that will would never tend to evil unless there were some ignorance or error in the reason. Hence it is written (Prov. 14:22): They err that work evil. (ST 1a2ae, q. 77, a. 2)10

Leibniz agrees: ‘‘We can only will what we think good’’ (NE 2.212.21.19/RB 180). In De veritate, Thomas writes that since ‘‘evil happens only because the evil is presented to it under the aspect of a good . . . evil is involuntary’’ (q. 24, a. 8). Leibniz concurs, writing in his ‘‘Remarks’’ on King: ‘‘Errors are therefore, absolutely speaking, never voluntary’’ (T k25/H 435). Both thinkers insist that the will can only be moved by ends recognized as good by the intellect; from this it follows that choosing evil as evil is conceptually impossible. Interpreting the ‘‘will as rational appetite’’ doctrine robustly, as Thomas and Leibniz do, yields the truth of (some form of ) intellectualism analytically. We have now pushed the original question— why does Leibniz subscribe to intellectualism—back to the question: why does Leibniz think that the will is a rational appetite? I think we can distinguish at least three important Leibnizian motivations. After chastising Descartes for taking refuge in mystery when confronted with the conflict between God’s preordination and human freedom, Leibniz writes: There should be no contradiction even in the mysteries of faith; much less so in the mysteries of nature. So if we wish to stand out as philosophers, we must once more take up the arguments. (G 4:363/L 389)

Philosophical theology raises some of the most difficult questions in all of philosophy, and this is one reason Leibniz is drawn to it. The same can be said for providing an account of human nature—Leibniz’s philosophical program is astonishingly ambitious—but Leibniz takes up the arguments, buoyed by his confidence that ‘‘the foundations are everywhere the same’’ (NE 4.17.16/RB 490). Nor is he willing to make his analysis easier by admitting mysteries; the workings of the world and the human mind are ultimately explicable. This refusal to countenance brute facts manifests itself in his psychology with the demand that cognitive and

240 Leibniz: Nature and Freedom

volitional states and acts have sufficient causes. Leibniz had several objections to the indeterminism required by the Molinists and Arminians, but perhaps his deepest criticism is that libertarianism posits a mystery at the very heart of the theory of the self and its agency. (This is particularly damning since agency is at the heart of Leibniz’s theory of substance.) The will requires a sufficient cause; since willing to will would lead to a vicious regress: [W]e do not will because we will to will, but because our nature is to will what we believe best. And this belief does not come from our will, but from the nature of things, or the state of our mind. (A 6.4:1408)

Since the end or telos of freedom is to act in accordance with the good on the basis of a distinct understanding of the true, it is fitting and natural that the will is a rational appetite. This last point introduces the second reason Leibniz maintains that the will tracks the intellect. Leibniz champions various conceptions of freedom in different contexts. At the center of these related conceptions of agency is the kind of freedom enjoyed by God, the paradigm of freedom. God is free because he acts from reasons in accordance with the genuine good. Leibniz sometimes calls this notion the freedom of understanding. Freedom is a thoroughly teleological notion—its value lies not in the (to Leibniz) impossible ability to choose, willy-nilly, anything at all; it is a perfection because it is the capacity to choose, through reasons, what we should choose. To be determined by the good is true freedom, ideal agency. As Leibniz puts it in Du Franc Arbitre: ‘‘I say that freedom to will whatever we want is an impossible thing . . . true freedom of the mind consists in recognizing and choosing the best’’ (A 6.4:1408). Leibniz believes that we have been created in the image of God, insofar as finiteness permits.11 A crucial aspect of the imitatio dei doctrine, as understood by Leibniz, is that our wills act on the apparent good, just as God’s will acts on the actual good. If the will does not pursue the best as perceived by the intellect, we would not be beings that are rational by nature. To act against our perception of the best would be a symptom of madness, not sanity and freedom. The last reason Leibniz gives for maintaining that the will is a rational appetite is that denying this truth ‘‘subverts all the principles of morality’’ (A 6.3:135). Leibniz’s point is that denying that the will tracks the representation of the good renders our choices random by severing the connection between choice and our reasons for actions. Such ‘‘liberty’’ would be blind and brutish, rendering ascriptions of moral responsibility pointless. We have now seen Leibniz’s defense of his claim that the will is moved to action by the intellect. What, however, does the will do? The will is that faculty of the mind whereby some of its ideas are converted into actions. Appetition, the source of any monadic change (Mon 15), is driven and rendered determinate through perception by being an endeavor toward a specific end that is represented as a true good (NE 2.21.36). Leibniz’s account of volition, however, must satisfy the doctrine of spontaneity and do justice to our commonsense intuition that our desires and intentions interact in complex ways with our environment. Commenting on these points in his ‘‘Remarks’’ on King, Leibniz writes:

The Intellectual Source of Sin 241 [E]ven though an active substance is determined only by itself, it does not follow that it is not moved by objects: for it is the representation of the object within it which contributes towards the determination. Now the representation does not come from without, and consequently there is complete spontaneity. Objects do not act upon intelligent substances as efficient and physical causes, but as final and moral causes. (T k20/H 427)12

Leibniz gives different definitions of the will in different places, a generous sample of which is contained in the papers and letters collected by Grua. In a work entitled On the Elements of True Piety, or the Love of God for All (henceforth True Piety), Leibniz writes: The will is a judgment about good and bad. . . . Judgment is practical reason [or thought], or reason with a striving to action. . . . If you substitute the definition of ‘judgment’ for the definition of ‘will,’ you will see that ‘Will is the thought of good and bad, linked with a striving to action.’ This agrees with those who hold that the will is the last stage of deliberation. (A 6.4:1360; emphasis in the original)

And in a text On Postulates, Leibniz writes: The will is the striving of an intelligent being. Or the will is the judgment about good and bad. . . . The two definitions coincide, because God formed the nature [of humans] to be rational, with the striving of an action consequent on that agent’s opinion of good and evil. (A 6.4:2820)

These definitions, both of which Grua dates speculatively from 1678 or 1679, are quite striking. Traditionally, the will is restricted to the appetitive side of the mental divide, with reason restricted to the side of the intellect. In True Piety and Postulates, however, Leibniz identifies the will as a judgment, which is practical reason. Will is also characterized as ‘‘the thought of good or bad’’ and ‘‘the judgment about good and bad.’’ In these texts, Leibniz identifies volition or willing with a value judgment. Willing an action is identical to judging that action to be the best available. To be sure, the will consists of thinking or judging that is a conatus agendi, a striving or impulse to act. So even in these passages, the will is a form of thought with special appetitive properties, unlike the thoughts of mere reflection, contemplation, and imagination. After the definition in True Piety, Leibniz continues: If someone truly prefers to say that will is not a judgment about good and evil, but instead is that striving to act that follows immediately from judgment, I will not argue with him, provided that he acknowledges that the striving arises from judgment. For the striving to act which does not arise from thought is brutish. (A 6.4:1360)

Leibniz is aware that his True Piety and On Postulates definitions are heterodox, and is willing to accede to traditional terminology so long as it is recognized that the will is the immediate consequence of the intellectual judgment.13 As he puts it in On Postulates, the two ways of talking ‘‘coincide’’ because God has made humans rational, which means that we, by our very nature, strive for that which we perceive to be good. I take these texts to suggest that at least into the late 1670s Leibniz was flirting with a fairly potent form of intellectualism. I know of no later texts in which Leibniz intellectualizes the will to this extent. Leibniz’s considered view is stated

242

Leibniz: Nature and Freedom

in Theodicy 311: ‘‘But whatever perception one may have of the good, the effort to act in accordance with the judgement, which in my opinion forms the essence of the will, is distinct from it’’ (emphasis added). I believe that Leibniz comes to see that a potent form of intellectualism such as that defended in his early writings faces grave difficulties in explaining cases of incontinence and in assigning culpability for sin. Call strong intellectualism the view that sin only involves cognitive error, a breakdown of practical reason attributable only to reason. Call weak intellectualism the view that while sin always involves cognitive error, it need not only involve cognitive error. Plato’s (and Aristotle’s) Socrates is a strong intellectualist. I think it is clear that by the early 1700s, when working on the New Essays, Leibniz defends weak but not strong intellectualism. Before turning to the details of his account of error and sin, it is worth comparing Leibniz’s understanding of the will to Luther’s. We have seen that Leibniz links the fact that our wills are structured so that we act on the apparent good, just as God acts on the actual good, to the imitatio dei doctrine. His interpretation of this doctrine is important and controversial. As imitators of God, we share God’s cognitive structure insofar as our finiteness permits. This means that our wills are free. Unlike thinkers such as Augustine and Luther who stress that our wills are so subverted by original sin that they are wholly enslaved, Leibniz generally emphasizes those aspects of our being that are godlike. Though by confession a Lutheran, Leibniz rejects out of hand Luther’s denial of free will.14 In Tractatio de Deo et Homine, Leibniz directly confronts the question of our postlapsarian faculties. His analysis is instructive: Now it must be seen, what of power remains in corrupted nature, i.e., what remains of the divine image. This consists in two things: in the light of the intellect, and in freedom of the will. (G 4:36)

In Principles of Nature and Grace 14, Leibniz sums up his investigation: the human mind ‘‘imitates in its realm and in the small world in which it is allowed to work, what God does in the large world’’ (AG 212).15 At our core, we are divine.

Error and Sin Thinkers such as Augustine, Luther, and Calvin have little (initial) trouble explaining sin: it is our natural state. Leibniz, on the other hand, faces tremendous difficulties—after all, how can ‘‘little gods’’ sin? While Leibniz has a detailed metaphysical account of why there are creatures that sin in the best of all possible worlds, my interest here is not with the theodicean project (as important as that is), but rather with the psychological explanation of sin embedded in his theory of human action. Like Thomas, Leibniz holds that we are constituted so as to pursue the good, in which lies our true happiness.16 In contrast to the later Kantian tradition, Leibniz holds that the demands of genuine happiness coincide with the claims of religion and morality. Truly enlightened behavior is self-interested and seeks the good; within Leibniz’s eudaimonistic ethics, the demands of the good are never antithetical to the rational pursuit of self-interest.

The Intellectual Source of Sin 243

Leibniz dedicates an immense amount of intellectual energy to working through problems concerning free will, moral psychology, and sin. Not surprisingly, he says different things concerning these topics, in different words, in different contexts. It is worth noting at the outset, however, that Leibniz is committed to providing an explanation of sin that is located squarely within his larger account of human error. Sin is one (particularly important) instance of a failure of reason, and his explanation will appeal to the same sorts of features necessary for understanding cognitive error in general. Sin, being an instance of a free action, can truck with no mystery. The fundamental commitments of his psychology yield the following groundfloor account of moral failure. If an agent acts at all, she will act in accordance with her perception of the best available action. Since the will is determined by the representation of the good, wrongdoing is explicable by the inherent confusion and inaccuracy of our perceptions or representations of the good. We can state this view as the conjunction of the following: the will cannot move towards some possible action x unless the understanding has judged x to be good; and, if the understanding has judged x to be the best, the will moves toward x. Nothing in this account, of course, entails that x is done. For Leibniz, a judgment by the intellect that some proposed action is good is action-guiding.17 Unlike Hume, Leibniz holds that it is a conceptual truth that if an agent judges x to be good, she will have some motivation to x. Likewise, it is a conceptual truth that if an agent judges x to be good, she has a reason (or reasons) that objectively justifies xing. Since the judgment that x is good provides motives and reasons, but not necessarily sufficient motives or reasons, Leibniz is, in contemporary terms, a weak motive and reason internalist.18 Error is the result of misapprehension, and misapprehension has several causes, all of them rooted in our essential embodiment. Leibniz holds that minds never lose their bodies.19 As physical creatures, we have passions, which are drives and appetites for various states of affairs that are conducive to our well-being and pleasurable to us as corporeal beings in a physical world. The passions are mental states that are essentially linked to bodies; they are helpful when prompting us to satisfy our basic needs as human animals, and dangerous when prompting us to satisfy excessive or immoral desires. When Leibniz speaks psychologically, the passions refer to the sensuous appetites, emotions, and nonrational parts of our nature. However, his metaphysical analysis is that these appetitive and aversive states are confused and indistinct perceptions or thoughts (NE 1.2.11; NE 2.21.35, 72). Since we are beings with organic bodies, we will always represent the universe from a particular point of view (C 15/P 176). Since any particular perspective within the universe of created spirits will be inevitably distorted (or ‘confused,’ in Leibniz’s preferred terminology), the view sub specie aeternitatis is forever denied us, as is omniscience. Not surprisingly, then, Leibniz often attributes our cognitive opacity to the pernicious influence of the passions, claiming that we are free in proportion to our mastery over our passions. The following passages are typical, the first from a 1689 note, De Dispositionibus Internis, the second from a 1698 letter to Molanus:

244 Leibniz: Nature and Freedom You call us free insofar as we reason justly, and slaves when we are mastered by our passions. . . . (Grua 327) [I]t must be said that we are free in such a way that we [know] hare determined to followi the perfection of our nature, which is to say reason, but that we are slaves in such a way when we follow passion. . . . (Grua 481)20

All embodied beings, including angels and the blessed, are often confused and unaware of much that is in their perceptual manifold. Indeed: [God] was acting in our interests when he brought it about that we are often ignorant and subject to confused perceptions—so that we can act the more quickly by instinct, and not be troubled by excessively distinct sensations of hosts of objects which, necessary though they are to nature’s plan, are not entirely agreeable to us. (NE I2.10.6/RB 165)

In the New Essays, Leibniz takes the possibility that there might be no upper limit to our intellectual enlightenment to be an advantage: I am inclined to believe that it [intellectual pleasure in our increasing knowledge] can increase ad infinitum, for we do not know how far our knowledge and our organs can be developed in the course of the eternity which lies before us. (NE 2.21.41/RB 194)

The passions contribute to our well-being and destruction because they are powerful; Leibniz notes in passing that intense joy has killed people (NE 2.20.8/RB 167). In his most careful psychological writings, Leibniz depicts these appetitive and aversive states, not as beliefs, but as ‘‘endeavours—or rather modifications of endeavours—which arise from beliefs or opinions and are accompanied by pleasure or displeasure’’ (NE 2.20.8/RB 167).21 If passions are not cognitive, how do they disrupt the rule of reason? Having rejected the Stoic theory of the passions as false beliefs, Leibniz cannot claim that they derail deliberate processes by posing as potential candidates for belief and assent. Being impulses with no cognitive content, they cannot influence the content of any particular perception or judgment; to claim they can would ‘‘rationalize’’ the will and lead down the garden path to strong intellectualism. Let us investigate this question by asking: what is necessary for right reasoning? For humans, properly exercising theoretical or practical reason requires near ideal empirical conditions under which to deliberate: we need time to consider the various propositions or alternative principles of action before us; we need to be rested, our basic bodily needs met, our emotions stable, our memory reliable, and so on. Violate any of these conditions, and our chances of achieving theoretical or practical rationality plunge dramatically. One way, then, that the passions disrupt rationality is by truncating theoretical completeness. Embodied beings experience powerful drives—working through Kant’s transcendental deduction is difficult; working through it when famished or exhausted is all but impossible. In the grip of strong appetitive states driven by our bodies, we often fail to consider all the relevant alternatives, cutting short the completeness of our deliberation.22 Reflecting on Leibniz’s reaction to Locke’s treatment of akrasia also provides help in understanding how the passions unseat the rule of reason, without

The Intellectual Source of Sin 245

themselves being cognitive. In the first edition of the Essay Locke embraced the traditional doctrine that the will is determined by the greatest good perceived. Confronted with the case of the drunkard who, cognizant of the greater good of sobriety, still cannot resist ‘‘a glass of Wine, or the idle chat of a soaking Club,’’ Locke saves the phenomenon and abandons the traditional theory (Essay 2.21.35). Leibniz comments: There is merit and substance in these thoughts. However, I would not want them to encourage people to believe that they should give up the old axioms that the will pursues the greatest good, and flees the greatest evil, of which it is sensible. The neglect of things that are truly good arises from the fact that, on topics and in circumstances where our senses are not much engaged, our thoughts are for the most part what we might call ‘blind’—in Latin I call them cogitationes caecae. (NE 2.21.35/RB 185)

Leibniz holds that ‘‘created minds and souls . . . cannot reason without symbols’’ (NE 2.21.73/RB 211). Some of our reasoning is purely symbolic in nature, as when we reason in mathematics and geometry, letting placeholders ‘‘stand in’’ for the mathematical or geometric objects themselves. Reasoning through symbols is often indispensable—if one had to turn one’s mind’s attention to each mathematical object in a long proof, progress would be tedious, if possible at all. We often reason analogously when thinking about God or virtue, using words as place-holders for the ideas denoted by these words. When our cognition remains only at the verbal level, such thinking is ‘‘empty of perception and sensibility’’ (NE 2.21.35/RB 186). The problem with such sensibly shorn thinking in practical reasoning is that it packs little motivational punch for anyone lacking a robust moral acculturation.23 As Leibniz puts it, ‘‘the finest moral precepts and the best prudential rules have weight only in a soul which is as sensitive to them as to what opposes them’’ (NE 2.21.35/RB 186). Since passions ‘‘are accompanied by pleasure or displeasure’’ (NE 2.20.10/RB 167), it is no wonder that in such cases, ‘‘the morally good is mastered by the agreeably good which makes more impression on souls when they are disturbed by the passions’’ (T 154/H 220). Since the true good ensures our happiness, or lasting joy, we are inclined to seek joy and avoid sorrow or pain by our very nature (NE 1.2.3). Leibniz often compares the conative drive of the soul toward the good with the force of a rock falling toward the center of the earth, which, taking the most direct route downward, shatters on protruding rocks.24 This is Leibniz’s diagnosis of our state when, under strain from appetition and passions, we mistake a temporary pleasure for a lasting one, and exchange happiness for misery. In a short work entitled On Tendencies, Leibniz writes that this tendency to real happiness: is not able to be firmly and originally assigned as the object of our tendencies, any more than a stone could be assigned the optimal path to the center, without hitting and shattering on the way down. (Grua 487)

Leibniz does not state why this world could not contain such optimizing laws— perhaps such a possible world would fall short in producing as much variety compatible with a simplicity of laws—but adds: ‘‘rising above the original

246 Leibniz: Nature and Freedom

tendencies of man or his first nature, reason and experience supervene to moderate, which is absent in the stone’’ (Grua 487).25 At times Leibniz talks of the subversion of reason by passion as instances of deception, albeit cases of self-deception: But free will tends towards good, and if it meets with evil it is by accident, for the reason that this evil is concealed beneath the good, and masked, as it were. (Grua 487)26

We can put Leibniz’s various remarks about how sin occurs in a paradigmatic case involving the passions into outline form: (1) At time t1, under fairly optimal epistemic conditions, agent S knows that xing is wrong. (2) At time t2, under the influence of the passions, S is motivated to x. (3) At time t2, S’s all-things-considered judgment concerning xing is occluded. (4) At time t3, still impassioned, S (mistakenly) judges that the good (or pleasure) involved in xing outweighs the bad. (5) At time t4, still impassioned, S chooses to x.

S is operating under ‘‘fairly optimal epistemic conditions’’ when he is in a state of (rough) emotional equilibrium, his basic bodily needs are satisfied, and he is attending to all of his knowledge concerning x. Were S to act at t1, he would avoid x, for one always move[s] towards what one finds good and away from what one finds bad, the endeavour arising immediately out of one’s awareness of those things. This definition has as a corollary the famous axiom that from will and power together, action follows; since any endeavour results in action unless it is prevented. (NE 2.21.5/RB 172–173)

It is passion that explains how an agent can go from knowing some action is wrong, in (1) of the foregoing schema, to a state in (3) where such knowledge is merely dispositional. (2) is the stage where passions wreak their damage by causing slippage between optimal judgment and action. The resultant intellectual occlusion can be caused by a number of factors: the influence of ‘‘minute insensible perceptions’’ (NE 2.21.36/RB 188), ‘‘the (often undetectable) traces left behind by earlier perceptions’’ (NE 2.21.12/RB 177), and confused thoughts (NE 2.21.35). These factors, combined with the strong desires accompanying the passions, distract or misdirect the intellect’s focus from the wrongness of x to the pleasure anticipated from xing. The potent promise of pleasure emanating from xing, together with the temporary repression or inattention to some of the morally salient features S knew at t1, results in the misjudgment of (4), which erroneously weighs present pleasure, or joy, over lasting pleasure, or happiness. Three features are striking in Leibniz’s account. First, with some modifications, it is Thomas’s explanation for sin. Second, while the passions are indicted in the mental processes that end in sin, all sin involves a mistaken practical conclusion; moral evil ineluctably involves ignorance or error. Third, Leibniz’s analysis of sin relies on the fact that decision making for finite beings takes time.

The Intellectual Source of Sin

247

This essential temporality of practical reasoning plays a key role in his analysis of how we fail to follow reason’s rule.

How to Achieve Our Conversion We are now in position to investigate whether the will, in Leibniz’s psychological model, does any real work. What are we to make of passages such as the following in the New Essays? For we can only will what we think good, and the more developed the faculty of understanding is the better are the choices of the will. And, in the other direction, in so far as a man wills vigorously, he determines his thoughts by his own choice instead of being determined and swept along by involuntary perceptions. (NE 2.21.19/RB 180)

Since Leibniz is emphatic that the will is subordinate to the intellect, his claim that willing vigorously determines thought raises questions of consistency. Leibniz’s advice on moral therapy and spiritual enlightenment in the New Essays is instructive on this question. Freedom of the understanding is the heart of Leibniz’s positive analysis of freedom. We have been created to be responsive to reasons; we are beings with judgment-sensitive attitudes and responses. When we act in accordance with (true) reason, we are most free. Freedom consists in the mastery of (true) reason; by acting on distinct knowledge we ‘‘achieve our own conversion,’’ whereas by acting primarily on the basis of confused perceptions and passions we secure our own perversion (NE 2.21.23/RB 182; T 310–311). Spirits not only mirror the world, but mirror God in being self-conscious and capable of understanding necessary and eternal truths. As spirits, we are most true to our natures when the self-mastery of reason is operative. Self-mastery involves tightening the connection between (true) judgment and action. Leibniz’s counsel on minimizing the slippage between (true) judgment and action focuses on two fronts: increasing the vividness and vivacity of ideas of the good, and moral training and practice (which may or may not involve the sensory embellishment of key ideas). This last point falls within the province of the will. Leibniz maintains that many ostensible cases of incontinence involve people who lack real or robust knowledge of the good (NE 2.21.39). Such people have doubts, either because they never understood the reasons or proofs for true religion or immortality, or because they no longer remember (or misremember) such proofs. Since belief in morality, for such people, lacks not only certainty but even probability, their actions fail as evidence for the existence of weakness of will. Others ‘‘have the idea of an absent good or evil, but only very faintly, so it is no wonder that it has almost no influence on them’’ (NE 2.21.35/RB 186). The faint (or blind) idea of x is never sufficient for knowledge of x; when employing only blind thoughts, ‘‘if we prefer the worse it is because we have a sense of the good it contains but not of the evil it contains or of the good which exists on the opposite side’’ (NE 2.21.35/RB 186). While blind thoughts or symbols are often useful in

248 Leibniz: Nature and Freedom

reasoning in formal domains, the results of such symbolic calculation only constitute knowledge when the reasoner thoroughly understands the denotations of the signs, which Leibniz denies in cases of purely blind notions of God and the good. When our ideas of God, moral principles, and evil are accurate (as far as they go) but blind, Leibniz recommends a two-stage approach: reviewing the validity of the reasoning or proof and then (when possible) fleshing out the principle or truth in vivid sensory detail. The first exercise guards against mistakes and lapses in memory, a malady Leibniz thinks is more widespread than is commonly supposed, and in some sense helps ‘‘anchor’’ the truth for us. The second exercise magnifies the motivational pull of these truths—the idea being to get the passions working for our salvation and (true) happiness. In this connection Leibniz tells the tale of the Prince of Assassins drugging his soldiers and transporting them to awaken in a beautiful (if sham) Muslim paradise so they would not fear death in carrying out their dangerous assignments (NE 2.12.37).27 Leibniz admits that some central ideas of religion and morality are resistant to sensory vivification—the idea of God being a case in point—but takes hope in the fact that people can be strongly motivated by goals that lack significant sensory content. When the ambitious or miserly acquire honours or riches they get but feeble pleasures from them and often almost none at all, because their thoughts are always on the next move. Nature the Architect did very well, it seems to me, making men so sensitive to things which have so little effect on the senses. If we were incapable of becoming ambitious or miserly, it would be hard for us—in the present state of human nature—to become virtuous and rational enough to work towards our own perfection in face of the present pleasures which distract us from it. (NE 2.21.65/RB 205; see also NE 2.21.37)

The second approach to tightening the connection between (true) judgment and action focuses squarely on moral training. In the Preface to the New Essays Leibniz claims that in general his system is closer to Plato’s, while Locke’s is closer to Aristotle’s, ‘‘although each of us parts company at many points from the teaching of both of these ancient writers’’ (RB 47–48). This is one such departure. In the Phaedo, Plato writes that ‘‘the soul of the philosopher most disdains the body, flees from it and seeks to be by itself ’’ (Phaedo 65d). Plato views the body and its senses as a prison that, by its very nature, impedes our pursuit of the true and the good. Since full philosophical success appears to require death, we are enjoined to disdain the body and practice for death. Unlike the Platonist, Leibniz acknowledges the value of the passions and thus the legitimacy of satisfying, in temperance, bodily appetites (G 7:86/L 426). Leibniz stresses the importance of moral education. Ideally, the training and inculcation toward virtue begins at a young age (Nicomachean Ethics 1103b23–26; NE 2.21.35). Like Aristotle, Leibniz stresses understanding why moral (and in the latter’s case, religious) truths are true, at least in cases of the morally immature. Understanding the reasons helps ground behavior. Like Aristotle, Leibniz recognizes that ethical education consists of both cognitive and noncognitive elements, and thus that the development of practical wisdom and virtue requires training our

The Intellectual Source of Sin 249

nonrational dispositions. Like Aristotle, Leibniz preaches proper habituation—for habit molds character—exemplified in the virtuous person, for whom it is second nature for the passions and desires to follow the rule of reason (Nicomachean Ethics 1102b13–28, 1105a27–b1; NE 2.21.35, 67). We become virtuous by doing virtuous actions (Nicomachean Ethics 1103a19–25). Habit is what aligns our appetitive element, the will, whose object is too often present pleasure, with the proper object of the will and intellect, the true good, leading to happiness (Nicomachean Ethics 1099a13–15, 1156b16–17; NE 2.21.67). Practice and habit have cognitive and motivational import; acting well, in the fullness of time, fully integrates conative and rational elements, allowing spirits to love and imitate the author of all good, as they should, finding pleasure in the consideration of his perfections according to the nature of genuinely pure love, which takes pleasure in the happiness of the beloved. (Mon 90/AG 224)

The lover of God and the good has achieved her end: perfect harmony between her affective states and reason. All of her pleasures and emotions center on the good; having achieved virtue and practical wisdom, she can no longer sin.28 This program of ethical education and moral therapy is predicated on the habit of exercised will and reason. In his critical remarks on Descartes’ Principles, Leibniz comments that there is one thing we recognize to be within the power of will—to command attention and exertion. And so the will, though it does not bring about any opinion in us, can nevertheless contribute to it obliquely. Thus it happens that men often finally come to believe what they will to be true, after having accustomed the mind to attend most strongly to the things which they favor. (G 4:356–357/ L 384–385)

Several pages later, he expands on the mission of the will ‘‘to prepare the mind well in advance’’ (G 4:362/L 388). This involves attending to experience, ours and others,’ so as to steer away from circumstances conducive to danger, and ‘‘training the mind to follow a definite series and method when thinking, so that later the required attitude offers itself spontaneously, as it were’’ (G 4:362/L 388). The strategy of dramatic rehearsal, in full rational equilibrium, is Leibniz’s principal therapeutic practice for shaping the kind of person who thinks and acts from a settled disposition of reason and virtue. Leibniz holds ‘‘there is as much connection or determination amongst thoughts as amongst motions’’ (NE 2.21.13/ RB 178); further, we cannot ‘‘will to will’’: rather, ‘‘we can only will what we think good’’ (NE 2.21.19/RB 180). We know from experience under which circumstances we are especially susceptible to the perturbations of the passions and sensual enticement. Our only chance, under these conditions, is for ‘‘the mind [to] be prepared in advance, and be already stepping from thought to thought, so that it will not be too much held up when the path becomes slippery and treacherous’’ (NE 2.21.47/RB 195–196). Such a practitioner, because of her training and fixed habits, is able to rise above the cacophony of emotional turbulence, suspend her appetitive drives, and ask where reason guides. Reflecting upon the competing practical judgments recognized by the intellect, she is able to evaluate which

250 Leibniz: Nature and Freedom

choice involves what reason asserts to be a true good. It is through engaging in metajudgment that we become masters of ourselves, and can bring it about that we have certain thoughts and that when the time comes we shall will according to our present preference and according to reason’s decrees. (NE 2.21.47/RB 196)

In the New Essays, Leibniz never veers from insisting that we can transcend our vulnerability to the power and irrational aspects of the passions if we have skillfully trained our affective and conative elements—the power to suspend strong desires and reconsider practical conclusions ‘‘always takes place in determinate ways and never without reasons—never by the fictitious principle of total indifference or equilibrium’’ (NE 2.21.47/RB 198). The reflexive power of reason, coupled with sure habits of virtue, result in the self-rule of rational autonomy.

Culpability Weak intellectualists deny what we might term strong or synchronic akrasia: acting at time t1 contrary to what an agent judges to be right at time t1. However, weak intellectualists do countenance weak or diachronic akrasia: acting at time t2 contrary to what an agent judged to be right at time t1. I take Aristotle, Thomas, and Leibniz to be exemplars of this tradition. Each is sensitive to the temporal dimension of decision making, and explains conflicts between an agent’s action and values, judgments and knowledge by distinguishing types of knowledge an agent can attend to and neglect in a period of deliberation. Thus while Medea knew, when in a period of rational and emotional equilibrium, that murdering her children was wrong, under the influence of the passions she was no longer attending to that piece of knowledge; at the fatal moment, it was merely dispositional, since she was focusing on the good of revenge and retribution. Serious problems remain. First, ignorance often excuses; total ignorance usually exculpates. Second, in the sense of ‘‘freedom of will’’ that I have called the heart of Leibniz’s account of freedom, one is free in inverse proportion to the ‘‘imposition or constraint . . . which the passions impose’’ (NE 2.21.7/RB 175). Yet this entails that we are least free when we sin. Philosophically and theologically, this doctrine leads to the rocks. Leibniz has several replies we should distinguish in answering the first problem. First, nondispositional ignorance which we possess and cannot reasonably be expected to lack, is exculpatory. Leibniz stresses in several passages the ‘‘weakness of the human mind [which] arises from a lack of attention and memory and cannot be completely overcome’’ (G 4:356/L 384; see also G 4:362/L 388). Nor can we ever escape what Leibniz calls the ‘‘involuntary thoughts’’ that come from our senses (NE 2.21.12/RB 177). Second, Medea’s ignorance of the wrongness of murder, being merely dispositional, is culpable. She failed both to call forth knowledge she possessed dispositionally, and to apply that knowledge to the case at hand. Since by her past actions she created her intemperate character, she is blameworthy for the actions that follow

The Intellectual Source of Sin

251

from it. The reason for Medea’s action lies in her history; she acted in the past in ways that brought her to respond as she does now, and therein lies her responsibility. Leibniz’s whole program of disciplined self-transformation, in which our appetites and passions are trained to harmonize and seek the object of reason, depends upon the truth of determinism. As Leibniz puts it, ‘‘we achieve our own conversion or perversion depending on what our experience has been’’ (NE 2.21.23/RB 182). Leibniz’s theory of freedom does entail that we are least free (in terms of the freedom of will under discussion) when we sin. However, properly understood, this is not surprising. To say that we are least free when we sin is simply to say that we are irrational when we sin. Since acting freely for Leibniz is acting rationally (on the basis of the perception of the good), a failure of rationality is a failure to act in as free a manner as we can. Because freedom of this type does admit of degrees, to say we are less free does not entail that we are not responsible for our lack of selfmastery and rationality. The people we have made ourselves in the past determine whether we remain in control, and act with reason, or become the passive objects of the passions and sensory appetites. Leibniz’s approach couldn’t be more different from that of his libertarian adversaries. Their insistence on the privileges of an unmoved mover severs the dependence of the will on the intellect. Leibniz turns this objection on its head—if our intellectual perception was always pellucid, and our will was determined by the intellect’s judgment, we would always be free. I have argued that Leibniz is a thinker who follows Thomas in maintaining that the will pursues what the intellect judges to be good. Moral evil always involves a cognitive error. Within this tradition, for any goods x and y, we cannot both believe that x is better than y at time t1, and choose y at t1. However, for any goods x and y, we can both believe that x is better than y at t1, and choose y at t2. In the latter case, however, we actually believe that y is better than x at t2. Leibniz was well aware that we sometimes fail to carry out our intended actions. The booming, buzzing phenomenal world, along with our own appetites, impinge all too frequently on our theoretical and practical deliberations. In Theodicy 311, Leibniz puts the point as follows: As for the parallel between the relation of the understanding to the true and that of the will to the good, one must know that a clear and distinct perception of a truth contains within it actually the affirmation of this truth: thus the understanding is necessitated in that direction. But whatever perception one may have of the good, the effort to act in accordance with the judgement, which in my opinion forms the essence of the will, is distinct from it. Thus, since there is need of time to raise this effort to its climax, it may be suspended, and even changed, by a new perception or inclination which passes athwart it, which diverts the mind from it, and which even causes it sometimes to make a contrary judgment. Hence it comes that our soul has many means of resisting the truth which it knows, and that the passage from mind to heart is so long. (H 314)29

When the suspended action was in accordance with the perception of the actual best, it is a tragedy that ‘‘the passage from mind to heart is so long’’ (H 314).30 Conversely, suspending when the intended action was a misperception of the actual best is all for the good.

252 Leibniz: Nature and Freedom

We can now appreciate Leibniz’s claim that, even though moral error ineluctably involves cognitive failure of some sort, [i]n truth, true happiness requires less knowledge but greater strength and goodness of will, so that the dullest idiot can achieve it just as easily as can the cleverest and most educated person. (NE 2.21.66/RB 207)

Leibniz’s point is that true happiness requires far less intellectual facility than is involved in the pure or applied sciences. He is calling attention to the fact that all the intellect in the world will not help the brilliant person with poor habituation. Often the root of one’s moral lapse lies more in a failure to attend to knowledge that one already knows, keeping it steadily in view right through action.

Summary While much remains unstudied, this propaedeutic suggests a Leibniz who conceives of human nature as both godlike and weak, and who implicates the intellect and character in his diagnosis of moral failing. This Leibniz stresses our essential embodiment to an extent not usually recognized, at one point wondering ‘‘how far our knowledge and our organs can be developed in the course of the eternity which lies before us’’ (NE 2.21.41/RB 198). This Leibniz sees himself as working within and modifying an Aristotelian tradition of philosophical psychology that runs through Thomas. In the texts under consideration, Leibniz attempts to provide an account of psychological conflict that does not involve disunifying the mind, and that connects in deep and important ways to core doctrines in his metaphysics. Because Leibniz’s account of moral error is at the intersection of so many fundamental problems and conceptual pressures, it occupies a point of maximal stress within his attempt to articulate our nature in an intelligible world. As such it represents fertile ground, both for understanding his motivations for some of his most basic assumptions, and for assessing the overall coherence of his psychological, theological, and metaphysical principles.

Notes Dan Kervick provided rich philosophical engagement on a very preliminary version of this essay that was presented at the University of Massachusetts Homecoming Conference of October 1999. Comments from colleagues at Iowa State University, in particular those from Joseph Kupfer and William Robinson, improved an intermediate version. The essay owes the most to Donald Rutherford, who first commented on it at the Central Division of the American Philosophical Association in April 2000. Then and later he has been a constant source of incisive criticism and encouragement. My thinking about Leibniz on the metaphysics of free will and moral psychology has been most affected by Robert Sleigh, who first awakened me to the richness of Leibniz’s writings in these areas. 1. Ovid 1958: 7.20–21. 2. See Protagoras 345e, 355, 358c; Meno 77e. 3. All translations are from Plato 1997. 4. Van Limborch is paraphrasing Molina’s famous definition of freedom (1953: 14). 5. Citations to Turretin’s Institutio are to topic, question, and article. 6. My thanks to Robert Sleigh, whose translations of the Confessio Philosophi and Du Franc Arbitre I use in this essay.

The Intellectual Source of Sin

253

7. For other such claims, see A 6.4:1376, 2552. 8. Leibniz also quotes Medea’s words in T 297. 9. For some of the recent literature on the Private Miracle passage, see Paull 1992; Borst 1992; Murray 1995; Davidson 1998; Sleigh, Chappell, and Della Rocca 1998; Sleigh 1999. 10. See also De malo, q. 16, c. 5. 11. See, for example, Grua 299–300. 12. See also T 291. 13. In On the Affections, a text from the same period, Leibniz provides a much more orthodox definition, which is the one he retains in later writings: ‘‘Willing is the striving which follows from the judgment of good and bad that is present in the agent’’ (Grua 512; emphasis added). 14. Leibniz attributes Luther’s mistaken denial of free will to the poor definitions of freedom taught in the schools of his day (Grua 227). In the ‘‘Preliminary Dissertation’’ to the Theodicy Leibniz provides a gloss on Luther’s and Melanchthon’s antagonism to philosophy (T pd12). 15. See Mon 83. 16. See T 309. 17. See NE 2.21.5; T 336; LC 5.8. Hume forcibly argues for the opposing view: ‘‘reason is perfectly inert, and can never either prevent or produce any action or affection’’ (Treatise 294). 18. For a guide to this terminology, see Brink 1989: 39. By contrast, God has arranged the apprehension of the good by the blessed to be necessarily efficacious. Leibniz also maintains that should any of us in via have ‘‘a very clear recognition of the best’’ such perspicacity is sufficient for action (T 310/ H 313). When discussing these matters, Leibniz attempts to balance his claim that we are little gods while maintaining the essential distance between God and (embodied) creatures, since all creatures have ‘‘some confused perception mingled with distinct knowledge’’ (T 310/H 314). 19. See NE Preface/RB 58; NE 2.21.73; T pd10; T 124. I am not taking a stand on how to interpret this claim within the idealism of Leibniz’s mature metaphysics. In metaphysical rigor, it is false to call finite minds ‘‘embodied beings’’; rather, we are minds that never lack bodies, where this last phrase is understood to entail some ontological reduction. In what follows, I refer to humans as ‘‘embodied beings’’ (and like cognate descriptions), exchanging ontological precision for ease of expression. 20. Text within brackets was deleted by Leibniz, while text within carets was added. For similar claims regarding the slavery of the passions, see G 4:362/L 388; NE 2.21.8; T 289. 21. See also T 310, where he writes that ‘‘confused perceptions . . . beget passions’’ (H 314). 22. See, for example, NE 2.21.32, 62. 23. See, for example, T 311. 24. See Grua 15, 484; NE 2.21.36. 25. For a compelling guide to interpreting the criteria of variety of phenomena and economy of laws, see Rutherford 1995a: 22–45. 26. See also NE 2.21.22. 27. Public health agencies in the United States are now spending millions on images designed to add flesh and bones to the statistics regarding the danger of smoking. For some of the psychological literature on the ‘‘vividness’’ effect, see Taylor and Thompson 1982: 155–181, and Eagly and Chaiken 1993: 279–280. I am indebted to Peter Vranas for these references. 28. Compare with Nicomachean Ethics 1152a1–8. 29. For other texts in which Leibniz talks about the suspension of actions, see Grua 384, 324; NE 2.21.35, 47; T 64; LC 5.11. 30. Consider the well-known letter to Coste of December 19, 1707: ‘‘God or a perfectly wise person will always choose the best that they know of, and if one side is better than the other, they would choose neither one nor the other. The passions often take the place of reason in other intelligent substances, and we can always assert, with respect to the will in general, that choice follows the greatest inclination (by which I understand both passions and reasons, true or apparent). . . . Thus when God chooses, it is by reason of the best, and when a person chooses, it is the option that struck him the most. If he chooses what he sees as less useful and pleasant in some respects, perhaps it becomes more agreeable to him through a whim, or contrariness, or for similar reasons which belong to a depraved taste; these are determining reasons, even though they are not conclusive reasons. And we will never find a contrary reason’’ (G 4:401–402/AG 195).

This page intentionally left blank

Bibliography

Adams, Robert M. 1994. Leibniz: Determinist, Theist, Idealist. New York: Oxford University Press. Alexander, Archibald. 1898. Theories of the Will in the History of Philosophy. New York: Scribner. Angeles, Peter A. 1981. Dictionary of Philosophy. New York: Harper and Row. Antognazza, Maria Rosa. 1999. Trinita` e Incarnazione: Il rapporto tra filosofia e teologia rivelata nel pensiero di Leibniz. Milan: Vita e Pensiero. Arendt, Hannah. 1978. The Life of the Mind: Willing. Orlando, Fl.: Harcourt Brace. Aristotle. 1984. The Complete Works of Aristotle. Ed. Jonathan Barnes. Princeton: Princeton University Press. Arthur, Richard. 1989. ‘‘Russell’s Conundrum: On the Relation of Leibniz’s Monads to the Continuum.’’ In An Intimate Relation: Studies in the History and Philosophy of Science, ed. J. R. Brown and J. Mittelstraß. Dordrecht: Kluwer, 171–202. —. 1998. ‘‘Infinite Aggregates and Phenomenal Wholes.’’ Leibniz Society Review 8: 25–45. —. 1999a. ‘‘Infinite Number and the World Soul: In Defence of Carlin and Leibniz.’’ Leibniz Review 9: 105–116. —. 1999b. ‘‘The Transcendentality of p and Leibniz’s Philosophy of Mathematics.’’ Proceedings of the Canadian Society for History and Philosophy of Mathematics 12: 13–19. —. 2001. ‘‘Leibniz on Infinite Number, Infinite Wholes, and the Whole World: A Reply to Gregory Brown.’’ Leibniz Review 11: 103–116. Bassler, O. Bradley. 1998. ‘‘Leibniz on the Indefinite as Infinite.’’ Review of Metaphysics 51: 849–874. Beeley, Philip. 1999. ‘‘Mathematics and Nature in Leibniz’s Early Philosophy.’’ In The Young Leibniz and His Philosophy (1646–1676), ed. S. Brown. Dordrecht: Kluwer, 123–145. —. 2002. ‘‘Leibniz on Wachter’s Elucidarius cabalisticus: A Critical Edition of the SoCalled ‘Re´futation de Spinoza.’ ’’ Leibniz Review 12: i–viii, 1–14. Belaval, Yvon. 1962. Leibniz: Initiation a` sa philosophie. Paris: Vrin. Bellarmino, Roberto. 1873. Opera Omnia. Ed. Justinus Fevre. Paris: Minerva.

255

256 Bibliography Bennett, Jonathan. 1976. Linguistic Behaviour. Cambridge: Cambridge University Press. Reprint: Indianapolis: Hackett, 1989. —. 1991a. ‘‘Analysis without Noise.’’ In Mind and Common Sense, ed. Radu Bogdan. Cambridge: Cambridge University Press, 15–36. —. 1991b. ‘‘Folk-Psychological Explanations.’’ In The Future of Folk Psychology, ed. John D. Greenwood. Cambridge: Cambridge University Press, 176–195. —. 2001. Learning from Six Philosophers. 2 vols. Oxford: Oxford University Press. Bobro, Marc, and Kenneth Clatterbaugh. 1996. ‘‘Unpacking the Monad: Leibniz’s Theory of Causality.’’ Monist 79: 408–425. Bode´u¨s, Richard. 1993. Leibniz-Thomasius, Correspondence, 1663–1672. Paris: Vrin. Bolton, Martha. 2004. ‘‘Leibniz to Arnauld: Platonic and Aristotelian Themes on Matter and Corporeal Substance.’’ In Lodge 2004. Borst, Clive. 1992. ‘‘Leibniz and the Compatibilist Account of Freedom.’’ Studia Leibnitiana 24: 49–58. Bourke, Vernon J. 1964. Will in Western Thought. New York: Sheed and Ward. Brandom, Robert B. 1981. ‘‘Leibniz and Degrees of Perception.’’ Journal of the History of Philosophy 19: 447–479. Brink, David O. 1989. Moral Realism and the Foundations of Ethics. Cambridge: Cambridge University Press. Brown, Gregory. 1998. ‘‘Who’s Afraid of Infinite Numbers? Leibniz and the World Soul.’’ Leibniz Society Review 8: 113–125. —. 2000. ‘‘Leibniz on Wholes, Unities and Infinite Number.’’ Leibniz Review 10: 21–51. Carlin, Laurence. 1997. ‘‘Infinite Accumulations and Pantheistic Implications: Leibniz and the ‘Anima Mundi.’ ’’ Leibniz Society Review 7: 1–24. Carriero, John. 1993. ‘‘Leibniz on Infinite Resolution and Intra-Mundane Contingency.’’ Studia Leibnitiana 25: 1–26. Cassirer, Ernst. 1902. Leibniz’ System in seinen wissenschaftlichen Grundlagen. Darmstadt: Wissenschaftliche Buchgesellschaft. Reprint, Hildesheim: Georg Olms, 1962. Castan˜eda, Hector-Neri. 1990. ‘‘Leibniz’s View of Contingent Truth in the Late 1680’s.’’ In Heinekamp, Lenzen, and Schneider 1990: 255–272. Clarke, Samuel. 1998. A Demonstration of the Being and Attributes of God. Ed. Ezio Vailati. Cambridge: Cambridge University Press. Couturat, Louis. 1901. La Logique de Leibniz, d’apre`s des documents ine´dits. Paris: Felix Alcan. Reprint, Hildesheim: Georg Olms, 1961. —. 1902. ‘‘Sur la Me´taphysique de Leibniz.’’ Revue de me´taphysique et de morale 10: 1–25. Reprinted in Frankfurt 1972. Cover, J. A., and John O’Leary-Hawthorne. 1999. Substance and Individuation in Leibniz. Cambridge: Cambridge University Press. Cover, J. A., and John Hawthorne. 2000a. ‘‘Infinite Analysis and the Problem of the Lucky Proof.’’ Studia Leibnitiana 32: 151–165. —. 2000b. ‘‘Leibnizian Modality Again: Reply to Murray.’’ Leibniz Review 10: 87–103. Cudworth, Ralph. 1996. A Treatise Concerning Eternal and Immutable Morality. Ed. Sarah Hutton. Cambridge: Cambridge University Press. Davidson, Jack. 1996. ‘‘Untying the Knot: Leibniz on Divine Foreknowledge of Future Contingents.’’ History of Philosophy Quarterly 12: 89–116. —. 1998. ‘‘Imitators of God: Leibniz on Human Freedom.’’ Journal of the History of Philosophy 36: 387–411. Deleuze, Gilles. 1993. The Fold: Leibniz and the Baroque. Trans. Tom Conley. Minneapolis: University of Minnesota Press.

Bibliography

257

Dennett, Daniel C. 1989. ‘‘True Believers: The Intentional Strategy and Why It Works.’’ In The Intentional Stance. Cambridge, Mass.: MIT Press, 13–35. Dola, Ludovicus a. 1634. De Modo Conjuntionis Concursuum Dei et Creaturum. Lyon. Eagly, Alice, and Shelly Chaiken. 1993. The Psychology of Attitudes. Fort Worth, Tex.: Harcourt Brace. Earman, John. 1989. World Enough and Space-Time. Cambridge: Cambridge University Press. Feynman, Richard, Robert B. Leighton, and Matthew Sands. 1963. The Feynman Lectures on Physics: Mainly Mechanics, Radiation, and Heat. Reading, Mass.: Addison-Wesley. Fichant, Michel. 1974. ‘‘La ‘re´forme’ leibnizienne de la dymanique, d’apre`s des textes ine´dits.’’ Studia Leibnitianna Supplementa 13: 195–214. Ficino, Marsilio. 1559. Theologia Platonica, de immortalitate animorum. Paris. Reprint, Hildesheim: Georg Olms, 1995. Frankfurt, Harry (ed.). 1972. Leibniz: A Collection of Critical Essays. Notre Dame, Ind.: University of Notre Dame Press. Freddoso, Alfred J. 1986. ‘‘The Necessity of Nature.’’ Midwest Studies in Philosophy 11: 215– 257. Friedmann, Georges. 1962. Leibniz et Spinoza. Rev. ed. Paris: Gallimard. Futch, Michael J. 2001. ‘‘Time and Causation in Leibniz’s Metaphysics.’’ Ph.D. diss., Emory University. Garber, Daniel. 1982. ‘‘Motion and Metaphysics in the Young Leibniz.’’ In Hooker 1982: 160–184. —. 1985. ‘‘Leibniz and the Foundations of Physics: The Middle Years.’’ In The Natural Philosophy of Leibniz, ed. Kathleen Okruhlik and James Brown. Dordrecht: Reidel, 27–130. —. 1992a. Descartes’s Metaphysical Physics. Chicago: University of Chicago Press. —. 1992b. ‘‘Review of Catherine Wilson, Leibniz’s Metaphysics: A Historical and Comparative Study, and Robert C. Sleigh, Jr., Leibniz and Arnauld: A Commentary on their Correspondence.’’ Journal of Philosophy 89: 151–165. —. 1995. ‘‘Leibniz: Physics and Philosophy.’’ In Jolley 1995: 270–352. Garber, Daniel, and Michael Ayers (eds.). 1998. The Cambridge History of SeventeenthCentury Philosophy. 2 vols. Cambridge: Cambridge University Press. Garrett, Don. 1997. Cognition and Commitment in Hume’s Philosophy. Oxford: Oxford University Press. Gentz, William H. (ed.). 1986. The Dictionary of Bible and Religion. Nashville: Abingdon Press. Gibbard, Allan. 1990. Wise Choices, Apt Feelings. Cambridge, Mass.: Harvard University Press. Ginet, Carl. 1995. ‘‘Reasons Explanation of Action: An Incompatibilist Account.’’ In O’Connor 1995: 69–94. Grosholz, Emily. 1991. Cartesian Method and the Problem of Reduction. Oxford: Clarendon Press. Grua, Gaston. 1953. Jurisprudence Universelle et The´odice´e selon Leibniz. Paris: Press Universitaires de France. Reprint, New York: Garland, 1985. —. 1956. La Justice humaine selon Leibniz. Paris: Presses Universitaires de France. Reprint, New York: Garland, 1985. Hacking, Ian. 1974. ‘‘Infinite Analysis.’’ Studia Leibnitiana 6: 126–130. Hardin, C. L. 1988. Color for Philosophers: Unweaving the Rainbow. Indianapolis: Hackett. Hare, R. M. 1952. The Language of Morals. Oxford: Oxford University Press. —. 1963. Freedom and Reason. Oxford: Oxford University Press.

258 Bibliography —. 1981. Moral Thinking: Its Levels, Method and Point. Oxford: Oxford University Press. Hartz, Glenn. 1992. ‘‘Leibniz’s Phenomenalisms.’’ Philosophical Review 101: 511–549. —. 1998. ‘‘Why Corporeal Substances Keep Popping Up in Leibniz’s Later Philosophy.’’ British Journal for the History of Philosophy 6: 193–207. Heinekamp, Albert. 1969. Das Problem des Guten bei Leibniz. Bonn: H. Bouvier. Heinekamp, Albert, Wolfgang Lenzen, and Martin Schneider (eds.). 1990. Mathesis rationis: Feschrift fu¨r Heinrich Schepers. Mu¨nster: Nodus. Hobbes, Thomas. 1839. The English Works of Thomas Hobbes of Malmesbury. 11 vols. Ed. William Molesworth. London: John Bohn. Reprint, Aalen: Scientia Verlag, 1966. Hobbes, Thomas. 1994. Leviathan, with Selected Variants from the Latin Edition of 1668. Ed. Edwin Curley. Indianapolis: Hackett. Hobbes, Thomas, and John Bramhall. 1999. Hobbes and Bramhall on Liberty and Necessity. Ed. Vere Chappell. Cambridge: Cambridge University Press. Hofmann, Joseph E. 1974. Leibniz in Paris, 1672–1676: His Growth to Mathematical Maturity. Cambridge: Cambridge University Press. Hooker, Michael (ed.). 1982. Leibniz: Critical and Interpretive Essays. Minneapolis: University of Minnesota Press. Ishiguro, Hide´. 1990. Leibniz’s Philosophy of Logic and Language. 2nd ed. Cambridge: Cambridge University Press. Izquierdo, Sebastian. 1670. Opus theologicum iuxta atque philosophicum de Deo uno. Rome. Jalabert, Jacques. 1981. ‘‘Leibniz et Malebranche.’’ Les e´tudes philosophiques, 279–292. Jolley, Nicholas (ed.). 1995. The Cambridge Companion to Leibniz. Cambridge: Cambridge University Press. —. 1998. ‘‘Causality and Creation in Leibniz.’’ Monist 81: 591–611. Kent, Bonnie. 1995. Virtues of the Will. Washington, D.C.: Catholic University Press. Knebel, Sven. 1991a. ‘‘Necessitas moralis ad optimum.’’ Studia Leibnitiana 23: 3–24. —. 1991b. ‘‘Scientia Media.’’ Archiv fu¨r Begriffsgeschichte 34: 262–294. —. 1992. ‘‘Necessitas Moralis ad Optimum (III): Naturgesetz und Induktionsproblem in der Jesuitenscholastik wa¨hrend des zweiten Drittels des 17. Jahrhunderts.’’ Studia Leibnitiana 24: 182–215. Kraye, Jill (ed.). 1997. Cambridge Translations of Renaissance Philosophical Texts. Vol. 1. Cambridge: Cambridge University Press. Kulstad, Mark. 1977. ‘‘Leibniz’s Conception of Expression.’’ Studia Leibnitiana 9: 55–76. —. 1982. ‘‘Some Difficulties in Leibniz’s Definition of Perception.’’ In Hooker 1982: 65–78. —. 1990. ‘‘Appetition in the Philosophy of Leibniz.’’ In Heinekamp, Lenzen, and Schneider 1990: 133–151. —. 1994. ‘‘Did Leibniz Incline towards Monistic Pantheism in 1676?’’ In Leibniz und Europa: VI. Internationaler Leibniz-Kongress (conference proceedings). Hannover, teil 1, 424–428. —. 1996. ‘‘Spinoza’s Demonstration of Monism: A New Line of Defense.’’ History of Philosophy Quarterly 13: 299–316. —. 1999. ‘‘Leibniz’s De Summa Rerum: The Origin of the Variety of Things, in Connection with the Spinoza-Tschirnhaus Correspondence.’’ Studia Leibnitiana Supplementa 34: 69–85. —. 2000. ‘‘Pantheism, Harmony, Unity and Multiplicity: A Radical Suggestion of Leibniz’s De Summa Rerum.’’ In Lamarra and Palaia 2000: 97–106. —. 2001. ‘‘Leibniz’s Early Argument That All Things Are One in Relation to Descartes’s Notions of Real and Modal Distinction.’’ Nihil Sine Ratione: VII. Internationaler Leibniz-Kongress (conference proceedings). Berlin, teil 2, 663–668.

Bibliography 259 Lamarra, Antonio, and Roberto Palaia (eds.). 2000. Unita` e molteplicita` nel pensiero filosofico e scientifico di Leibniz. Florence: Olschki. Levey, Samuel. 1998. ‘‘Leibniz on Mathematics and the Actually Infinite Division of Matter.’’ Philosophical Review 107: 49–96. —.2003. ‘‘The Interval of Motion in Leibniz’s Pacidius Philalethi.’’ Nouˆs 37: 371–416. Levine, Michael P. 1990. ‘‘Divine Unity and Superfluous Synonymity.’’ Journal of Speculative Philosophy 4: 211–236. —. 1992. ‘‘Monism and Pantheism.’’ Southern Journal of Philosophy 30: 95–98. —. 1994. Pantheism: A Non-Theistic Concept of Deity. London: Routledge. Locke, John. 1982. The Correspondence of John Locke. 8 vols. Ed. E. S. De Beer. Oxford: Clarendon Press. Lodge, Paul (ed.). 2004. Leibniz and His Correspondents. New York: Cambridge University Press. McGuire, J. E. 1976. ‘‘‘Labyrinthus Continui’: Leibniz on Substance, Activity, and Matter.’’ In Motion and Time, Space and Matter, ed. Peter K. Machamer and Robert G. Turnbull. Toronto: University of Toronto Press, 290–326. McCracken, Charles J. 1983. Malebranche and British Philosophy. Oxford: Oxford University Press. Mancosu, Paolo. 1996. Philosophy of Mathematics and Mathematical Practice in the Seventeenth Century. Oxford: Oxford University Press. Mandelbrot, Benoit. 1982. The Fractal Geometry of Nature. New York: Freeman. Mates, Benson. 1986. The Philosophy of Leibniz: Metaphysics and Language. New York: Oxford University Press. Melanchthon, Philip. 1969. Loci Communes. In Melanchthon and Bucer. Ed. Wilhelm Pauck. Trans. Lowell J. Satre. Library of Christian Classics, vol. 19. Philadelphia: Westminister Press. Mercer, Christia. 1999. ‘‘Leibniz and Spinoza on Substance and Mode.’’ In The Rationalists, ed. D. Pereboom. Lanham, Md.: Rowman and Littlefield, 273–300. —. 2000. ‘‘God as Both the Unity and Multiplicity in the World.’’ In Lamarra and Palaia 2000: 71–95. —. 2001. Leibniz’s Metaphysics: Its Origins and Development. Cambridge: Cambridge University Press. —. 2004. ‘‘Leibniz and His Master: The Correspondence with Jakob Thomasius.’’ In Lodge 2004. Molina, Luis de. 1953. Liberi Arbitrii cum Gratiae Donis, Divina Praescientia, Providentia, Praedestinatione et Reprobatione Concordia. Ed. J. Rabeneck. On˜a: Colegio Ma´ximo S. I. Morfino, Vittorio. 1996. ‘‘Lo Spinozismo di Leibniz: Linee per una ricostruzione della storia della questione.’’ Annali della Facolta` di Lettere e Filosofia dell’Universita` degli Studi di Milano 49/3: 55–81. Murray, Michael. 1995. ‘‘Leibniz on Divine Foreknowledge of Future Contingents and Human Freedom.’’ Philosophy and Phenomenological Research 55: 75–108. —. 1996. ‘‘Intellect, Will, and Freedom: Leibniz and His Precursors.’’ Leibniz Society Review 6: 25–60. —. 2000. ‘‘Critical Review of Cover and Hawthorne on Leibnizian Modality.’’ Leibniz Review 10: 73–86. O’Connor, Timothy (ed.). 1995. Agents, Causes, and Events. New York: Oxford University Press. Ovid. 1958. The Metamorphoses. Trans. Horace Gregory. New York: Viking. Owen, H. P. 1971. Concepts of Deity. London: Macmillan.

260 Bibliography Parkinson, G.H.R. 1965. Logic and Reality in Leibniz’s Metaphysics. Oxford: Clarendon Press. —. 1970. Leibniz on Human Freedom. Studia Leibnitiana, Sonderheft 2. Wiesbaden: Franz Steiner Verlag. —. 1978. ‘‘Leibniz’s Paris Writings in Relation to Spinoza.’’ Studia Leibnitiana Supplementa 18: 73–89. Paull, R. Cranston. 1992. ‘‘Leibniz and the Miracle of Freedom.’’ Nous 26: 218–235. Pen˜afiel, Leonardo de. 1663. Tractatus et disputationes in primam partem D. Thomae: De deo uno et trino. Vol. 1. Lima. Phemister, Pauline. 1991. ‘‘Leibniz, Freedom of Will and Rationality.’’ Studia Leibnitiana 23: 25–39. —. 1999. ‘‘Leibniz and the Elements of Compound Bodies.’’ British Journal for the History of Philosophy 7: 57–78. Philo. 1993. The Works of Philo. Trans. C. D. Yonge. Peabody, Mass.: Hendrickson. Plato. 1997. Complete Works. Ed. John M. Cooper. Indianapolis: Hackett. Plotinus. 1990. Enneads. Trans. A. H. Armstrong. Cambridge, Mass.: Harvard University Press. Poser, Hans. 2000. ‘‘Ens et unum convertuntur: Zur Leibnizschen Einheit der Monade.’’ In Lamarra and Palaia 2000: 271–283. Proclus. 1963. The Elements of Theology. Ed. and trans. E. R. Dodds. Oxford: Clarendon Press. Pufendorf, Samuel. 1934. De Jure Naturae et Gentium Libri Octo. Trans. C. H. Oldfather and W. A. Oldfather. Oxford: Clarendon Press. Pyle, A. J. 1995. Atomism and Its Critics. London: Routledge. Riley, Patrick. 1996. Leibniz’ Universal Jurisprudence: Justice as the Charity of the Wise. Cambridge, Mass.: Harvard University Press. Robinet, Andre´. 1957. Correspondance Leibniz-Clarke: Pre´sente´ d’apre`s les manuscrits originaux des bibliothe`ques de Hanovre et de Londres. Paris: Presses Universitaires de France. Ross, George MacDonald. 1984. Leibniz. New York: Oxford University Press. Rowe, William. 1995. ‘‘Agent Causation.’’ In O’Connor 1995: 173–200. Rozemond, Marleen. 1997. ‘‘Leibniz on the Union of Body and Soul.’’ Archiv fu¨r Geschichte der Philosophie 79: 150–178. Russell, Bertrand. 1900. A Critical Exposition of the Philosophy of Leibniz, with an Appendix of Leading Passages. London: Allen and Unwin. —. 1913. ‘‘On the Notion of Cause.’’ Proceedings of the Aristotelian Society 13: 1–26. Rutherford, Donald. 1993. ‘‘Natures, Laws, and Miracles: The Roots of Leibniz’s Critique of Occasionalism.’’ In Causation in Early Modern Philosophy, ed. Steven Nadler. University Park: Pennsylvania State University Press, 135–58. —. 1995a. Leibniz and the Rational Order of Nature. New York: Cambridge University Press. —. 1995b. ‘‘Metaphysics: The Late Period.’’ In Jolley 1995: 124–175. —. 1996. ‘‘Demonstration and Reconciliation: The Eclipse of the Geometrical Method in Leibniz’s Philosophy.’’ In Leibniz’s ‘‘New System’’ (1695), ed. Roger S. Woolhouse. Florence: Olschki, 181–201. —. 2001. ‘‘Leibniz and the Stoics: The Consolations of Theodicy.’’ In The Problem of Evil in Early Modern Philosophy, ed. Elmar J. Kremer and Michael J. Latzer. Toronto: University of Toronto Press, 138–164. —. 2003. ‘‘Patience sans Espe´rance: Leibniz’s Critique of Stoicism.’’ In Hellenistic and Early Modern Philosophy, ed. Jon Miller and Brad Inwood. New York: Cambridge University Press, 62–89.

Bibliography 261 —. 2004. ‘‘Idealism Declined: Leibniz and Christian Wolff.’’ In Lodge 2004. Schepers, Heinrich. 2000. ‘‘Die Polarita¨ten des Einen und des Vielen im Begriff der Monade.’’ In Lamarra and Palaia 2000: 171–184. Schneewind, J. B. 1998. The Invention of Autonomy: A History of Modern Moral Philosophy. Cambridge: Cambridge University Press. Schneiders, Werner. 1978. ‘‘Deus subjectum: Zur Entwicklung der Leibnizschen Metaphysik.’’ Studia Leibnitiana Supplementa 18, vol. 2: 21–31. Scotus, John Duns. 1950. Ordinatio Prologus. Opera Omnia, vol. 1. Vatican City: Typis Polyglottis Vaticanis. Seidler, Michael. 1985. ‘‘Freedom and Moral Therapy.’’ Studia Leibnitiana 17: 15–35. Sleigh, R. C., Jr. 1990. Leibniz and Arnauld: A Commentary on their Correspondence. New Haven: Yale University Press. —. 1999. ‘‘Leibniz on Freedom and Necessity: Critical Notice of Robert Adams’s Leibniz: Determinist, Theist, and Idealist.’’ Philosophical Review 108: 245–276. Sleigh, R. C., Jr., Vere Chappell, and Michael Della Rocca. 1998. ‘‘Determinism and Human Freedom.’’ In Garber and Ayers 1998, 2:1195–1278. Sleigh, R. C., Jr., and Christia Mercer. 1995. ‘‘Metaphysics: The Early Period to the Discourse on Metaphysics.’’ In Jolley 1995: 67–123. Sousa, Jeronimo de. 1680. Futororum contingentium Polysophia seculis decretis omnibus et scientia media ad mentem Doctoris Subtilis. Paris. Stahl, Daniel. 1655. Compendium Metaphysicae in XXIV Tabellas redactum. Jena. Stein, Ludwig. 1890. Leibniz und Spinoza. Berlin: Reimer. Stier, Johann. 1641. Praecepta Metaphysica. 2nd ed. Erfurt. Taylor, Charles. 1964. The Explanation of Behaviour. London: Routledge. Taylor, S. E., and S. C. Thompson. 1982. ‘‘Stalking the Elusive ‘Vividness’ Effect.’’ Psychological Review 82: 155–181. Thomasius, Jakob. 1665. Schediasma Historicum. Leipzig. —. 1676. Exercitatio de Stoica Mundi Exustione. Leipzig: F. Lanckisius. Turretin, Francis. 1992. Institutes of Elenctic Theology. 3 vols. Ed. James Dennison, Jr. Trans. George Giger. Phillipsburg, N.J.: P & R. Vailati, Ezio. 1990. ‘‘Leibniz and Locke on Weakness of Will.’’ Journal of the History of Philosophy 28: 213–228. —. 1997. Leibniz and Clarke: A Study of Their Correspondence. New York: Oxford University Press. Van Inwagen, Peter. 1995. ‘‘When Is the Will Free?’’ In O’Connor 1995: 219–238. Watson, Gary (ed.). 1982. Free Will. Oxford: Oxford University Press. —. 1996. ‘‘Two Faces of Responsibility.’’ Philosophical Topics 24: 226–248. Weidling, Christian. 1696. Philosophia Juridica, qua Praecepta Logica, Metaphysica, Ethica, Politica, ac Disputandi Artificium, Remotis otiosis Doctrinis & Spinosis Controversiis. Leipzig. White, Michael J. 1992. ‘‘The Foundations of the Calculus and the Conceptual Analysis of Motion: The Case of the Early Leibniz.’’ Pacific Philosophical Quarterly 73: 281–313. Williams, Bernard. 1978. Descartes: The Project of Pure Enquiry. London: Penguin. Wilson, Catherine. 1987. ‘‘De Ipsa Natura: Sources of Leibniz’s Doctrine of Force, Activity and Natural Law.’’ Studia Leibnitiana 19: 148–172. —. 1989. Leibniz’s Metaphysics: A Historical and Comparative Study. Princeton: Princeton University Press. Wilson, Margaret (ed.). 1969. The Essential Descartes. New York: New American Library. —. 1978. Descartes. London: Routledge and Kegan Paul.

262 Bibliography Wittgenstein, Ludwig. 1958. The Blue and Brown Books. 2nd ed. New York and Evanston: Harper and Row. Woolhouse, R. S. 1988. ‘‘Leibniz and Occasionalism.’’ In Metaphysics and Philosophy of Science in the Seventeenth and Eighteenth Centuries, ed. R.S. Woolhouse. Dordrecht: Kluwer, 165–183. —. 1993. Descartes, Spinoza, Leibniz: The Concept of Substance in Seventeenth Century Metaphysics. London: Routledge. —. 2000. ‘‘Pre-established Harmony between Soul and Body: Union or Unity?’’ In Lamarra and Palaia 2000: 159–170. Wright, Larry. 1976. Teleological Explanations: An Etiological Analysis of Goals and Functions. Berkeley: University of California Press.

Index

accommodation, 9, 15, 108–19 Adams, R.M., 14, 16–17, 18 n.5, 29, 33–38, 84–87, 139–41, 206 akrasia, 148, 204, 234–35, 239, 244, 250 appetition, 138–44, 146–50, 152–54, 164–68, 170–74, 184–87, 205, 223–24, 238–40 Aquinas, T., 204–5, 239, 242, 246, 250–52 Arendt, H., 230 n.2 Aristotelianism, 55–56, 121–22, 130, 135, 219, 225, 252 Aristotle, 71, 92 n.4, 175, 192 n.4, 234, 239, 242, 248–50 Arminians, 235–36, 240 Arnauld, A., 15, 20, 44–53, 62, 65, 121, 133 n.2 Augustine, 19 n.18, 242 Bayle, P., 20, 124–25, 170–73 Bennett, J., 15–16 Bolton, M., 66 n.3 Bourke, V., 230 n.2 Boyle, R., 128–29 Bramhall, J., 198, 218 Brandom, R., 150 Calvin, J., 218, 242 causation efficient, 10–11, 15, 136–43, 150, 166, 187 final, 10–11, 15–16, 136–43, 150, 166, 168–69, 184–85, 187, 192 n.4 Clarke, S., 128–29, 182, 186, 218, 232 n.18 compatibilism, 17, 194–96, 202, 213–14, 215 n.19, 236 compossibility, 14–15, 108–10, 117, 119 contingency, 3–4, 16, 158–59, 161–65, 175–76, 188–91, 202–4, 226–28 contingents, conditional future, 209–13

continuum, 14, 18 n.4, 70, 73–78, 80, 82–83, 85, 90–91 Cordemoy, G., 96–97 Couturat, L., 3, 18 n.2,n.5 Cover, J., 40–41 n.33, 215–16 n.27 Cudworth, R., 218, 230 n.6, 231 n.11 Davidson, J., 17–18 Deleuze, G., 120, n.7 Dennett, D., 145 Des Bosses, B., 181 Descartes, R., 5–9, 25–26, 28–29, 41 n.41, 69–70, 73–75, 89, 112, 121–22, 125, 129, 135, 137–38, 237, 239 determinism, 17, 141–42, 182, 185, 189–92, 196, 214, 229, 251 causal, 191, 201, 213, 217 psychological, 17, 200–8 De Volder, B., 130–32 distinction, 13, 23, 36–38 modal, 25, 28–31, 35–38, 43 n.70 real, 13, 25, 27–38 Dole, L., 212–13 Dominicans, 209–10, 223, 225, 235 Eckhard, A., 23–26 entelechy, 90, 160, 163–66 Epicurus, 5, 96 error, 18, 234, 242–43, 251 expression, 3, 14–15, 84–85, 87, 96, 100–5, 108–17, 123 extension, 8, 14, 69–70, 73–74, 81–87, 89–90, 99–100, 126, 128 Fardella, M., 97, 106 Ficino, M., 54–57, 63–64, 67 n.8

263

264 Index force derivative, 131–32, 163–65 primitive, 15, 89, 99–100, 163–65 Franciscans, 203–4 Freddoso, A., 219 freedom, 4, 12, 16–18, 141–42, 156–60, 174–77, 181–82, 185–92, 194–214, 217–29, 235–40, 242–44, 246–47, 250–51

Levey, S., 14 Levine, M., 22 Levy-Bruhl, L., 121, 124 libertarianism, 194–96, 201, 204, 207, 211, 213–14, 235–37, 240, 251 Locke, J., 126, 147–49, 218, 230 n.6, 244–45, 248 Luther, M., 242, 253 n.14 Lutherans, 218, 235, 242

Garber, D., 14 Garrett, D., 232–33 n.24 God, 6, 32–33, 118–19, 120 n.6, 136–38, 147, 184, 191, 194–96, 215 n.12, 235–37, 247–49 and causation, 7–9, 121, 131–33, 133 n.3, 142 and choice, 5, 9, 127–29, 181, 185, 187–89 and mind, 11, 55–57, 59, 63 as one and many, 21–27, 31–37, 42 n.53 and sin, 17, 196, 208–14, 236 Granado, D., 182, 192 n.9, 203 Greenberg, S., 17–18 Grua, G., 181

Malebranche, N., 7–9, 15, 121–25, 127–28, 133, 133 n.3 matter, 5–7, 10–11, 14, 70, 73, 75–76, 78, 80, 82, 89–90, 95–106, 126, 128, 130 mechanism, 4, 99, 135–38, 140–42, 145–47 Melanchthon, P., 235 Mercer, C., 13–14, 27, 33, 35–38, 39 n.16, 42 n.52 mind, 7–9, 14, 52–53, 100–1, 105–6, 142–43, 191–92, 217–18, 223–29 miracle, 7, 15, 121–22, 124–27, 129, 133 Molina, L. de, 17, 202, 218, 222–23 Molinism, 17, 218–29, 236–37, 240 monads, 9–11, 16, 22–23, 84–85, 95, 98–99, 130–33, 138–52, 163–68, 170–74, 184–85, 223, 238 monism, 9, 13, 21–28, 31–34, 37–38, 39–40 n.25, 40 n.26, n.31, n.32 motion, 7–8, 11, 14–15, 69–92, 102–4, 112–16, 131, 184, 187–88 Murray, M., 16–17, 182, 186, 230 n.3

happiness, 12–13, 242, 245–46, 248–49, 252 Hare, R.M., 154 harmony, 6, 9–16, 19 n.14, 76, 108–11, 115–19 of kingdoms of nature and grace, 11–13 preestablished, 3, 7–8, 10, 14–15, 46, 67 n.23, 109, 132–33, 141–43, 145–46, 159–61, 167, 170, 174–76, 184, 186, 197–98, 228 universal, 9, 20–21, 108–9, 135, 167 Hawthorne, J., 40–41 n.33, 215–16 n.27 Hobbes, T., 4–5, 19 n.9, 168, 183, 218, 230 n.6 Hume, D., 124, 133 n.4, 243, 253 n.17 idealism, 11, 14, 19 n.17, 69, 84–85, 87, 91–92, 95–106 identity, 3, 7, 9, 13, 18 n.5, 22–27, 36, 46–56, 58, 61–62, 65 indifference, 17, 218–21, 223, 235–36 intellectualism, 18, 202–4, 214–15 n.10, 234–39, 242, 244, 250 Jalabert, J., 109–10 Jesuits, 17, 181–82, 203, 205–7, 209–11, 213, 218, 223, 225, 235–36 Jolley, N., 14–15 justice, divine, 5–6, 12–13, 19 n.10, 236 Kant, I., 11–12, 156–57, 185, 228, 232 n.22, 242, 244 King, W., 220 Knebel, S., 182 Kulstad, M., 13 Lamy, F., 125, 215 n.19 laws moral, 16, 184–85, 187, 201 natural, 6–8, 15, 115–16, 122–29, 185, 187–88

naturalism, 5–7, 9, 12 nature, 6–9, 12, 15, 78–83, 85, 87–89, 91, 125–27, 186, 188–89, 191–92 necessitarianism, 3, 127, 129, 203–7 necessity metaphysical, 16, 158, 176, 181–82, 186–91, 196–97, 203–4, 206, 208, 226–27 moral, 16–17, 181–92, 196, 201, 203–8, 213–14 physical, 182–83, 186–87, 196, 201–4, 206–8, 213–14 nominalism, 20–21 noncontradiction, principle of, 186, 190–91 occasionalism, 7–9, 15, 121–33 organism, 10, 136–37, 140, 142 pantheism, 13, 21–28, 31–38, 39–40 n.25, 40 n.31 Parkinson, G.H.R., 18 n.4, 23–24, 26, 34 Pelagianism, 236–37 perception, 9, 20, 80–82, 85, 88, 105, 138–42, 146, 150–53, 169, 171–74, 199–200, 205, 223–28 perfection, 15–16, 95, 100, 102–4, 183, 200, 206, 238, 240 Plato, 53–54, 56–57, 61, 65, 234–35, 242, 248 Platonism, 14, 36–37, 45, 53–57, 63–65, 66 n.4–5 Plotinus, 55–56, 66 n.6 Poser, H., 20 possible worlds, best of all, 5, 9–10, 13, 116, 184, 189, 242

Index 265 Proclus, 55–56 Pufendorf, S., 218 realism, 14, 69, 74, 84–92, 105–6 reason, 11–12, 16–18, 141–42, 148, 150–54, 158, 169, 173, 205, 213, 235–41, 243–51 Reformed theology, 195–96, 208, 235–36 Ruiz de Montoya, D., 182, 203 Russell, B., 3, 18 n.5, 121–22 Rutherford, D., 16, 125, 134 n.6 Schepers, H., 21 Schneewind, J.B., 228, 231 n.8, 233 n.24 scholasticism, 66 n.1, 73, 82, 106, 121, 129–30, 220, 223, 226, 229, 238–39 Scotus, J.D., 202 shape fractal, 14, 76–77, 84, 86–89, 91 precise, 14, 69–92 sin, 17–18, 208–10, 236–38, 242–243, 246, 251 Sleigh, R.C., 13–15, 44–53, 58, 61–62, 65, 66 n.2, 67–68 n.25, 84, 87–88, 91, 107 n.7, 108–11, 115, 123, 206–7, 222 Snell, W., 137–38 Socrates, 53–54, 234–35, 242 soul, 5–6, 8–11, 45, 51–57, 62–64, 103–6, 159–60, 166–74, 234–35, 238 Sousa, J. de, 204, 207 Spinoza, B., 3–9, 12–13, 21–22, 25, 31, 40 n.31, 109, 119, 125, 129, 131, 140–41, 156, 158, 183, 206 spontaneity, 13, 16–17, 49–52, 58, 156–77, 195–201, 208, 213–14, 226–27, 236, 240 agent, 16, 157, 159–61, 168–74, 176–77 monadic, 16, 157, 159–68, 171, 173, 175–76 Stoics, 5–6, 19 n.10, n.18, 244 Sua´rez, F., 129, 218–22, 224–28, 230–31 n.7, 231 n.10, 232 n.22

substance, 8–9, 11, 13–15, 22–23, 25–33, 44–66, 95–106, 108–13, 115–19, 122–23, 131–32, 159–65, 175–76, 185–86, 196–97, 199–201, 207, 214 corporeal, 10–11, 13–14, 20, 38 n.2, 46, 48, 52, 55–56, 60, 62, 69–70, 73, 84–91, 95, 97–99, 101, 103–106, 142 created, 7, 15, 102, 121, 127, 238 simple, 9–11, 20, 84, 91, 95, 98, 105, 115, 131–32, 184–85, 187, 197, 226–28 sufficient reason, principle of, 3, 17, 134 n.6, 190–92, 211–12, 221–23, 225 superintrinsicalness, 49–52, 58 teleology, 15–16, 136–50, 153, 166–71, 173–74, 238–40 theodicy, 5, 12, 236 Thomasius, J., 66 n. 5 truth, 3–4, 11, 16, 188–92, 209–13, 235, 238, 243, 247–48 Turretin, F., 195–96, 236–37 union, hypostatic, 14, 59–62 unity, 13–14, 20–22, 25, 35–36, 44–66, 96–97 Vailati, E., 126, 129, 232 n.18 Van Inwagen, P., 217 Van Limborch, P., 235 voluntarism, 127, 129, 202–3, 235–37 Watson, G., 229 n.1 will, 16–17, 123, 148–49, 156–57, 159–60, 168–70, 172, 174–77, 196, 201–14, 217–29, 234–52 divine, 17, 125–26, 128, 173, 209–12, 214 Wilson, C., 14–15 wisdom, divine, 7, 127, 134 n.8 Wittgenstein, L., 120, n.5 Woolhouse, R.S., 122 Wright, L., 137